ARC

PHOTOVOLTAICS
CENTRE OF EXCELLENCE

2010/11
ANNUAL REPORT
ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

CONTENTS
1. DIRECTORS’ REPORT......................................................................... 2

2. HIGHLIGHTS...................................................................................... 4

3. STAFF LIST....................................................................................... 10

4. RESEARCH

4.1. INTRODUCTION TO RESEARCH............................................... 14

4.2. FACILITIES AND INFRASTRUCTURE........................................ 16

4.3. FIRST GENERATION: WAFER-BASED PROJECTS.................... 21

4.4. SECOND GENERATION: SILICON, ORGANIC AND OTHER

“EARTH ABUNDANT” THIN-FILMS......................................... 38

4.5. THIRD GENERATION: ADVANCED CONCEPTS........................61

4.6. PHOTONICS AND DEVICE CHARACTERISATION...................100

5. EDUCATION....................................................................................106

6. ORGANISATIONAL STRUCTURE....................................................120

7. FINANCIAL SUMMARY...................................................................122

8. PUBLICATIONS...............................................................................124

1
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

1. DIRECTORS’
REPORT
Photovoltaics involves the direct conversion of light, normally sunlight, into electricity
when falling upon devices known as solar cells. Silicon is the most common material
used to make these photovoltaic cells, similarly to its key role in microelectronics.

The Australian Research Council (ARC) Photovoltaics use with normal retail electricity prices in many
Centre of Excellence commenced at the University parts of the world. This is expected to lead to a
of New South Wales (UNSW) on 13th June, 2003 period of self-sustaining growth with photovoltaics
with funding initially until 31st December 2007. positioned to become one of the world’s major
Upon review in 2005 and 2006, this funding was industries of the 21st century. Even in Sydney where
extended until 31st December 2010. Criteria for “cheap” electricity is supplied by power stations
such extension included “progress of the Centre built on coal fields, photovoltaics is coming close
towards becoming independent of the ARC Centres to being competitive with retail electricity prices
of Excellence Scheme for funding”. With the rapid without subsidies.
growth of the photovoltaics industry since the
Most present photovoltaic sales are of “first-
Centre’s commencement and the Centre’s strong
generation” solar cells made from silicon wafers,
and seminal links to this industry, the Centre was
similar to the wafers used in microelectronics.
clearly in a much stronger position for independent
The Centre’s world-leadership with these “first-
operation beyond 2010 than many others in its
generation” devices, with international records
funding cohort. Accordingly, while remaining under
for the highest-performing silicon cells in most
the ARC Centres of Excellence umbrella beyond
major categories (including the outright highest-
2010 in recognition of the quality of its research,
performing cells and modules) was recently
the Centre will now be dependent upon industry-
extended with the demonstration of the first 25%
related funding for research with near-term
efficient silicon solar cell. First-generation Centre
outcomes and upon more academically-orientated
research addresses the dual challenges of reducing
ARC and international schemes for its long-
cost and further improving efficiency. The rapid
term research.
growth of the industry is generating widespread
The Centre‘s mission remains to advance silicon interest in ongoing innovations of the Centre’s
photovoltaic research on three separate fronts, as first generation technology with several distinct
well as to apply these advances to the related field technologies now in large-scale production and
of silicon photonics. The educational activities of the additional licences signed during 2010.
former Key Centre for Photovoltaic Engineering are
Silicon is quite brittle so silicon wafers have to be
also integrated into the Centre.
reasonably thick, a fraction of a millimetre, to be
Over the period of funding, photovoltaics has sufficiently rugged for processing into solar cells
become the world’s most rapidly growing with reasonable yield. Without this mechanical
energy source, with markets increasing at a constraint, silicon would perform well even if very
compounded rate of above 40%/year over this thin, even 100 times thinner than present wafers.
period. The electricity generating capacity of new Centre researchers have pioneered an approach
photovoltaic product installed in 2006 exceeded where very thin silicon layers are deposited directly
new nuclear power capacity for the first time, with onto a sheet of glass with the glass providing
the gap widening every year since. In Germany, the required mechanical strength. This “second-
photovoltaics has been the largest source of new generation” approach gives enormous potential cost
electricity generating capacity for each of the savings. Not only are the costly processes involved
last four years and became the second largest in making wafers no longer required, but also there
such source across Europe in 2010, ahead of wind is an enormous saving in silicon material. Cells also
generators and behind only gas turbines. Reducing can be made more quickly over the entire area of
prices, at least partly traceable to past Centre large glass sheets. The Centre is at the forefront
initiatives, means the technology is reaching “retail of international research with such “second-
grid parity”, where the cost of electricity generated generation”, silicon based approaches, with the
using photovoltaics can compete at the point of first commercial product from “spin-off”, CSG Solar,

2
several megawatt fields now operating in Europe. our experience with solar cells, using light to
During 2010, with support from an Australian produce electricity, to the reverse problem of
ARC
Research Council Linkage Grant, collaborative work engineering silicon devices that use electricity to
continued with CSG Solar aimed at further reducing produce light. The Centre holds the international PHOTOVOLTAICS
costs. The project is investigating electron-beam record for the light emission performance from CENTRE OF
evaporation of silicon onto the glass substrate, a bulk silicon, in both electroluminescent and EXCELLENCE
much quicker process than the plasma-enhanced, photoluminescent devices. Emphasis is now upon 2010/11
chemical deposition processes used to date, and exploiting our expertise in silicon light emission to ANNUAL REPORT
also diode laser processing of the deposited films. develop new techniques for silicon wafer and cell
characterisation. A Centre “spin-off” company, BT
Steps to broaden “second generation” activities from
Imaging, consolidated its position as the premium
silicon to both silicon and carbon were boosted by
equipment supplier internationally using this
the appointment of a new academic staff member,
approach during 2010.
Dr Ashraf Uddin, commencing on-duty early in
2009, with this organic solar cell work supported In addition to these four research strands, the
by an ARC Discovery Grant. The activities were activities of the former Key Centre for Photovoltaic
further broadened in 2010 by the award to Centre Engineering have been integrated into the ARC
Postdoctoral Fellow Dr Xiaojing Hao of an Australian Centre of Excellence. The seventh year of students
Solar Institute Fellowship to develop “earth from the Bachelor of Engineering (Photovoltaics
abundant” CZTS (copper-zinc-tin-sulphide) solar and Renewable Energy) program graduated
cell technology. By combining these four relatively during 2010. This program has been enormously
benign and abundant elements in appropriate successful, attracting some of the best and brightest
proportions, a tetrahedrally co-ordinated students entering the University and providing
“synthetic silicon” compound can be produced with the human resources to fuel the recent growth of
some advantages over silicon, such as stronger the industry. The fifth year of students have now
light absorption. graduated from the Centre’s second undergraduate
program, leading to a Bachelor of Engineering
The “second generation” thin-films have a large
(Renewable Energy).
potential cost advantage over the silicon wafer-
based approach, due mainly to reduced material A significant relevant development during 2010
costs. In large enough production volumes, even was the blossoming into full scale operation of the
these reduced material costs will dominate thin- Australian Solar Institute, of which the Centre is a
film costs. This has led to the Centre’s interest founding member given its key role in the Institute’s
in advanced “third-generation” thin-film solar establishment. During the year, the Institute
cells targeting significant increases in energy- announced the outcome of its Round 1 and Round
conversion efficiency. Higher conversion efficiency 2 call for applications with some notable successes
means more power from a given investment in for the Centre. The difficulty for the Centre with this
materials, reducing overall power costs. The Centre’s project based funding is the uncertain long-term
experimental program in this area is concentrating future of the Institute.
on “all-silicon” tandem solar cells, where high
We thank all those who contributed to the Centre’s
energy-bandgap cells are stacked on top of lower-
success during 2010. The present continues to
bandgap devices. The silicon bandgap is controlled
be a most exciting time for photovoltaics. More
by quantum-confinement of carriers in small silicon
are recognising the possibility of a future where
quantum-dots dispersed in an amorphous matrix of
solar cells provide a significant part of the world’s
silicon oxide, nitride or carbide. Cells based on “hot”
energy needs, without the environmental problems
carriers are also of great interest since they offer the
and escalating costs associated with present
potential for very high efficiency from simple device
approaches. The impending transition to “grid
structures. Although their implementation poses
parity” is expected to make the coming decade
daunting challenges, considerable progress on
a time of rapid change for the industry, with an
addressing these was made during 2010.
accelerated pace of adoption of newer, lower cost
The final Centre research strand involves silicon technologies pioneered by the Centre.
photonics where the emphasis is upon using

Professor Stuart R. Wenham Professor Martin A. Green

Director Executive Research Director

3
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Associate Professor Thorsten Trupke accepts

the Berthold Leibinger Innovationspreis in
Ditzingen, Germany.

2. HIGHLIGHTS now Suntech COO (Chief Operating Officer),

Suntech Becomes World No. 1 in 2010 Stuart Wenham, part-time Suntech CTO (Chief
UNSW Linked Companies Also Nos. 2, 4 and 5 Technology Officer), Senior Vice Presidents Shijun
Cai and Guangchun Zhang, as well as Dr Jingjia Ji
who heads the company’s research division and
Suntech, the first of the UNSW-linked photovoltaic companies to Dr Jingbing Zhu, head of quality control.
commence production in China, officially opened its first production
Despite Suntech’s outstanding success, the impact of the Centre and
line in September 2002. Suntech also had the distinction of being the
its programs has been much more widespread. Number 2 on the list of
first privately-owned Chinese company to list on the New York Stock
manufacturers with over 1.4 GW produced in 2010 was JA Solar, with
Exchange (NYSE) when it did so in 2005. Already by 2004, Suntech had
the initial cell production line for this company established by former
moved into the ranks of the world’s “top ten” manufacturers and has
Centre researchers, Ted Szpitalak, Dr Bruce Beilby and Dr Ximing Dai.
moved steadily up this list ever since. In 2010, Suntech claimed the
JA Solar was the third major company Ted had helped establish in
number 1 position in the rapidly expanding market with close to 1.6 GW
manufacturing, the previous two being Suntech and Sunergy, with the
of photovoltaic product manufactured during the year.
latter founded by former Centre researchers and students, Drs Jianhua
Although based in China, Suntech has both a large portion of Australian Zhao and Aihua Wang. Dr Dai was Deputy General Manager and CTO
ownership and a management structure dominated by Australians. Dr of second-placed JA Solar at the time of the company’s initial listing on
Zhengrong Shi, Suntech’s founder, Chairman and CEO (Chief Executive NASDAQ in 2007.
Officer) is an Australian citizen. Other senior staff formerly involved
with UNSW and/or its “spin-off”, Pacific Solar, include David Hogg,

4
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Former student and postdoctoral An electron beam induced current (EBIC) image of the
researcher Dr Zhengrong Shi on the laser doped region of a LDSE solar cell superimposed
cover of Photon International as upon a scanning electron micrograph (SEM) of the
founder and chief of the world’s same region. The inset shows Dr Brett Hallam, a key
No. 1 solar company. member of the team producing the new result.

Suntech Highlights new cell efficiency record

Number 4 on the list and another member of the 19.3% Laser Doped Solar Cell
“gigawatt” club with 1.1 GW produced in 2010 was A new solar cell efficiency record of 19.3%
Yingli Green Energy Holding. Dr Guoxiao Yao, who has been independently confirmed by the
was awarded his PhD from UNSW in 2005, was CTO Fraunhofer Institute in Germany for a large area
at the time of the company’s listing on the NYSE (157 cm2) laser doped, selective emitter (LDSE)
in 2007. Yingli’s present CTO is Dr Dengyuan Song, cell fabricated by the Centre using cells largely
who also received his PhD from the Centre in 2005. processed on a standard Centrotherm screen-
Number 5 on the list was Trina Solar with 1.1 GW printing production line. The only process
produced. Former PhD student, Dr Mohan variation apart from a lighter junction diffusion
Narayanan, was CTO at the time of the company’s was to replace the front surface screen-printed
listing on NASDAQ in 2006. Dr Narayanan is now metal with the Centre’s laser doping process to
simultaneously pattern the cell antireflection Associate Professor Thorsten Trupke
CTO of Hanwha Solar One (formerly Solarfun), with with co-awardee Dr Robert Bardos
0.5 GW of production in 2010. Another former PhD coating and create localised doped regions for the on campus.
student, Fei Yun was CTO of Solarfun at the time of subsequent self-aligned process of light-induced
plating of Ni/Cu. Key members of the team from Photouminescence
its initial listing on NASDAQ in 2006. Fei is now Vice-
awards
President, Technology at LDK Solar, another major the Centre responsible for this result, Brett Hallam,
player in the solar field. Stuart Wenham, Chee Mun Chong, Matt Edwards
and David Jordan, plan to present the new results at
UNSW-linked companies therefore form an amazing the 37th IEEE Photovoltaics Specialists Conference in
4 of the top 5 in 2010. The remaining company Seattle in June 2011.
making the list of the top 5 was former “Number
One”, First Solar (USA, Malaysia, Germany). With further optimisation of the cell design, the
laser doped solar cell is expected to become
Numerous other of the world’s major photovoltaic the first low-cost cell capable of energy
companies have former Centre members in key conversion efficiency above 20% to enter full-
technical positions. The Centre estimates the scale production.
number as over 30, with the better known including
SunPower (USA, Philippines, Malaysia), Sunergy
Photoluminescence Awards
(China), SolarWorld (Germany), Hyundai Solar
(Korea), Gintech (Taiwan), TopCell Solar (Taiwan), The Centre has pioneered the use of
CSG Solar (Australia, Germany), Shinsung (Korea), luminescence imaging technology for the
Sulfurcell (Germany), LG Electronics (Korea), characterisation of silicon ingots, wafers and
Innovalight (USA), Chinalight Solar (China), E-Ton cells with this technology now commercialised
Solar (Korea), LDK Solar (China), Topco (Taiwan), by “spin-off” BT Imaging. This work continues
MAGI Solar (China), Corum Solar (Taiwan), Apollon to attract substantial international attention
Solar (France), Sunrise Global (Taiwan) and REC with the work of the two original developers
Solar (Norway, Singapore). of the technology and the “B” and the “T” in Series resistance map of a silicon
BT Imaging, Deputy Director Associate Professor solar cell showing features due
Dr Zhengrong Shi recently nicely summarised Thorsten Trupke and Dr Robert Bardos, were to non-uniform processing
Australia’s impact in photovoltaics in his article “Can honoured in July by the award of the Berthold temperatures. The image was
Australia Save the World” (ATSE Focus, February created by the new technique
Leibinger Innovationspreis. The photo shows
developed by Dr Henner
2011) when he observed that, in the past, the UK Associate Professor Trupke accepting the prize in Kampwerth (inset) during his
had shown the way to use coal for energy, the Ditzingen, Germany. The Berthold Leibinger Stiftung doctoral thesis, resulting in his
USA had shown how to harness atomic power but gives this award every two years for “excellent selection as 2010 Student Inventor
Australia and China were showing the world how to of the Year.
research and development work on the application or
best use solar. generation of laser light”. 2010 student inventor of
the year

5
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Left: Artist’s impressions of the new Solar Industrial

Research Facility (SIRF) due to be completed by mid-2011
which will showcase the Centre’s near-term technology.
Above: Artist’s impressions of the new Tyree Energy
Technologies Building that will house Centre research and
teaching in the future.

new photovoltaic buildings

Later in the year, Associate Professor Trupke was (photo), herself no stranger to receiving awards.
also chosen as the UNSW 2010 Inventor of the In March, Nicole was awarded the Australia Youth
Year in recognition of the innovative features and Trust’s Tom Harvey Award for Citizenship for 2010 to
commercial significance of this work, following in recognise outstanding work of a young Australian in
the footsteps of Centre Director Stuart Wenham, development work.
who won this title in 2009. Also recognised in the
2010 Awards was Dr Henner Kampwerth as 2010 New Photovoltaic Buildings
Student Inventor of the Year, for his doctoral thesis
During 2010, construction began on two new
work on imaging series resistance in silicon cells.
buildings on the UNSW Kensington campus
This work was supervised by Associate Professor
destined for the Centre’s future use.
Trupke and Professor Martin Green.
The first of these is the Solar Industrial Research
Suntech Establishes 450 MW Facility (SIRF) on the upper campus, due to be
Pluto Capacity completed in mid-2011. This impressive, stylish
building will house a full-scale, state-of-art solar
Suntech has announced the company has
Dr Zhengrong Shi (left), CEO and cell production line, although not with the level of
Chairman of Suntech Power with now established 450 MW capacity of its high
automation required for commercial throughputs.
Professor Martin Green and Pluto performance Pluto cell, jointly developed with the
A special area is also set aside for applying new
module. Centre. This corresponds to close to $1 billion/year
processing steps to cells otherwise processed on
450 MW Pluto capacity in product in today’s market.
the main line.
The Pluto technology represents a low cost
The SIRF facility will be used to demonstrate
approach to fabricating the Centre’s world-record,
the next generation of the Centre’s wafer-based
25% efficient silicon cell. In 2009, early
technology at large scale, accelerating the transfer
production cells were used in a solar module
of this technology into production and increasing
that surpassed the previous world record
the scope and effectiveness of licensing to the
of 15.5% for multicrystalline cells. In 2010,
rapidly growing industry. The facility will also play
Suntech improved this module efficiency to
a key role in Centre research and in the training
17%. The strengths of the technology were
of Centre students as well as, via short courses,
recognised at the 2010 Energy Institute
the huge numbers of external engineers and
Awards in London when Suntech/UNSW were
technicians entering this industry.
jointly awarded the Technology Award in
recognition of the favourable cost ratio of Pluto The second new Centre building is the Tyree Energy
technology “with manufacturing capacity now Technology Building (ETB) being purpose-built
well above any other new silicon cell technology on the lower campus to house the Centre’s main
commercialised worldwide”. laboratories. These laboratories are expected to be
moved from their present site during 2012. The new
Also at the same Awards, Centre Executive
$155 million building is supported by Sir William
Doctoral student Nicole Kuepper Research Director Martin Green received the
in December accepting the 2010 Tyree as well as by a grant of $75 million from the
Award for Individual Achievement for “unparalleled
Energy Institute Awards in London Australian Government’s Education Investment
contributions to solar cell design, incitement of major
on behalf of Suntech/UNSW as Fund. This grant was awarded after a presentation
well as the Individual Achievement investment in solar resulting in vastly improved
by Centre researchers, Professors Stuart Wenham
Award on behalf of Professor Martin performance and reduced costs, greatly increasing
and Martin Green as well as doctoral student, Nicole
Green. Earlier in the year, Nicole expectations of what is commercially possible to meet
received the Australia Youth Trust’s Kuepper, to the Selection Committee emphasising
growing energy demands”.
Tom Harvey Award for Citizenship the past achievements of the Centre and the impact
for 2010. Accepting the two awards on behalf of the Centre, of the new building on boosting Centre activities.
technology awards Suntech and Professor Green, was Nicole Kuepper Co-located in the building will be other UNSW

6
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT
Dr Zi Ouyang – Widely cited
first author. Professor Martin Green receiving
the Eureka Prize for Leadership in
Research from Dr Megan Clark, Chief
Executive, CSIRO.

Professor Stuart Wenham (top)

and Dr Richard Corkish (above)
(accepting award on behalf of
Dr Supriya Pillai – First author of top Chris McGrath) at the Engineering
other awards
cited paper. Excellence Awards.

energy-related activities such as the Australian Year Award at the Sydney Division’s Engineering
Energy Research Institute (AERI). Excellence Awards.

Other Awards Undergraduate Awards

Following on from a best paper award at the 24 th
In March, the Dean of Engineering, Professor
European Photovoltaic Solar Energy Conference Graham Davies, awarded Dean’s Awards to eight
in 2009, then doctoral student Zi Quyang was first of the Centre’s top undergraudate students for
author on another paper “Effective light trapping in their academic achievements in 2009.
polycrystalline silicon thin-film solar cells by means
Included in this group (see photo) was Bai Xue
of rear localized surface plasmons”, published in
who, this year, based on her achievements over
July 2010 which was one of the “monthly top 20
the entire photovoltaic engineering course was
most-downloaded articles last year” in Applied
awarded the University Medal, the University’s Left to right: Professor Graham
Physics Letters. Davies, Yu Feng, Shen-Long Ooi, Xue
award for the highest academic achievement in the
Bai, Professor Richard Henry, John
The Centre’s role in stimulating the current graduating cohort. Durrant, Benjamin Hughes, Vincent
high level of interest in applying plasmonics to Allen, Dr Richard Corkish, Dongchen
Another Centre student, Alice Lang, has been
photovoltaics has been recognised by the intense Lan. Absent: Dong Lin
awarded a prestigious Rhodes Scholarship. “Alice
interest in its seminal paper in this field, “Surface
has really worked hard in the Bachelor of Engineering
plasmon enhanced silicon solar cells”, authored by
in Photovoltaics and Solar Energy Engineering degree
Supriya Pillai (pictured), Kylie Catchpole, Thorsten
program and it is a real pleasure to see her rewarded
Trupke and Martin Green, published in the Journal
like this”, said Dr Richard Corkish, Head of School of
of Applied Physics in May 2007. This has not only
Photovoltaic and Renewable Energy Engineering.
been by far the Journal’s most highly cited paper
“The engineering degrees that we offer attract some
of the 3,470 papers published in 2007, but also has
fantastic students. This Rhodes Scholarship couldn’t
been on the monthly list of the Journal’s 20 most
have gone to a more capable and deserving student”.
downloaded papers (all years) every month since,
including each month of 2010. Professor Stuart Wenham and Bai
New Books Xue, winner of the University Medal.
In August 2010, Executive Research Director
Following on from the translation of the Centre’s undergraduate awards
Professor Martin Green received the Eureka Prize for
highly regarded textbooks into simplified
Leadership in Research, one of the most prestigious
and traditional Chinese in 2008 and 2009, the
awards in Australian science.
Centre was invovled in publishing two more
In September, it was the turn of Centre Director, translated versions in 2010. Martin Green’s
Professor Stuart Wenham who was named 2010 text “Solar Cells” was translated into simplified
Professional Engineer of the Year by the Sydney Chinese during 2010 while his text was also
Division of Australia’s peak engineering body, reproduced in Japanese translation during
Engineers Australia. Chris McGrath, a graduate from the year. There are now nine different texts in
the School, also won the student Engineer of the total published by the Centre. The translation
into Korean of “Applied Photovoltaics” is also
in progress.

new books published

by centre

7
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Dr Richard Corkish (left) and Professor Martin Green (Left to Right): Dr Gavin Conibeer, Dr Richard Corkish,
(middle) talking with Minister Wan in the laboratory Professor Martin Green, Dr Mary Archer,
area where the Centre’s world-record 25% efficient Dr Sergey Varlamov, Dr Anita Ho-Baillie, Professor
cells are fabricated (Credit: NSW Office for Science and Stuart Wenham and Dr Ashraf Uddin.
Medical Research ).

ranging from incremental efficiency improvements

in commercial technology to Third Generation
PV theory and its implementation. Dr Archer is
working on a revised edition of her monumental
reference work “Clean Electricity from Photovoltaics”
in conjunction with Centre Executive Research
Director Professor Martin Green.

Chinese Minister of Science and

Technology Visit
Also early in 2011, The Chinese Minister of Science
and Technology, Dr Wan Gang, met with Professor
Martin Green and Dr Richard Corkish. The Minister
visited SIRF, toured the Centre labs where the
Centre’s world-record 25% efficient solar cell was
developed and spoke with Chinese postgraduate
students about their work. At a subsequent NSW
Government-organised Innovative Materials
Roundtable, Minister Wan said, “I am impressed by
the commitment shown by Australian Universities to
Left to Right:
• Dr Matt Edwards High Profile these new technologies and that so many Chinese
graduates have benefited from this expertise. In
UNSW, Technology Transfer Team
• Mr Valantis Vais Visitors particular, I should point out that the pioneering
Renewable Energy program set up by UNSW over 10
UNSW Co-op Scholar
• The Hon. Kristina Keneally MP NSW Premier Visits Centre years ago and led by Prof Martin Green is an excellent
Premier of NSW example of Australian innovation and shows how
Following on from visits to the Centre in 2009
• Prof. Stuart Wenham Australia and China can work together to create new
of the President of Korea, President Myung-
UNSW, Director, Photovoltaics fields of technology and opportunity”.
Centre of Excellence back Lee and the soon-to-be Prime Minister of
• Ms Adeline Sugianto Australia, Julia Gillard, the then-NSW Premier, The Also attending with Minister Wan was the newly
UNSW, Technology Transfer Team Honourable Kristina Keneally, visited the Centre in appointed Chinese Consul-General to Sydney,
May to announce that the NSW Government has Duan Jielong, on his first official visit to an
NSW Premier Visits Centre
pledged $120 million to support and attract solar Australian university.
energy projects to the State in conjunction with
the Commonwealth Government’s Solar Flagships
Program. “This is about leading the nation on solar
energy projects. Australia is a sun drenched nation and
we must harness this boundless energy,” Ms Keneally
said during her announcement.

Baroness Dr Mary Archer Visits Labs

Early in 2011, Baroness Dr Mary Archer, a specialist
in solar power conversion, visited the Centre.
While here, the Centre’s Deputy Directors gave
presentations in the Centre’s research strands

8
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

3. STAFF LIST
Director Postgraduate Co-ordinator
Stuart Ross Wenham, BE BSc PhD UNSW, FTS, FIEEE, FIEAust • Alistair B. Sproul, BSc (Hons) Sydney, PhD UNSW
(Scientia Professor)
Associate Professors
Executive Research Director • Gavin Conibeer, BSc MSc London, PhD Southampton
Martin Andrew Green, BE MEngSc Qld., PhD DSc(Hon) McMaster, DEng • Alistair B. Sproul, BSc (Hons) Sydney, PhD UNSW
UNSW, FAA, FTSE, FIEEE (Scientia Professor and Federation Fellow) • Thorsten Trupke, PhD Karlsruhe

Deputy Directors Senior Lecturers

• Gavin Conibeer, BSc MSc London, PhD Southampton • Stephen Bremner, BSc(Hons), PhD UNSW
• Thorsten Trupke, PhD Karlsruhe • Richard Corkish, BEng (Com) RMIT, PhD UNSW
• Ashraf Uddin, BSc (Hons) MSc Dhaka U, Bangladesh; • Evatt Hawkes, BSc (Hons) BEng (Hons) W. Aust., PhD Cantab.
PhD Osaka U, SMIEEE
• Alison Lennon, BSc(Hons) Sydney, PhD Sydney, PhD UNSW
• Sergey Varlamov, PhD Moscow
• Geoffrey J. Stapleton, BE UNSW
Business, Technology & Operations Manager • Ashraf Uddin, BSc(Hons) MSc Dhaka U, Bangladesh;
PhD Osaka U, SMIEEE
Mark D. Silver, BE UNSW, GMQ AGSM
• Muriel Watt, BSc N.E., PhD Murdoch
Undergraduate Co-ordinator
Lecturers
Ashraf Uddin, BSc (Hons) MSc Dhaka U, Bangladesh;
• Anna Bruce, BE(Hons) PhD UNSW
PhD Osaka U, SMIEE

10
• Merlinde Kay BSc(Hons) PhD UNSW • Omaki Kazuo
• Martha Lenio, B.A.Sc, Uni. of Waterloo • Nicole Kuepper
(until February 2010)
ARC
• Pei Hsuan Lu
• Santosh Shrestha, PhD UNSW PHOTOVOLTAICS
• Ly Mai, BE UNSW
• Long Seng To, BE(Hons) BA UNSW • Berhard Mitchell
CENTRE OF
• Zi Quyang (until November 2010) EXCELLENCE
Senior Fellows 2010/11
• James Rudd
• Anita W.Y. Ho-Baillie, BE PhD UNSW (returned • Lei “Adrian“ Shi ANNUAL REPORT
March 2010)
• Adeline Sugianto
• Dirk König, Dipl. Ing. (EE) Dr. rer. nat. (Ph) Chemnitz
• Yuguo Tao
• Sergey Varlamov, PhD Moscow
• Budi Tjahjono
Research Fellows • Kalaimagal Valliappan
• Patrick Campbell, BSc BE PhD UNSW • Ned Western
• Matthew Edwards, BE (Hons 1) PhD UNSW • Guangqi Xu
• Shujuan Huang, BE ME Tsinghua Uni., • Yao Yu
PhD Hiroshima Uni. • Xiaohan Yang
• Henner Kampwerth, MSc Freiburg, PhD UNSW • Tian Zhang
(since July 2010)
Technical Support Staff
• Emily Mitchell, BE PhD UNSW
(since November 2010) • Kian Fong Chin, BE QUT, MEngSc QUT,
• Ivan Perez-Wurfl, PhD Uni. of Colorado, Boulder MEngSc UNSW
• Thomas Soderstrom, PhD Uni. Of Neuchatel, • Bany Jaya, BE UNSW
Switzerland • Robert Largent, AS USA
• Nicholas Shaw, BE UNSW, PhD UNSW
Post-Doctoral Fellows • Lawrence Soria, AssocDipCompAppl W’gong
• Xiaojing Hao, BE, MSEE Northeastern Uni, China • Zhu Shi Yang
• Supriya Pillai, B. Tech (India), PhD UNSW • Alan Yee, BE UTS
• Sangwook Park (until March 2010)
Casual Staff
Adjunct Fellow • Suprabhat Basnet
Didier Debuf, BE ME PhD UNSW • Angela Jin
• Alex Lee
Visiting Professors/Fellows • Clara Mazzone
• Dr. Robert Bardos • Xian Ni, MPhil UNSW
• A/Prof Chee Mun Chong • Jonathan Pye
• Dr Dong-Jin Kim, Presidential Committee, Korean • Alexandra Ryan
Government (December 2010 – May 2011)
• Erdinc Saribatir
• A/Prof Soon Kweon, Chungju National University,
• Andrew See
Korea (January 2010 – January 2011)
• Lei Shi
• Darcy Wentworth, BSc BE Sydney U., MSc Colorado
• Yuguo Tao
Sch. of Mines, MSc UNSW
• Dean Utian
• Prof. Bingqing Zhou, Inner Mongolia Normal
University, China (until July 2010)
Student Administration Manager
Visiting Researchers Kimberly Edmunds, MEdLead UNSW
• Yukiko Kamikawa
Financial Officer
(JSPS Fellow, AIST, Tsukuba, Japan)
• Mallar Ray, Bengal Eng. Sci. University Julie Kwan
(3 weeks in May 2010)
Administrative Staff
Professional Officers • Danny Lin Chen, BSc UNSW
• Mark Griffin, BE UNSW • Mathew Cheung
• Yidan Huang, BSc Anhui • Sue Edwards
• Hamid R. Mehrvarz, PhD UNSW • Joyce Ho
• Tom Puzzer, BSc PhD UNSW • Jill Lewis
• Bernhard Vogl • Melissa Tang

Research Assistants PhD Students

• Nino Borojevic • Pasquale Aliberti
• Catherine Chan • Yael Augarten, BSc UNSW
• John Durrant • Shelley Bambrook
• Brett Hallam • Jose Bilbao
• Ziv Hameiri • Nino Borojevic, BE UNSW
• Phillip Hamer • Kristen Casalenuovo
• Jianshu Han • Catherine Chan
• Jialiang Huang • Obulesu Chatakonda
• Guangyao Jin • Jessie Copper
• Anahita Karpour • Hongtao Cui

11
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

• Dawei Di • Johnson Wong (until September 2010)

• Bonneville Eggleston • Bo Xiao
• Jack Fan • Guangqi Xu
• Peter Gress • Chaowei Xue
• Brett Hallam • Xiaohan Yang
• Ziv Hameiri, BSc BGU • Yang Yang
• Phillip Hamer • Yu Yao
• Jianshu Han • Jae Sung Yun
• Andy Hsieh • Bo Zhang
• Jialiang Huang • Haixiang Zhang
• Guangyao Jin (until September 2010) • Haoyang Zhang
• Craig Johnson
• Nicole Kuepper, BE UNSW Masters Students
• Taraka Kukunuri • Phillip Cook
• Martha Lenio, B.A.Sc, Uni. of Waterloo (until March • Telia Curtis
2010) • Mark Griffin
• Sammy Lee • Do Soon Jung
• Hua Li • Miga Jung
• Wei Li • Anahita Karpour
• Xiaolei Liu • Kyung Hun Kim
• Doris Lu • Taekyun Kim
• Sarah Macleod • Kenneth McLean
• Ly Mai, BE UNSW • Gerald Parnis
• Bernhard Mitchell • Shahla Zamani Javid
• Amir Nashed • Haixiang Zhang
• Zi Ouyang (until September 2010) • Tian Zhang
• Shyam Pasunurthi
Practicum Students
• Robert Patterson
• Yuanjiang Pei • Stephan Lehnen, RWTH, Aachen, Germany (until
March 2010)
• Binesh Puthen Veettil
• Stephan Michard, RWTH Aachen, Germany
• Fang Qi
(September 2009 – February 2010)
• John Rodriguez
• Manuel Morgano, Milano-Bicocca University, Italy
• Yong Heng So (October 2010 – January 2011)
• Adeline Sugianto, BE UNSW • Chandraprasad Rmachandran, UNSW (July –
• Yuguo Tao December 2010)
• Anthony Teal • Dolf Timmerman, University of Amsterdam, The
• Budi Tjahjono, BE UNSW Netherlands (September – December 2010)
• Long Seng To • Hsuan-Wen Wang, National Central University,
• Lara Treiber Taiwan (March 2009 – March 2010)
• Chao-Yang Tsao
• Zhenyu Wan
Taste of Research Summer Scholars
• Stanley Wang, BE, MEngSc UNSW • Xin Rui An
• Kee Soon Wang • Cong Chen
• Qian Wang • David Cordiner

12
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

• Solomon Freer • Ricky Li Qi Gu • Chao Shen

• Alexander Krisman • William Edwin Hargreaves • Dong Dong Shen
• Ruyou Li • Joel Jonathan Hohn • Lihui Song
• Zhongtian Li • Sueanne Lit Hunter • Eden James Tehan
• Zhong Lu • Gilbert Allen Jin • Lin Tian
• Desiree Surjadi • Haini Jin • Andrew Thai
• Xia Yan • Sara Louise Johnston • Chee Hong Thoo
• Chuanxi Yang • Douglas Jones • Wendy Truong
• Yuebin Zhao • Mattias Klaus Juhl • Valantis Vais
• Chao-Li (Philip) Kao • Ian Thomas Van Eerden
Master’s Coursework • Joseph Michael Kassouf • Taru Riikka Veijalainen
Chandraprasad Ramachandran • Harry Lam • James Dean Von Dinklage
• Alice Virginia Lang • Kan Wang
Undergraduate Thesis Students
• Andrew Lem • Ki Yun (Leo) Wang
• Padraic Nolan Aicken
• Ruoyu Li • Ned Western
• Soravit Angsuvarungsri
• Zhe Li • Ewan Robert Willcox
• Daniel Thomas Atkins
• Dong Lin • Eleanor Elizabeth Wood
• Mohammad Taufik Azami
• Xi Lin • Matthew Wright
• Xue Bai
• Tianyue Liu • Tian Xia
• Nicholas Daniel Boerema
• Ziheng Liu • Bo Xiao
• Anna Mary Cain
• Lui Lo • Shouyi Xie
• Kah Howe Chan
• Shunmian Lu • Heli Xu
• Cong Chen
• Meng Luo • Tian Xu
• Jian Chen
• Linzhi Ma • Buyi Yan
• Xu Yang Chen
• Henry Makgawinata • Chien-Jen Yang
• Yang Chen
• Soleman Mohammad Massoud • Xinda Yang
• Fang Cheng
• Praditya Mattjik • Zihao Yang
• Leroy Clunne-Kiely
• Clara Francesca Mazzone • Jizhong Yao
• Jie Cui
• Alexander Minchin • Yao Yao
• Claire Elizabeth Disney
• Fabian Johannes Muschalik • Cao Yu
• Michael Robert Dodds
• Viren Nathoo • Lan Yu
• Robert William Dumbrell
• Adam Sen (Ann) Ong • Tao Zhan
• Sarah Eager
• Shen Long Ooi • Lei Zhang
• Hua Fan
• Matthew Phillip O‘Regan • Tian Zhang
• Yu Feng
• Chen Pan • Xinyu Zhang
• Gerardo Flores Torres
• Jaga Park Ross • Xiang Zhao
• Alistair Forster
• Rochelle Tara Phillis • Xue Ying Zhao
• Zhen Gao
• Jessada Pothipakdi • Yinghui Zheng
• Daniel Scott Gilbert
• Helga Jane Prochaska • Xi Zhu
• Thomas Maxwell Glyde
• Jonathan Andrew Pye • Yong Zhu
• Linfei Gu
• Zixuan Qiu

13
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

4. RESEARCH
4.1 Introduction to Research
Photovoltaics, the direct conversion of sunlight to “buried-contact” solar cell, the first of the modern
electricity using solar cells, is recognised as one of high-efficiency cell technologies to be successfully
the most promising options for a sustainable energy commercialised (Fig. 4.1.1), the Centre has
future with the photoovltaics industry poised developed several other high-efficiency processes
to become one of the world’s largest of the 21st in commercial production or close to this, based on
century. The ARC Photovoltaics Centre of Excellence Centre innovations in laser and ink-jet processing.
“First-generation” wafer-based commenced in mid-2003, combining previous
Wafers are expensive and need quite careful
technology (BP Solar Saturn disparate strands of work supported under a variety
encapsulation, since brittle and also thermally
Module, the photovoltaic product of programs, into a coherent whole addressing
manufactured in the highest mismatched to the glass coversheet, making
the key challenges facing photovoltaics, as well
volume by BP in Europe, using first-generation technology reasonably material-
as “spin-off” applications in microelectronics and
UNSW buried-contact technology). intensive. Several companies worldwide are
optoelectronics. The Centre was funded by the ARC
Figure 4.1.1 commercialising “second-generation” thin-film cell
until December 2010 and has since been funded
technology based on depositing thin layers of the
under a variety of other schemes.
photoactive material onto supporting substrates
The Centre’s photovoltaics research is or superstrates, usually sheets of glass (Fig. 4.1.2).
divided into three interlinked strands Although materials other than silicon are of interest
addressing near-term, medium-term for these films, silicon avoids problems that can
and long-term needs, respectively. The arise with these more complex compounds due
present photovoltaic market is dominated to stability, manufacturability, moisture sensitivity,
by “first-generation” product based on toxicity and resource availability issues. CSG Solar,
silicon wafers, either single-crystalline a partner in the Centre, has commercialised an
as in microelectronics (Fig. 4.1.1) or a approach pioneered by Centre researchers that
lower-grade multicrystalline wafer. This is unique in that it is based on the use of the
market dominance is likely to continue for same high quality silicon used for first-generation
at least the next decade. First-generation production, but deposited as a thin layer onto glass.
production volume is growing rapidly,
As well as its collaborative activities with CSG
with the technological emphasis upon
Solar, the Centre currently maintains a largely
streamlining manufacturing to reduce costs
independent program addressing alternative
Example of “second-generation”
while, at the same time, improving the
solutions to those adopted by CSG Solar for
thin-film technology (module energy conversion efficiency of the product. Also
producing high-performance “silicon-on-glass”
fabricated on CSG Solar’s German important is the reduction of the thickness of the
production line, based on thin-films solar cells. The main emphasis of both is the
starting silicon wafer without losing performance,
of polycrystalline silicon deposited development of lower-cost approaches (such as
to save on material use.
onto glass, again UNSW-pioneered deposition by evaporation rather than PECVD). The
technology). The Centre’s first-generation research is focussed Centre also commenced activities on carbon-based,
Figure 4.1.2 on these key issues. Building upon the success of organic solar cells during 2009 and a program on

14
 ARC
PHOTOVOLTAICS
CENTRE OF
Efficiency and cost projections for
first-, second- and third-generation
EXCELLENCE
photovoltaic technology (wafers, 2010/11
thin-films and advanced thin- ANNUAL REPORT
films, respectively).
Second-generation thin-film
technology has a different cost
structure as evident from this figure.
Production costs per unit area are
a lot lower, since glass or plastic
sheets are a lot less expensive
than silicon wafers. However, likely
energy-conversion efficiencies are
lower (6-15%). Overall, this trade-off
produces costs/watt estimated as
about 2 times lower than those
of the wafer product, in large
production volumes.

Figure 4.1.3

other “earth abundant” materials, particularly CZTS bandgap-engineering using nanostructures

(Cu2ZnSnS4), in 2010. was selected as the most promising for
relatively near-term implementation (Fig. 4.1.4).
At the present time, second-generation thin-films
This involves the engineering of a new class of
are entering the market in increasing quantities.
mixed-phase semiconductor material based
Large-scale commercialisation of thin-film product
on partly-ordered silicon quantum-dots in
leads to a completely different manufacturing cost
an insulating amorphous matrix. The general
structure compared to the wafer-based product.
CZTS material system may also have some
However, costs again are increasingly becoming
potential here due to the range of bandgaps
dominated by material cost as production increases,
accessible. Even tandem cells built on silicon
for example, by the cost of the glass sheet on which
wafers could fall into the targetted cost zone
the cells are deposited.
in Fig. 4.1.3. Photon up-conversion as a way of
More power from a given investment in material “supercharging” the performance of relatively
can be obtained by increasing energy-conversion standard cells forms a second line of research.
efficiency. This leads to the possibility of a third- A third is the investigation of schemes for
generation of solar cell distinguished by the fact implementing hot-carrier cells. Both of the
that it is both high-efficiency and thin-film. To latter may prove to be particularly well suited
illustrate the cost leverage provided by efficiency, to organic solar cells.
Fig. 4.1.3 shows the relative cost structures of the
The fourth Centre strand of silicon photonics Conceptual design of an all-silicon
three generations being studied by the Centre. This
draws upon elements of all three of the tandem cell based on Si-SiO2 (or
figure plots efficiency against manufacturing cost, Si-Si3N4 or Si-SiC) quantum dot
photovoltaic strands. A by-product of this work
expressed in US$/square metre. First-generation superlattices. Two solar cells of
has been the development of techniques based
technology has relatively high production cost per different bandgap controlled by
on silicon light-emission for characterising both quantum dot size are stacked
unit area and moderate likely efficiencies at the
completed devices, particularly solar cells, as well as on top of a third cell made from
module level (13-20%). The dotted lines in Fig. 4.1.3
silicon wafers at different stages of processing (Fig. bulk silicon.
show the corresponding cost/watt, the market
4.1.5). Developing this approach to its full potential
metric. Values below US$1/watt seem increasingly Figure 4.1.4
has formed an increasingly large part of the Centre’s
feasible by improving the efficiency while reducing
photonics program.
manufacturing cost, but this is the likely limit of the
first-generation approach.

The third-generation is specified as a thin-film

technology, which therefore has manufacturing
costs per unit area similar to second-generation,
but is based on operating principles that do
not constrain efficiency to the same limits as
conventional cells (31% for non-concentrated Schematic representation of a
sunlight for the latter). Unconstrained PL imaging system. An external
thermodynamic limits for solar conversion are much light source is used to illuminate
the silicon wafer or solar cell
higher (74% for non-concentrated light, giving an homogeneously. The luminescent
idea of the scope for improvement). If a reasonable emission (red arrows) from the
fraction of this potential can be realised, Fig. 4.1.3 sample is captured with a sensitive
suggests that third-generation costs could be lower CCD camera. This technology was
commercialised by Centre “spin-off”
than second-generation by another factor of 2 to 3.
BT Imaging during 2008.
Of the third-generation options surveyed by Centre Figure 4.1.5
researchers, “all-silicon” tandem cells based on

15
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

4.2 Facilities and

Infrastructure
The ARC Photovoltaics Centre of Excellence is X-ray diffraction, Raman spectroscopy, AFM and
located at the Kensington campus of the University surface analysis equipment. TEM, ellipsometry and
of New South Wales, about 6 km from the heart of a femtosecond time resolved photoluminescence
Sydney and close to its world famous beaches measurement system are also regularly accessed at
including Bondi, Coogee and Maroubra Sydney University.
(Fig. 4.2.1).
During 2010, the development and acquisition of
Organisationally, the Centre of Excellence laboratory equipment and infrastructure continued
is located within the School of Photovoltaic with specific details about significant additions
and Renewable Energy Engineering (SPREE) found under the laboratory headings which follow.
within the Faculty of Engineering. The Centre
The Centre of Excellence has three computer
of Excellence has a large range of laboratory
networks. Both Kensington and Bay St Botany each
facilities (Fig. 4.2.2). These include the Bulk
have a research/admin network. The third network
Silicon Research Laboratories, the Device
is for students enrolled in the Undergraduate
Characterisation Laboratory, the Optoelectronic
Degree Courses in Photovoltaics and Renewable
Research Laboratories, the Thin-Film Cell
Energy. In 2010, the Kensington Research and
Laboratory, the Industry Collaborative
Administrative network grew by 20% to consist of
Laboratory, Inkjet Processing Laboratory and
5 file servers, 1 terminal server, 1 intranet server,
Organic Photovoltaic (OPV) Laboratory. Off
2 Internet web-servers (hosted on Central IT
campus the Centre has a Thin-Film Cleanroom
infrastructure), 2 licence servers (local and Central)
facility at Botany, 5km south-west of the main
and over 170 client workstations. At Kensington,
campus. Another important resource is the
an additional 31 computers are dedicated to
Semiconductor Nanofabrication Facility (SNF)
the computer control of laboratory and other
jointly operated by the Faculty of Science and
equipment. The Centre also has a 3 server computer
the Faculty of Engineering.
controlled SCADA and PLC network for equipment
Additional equipment commonly used for solar control and monitoring the research laboratories
cell work is found at the new MicroAnalytical and related infrastructure. Bay St Botany has 9
Research Centre (MARC) and elsewhere on the dedicated equipment controllers and the research/
University campus. Included in this category admin network supports 11 workstations.
Centre of Excellence location
in Sydney. are TEM/SEM electron microscopes, focused ion
The computer resources are used for measurement,
beam (FIB) specimen preparation equipment,
Figure 4.2.1 modelling/simulations, equipment control,

16
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Layout of laboratories
and other facilities
within the Electrical
Engineering
Building, Kensington.

Figure 4.2.2

document control, laboratory design support, and Buried Contact Cell Laboratories. Total
Internet access, general administrative purposes laboratory processing space of over 580
and maintaining the Centre’s presence on the m2 (including the Industry Collaborative,
Internet. InkJet Processing Development, Device
Characterisation and Optoelectronic
Each postgraduate research student is allocated
Research Labs) is located in the Electrical
a dedicated personal computer and has access to
Engineering Building and is serviced
shared computer resources.
with filtered and conditioned air,
The Centre upgraded Linux (Beowulf ) 64bit appropriate cooling water, processing
cluster has an estimated computational power of gases, de-ionized water supply, chemical
2 Teraflops to support the demanding Density- fume cupboards and local exhausts.
Functional-Hartree-Fock and molecular dynamical There is an additional 819 m2 area for
computations. A second 128 core cluster has the accommodation of staff, research
been installed for complementary investigations students, school office and laboratory
CNC Laser Scribe Tool.
of energy efficiency and renewable energy via support facilities. Another 480 m2 of combined roof
computational fluid dynamics techniques. space accommodates fixed PV arrays and 361m2 Figure 4.2.3
of accessible outdoor experimental space. Off site,
The Student Computer network comprises of
areas totalling 700 m2 are used for the storage of
two computer laboratories of total 78 m2 with 5
chemicals and equipment spare parts and a new
servers, a UNSW custom CSE computer based IPQ
100 m2 area for a nickel sintering belt furnace facility
router and 34 workstations. Students enrolled
at 78 Bay St Botany.
in the Undergraduate Degree Courses use these
computers for computer-related coursework and The laboratories are furnished with a range of
Internet access. There is also an Internet capable processing and characterisation equipment
web-server that gathers and displays data collected including 23 tube diffusion furnaces, 6 vacuum
from solar arrays on the roof of the building. These evaporation deposition systems, a laser-scribing
data can be viewed using a web browser and can be machine, 2 laser doping machines, rapid thermal
made available for Internet access. annealer, four-point sheet-resistivity
probe, quartz tube washer, silver/
Students also have access to over 200
nickel and copper plating facility, visible
additional workstations made available through
wavelength microscopes, 3 wafer mask
collaboration with the School of Computer Science
aligners, spin-on diffusion system,
and Engineering.
photoresist and dopant spinners, electron
The Laboratory Development and Operations Team beam deposition system, metallization
develops and maintains core Centre and laboratory belt furnace, nickel sinter belt furnace,
facilities. During 2010, the team, under the manual and automatic screen printers
leadership of Mark Silver, comprised of an additional and a laboratory system control and data
5 equivalent full-time and 11 casual employees, acquisition monitoring system.
including: electrical engineers, a computer/
The laser scribe tool, shown in Figure 4.2.3,
network manager, electronic/computer/laboratory
has a 20 watt Nd:YAG laser for infrared
technicians and administrative staff.
operation (1064 nm) and an optional
frequency doubler for green operation Wafer Mapping Tool.
Bulk Silicon Research Laboratories (532 nm). The work stage is CNC controlled allowing
The Centre houses the largest and most 1 micron positional accuracy and table speeds Figure 4.2.4
sophisticated bulk silicon solar cell research facility approaching 25 cm/second across an area of 15
in Australia, incorporating both the High Efficiency cm by 15 cm. The tool is used primarily for Buried

17
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Remote plasma PECVD machine.

Figure 4.2.7

Contact solar cell fabrication, cutting 35-micron Optoelectronic Research Laboratories

wide laser grooves as deep as 100 microns into This facility has six optical benches and several
silicon wafers and scribing wafers in preparation for visible and near-infrared semiconductor diode
cleaving. It can also be used to cut other suitable lasers along with other optical and electrical
materials, such as stainless steel. instrumentation. The facility is used for
photoluminescence (PL) and electroluminescence
Device Characterisation Laboratory measurements in the visible and infrared spectral
This laboratory is located on the lower ground floor range up to wavelengths of 2500nm, luminescence
Flash Cell Tester. of the Electrical Engineering Building. Associated experiments with simultaneous two-colour
with it is the Optoelectronic Research Laboratories, illumination, quasi steady state photoluminescence
Figure 4.2.5
reception area, seminar room, offices for Centre staff lifetime measurements and Sinton lifetime testing
interacting with the public and industry, including with the conventional flash-light replaced by a high-
the Business & Technology Manager, Finance power light emitting diode array.
Officer, External Relations and Design Assistance
This facility also houses the Centre’s world-leading
Division Manager.
first photoluminescence imaging system. Other
The Device Characterisation Laboratory houses equipment includes a silicon CCD camera (for
characterisation equipment including “Dark sensitive PL measurements), spectroscopic PL
Star”, the Centre’s station for temperature systems, PL lifetime measurement unit, and LBIC
controlled illuminated and dark current-voltage measurement system. Areas separate from the
measurements, the Centre’s Fourier-transform Device Characterisation Laboratory were necessary
infrared spectroscopy system (FTIR), frequency in order to meet stringent standards for safe laser
dependent impedance analyser, ellipsometer, use. It shares cryogenic cooling equipment with the
Sinton photoconductance lifetime equipment, Device Characterisation Laboratory.
wafer probing station, open circuit voltage versus
illumination measurement system (Suns-Voc), 4 point Thin-Film Cell Laboratory
resistivity probe, spectral response system and This 40 m2 laboratory is equipped with a range of
Optical characterisation bench.
spectrophotometer with integrating sphere. equipment for thin-film deposition and patterning,
Figure 4.2.6
In 2008, a multi function wafer mapping tool, shown including a plasma-enhanced chemical vapour
in Figure 4.2.4, was installed. This unit provides deposition (PECVD) system, a sputtering system,
state of the art capability for the measurement larger area plasma etcher, a reactive ion etcher (RIE),
of carrier lifetime, bulk resistivity, emitter sheet a resistively heated vacuum evaporator, helium
resistance and Light Beam Induced Current (LBIC) leak detector and an optical microscope with
on wafer samples. digital image acquisition system. Also used by the
laboratory is an electron-beam vacuum evaporator
Additions in 2009 included a high speed for silicon which is physically located within the
commercial flash cell tester with 156 mm square Bulk Silicon Research Laboratory. This Si evaporator
wafer capability as shown in Fig 4.2.5, a replacement is also equipped with an ionizer unit and a sample
UV/VIS/NIR spectrophotometer and a commercial heater, enabling fast-rate Si homoepitaxy at
Luminescence Inspection System (LIS) from spin off temperatures of about 500-600°C by means of ion-
company BT Imaging. assisted deposition (IAD). The IAD is also equipped
with a residual gas analyser to permit real time
process monitoring. Other equipment of use in thin-
film projects is located within the Semiconductor
Nanofabrication Facility.

18
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Thermal and e-beam Evaporator.

Figure 4.2.8

The PECVD system, shown in Fig. 4.2.7, has a 40 x nitrogen environment belt furnace capable of nickel
20 cm2 process platen and can handle large-area sintering 156 mm square wafers and larger on its 25
silicon wafers as well as smaller pieces. Two types inch wide belt was installed.
of plasma excitation (remote microwave and direct
RF) are available. The machine is used for the Inkjet Processing Development Lab
low-temperature deposition of thin dielectric films
This laboratory houses the Centre’s inkjet printing
(silicon nitride, silicon dioxide, silicon oxy-nitrides)
development systems. The laboratory is used to
and of amorphous silicon. The dual-cylinder,
develop solar processing “inks” (chemical solutions)
remote microwave plasma source produces
and for printing them onto a solar wafer under
excellent-quality silicon nitride and silicon dioxide
computer control. The aim is to develop low cost
films, with precise control over the stoichiometry
processing techniques for creating fine structures in
at temperatures up to 500°C. Amorphous and
solar cells. It is anticipated that the inkjet printing is
microcrystalline silicon films can also be deposited
capable of replacing processes such as laser scribing
in the system.
and photolithography for forming fine patterns
In 2008, the thermal evaporator was upgraded to for contacting but at a cost of at least 10 times
also support e-beam evaporation. This capability cheaper. Equipment includes two ink jet material
greatly expands the range of materials which can be deposition printing systems, capable of depositing
Inkjet development systems.
deposited and is of great interest to 3rd generation a wide range of materials onto different substrates,
researchers. In 2009 a new 4 point probing station a surface tension meter and viscometer. In 2009 a Figure 4.2.9
was added to the laboratory. In 2010 the computer commercial aerosol deposition system capable of
support system for the optical microscope was very fine scale patterning was purchased.
upgraded to improve the useability of the digital
camera fitted. Semiconductor Nanofabrication Facility
The Centre also owns equipment within, and has
Industry Collaborative Laboratory access to, the Semiconductor Nanofabrication
This 120 m2 laboratory houses equipment needed Facility (SNF) at the University. This is a joint facility
for many of the industry-collaborative research shared by the Faculties of Science and Engineering
activities including the Buried-Contact Solar Cell and houses a microelectronics laboratory and a
group. The laboratory equipment includes a belt nanofabrication laboratory for e-beam lithography.
furnace; a state of the art laser micromachining The SNF provides an Australian capability for the
tool; a new PECVD deposition system (located in fabrication of advanced nanoscale semiconductor
the adjacent Thin Film Solar Cell Laboratory); a devices and their integration with microelectronics.
TiO2 spray deposition station; a high temperature SNF research projects form an integrated effort to
semiconductor muffle furnace, manual screen fabricate innovative semiconductor nanostructures
printer and a fully automatic production scale using the latest techniques of electron beam
screen printer. In 2008, a new 532nm green laser patterning and scanning probe manipulation.
was added and customised in-house to increase the A major applied objective of the facility is the
capability for processing laser doped solar cells. development of a prototype silicon nuclear spin
quantum computer. The capabilities of this facility
In 2009, new additions included a fast metallization were expanded to house the, NCRIS funded,
firing furnace capable of processing 156 mm square Australian National Nanofabrication Facility. In
wafers, a light induced plating system, spin on 2010 the facility added a new large area e-beam
doper and mirror steered scanning laser for laser lithography system and new e-beam evaporator for
doping. In 2010 a new wafer spinner was added non MOS-compatible metals deposition.
to the laboratory. Off site at 78 Bay St Botany, a

19
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Bay Street Clean Room Sputter Machine. UHV Electron Beam Evaporator.

Figure 4.2.10 Figure 4.2.11

Thin-Film Clean Room facility in In 2009, an optically assisted IV measurement

Bay Street, Botany system was installed for the determination of static
hot carrier distributions. The LPCVD computer
In 2003, the Centre added a 120 m2 cleanroom
control system was also upgraded to provide
facility in Bay Street, Botany to its infrastructure,
enhanced throughput. A line beam diode laser
greatly improving its experimental capabilities in
system was also installed, outside the clean room,
the area of thin-films. This cleanroom is equipped
for work on low thermal budget defect annealing.
with several fume cupboards, two tube furnaces,
an electron-beam vacuum evaporator, a thermal In 2010, the first stage of a multi-chamber
vacuum evaporator, a glass washing machine, a sputter system has been ordered to allow our 3rd
rapid thermal processing (RTP) machine and a Generation Group to carry out significantly different
5-chamber cluster tool. The cluster tool presently functions on a sample without exposing it to the
features four plasma-enhanced chemical vapour atmosphere between each process. Processes
deposition (PECVD) chambers and one lamp- include sputter-depositing different dopants in
heated vacuum annealing chamber. The PECVD separate chambers, vacuum annealing, also PECVD
chambers enable the low-temperature deposition layers. This purchase will be “Stage I” in that one
of dielectric films (silicon oxide, silicon nitride, chamber will be purchased; a loadlock chamber
etc) and amorphous silicon films (either n- or and other peripherals have been configured from
p-doped or undoped). Furthermore, samples decommissioned stock. Stage II will involve the
can be hydrogenated by PECVD using hydrogen purchase of a second processing chamber and a
plasma at substrate temperatures of up to 480°C. transfer/loadlock chamber.
During 2004, the Centre purchased a low-pressure
chemical vapour deposition (LPCVD) system, an The addition of computer control to our
infrared NdYAG laser scriber and a box furnace for hydrogenation tool provides more reliable
sample annealing. The LPCVD system is capable of process repeatability and accuracy of set process
depositing doped crystalline silicon on glass and parameters. The processor inputs all process
with its additional remote plasma source is currently parameters for each “layer” defining each step
engaged in hydrogenation work. The NdYAG before starting the process. This upgrade also
laser, located outside the clean room, is used for frees processors’ time in that they need not be in
scribing silicon films and other suitable metal and attendance to change settings when beginning
dielectric materials. each process layer.

In 2006, a state of the art multi-target sputter Organic Photovoltaic Laboratory (OPV)
machine was installed. In 2008 this vacuum tool
In 2010 fit out of a new OPV laboratory in the
was upgraded from four to five separate targets.
Chemical Sciences Building commenced and
Each target is able to be operated independently
OPV equipment currently located in other Centre
of one and other, allowing users to cosputter a
laboratories is being integrated into the new facility.
thin film from more than one target and deposit
A recent arrival has been the inert atmosphere, four
multilayers without breaking vacuum. Three power
glove boxes with integrated spin coater. The new
supplies are available with substrate biasing. The
laboratory includes the Centre’s first fume cupboard
custom made system can handle substrates up to
with integral hydrofluoric acid scrubber as well as
150mm x 150mm. Excellent film purity is assured
support space for vacuum evaporation, annealing
as the system incorporates a load-lock. Computer
and sample storage.
control can be used for most operations, including
substrate heating, allowing precise multilayers to be
deposited repeatedly.

During 2007, a new computer controlled multi

pocket, multi source UHV electron beam evaporator
system was delivered. This equipment provides the
capability for high “industrial” rate silicon and other
thin film deposition.

20
4.3 First Generation
Wafer Based Projects
ARC
Research Team • K. Valliappan (Research Assistant) PHOTOVOLTAICS
• L. Zhang (Research Assistant) CENTRE OF
University Staff • Y. Yao (PhD student and Research EXCELLENCE
• S. R. Wenham (Centre Director) Assistant)
2010/11
• M. A. Green (Executive • L. Gu (Research Assistant)
ANNUAL REPORT
Research Director) • N. Western (Research Assistant)
• T. Trupke (Deputy Director – • Mattias Klaus Juhl (Research Assistant)
Photoluminescence Imaging) • Xi Zhu (Research Assistant)
• A. Lennon (Group Leader – Inkjet and • X. Bai (Research Assistant and
Plating Technologies) undergraduate thesis)
• A. Ho-Baillie (Group leader - High • Y. Yeung (Research Assistant)
Efficiency Cells) • Bo Xiao (PhD student)
• J. Roderiguez (PhD student)
Postdoctoral Fellows
• B. Mitchell (PhD Student)
• H. Mehrvarz (High efficiency Cells)
• Li Hua (PhD student)
• M. Edwards (Tech Transfer)
• D. Lu (PhD student)
• B. Tjahjono (Core Research)
• Y. Yang (PhD student)
• L. Mai (Tech Transfer)
Technology Transfer Team
Postgraduate Research Students
and Research Assistants Managers
• M. Lenio (Research Assistant and • D. Jordan (Team leader)
PhD student)
• C.M. Chong (Deputy Team Leader
• Z. Hamieri (PhD student) – Devices)
• N. Kuepper (Research Assistant and • M. Edwards (Deputy Team Leader –
PhD student) Program Manager)
• S. Wang (PhD student) • S. Wenham (Technical Director)
• N. Borojevic (Research Assistant and
PhD student) Team Members
• S. Zamani Javid (Research Assistant and • A. Sugianto
Masters student) • N. Kuepper
• P. Hamer (Research Assistant and • B. Hallam
PhD student)
• C. Chan
• A. Sugianto (Research Assistant and
• A. Lennon
PhD student)
• B. Tjahjono
• B. Hallam (Research Assistant and
PhD student) • L. Mai
• A. Karpour (Research Assistant and • C. M. Chong
PhD student) • X. Bai
• D. Wang (PhD student) • P. Hamer
• A. Han (Research Assistant) • E. Mitchell
• C. Chan (Research Assistant and
NewSouth Innovations
PhD student)
• G. Xu (Research Assistant and • N. Simpson
PhD student) • A. Prib

4.3.1 High Efficiency Silicon Cells

Alternate Rear Reflectors for Silicon

Si wafer
Solar Cells
Si wafer
During 2010, the high efficiency group conducted Passivation
studies into improved planar rear reflectors and also Dielectric stacks Padding laye
into a scattering rear reflector for silicon solar cells Metal
Metal
as shown in Fig. 4.3.1.1. The aim of this work is to (a) Planar rear reflector made
increase reflectance over a broad angular range and up of dielectric stacks and
(b) scattering rear reflector
to enhance light trapping properties, respectively.
made up of silver plasmonic
The effect of varying refractive index of the middle nanoparticles sandwiched
Si wafer between passivating and
layer of a dielectric stack (DS) of SiO2/SiNx/SiO2
Si wafer padding dielectric layers.
has been studied. The thicknesses of each layer in Passivation layer
the DS are kept at ¼ wavelength thick
Dielectric while the
stacks Padding layer Figure 4.3.1.1
refractive index of the middle layer is varied from Metal
Metal

21
 ARC
PHOTOVOLTAICS
CENTRE OF
Rb and IQE under Lambertian light distribution
EXCELLENCE
Rb and IQE under perpendicular incident light

1 1.2 1.2

2010/11 0.95 1
0.99

1
0.97
ANNUAL REPORT 0.9 R (n=1.5) 0.8 0.95 0.8

Rear reflectance
Rear reflectance
R (n=2.0) R (SiO2 / Al)
R (n=2.6) 0.93 R (SiO2 / Ag)

IQE
IQE
0.85 R (n=3.0) 0.6 R (optimised DS / Al) 0.6

R (n=3.2) 0.91 R (optimised DS / Ag)

0.8 IQE (n=1.5) 0.4 IQE (SiO2 / Al) 0.4
IQE (n=2.0) 0.89 IQE (SiO2 / Ag)
IQE (n=2.6) IQE (optimised DS / Al)
0.75 IQE (n=3.0) 0.2 IQE (optimised DS / Ag) 0.2
0.87
IQE (n=3.2)
0.7 0 0.85 0
900 950 1000 1050 1100 1150 1200 1250 900 950 1000 1050 1100 1150 1200 1250
Wavelength (nm) Wavelength (nm)

R (left axis) and IQE (right axis) under normally R and IQE under a Lambertian distribution of light for
incident light for a planar silicon solar cell with a silicon solar cell with various types of rear reflectors
dielectric stack plus metal rear reflector. The including the standard SiO2/metal and the optimised
effect of varying the refractive index of the dielectric stack (DS)/metal rear reflectors.
middle layer of the dielectric stack (DS) is shown.
Figure 4.3.1.3
Figure 4.3.1.2

Measured EQE of un-textured solar cell

1.00E+02
(4.3.1.1)
9.00E+01
8.00E+01
7.00E+01 EQE enhancement
(4.3.1.2)
6.00E+01
EQE (%)

5.00E+01
Original solar cell
4.00E+01
Solar cell only with passivation layer
3.00E+01
2.00E+01 Solar cell with passivation layer and Ag plasmon

1.00E+01
0.00E+00 (4.3.1.3)
300 400 500 600 700 800 900 1000 1100 1200
wavelength (nm)
Current density Jsc (mA/cm2) was calculated
by integrating the product of IQE and photon
flux, from 300nm to 1300 nm (assuming zero
EQE of a planar solar cell with 1.5 to 3.2. Such stacks with a rear contacting metal front surface reflection). The standard AM1.5G
standard rear reflector (SiO2/Al) layer applied onto a planar cell result in internal spectrum has been used. Table 4.3.1.1 shows the
structure and after the application improvement of Jsc as a result of applying DS/metal
reflectance (R) and internal quantum efficiencies
of Ag nanoparticles (prior to
the application of padding and (IQE) shown in Fig. 4.3.1.2. Figure 4.3.1.3 compares rear reflectors (over the standard SiO2/Al reflector)
metal layers). the R and IQE of standard SiO2/metal reflectors to on both planar and textured solar cells.
optimised DS/metal reflectors when applied to cells
Figure 4.3.1.4
with textured front surface.

The calculation of R and the IQE for Figures 4.3.1.2 (4.3.1.4)

and 4.3.1.3 is carried out as follows. All photo-
generated carries are assumed to be collected Another promising scattering rear reflector
Spectrally sensitive mirrors
by the solar cell (which is roughly the case in structure involves plasmonic scattering from Ag
the Centre’s PERL cells). AB is the absorption nanoparticles. The idea is to scatter the unabsorbed
in the bulk region, AT refers to the cell’s total long wavelength light back into the silicon bulk,
absorption (AT=AB+AR, with AR defined as rear thereby randomising the reflected light directions
reflector absorption). IQE equals the ratio of to enhance light trapping. This can be particularly
AB and AT as indicated in Eq. 4.3.1.1. Formulae important from the first double-pass of light across
for AB and AT are given elsewhere [4.3.1.1]. a cell with pyramidal texture since a large fraction
R is the wavelength dependent internal is lost when internally striking the surface of a
Sunlight
rear reflectance. It can be simply calculated pyramid of opposite slope to that coupling it in.
for perpendicularly incident light, or be a Light travelling past the Ag nanoparticles will be
weighted value of a Lambertian angular light reflected back by the padding layer/metal interface
distribution as shown in Eq. 4.3.1.3. T+ is the as shown in Fig.4.3.1.1 (b), and be scattered for
Solar cells of
different band gaps fraction of light transmitted to the rear surface the second time. Fig. 4.3.1.4 shows the preliminary
and T- is the fraction of light reflected from the results of applying such structure on a silicon solar
rear side transmitted to the front surface. An cell prior to the application of the padding layer.
Multi-junction cell concept using
approximation for T+ T- is used to simplified formula The application of such a nanoparticle reflector
spectral splitting.
in Eq. 4.3.1.2, where a is the absorption coefficient involves the removal of existing back metal layers
Figure 4.3.1.5 and the thinning of the rear passivating SiO2
which varies with wavelength and W is the wafer
thickness. layer followed by the growth of Ag plasmonic

22
 Refractive index of middle layer in DS

Jsc improvement n=2.0 n=2.6 n=3.0 n=3.2

ARC
PHOTOVOLTAICS
Under normal incident light 1.08% 1.97% 2.50% 2.65%
CENTRE OF
Under Lambertian incident light -0.03% 0.06% 0.15% 0.15% EXCELLENCE
2010/11
Jsc improvement after
applying dielectric stack (DS)/
ANNUAL REPORT
metal rear reflectors compared
to the standard SiO2/Al
reflector for both planar and
textured solar cells.

Table 4.3.1.1

40.7% III-V cell stack on Ge showing

the I-V curves of constituent
cells [4.3.1.2].

Figure 4.3.1.6

particles onto the thinned passivating layer. While

the EQE enhancement as shown in Fig. 4.3.1.4
is encouraging, it is anticipated that more EQE
gain is possible by adding the padding and metal
reflector layers.

Spectrum Splitting
During 2010, the group continued its work
on improving the efficiency of solar cells by
subdividing the broad solar spectrum into smaller
energy ranges and to convert each range with
a cell of appropriately matched bandgap (see
Figure 4.3.1.5).

The Centre is involved in a collaborative research

project involving Solar Systems and Spectrolab in The initial idea was to reflect some of the excess (a) Original “Power Cube” receiver
fully exploring the potential of spectrum splitting photons reaching the Ge cell onto a separate and concept. Light in approximately the
890-1100nm wavelength range is
in combination with the power tower concept, relatively inexpensive silicon cell array using an
reflected by a band-pass dielectric
for higher efficiency conversion of concentrated innovative “power cube” receiver as in Fig. 4.3.1.7(a). reflector onto a silicon array rather
solar power (CSP). A working prototype CSP power For the cell stack shown in Fig. 4.3.1.6, up to 0.085 than being dissipated as heat in
tower system of 140kW rating at Waterbridge, A/W of photon current could be reflected onto a the Ge cell; (b) Improved “Demi-
Cube” concept. Thermal design is
Victoria has demonstrated the highest sustained silicon cell where it would be converted at about
improved by integrating one array
conversion efficiency to date for any solar to electric 0.7V under concentration. This would add an extra and the reflector. By switching the
conversion system without yet the benefit of such 6% absolute to the combined cell performance position of the silicon and III-V array,
splitting [4.3.1.3]. We plan to demonstrate the taking it from 40.7% efficiency to above 46%. the filter design becomes far less
additional potential by the construction of a small critical as shown in Fig. 4.3.1.8.
Additional advantages would be that the
“power-cube” prototype capable of independent Figure 4.3.1.7
essentially lossless dicroic reflector used to give this
confirmation of energy conversion efficiency above
wavelength selective reflection could also be used
40%, including all optical losses.
to control the angular emission from the III-V cells,
The work is based on combining the only two increasing their voltage output (by 2-20 mV/cell for
commercial solar cell technologies for concentrated radiative efficiencies above 10%). Moreover, by
sunlight, silicon and III-V semiconductor stacks on controlling the precise wavelengths reflected Fig. 4.3.1.7 (b)
Ge substrates, with the latter shown in Fig. 4.3.1.6. during the day, spectral mismatch losses could R
The Ge substrate provides a good crystallographic be greatly reduced compared to the use of a 1.0
template for subsequent layers but Ge has too low a III-V array alone. An additional advantage is the
bandgap to be a good opto-electronic match to the reduced amount of waste heat to be removed
two cells on top. From the current-voltage curves from the receiver and the larger area available
shown, the Ge cell can be seen to be generating for its extraction.
Fig. 4.3.1.7 (a)
much more current than the other two cells. The
More detailed analysis of the receiver has
extra current is essentially wasted as heat in the
shown that the “demi-cube” approach of Fig. 0 900
Ge cell with the situation becoming worse as the 1100
4.3.1.7(b) is even more attractive. By integrating
airmass (AM) increases from the standard AM1.5
the array and reflector and switching the Simplification of filter design with
test condition shown (particularly in the mornings
position of the two arrays, the design of the new configuration.
and afternoons).
reflector is greatly simplified. Low UV and far-IR
Figure 4.3.1.8
absorption in the filter to avoid excessive heating

23
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Filter Wavelength Jsc III-V Top Cell Jsc III-V Middle Jsc III-V Bottom Current
Cut-offs (nm) (Am-2) Cell (Am-2) Cell (Am-2) Mismatch (%)
Without Filter 152.41 147.27 199.24 38.8
Design 1 932-1070 152.41 146.50 146.46 4.09
Design 2 930-1068 152.31 146.89 144.96 6.41
Design 3 930-1066 151.99 146.89 146.12 4.54

Effect of the use of dielectric

reflectors on reducing current
mismatch within the III-V device.

Table 4.3.1.3

Fixed Demi-Cube Structure Variable Demi-Cube Structure demi-cube structure will be

Angle Angle even greater if used within a
Seasons between Filter III-V cell Si cell Total cell Filter between III-V cell Si cell Total cell day as the changes in daily
in La III-V & Si cut-offs efficiency efficiency efficiency cut-offs III-V & Si efficiency efficiency efficiency spectrum content will be
Parguera cells (°) (nm) (%) (%) (%) (nm) cells (°) (%) (%) (%)
greater than those observed
Summer 33.9 874-1060 41.0 4.1 45.1 874-1060 33.9 41.0 4.1 45.1
in Figure 4.3.1.9. Table 4.3.1.3
Autumn 33.9 874-1060 40.2 4.1 44.3 894-1080 35.3 40.8 4.2 45.0
shows the effect of the use
Winter 33.9 874-1060 41.3 4.1 45.4 874-1060 33.9 41.3 4.1 45.4
Spring 33.9 874-1060 40.3 4.1 44.4 894-1080 35.3 40.9 4.2 45.1
of dielectric reflectors on
reducing current mismatch
at 500 suns (or higher) sunlight concentration is no within the III-V device.
Effect of varying angle between
Si and III-V arrays to maximise cell longer essential. Also, the requirements on reflector
output from the various seasonal design are considerably relaxed due to the greater References:
spectral contents shown in
ease of designing for high reflection rather than for 4.3.1.1 M.A. Green, “Lambertian light trapping in textured
Fig. 4.3.1.9(b). solar cells and light-emitting diodes: analytical
high transmission (Fig. 4.3.1.8), with design now
Table 4.3.1.2 solutions”, Progress in Photovoltaics: Research and
simplifying to that of a noncritical band pass filter.
Applications 10(4), pp. 235-241, 2002.
The extra cell area required is not a factor since the
4.3.1.2. Richard King, 51st Electronic Materials Conference,
larger array is formed from low June, 2009.
cost silicon cells.
4.3.1.3 J. Lasich et al., “World’s First Demonstration of a
140kWp Heliostat Concentrator PV (HCPV) system”,
Another advantage of the
34th IEEE Photovoltaic Specialists Conference,
“demi-cube” concept is that the Philadelphia, 7-12 June 2009.
angle between the Si and III-V
arrays can be varied to take 4.3.2 Industry Collaborative Research and
advantage of the non-ideal Commercialisation
reflection and transmission
properties of the dielectric 4.3.2.1 Introduction
reflector. As the incidence
The Centre’s Technology Transfer team (TTT) was
angle on the dielectric reflector
established in 2008 to accelerate the development
changes, the reflector cut-off
of the Centre’s commercially significant
wavelengths vary and so do
technologies and to carry out technology transfers
the slopes in the transition
to industry. This includes assisting companies to
region between reflection
establish and optimise the large scale production
and transmission. This allows
of UNSW photovoltaic technology. With the high
the reflector to capture the
demand for UNSW technology and new licences,
spectrum with the most
the TTT grew significantly during 2010 with the
suitable wavelength range for
inclusion of Phil Hamer, Xue Bai and Emily Mitchell.
the Si and III-V arrays as the
The already strong academic record of the TTT was
spectral content of sunlight
also further enhanced with an additional two team
varies daily and seasonally, as
members having been awarded University Medals
shown in Fig. 4.3.1.9.
Seasonal averaged direct spectral for being placed first in their respective university
contents for (a) Canberra and (b) Table 4.3.1.2 shows the effect of varying angle programs. This included Emily Mitchell who received
La Parguera. between Si and III-V arrays to maximise output from the University Medal for first place in Photovoltaic
Figure 4.3.1.9 various spectral contents using the spectra of Fig. Engineering at UNSW in 2007 prior to completing
4.3.1.9 as an example. Note the spectra shown in her PhD at the Fraunhofer Institute in Germany and
Figure 4.3.1.9 and used for Table 4.3.1.2 are seasonal Xue Bai who received the University Medal for being
averages only. Hence the advantage of variable placed 1st in Photovoltaic Engineering at UNSW in

24
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

TTT photo June 2010 with Matt Edwards overseas at

the time and new comers Emily Mitchell, Phil Hamer
and Xue Bai about to join the team.

Figure 4.3.2.1

2010. This takes the number of University Medallists 4.3.2.3 Commercially Relevant 150-200 �m
within the TTT to six at the end of 2010, but with Technologies 3mm patterned metal contact

another university medallist Nicole Kuepper being

phosphorus
temporarily on leave from the TTT while completing 4.3.2.3.1 Introduction
her PhD. Figure 4.3.1 shows the TTT in June 2010 With screen-printed solar cells continuing bulk of wafer
with members from left to right being D. Jordan, A. to dominate commercial manufacturing
Sugianto, S. Wenham, A. Lennon, B. Hallam, L. Mai, B. with well over 50% market share, the
Tjahjono and N. Kuepper. n ++
broad aim of this work has been to
p-type rear metal contact
develop, in conjunction with several
p+
4.3.2.2 Industry Collaborators and industry partners, the next generation
Licensees of Technology of screen-printed solar cell. In particular,
metal
The Centre has a large number of collaborators the fundamental limitations of the
and licensees including many cell manufacturers conventional screen-printed solar cell
Standard screen-printed solar cell.
working with the Centre to improve or develop new that have limited its performance for the
solar cell technology for commercialisation. Some last 30 years have been identified, and innovative Figure 4.3.2.2
of these fund the Centre quite generously and are approaches to redesigning the emitter and front
seen as being partners to NewSouth Innovations metal contact have been devised, developed and
(NSi), the commercial arm of the University that analysed in this work. In addition to overcoming
manages the entire intellectual property portfolio the current and voltage limitations imposed by
for the Centre. Interest from companies wanting the design shown below, a further aim of this work
to become commercial partners with the Centre or has been to retain compatibility with existing
license technology grew rapidly during the last two equipment and infrastructure currently used by our
years despite the international financial downturn industry partners for the manufacture of screen-
in late 2008. This appears to be due in part to the printed solar cells.
perceived world leadership of UNSW in this area, Despite the dominance of this technology,
but also due to the growing long-term importance this solar cell design shown in Fig. 4.3.2.2 has
placed on photovoltaics by a world struggling to significant performance limitations that limit the
deal with climate change, diminishing resources cell efficiencies to well below those achievable
and increasing energy requirements. Collaborators in research laboratories around the world. In
with the Centre in the first generation photovoltaics particular, the front surface screen-printed
area come predominantly from China, Germany, metallisation necessitates a heavily diffused emitter
the United States, Australia, South Korea and to achieve low contact resistance and also to
Taiwan and include Suntech-Power, Roth and Rau, achieve adequate lateral conductivity in the emitter
Guodian Solar, Centrotherm, Tianwei, Optomec, since the metal lines need to be widely spaced
Global Sunrise Energy, BP Solar, Sunergy, CSG Solar, compared to laboratory cells to avoid excessive
E-ton Solar, Spectraphysics, Schott Solar, Hyundi shading losses. Such cells therefore typically have
Heavy Industries, China Light, Corum Solar, Advent, emitters with sheet resistivities in the range of 50-60
Shinsung, JA Solar and Yunnan Tianda. Several of ohms per square, which inevitably give significantly
these companies are amongst the world’s largest degraded response to short wavelength light.
solar cell and equipment manufacturers. Good work in recent years by partners Dupont
and Ferro has improved paste formulations,
allowing ohmic contacts to be formed to more
lightly doped emitters in the vicinity of 60 ohms/
square while simultaneously lowering the bulk
resistivity of the fired paste. To further raise this

25
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Screen-printed fingers running

perpendicular to the heavily
diffused semiconductor fingers
where electrical contact is
made. A dielectric/AR coating
passivates the top surface and
isolates the metal from the
lightly diffused top surface.

Figure 4.3.2.3

sheet resistivity to above 100 ohms per square, as lightly doped emitter which can be diffused to
required for near unity internal quantum efficiencies 100 ohms/square, ensuring excellent response to
for short wavelength light, serious resistive losses the short wavelengths of light. In the laser doped
are introduced, both in the emitter and through regions, approximately half of the incident light is
the contact resistance at the metal to n-type lost due to absorption within the heavily doped
silicon interface. semiconductor fingers.

Furthermore, the conventional design for screen- This is defined as a 50% effective shading loss
printed solar cells has quite poor surface passivation of the semiconductor fingers. Using this design,
in both the metallised and non-metallised the effective shading loss of the semiconductor
regions. Even if good ohmic contacts could be fingers is only 0.5%, while the effective emitter
made to more lightly doped emitters, the large sheet resistivity of the emitter is 50 ohms/square.
metal/silicon interface area would significantly This latter figure results from the emitter sheet
limit the voltages achievable due to the high resistance of 100 ohms/square in parallel with the
levels of recombination in these regions and the laser doped lines which cover 1% of the area with a
corresponding contribution to the device dark sheet resistivity of 1 ohm/square and therefore also
saturation current. These voltage limitations are effectively contribute 100 ohms/square. This allows
not of major significance at the moment due to the the screen-printed lines to retain their normal
limitations imposed by the substrates. However, spacing, but with 99% of the emitter being lightly
in the future as wafer thicknesses are reduced to doped and therefore able to achieve near unity
improve the device economics and improved rear internal quantum efficiencies for short wavelengths
surface passivation is introduced, the cells will have of light. In addition, the passivating dielectric layer
the potential for improved open circuit voltages, not only passivates the lightly diffused surface so
but only provided the surfaces, including under the as to give near unity internal quantum efficiencies
metal, are well passivated. for short wavelength light, but it also isolates the
metal from these same regions to minimise the
4.3.2.3.2 Semiconductor Finger device dark saturation current. Importantly, the
Solar Cells silicon is only exposed at the semiconductor fingers,
with the screen-printed metal having been shown
To accommodate a top surface emitter sheet
to make excellent ohmic contact to the heavily
resistivity of at least 100 ohms per square, metal
phosphorus diffused silicon in these regions.
fingers need to be spaced no more than 1mm apart
Both thick oxides and silicon nitride layers, when
to avoid excessive sheet resistivity losses. Due to
used with appropriate pastes, appear to provide
the large width of screen printed metal lines or
adequate protection to the lightly diffused surface
100 microns or more, such a close spacing is not
regions, preventing the screen-printed metal from
possible without shading well over 10% of the cell
contacting the silicon.
surface. The concept of semiconductor fingers is
therefore introduced as shown in Figure 4.3.2.3. A typical cell design based on this concept uses
These semiconductor fingers are formed by laser semiconductor fingers 8 microns wide of sheet
doping the silicon surface while simultaneously resistivity 1-2 ohms/square and spacing of 0.8mm.
patterning the dielectric layer to expose the The top surface sheet resistivity is typically 100
heavily doped silicon surface. Developmental ohms/square, with the effective sheet resistivity
work with the laser doping process allows sheet in the direction parallel to the semiconductor
resistivities as low as 1 ohm/square to be achieved fingers typically 50-60 ohms per square. The screen
while simultaneously forming laser doped lines of printed metal lines are printed perpendicular
only 8 microns width. Spacing such lines 0.8mm to the semiconductor fingers with a width of
apart avoids significant resistive losses within the 110 microns and spacing of 2.2mm. The overall

26
 ARC
PHOTOVOLTAICS
CENTRE OF
120 EXCELLENCE
2010/11
100
ANNUAL REPORT

80

60 EQE
%

Reflection
40
IQE Spectral response of
semiconductor finger cell.
20
Figure 4.3.2.4
0
300 400 500 600 700 800 900 1000 1100 1200
Wavelength (nm)

effective shading loss of the combination of the The best semiconductor finger solar cells have fill
semiconductor fingers and the screen-printed factors of 79%, demonstrated with this structure
metal lines (not counting interconnect strips) is on large area devices of approximately 150cm2,
therefore 5.5%, comprising 5% from the metal plus verifying the effectiveness of this contacting
0.5% from the semiconductor fingers. This is typical scheme for minimising resistive losses. These
of conventional screen-printed cells, but with the cells also have near perfect response to short
greatly enhanced short wavelength response of wavelength light as shown below, leading to Jsc
these cells giving about a 4% advantage in short values of 36-37mA/cm2. Even though commercial
circuit current. p-type substrates are not capable of voltages
above about 640mV, Voc values approaching this
This design also has the advantage of being the
have been achieved in pilot production with this
equivalent of a self-aligned selective emitter design
technology, with corresponding efficiencies as high
with the metal only contacting the silicon in heavily
as 18.4%.
doped regions. This gives the cell top surface
structure much higher voltage capability although This concept of semiconductor fingers does not
improved rear surface passivation will be necessary appear to have been used in large-scale commercial
to capitalise on this. The biggest challenge in solar cell production, and has considerable appeal
turning this into a robust technology for large scale as it facilitates good conductivity within the
production is the very small metal/silicon interface emitter, but without the normal trade-off found
area which is less than 0.1% of the top surface. in screen printed cells. Normally, such regions of
This is sufficient for suitably low contact resistance good emitter conduction are located at the top
to facilitate high fill factors provided everything surface and therefore degrade the cell spectral
works properly. However variability in this contact response and current generating capability due to
resistance is a weakness in the design causing the corresponding extremely short minority carrier
efficiencies in pilot production of the technology diffusion lengths in such regions. This technology
to vary from 17 to 18.4%, with the average being is particularly well suited to multicrystalline silicon
below 18%. wafers that normally degrade when exposed to
prolonged thermal treatments.
The processing sequence for this technology is
easily retrofitted onto a standard screen-print line Another possible implementation of this
and is as follows: technology would appear to be the incorporation
of the laser chemical processing (LCP) techniques
1. Surface texturing
developed by the Fraunhofer Institute in Germany
2. Emitter diffusion (100 ohms per square)
in conjunction with the laser company Synova.
3. Rear surface etch plus edge isolation
This would allow the superior performance of
4. SiNx deposition by PECVD (top surface)
the laser grooved semiconductor finger solar cell
5. Semiconductor fingers by laser doping
to be combined with the simplicity and low cost
6. Rear metal plus front metal screen-printing
of the laser doped semiconductor finger solar
7. Firing of metal contacts cell. Both institutions have intellectual property
that would appear to provide significant benefits
Only step 5 deviates from standard homogeneous in combination.
emitter screen-printed solar cell fabrication, with
the overall fabrication appearing to be simpler
and shorter than those sequences proposed for
introducing selective emitter designs for screen-
printed solar cells.

27
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Cross-sectional SEM photos show: (a) discontinuities

in the Al-doped p+ layer; (b) a deep and uniform Al-
doped p+ layer.

Figure 4.3.2.5

4.3.2.3.3 Advanced Semiconductor have the potential to achieve open circuit voltages
Finger Solar Cells approaching 700mV.
As part of the core research of the Centre of The development of this technology will form part
Excellence funded by the ARC and UNSW, an of the ASI project involving Silex and Suntech. It is
enhanced version of the semiconductor finger solar particularly well suited to multicrystalline silicon
cell was designed and developed to overcome the wafers due to the avoidance of prolonged high
limitations of the standard semiconductor finger temperature thermal processes. Based on small area
solar cell described in Section 4.3.2.3.2. This involves test devices, the inclusion of rear surface passivation
plating 2 microns of silver (or Ni/Ag or Ni/Cu/Ag) to is expected to take efficiencies on multi material to
the screen-printed metal at the end of processing, over 18% with corresponding open circuit voltages
which simultaneously plates a similar thickness in excess of 650mV while on mono, efficiencies
to the semiconductor fingers when using light of 21% are anticipated with open circuit voltages
Photo of a hybrid screen-printed
induced plating as shown in Figure 4.3.2.4B. exceeding 670mV.
and plated solar cell showing the
three levels of metallisation. Each of The application of this plated metal has several
the two busbars effectively collect 4.3.2.3.4 Advanced N-Type Screen-
benefits. Firstly, it makes good ohmic contact
current from 8 small solar cells,
each with a tapered screen-printed to both the heavily doped silicon and also the Printed Solar Cells
line that carries the current from screen-printed metal, overcoming the high contact
the narrow plated lines to the resistance sometimes experienced by the standard 4.3.2.3.4.1 N-type Screen-Printed Cells
respective busbar/interconnect. semiconductor finger solar cell. This makes high fill- with Homogeneous Top Surface Diffusion
Figure 4.3.2.4b factors routinely achievable while solving the yield Increasing interest is being shown in n-type
problems of the original implementation. Secondly, CZ material as a means for avoiding the widely
the increased conductivity of the semiconductor reported defects associated with the high boron
fingers facilitates increasing the screen-printed and oxygen concentrations in p-type CZ material.
metal line spacing to typically 2-3cm. Although the In particular, screen-printed aluminium has been
corresponding width has to be increased to 400 used as a simple and cost-effective way to create an
microns, it has the benefit of allowing both tapering Al-alloyed rear emitter for such n-type CZ material,
and increased height to above 50 microns. especially in the n+np+ cell design with rear junction.
However low voltages for such structures reported
This reduces the shading loss of the screen-printed
in the range 617-627mV appear to be a severe
metal fingers from 5% down to about 1%, with
limitation of this approach since n-type CZ wafers
the corresponding Jsc increase taking efficiencies
should be capable of achieving much higher open
to above 19%. Interestingly the effective shading
circuit voltages.
loss of the semiconductor fingers was expected
to increase by 0.5% although in reality, the plated Discontinuities in the p+ layer have been
laser doped surface appears rough enough to identified as the main cause for such performance
scatter the reflected light so that almost half is degradation of this device. These discontinuities,
totally internally reflected at the glass/air interface as seen in Fig.4.3.2.5 (a), are isolated points where
and returned to the solar cell surface. Therefore the junction fails to form, usually created by non-
the effective shading loss of the semiconductor uniform wetting of the silicon by the Al during the
fingers remains virtually unchanged. Importantly, alloying process. Although their presence can be
by avoiding the contact between the screen-printed minimized by optimizing the firing process so as to
silver and the lightly doped silicon, the top surface allow uniform wetting of the surface to occur, they
design of this solar cell behaves very differently to cannot be completely avoided. In small quantities,
standard screen-printed contacts and appears to such non-uniformities have almost negligible
influence on the performance of the back surface

28
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

PL images illustrate the improvement in uniformity

and quality of the p+ layer achieved by the new Inverted form of the PERL (Passivated Emitter and Rear
firing process. (a) before and (b) after the low Locally diffused) solar cell developed at UNSW based
temperature treatment. on the use of N-type silicon.

Figure 4.3.2.6 Figure 4.3.2.7

field in conventional p-type cells. However, they can base being very thick, approaching
significantly degrade the quality of the Al-alloyed 200 microns, the high lifetime of the
emitter in n-type wafers by allowing Al to locally CZ n-type material allows the transistor
bypass the p+ region and directly contact the n-type to achieve moderate gain levels and
bulk via a Schottky barrier causing a non-linear therefore conduct large currents when
shunt. A new and modified firing process has illuminated by light that generates
therefore been developed to avoid the damage the necessary base current for the
from such non-uniformities. In this method, a transistor. The unusual consequence
patented low temperature solid phase epitaxial is that the apparent shunt resistance
growth process is employed after the conventional of the cell is very poor (low) when
standard spike firing to minimize the impacts of illuminated brightly, increasing to
these junction discontinuities so that a uniform and high values in the dark or even low
good quality junction as illustrated in Fig. 4.3.2.5 (b) illumination levels. The ramifications
can be achieved. of this for cell efficiency and fill-factor
are that values fall with reducing light
These improvements have facilitated a 15-20mV
intensity as is normally the case for Schematic of a laser doped
increase in Voc relative to those reported in the
shunted cells, but with values increasing again Al-alloyed rear junction n-type
literature and efficiencies over 17% for standard
for low illumination levels as the shunt problems solar cell.
screen-printed solar cell technology with
disappear as the phototransistors are deactivated. Figure 4.3.2.8
homogeneous n-type emitter (front surface field)
applied to n-type wafers. Figure 4.3.2.6 shows Even without such phototransistor shunting,
the improvement in the photoluminescent (PL) analysis of such homogeneous emitter n-type
response resulting from the improved firing and devices indicates that even higher voltages
formation of the rear contact and junction. are potentially achievable if not for the large
dark saturation current contribution from the
A common problem with the manufacture of
heavily doped phosphorus diffused top surface.
conventional screen-printed solar cells is minute
This emphasises the importance of moving to
amounts of aluminium paste accidently coming
the equivalent of a selective emitter design for
into contact with the cell front surface. This can
the front surface to facilitate both improved
happen when wafers are face down during the
short wavelength response as well as increased
printing of the rear or during wafer transfer/
device voltages.
handling. Aluminium paste contamination of such
surfaces and transfer belts happens relatively easily
in a production environment such as through a
4.3.2.3.4.2 N-type Screen-Printed Cells
broken wafer that has been Al printed, operators’ with the Equivalent of a Selective Emitter
contaminated gloves, tiny holes in the screen- Until recently, the Centre held the world record
printing screen etc. However such contamination (jointly with Stanford University) for the most
with the current rear junction n-type technology efficient n-type silicon devices with 22.7% efficiency.
creates unusual photoactivated shunts that are not The cell design was based on the inverted form
present in the dark. of the PERL (Passivated Emitter and Rear Locally
diffused) solar cell developed at UNSW and is
This is because a p-n-p transistor structure is formed
shown below. In this work, the cell design has
when the unwanted aluminium on the top surface
been adapted to accommodate the use of low cost
is fired into the n-type surface (with the second p-n
screen-printed solar cell processes involving the
junction of course being at the rear of the device).
alignment of the screen-printed front metal lines to
In this phototransistor, the lightly doped wafer
the heavily doped n+ regions to form the equivalent
forms the base of the transistor, and despite the
of a selective emitter on the front surface and the

29
 use of a screen-printed aluminium grid pattern on heavily diffused lines. The wafers were rinsed and
the rear to form the localised p+ regions during the submerged in 1% HF solution for 30sec to remove
ARC spike firing of the Al. the dopant source and any native oxide from the
PHOTOVOLTAICS laser doped lines. Lastly, light induced plating (LIP)
CENTRE OF This approach enables the achievement of 18%
was subsequently performed to deposit Ni, which
efficiency but requires a quite complicated
EXCELLENCE was sintered at 400oC, followed by Cu plating to
processing sequence.
2010/11 form the front contacts.
ANNUAL REPORT For comparison purposes, the screen-printed front
The performance improvement relative to the
contacts were replaced with laser doped contacts.
selective emitter screen-printed counterparts arises
With a top surface homogeneous emitter diffused
primarily from the reduced shading losses by the
to about 200 ohms per square, the self aligned
top surface metalisation although slightly higher
metalisation with heavy doping beneath the
voltages as high as 650mV, fill-factors and yields
metal contact was formed by melting the silicon
are also achieved, apparently due primarily to the
through the silicon nitride anti-reflection coating
reduced metal/silicon interface area and superior
in the presence of an n-type dopant source. This
alignment with the laser doped contacts.
laser doping process automatically damages the
overlying silicon nitride layer, facilitating direct A spectral response measurement was performed
plating of metal to the heavily doped regions as to investigate different regions in the cell. A very
a self-aligned top surface metal contact. The cell high value of more than 95% was maintained for
structure is shown below. This approach appears to internal quantum efficiency (IQE) from 580nm to
have significant advantages over the screen printing 960nm. However, there was a slight drop in the
technique such as self alignment to the locally short wavelength range, indicating that the front
diffused top surface, narrower metal surface passivation can be further improved.
lines and corresponding significantly
100
lower shading losses. Despite being a 4.3.2.3.5 Buried Contact Solar Cells
90
significantly simpler process than the
80 Despite this technology being commercialised more
screen-print selective emitter design,
70 than a decade ago, it remains a key technology for
excellent cell efficiencies of 18.6%
60 IQE collaborative research with industry and continues
have been achieved on large cell
(%) 50 to do well commercially with close to $1 billion of
EQE areas of 148.6cm2 using commercial
40 product now deployed in the field and many new
Reflectance grade CZ n-type wafers. Commercial
30 companies interested in its commercial potential.
partners have demonstrated even
20 New developments at the Fraunhofer Institute in
higher efficiencies, exceeding 19%
10
conjunction with the laser company Synova make
while simultaneously using only
the groove formation and doping a much simpler
0 industrial equipment.
360 480 600 720 840 960 1080
and lower cost process than incorporated into the
In this work, the cells were fabricated original technology implementation. The German
Wavelength (nm)
using industrial sized (125x125mm) company Rena is offering a turn-key production line
phosphorus-doped CZ n-type wafers for this technology although it appears at this stage
Spectral response of a typical laser
of 3ohm-cm resistivity and ~180mm that significant laser induced damage is making the
doped n+np+ solar cell.
thickness. A texturing process was performed in a achievement of high efficiencies difficult.
Figure 4.3.2.9 NaOH/Isopropanol based solution to form random
Never-the-less, this is expected to create
pyramids. A thin phosphorus diffused n+-layer with
significant new interest in the Buried Contact
a sheet resistivity of 200 ohm/sq serving as the front
technology, particularly with the apparent growing
surface field was created by a thermal diffusion at
acknowledgement of the benefits of the technology
850oC using liquid POCl3 source in a conventional
over existing screen-printed solar cell technology.
tube furnace. A chemical etch containing HF
In the 2006 European Inventor of the Year Awards,
and HNO3 was then applied to the
the Buried Contact technology contributed to
back of these wafers to remove the
Green and Wenham receiving a Top 3 Ranking (out
unwanted n+-layer from this surface.
of more than 200,000 inventions world-wide in the
A silicon nitride layer with a refractive
period 1990-2000) for inventors outside Europe. A
index of 2.05 and thickness of 75nm
further distinction for this technology is its listing
was subsequently deposited using a
amongst Australia’s Top 100 Inventions of the 20th
commercial Roth & Rau remote plasma
Century as determined by the Australian Academy
PECVD system to simultaneously form
of Technological Sciences and Engineering. The
an anti-reflective coating and provide
original design for this solar cell has the buried
passivation for the front surface as well
metal contacts on the top surface as shown in
as the bulk material.
Figure 4.3.2.10 although in more recent years
After screen-printing with a selected the interdigitated rear contact design applied to
Aluminium paste on the entire back n-type wafers shown in Fig. 4.3.2.11 has become
surface with a gap of around 2mm particularly popular and a focus of UNSW research.
from the edges, the wafers underwent Efficiencies of 20% have been demonstrated
an alloying process at 860oC in a on small area devices with efficiencies of 21%
Buried contact cell with buried
contacts on both front and conventional conveyor belt furnace to produce eventually believed achievable on large area
rear surfaces. the rear Al-alloyed emitter. A phosphorus dopant CZ material.

Figure 4.3.2.10
source was then spun onto the front surface,
followed by the laser doping process using a 532nm
wavelength Q-switch diode laser to create locally

30
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Laser doped selective emitter solar cell with self

Buried contact cell with interdigitated rear contacts. aligned plated metal contacts.

Figure 4.3.2.11 Figure 4.3.2.12

4.3.2.3.6 Laser Doped Selective Emitter Copper. Using conventional electroless plating
Solar Cells techniques which tend to plate conformally with a
uniform plating rate in all directions, the 10 micron
The benefits of a selective emitter have been well
wide laser doped line plates to a width of about
known and quantified for many years. The benefits
30 microns for a plating height of about 10
of heavy doping beneath the metal contacts
microns as shown in Figure 4.3.2.13 (a). In
contribute significantly to the high performance
comparison, the new light induced plating
levels achieved by technologies such as the Buried
techniques achieve line widths of only 24
Contact Solar Cells, the semiconductor finger
microns for even greater metal height as
solar cell, the point contact solar cells and the
shown in Fig. 4.3.2.13 (b).
world-record holding Passivated Emitter and Rear
Locally diffused (PERL) solar cell. The heavy doping Contact resistances below 0.001 ohmcm2 have
not only facilitates reduced contact resistance been demonstrated, leading to fill factors
between the metal and the silicon, but probably as high as 80% being achieved on devices
more importantly it shields the high recombination of approximately 150cm2 in area. This has
velocity metal/silicon interface from the active facilitated the achievement of efficiencies
regions of the cell. In addition, by restricting the above 19% on commercial-grade CZ p-type
heavily doped material to the immediate regions silicon. Of particular importance has been the
beneath the metal contact, little light absorption defect generation accompanying the laser
takes place in such regions thereby avoiding doping process, particularly in conjunction
problems with carrier collection from heavily doped with dielectric-coated, textured silicon
regions where the minority carrier diffusion lengths surfaces. Planar surfaces present minimal
are very short. challenge in terms of achieving near defect
free regions in the vicinity of the laser
Despite the commercial success of the Buried
melted regions. Textured surfaces however,
Contact Solar Cell described above, a potentially
particularly in conjunction with dielectric
more effective and simpler way of achieving a
coatings of significantly different thermal
selective emitter is by using laser doping to produce
expansion coefficient, have provided a
the heavily doped regions beneath the metal
significant challenge to match the low defect
contacts as shown in Fig. 4.3.2.12. Following top
densities achievable with planar devices.
surface emitter phosphorus diffusion to about 100
Important processing parameters in the
ohms per square and silicon nitride deposition,
optimization of the laser doping process
an n-type dopant source is applied or can even be
for textured surfaces have included laser
incorporated into the silicon nitride layer. A 532 nm
pulse envelope shape, pulse duration, pulse
NdYAG laser is used to melt the silicon to a depth
frequency, laser light frequency, laser power,
in the vicinity of a micron while simultaneously
beam focus as well as the type of dielectric and
releasing the n-type dopants into the molten
dopant source being used.
region. The molten silicon subsequently regrows
epitaxially, heavily doped with phosphorus. Just The TT team has had significant success
as importantly, the overlying silicon nitride layer helping various companies get this technology
is removed from the silicon surface in isolated into pilot production, often achieving higher
regions, facilitating direct plating to the exposed efficiencies and on larger area commercial (a) electroless conformal plating
of Cu onto a laser doped region
n++ surface. Electroless plating of Ni and Cu provides wafers, than has been achieved at UNSW. For and (b) Improved aspect ratio
a particularly effective self aligned metalisation example, 19.0% efficiency or higher has been for metal plated by the new light
scheme to provide metal lines wherever the laser achieved in pilot production at several companies, induced plating techniques where
doping was effected. The laser doped regions are four of which have chosen to do joint publications the height is 12 microns while the
typically 12 microns wide, leading to metal lines of width is 24 microns. (c) FIB photo
with UNSW documenting the achievements
of Cu plating using the new light
only 20 microns width after plating. including Sunrise Global Energy in Taiwan, Roth induced plating techniques as
and Rau in Germany, Centrotherm in Germany and for (b).
Innovative light induced plating techniques have
Shinsung in Korea.
been developed that facilitate the achievement of Figure 4.3.2.13
very high aspect ratios for the plated Nickel and

31
 25
80
20
ARC 70
15
PHOTOVOLTAICS 60
10

Reflectance (%)
CENTRE OF 50 400 550 700 850 1000

EXCELLENCE 40
Photolith. inverted pyramids
2010/11 30 I.J. inverted pyramids

ANNUAL REPORT 20
I.J. v-grooves

10

0
300 450 600 750 900 1050 1200
Wavelength (nm)

(a) Optical Microscope

photograph of ink jet Reflectance of silicon wafers
patterned lines in a SiO2 layer with different surface finish (i)
and (b) Scanning Electron inverted pyramids formed by
Microscope photograph of photolithography, (ii) inverted
the grooves formed in the pyramids patterned by ink jet
silicon surface following KOH printing and (iii) v-grooves
etching of (a). patterned by ink jet printing.

Figure 4.3.2.16 Figure 4.3.2.17

The preferred implementation of the laser doping new manufacturers, to take up the new LDSE
selective emitter (LDSE) technology also uses an technology in large scale production.
equivalent laser doping/plating combination on
Two other popular implementations of the laser
the rear surface using a boron doping source. In
doping technology with industry collaborators are
this cell design, the majority of both the front and
the bifacial structure using laser doped self-aligned
rear surfaces is well passivated using silicon nitride
contacts of opposite polarity on both surfaces and
although the preferred rear surface passivation
the interdigitated rear surface laser doped contacts
of the undiffused p-type surface uses somewhat
for rear junction n-type devices. Collaborative
different deposition parameters for best results.
research projects based on these cell designs have
Implied Voc values above 730mV at one-sun
been established with the aim of developing the
demonstrate the near perfect passivation achieved
30-40 micron diameter holes technologies to take cell efficiencies on n-type
with such surfaces. Following laser doping of the
formed by ink-jetting. CZ also to above 20% in large scale commercial
rear surface, even though plated contacts can be
production at some stage in the future.
Figure 4.3.2.14 used similarly to on the front surface, the preferred
rear contact is achieved through depositing Another use of the laser doping technology was
aluminium over the entire surface followed by a described in Section 4.3.2.3.4.2 as a replacement
low temperature sinter used to form good ohmic for screen-printed contacts with n-type CZ wafers
contact with the boron laser doped regions. In in conjunction with aluminium alloyed rear
this cell design, the aluminium layer provides an contact and junction. As reported above, excellent
excellent rear surface reflector. Even with standard efficiencies in the vicinity of 19% have also been
commercial grade p-type CZ wafers, based on achieved with this cell design on 148.6cm2 n-type
laboratory results it appears this technology will CZ wafers. A simpler patented version of this
achieve comfortably over 20% efficiency on full technology has also been developed which requires
sized commercial wafers when established in pilot no diffusion processes, no edge junction isolation
production in early 2011 with impressive open- and no thermal processes above 450 degrees
circuit voltages in the vicinity of 670mV. Celsius except for the aluminium alloying process
for several seconds. Never-the-less, the technology
The simpler version of the LDSE technology
is still able to achieve efficiencies above 18% on
equivalent to the pilot production established at
full-sized commercial substrates. The processing
Sunrise with aluminium alloyed rear surface, is far
sequence used is:
simpler to retrofit onto existing screen-printed
production lines, with 90% of existing equipment 1. Wafer texturing
retained. For several manufacturers such as 2. Silicon nitride deposition
Shinsung in South Korea, this new technology 3. Aluminium rear surface printing and firing
whereby the laser doped selective emitter 4. Laser doping front surface
combined with light induced plating of the metal 5, Ni/Cu light induced plating
contacts are used to replace the front surface
screen-printed metal is achieving significantly
(a) Optical Microscope photograph
higher efficiencies well in excess of 19% compared 4.3.2.3.7 Inkjet Technology for Solar Cell
of ink jet patterned holes in a SiO2
layer and (b) Scanning Electron to 18% for conventional screen-printed cells with Fabrication
Microscope photograph of the minimal if any cost increase per cell. Roth and Rau
inverted pyramids formed in the
have also secured the rights from UNSW to develop 4.3.2.3.7.1 Resist Based Method
silicon surface following KOH
etching of (a). and sell turn-key production lines based on the Inkjet technology has been an area of rapid
LDSE technology world-wide This is expected to development over the last decade, particularly for
Figure 4.3.2.15 make it far simpler for many companies, particularly printing. In recent years, its application has been
spreading to other fields, but as yet has had only

32
 80
25
70 ARC
20

60
PHOTOVOLTAICS
15
CENTRE OF
Reflectance (%)

50 10
400 550 700 850 1000 EXCELLENCE
40

Random upright pyramids

2010/11
30
I.J. inverted pyramids ANNUAL REPORT
20 I.J. v-grooves

10

0
300 450 600 750 900 1050 1200

Wavelength (nm)

(a) Reflectance of silicon wafers with different surface

finish (i) laboratory fabricated random, upright pyramids,
(ii) inverted pyramids patterned by ink jet printing and (iii)
v-grooves patterned by ink jet printing and (b) SEM of (iii).

Figure 4.3.2.18

minimal impact in photovoltaics. The first company

apparently to commercialise a photovoltaic
technology incorporating inkjet technology is CSG
Solar who during 2006 commenced production
of a thin-film technology that uses inkjet printing
of a corrosive material to etch patterns in a
resist layer to facilitate metal contacting to the
underlying silicon.

In the present work, the use of inkjet technology

has been expanded to encompass a range of solar
cell fabrication processes including texturing,
grooving, patterning of dielectric layers for metal
cheaper alternative to photolithographic based Honeycomb texturing (a) and
contacting, localized diffusions, etc. The techniques
processing. macrogroove texturing (b) of
developed to carry out these processes are uniquely multicrystalline silicon wafers
different to those used before. A non-corrosive These new approaches appear capable of achieving through the use of inkjet
plasticizer is inkjet printed onto a low cost resist similar device performance levels but with the technology.
layer, altering the chemical properties of the resist greatest challenge being to match the dimensions Figure 4.3.2.19
layer in these localized regions to make them of features in the resist patterning achievable with
permeable to etchants such as hydrofluoric acid photolithography. Test devices to date based on
(HF). This facilitates the patterning or etching of inkjet technology have achieved feature dimensions
underlying dielectrics or semiconductor material such as holes of 30-40 microns diameter as shown
to facilitate a range of semiconductor processes. in the matrix of holes below.
Importantly, the change in resist permeability is However, these dimensions need
a reversible process making it feasible to return to be reduced to about 10 microns
the resist to its original state after carrying diameter to fully match the
out processes on the underlying material. This performance levels demonstrated
also opens the option to partially reverse the with photolithographic based
permeability to reduce the hole or feature size processing. New and innovative
produced in the underlying material. inkjet printing techniques
have been recently developed
A particular exciting application of this inkjet
for further reducing the resist
technology work is for very high efficiency silicon
patterning dimensions. This work
solar cells. The University of New South Wales
is being greatly assisted by the
has held the world record for silicon solar cell
recent availability of the new 1
efficiencies for the last 15 years, initially with the
picolitre inkjet heads.
Passivated Emitter solar cell (PESC) and more
recently with the Passivated Emitter and Rear The described dielectric patterning
Locally diffused (PERL) solar cell. Despite the capabilities via inkjet patterning Schematic showing the principle
performance and achievements of these two can not only be used for defining localized diffusion of “Direct Etching” for patterning
dielectric layers.
technologies, neither has been used commercially, regions and locations for metal contacts, but also for
apart from for space cells, primarily due to the the formation if textured surfaces and light trapping Figure 4.3.2.20
sophistication, cost and complexity of the processes schemes. Examples of the latter are shown below as
involved. The photolithographic based processing is well as corresponding reflectance curves.
probably the main contributor to this. In this work,
Significant interest is also being shown in
inkjet printing techniques have been developed for
developing high efficiency approaches for
patterning low cost resist layers as a simple, much

33
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

(a) Optical image, and (b) Dektak profile of a groove

etched in a 260 nm thick thermal oxide grown on
a polished Si wafer by deposition of 10 layers of
aerosolised 10% (w/v) NH4F solution onto a polyacrylic
acid polymer layer of thickness ~ 2.2 microns. The
aerosol and sheath gas flow rates were 21 and 50 cm3/
min, respectively. The process velocity of the stage
was 10 mm/s and the platen was heated to 45ºC. The
groove width at surface ~35 microns.

Figure 4.3.2.21

involves fewer steps than photolithography and is

safer than existing immersion etching techniques
in that the corrosive etchant HF is only formed
in-situ on the surface to be etched. Furthermore,
the method uses small amounts of inexpensive
chemicals and produces significantly less hazardous
fluoride waste than existing immersion etching
methods.

Aerosol jet printing is a new deposition technology

being pioneered by Optomec, Inc. The technique
enables the finely-controlled deposition of an
(a) Optical image of grooves multicrystalline silicon wafers. Record performance aerosol, which is generated from a liquid, by using
etched in a 260 nm thick thermal multicrystalline silicon cells have in the past a sheath gas to constrict the aerosol into a fine jet
oxide grown on a polished Si benefited from photolithographically defined which is directed to the substrate. The technique
wafer by deposition of 25 layers
“honeycomb” texturing using an acidic isotropic has been previously used in applications such as
of aerosolised 10% (w/v) NH4F
solution onto a polyacrylic acid etch. Such structures have also been demonstrated printed electronics, fuel cells and displays.
polymer layer of thickness ~ 2.2 recently using inkjet technology as shown in Figure
mm. The aerosol and sheath gas 4.3.2.19. Such techniques appear to have significant The aerosol etching method has been used to etch
flow rates were 18 and 55 cm3/min, commercial appeal. groove structures in SiO2 which have been thermally
respectively. The process velocity grown on polished silicon wafers. By varying the
of the stage was 10 mm/s and the
aerosol and sheath gas flow rates the geometry
platen was heated to 45ºC. (b) 4.3.2.3.7.2 Direct Etching Via
Dektak profile of a section of the and depth of etching can be varied. For example,
Aerosol Jetting
grooves shown in (a). grooves (see Figure 4.3.2.21) were cleanly etched in
The key to high-efficiency silicon solar cells is the a 260 nm thick SiO2 layer using aerosol and sheath
Figure 4.3.2.22 ability to form small-area metal contacts to the gas flow rates of 21 and 50 cm3/min respectively
silicon through dielectric passivating layers. In this and depositing 10 layers of aerosolized 10% NH4F
work, a new method for patterned etching of SiO2, solution. These grooves were ~35 microns wide at
SiOxNy and SiNx dielectric layers the surface but only ~ 20 microns wide at the Si/
has been developed and patented. SiO2 interface.
The method uses Optomec’s M3D
aerosol jetting device to deposit a Even narrower structures can be etched by reducing
solution containing fluoride ions, the aerosol flow rate and increasing the sheath gas
according to an etching pattern, flow rate. Figure 4.3.2.22 shows the formation of
onto an acidic water-soluble central, more deeply, etched grooves which are 200
polymer layer formed over the nm deep and ~10 microns at the Si/SiO2 interface.
dielectric layer as indicated in the It is anticipated that the use of an even lower flow
schematic of Figure 4.3.2.20. The rate for the first, wetting layer will largely eliminate
deposited solution reacts with the wider, shallow etched region at the surface thus
the polymer layer, at the locations enabling very narrow grooves to be etched to a
where it is deposited, to form an depth of at least 200 nm.
etchant that etches the SiO2 and SiNx We have also been able to etch grooves 20-
under the polymer layer to form a 25 microns wide in 75 nm thick SiNx layers on
pattern of openings in the dielectric chemically-textured silicon surfaces. The etched
Aerosol jet-etching of a silicon
oxynitride layer to form lines of only
layer. After the pattern of openings is formed, the grooves can be used to form front or rear metal
15-20 microns. acidic water-soluble polymer and the etch residue contacts for silicon solar cells using metal plating,
are easily removed by rinsing in water. The method screen printing or evaporation deposition methods.
Figure 4.3.2.23

34
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Spectral response curve of a 19%

efficient Pluto solar cell.

Figure 4.3.2.24
Solar cell efficiencies of commercially
manufactured Pluto solar cells (a) in
chronological order and (b) as a histogram.

Figure 4.3.2.25

Compared to current inkjet implementations of 17.5% to 19.5%. These efficiencies have been
this direct etching method, aerosol jet etching independently confirmed by the Fraunhofer
results in faster etching (fewer layers required Institute in Germany. The main contributors to
to be deposited), smaller etched features and these increased efficiencies are: reduced shading
less variation in etched groove width over large loss (3%); improved short wavelength response
etching patterns. Although the current etched due to selective emitter (3%); improved Voc due
feature sizes are sufficiently small for many current to lightly diffused emitter and well passivated
front-contact and rear-contact silicon solar cells, it surfaces (3%); and improved fill-factor due to
is anticipated that the feature sizes can be further reduced resistive losses (3%).
reduced in the near future with refinement of the
In comparison to the semiconductor finger
jetting parameters and modification of the surface
technology, many of the losses associated with
polymer layer composition. Replacing the thermal Pluto production cell (right) with
the screen-printed metal contacts (shading, significantly reduced reflection
oxide or SiNx layer with SiOxNy also has beneficial
contact resistance, metal resistance, dark saturation losses compared to standard
consequences for the dimensions of the feature
current from the metal/silicon interface, etc.) have screen-printed solar cells (left).
sizes formed such as shown by the etched lines of
been minimised through the Pluto technology,
15-20 microns width shown in Figure 4.3.2.23. These Figure 4.3.2.26
facilitating an efficiency increase by about 1%
further reductions in etched feature size may enable
in absolute terms. The spectral response and
high-efficiency silicon solar cell designs, such as the
corresponding performance spread in production
PERL cell, to be commercially realised.
are shown in Figures 4.3.2.24 and 4.3.2.25
This area of work has attracted significant interest respectively.
and funding from industry, including being
In comparison to the laser doping technology, the
awarded a 3-year ARC Linkage grant for the design
main advantage is the elimination of most of the
and fabrication of high efficiency solar cells through
defects and associated recombination that exists in
the use of the described patented inkjet technology
the resolidified laser doped regions or in locations
from UNSW.
immediately adjacent to these. Consequently, Pluto
cells achieve reduced junction recombination and
4.3.2.3.8 PLUTO Technology therefore higher fill factors and Vocs leading to a
The Pluto technology is a low cost implementation performance increase of about 5%.
of the UNSW PERL (Passivated Emitter and Rear
Locally diffused) solar cell developed at UNSW. This technology is unique to Suntech with some
The PERL technology currently holds the world- of the intellectual property being owned solely
record for silicon solar cell performance with by Suntech. Intellectual property arising from the
25% efficiency being recorded as the new world collaborative research is jointly owned by UNSW
record for silicon solar cell efficiency during 2008. and Suntech, but with the latter having the right to
Through collaborative research with Suntech, use it in its own production. Suntech has publicly
two generations of Pluto technology have been announced that it has increased the manufacturing
developed. The first uses the standard front surface capacity of the Pluto technology to 450MW during
design of the PERL cell but with a screen-printed 2010. This will be further scaled during 2011 with
rear aluminium alloyed rear metal contact that the full 2GW production capacity to be retrofitted
simplifies the technology and allows it to be easily to the Pluto technology during 2012. Figure 4.3.2.26
retrofitted onto existing screen-printed solar cell shows a photo of a Pluto cell juxtaposed to a
lines. In comparison to standard screen-printed standard screen-printed solar cell.
solar cells made on the same production line, Pluto The second generation of Pluto technology
cells achieve an increase in performance of more uses the full PERL cell design by eliminating
than 10% by taking average efficiencies from about the rear screen-printed aluminium layer and

35
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Luminescence image of a
completed multicrystalline silicon
solar cell.

Figure 4.3.2.27

replacing it with a localised metal contact. It is carrier lifetime shown below for a multicrystalline
expected to take efficiencies to the 20-23% range silicon solar cell. The data acquisition times for
depending on wafer quality. It does however such measurements are of the order of typically
avoid the use of vacuum deposition processes and only one second per wafer, which is orders of
photolithographic techniques, thereby avoiding magnitude faster than any competing experimental
the expensive fabrication costs associated with techniques. High resolution images of other
most laboratory technologies. This technology is electrical parameters such as the shunt resistance
expected to follow the first generation of Pluto are also feasible.
technology into large scale production about 1-2
A prototype bench-top system established at
years later. Suntech is also the only company with
the Centre three years ago was in such heavy
the right to manufacture generation 2 Pluto devices.
demand by various UNSW research groups and for
The PLUTO technology has been also successfully consultancy projects that a new state-of-the-art
applied to multicrystalline silicon wafers and system was purchased from BTI during 2010. It has
implemented into large scale production with helped identify a variety of unexpected processing
efficiencies of 17-18% also independently problems and in developing new processing
confirmed by the Fraunhofer Institute. In mid sequences much more quickly than previously
2009, Suntech beat the 15.5% multicrystalline possible. In many cases the two dimensional
silicon module world-record efficiency held by information contained in PL images has given
Sandia National Laboratories in the US for 15 years. easy to interpret clues about the origin of specific
This was followed later in the year by Suntech’s problems. The ability to measure a large number
achievement of another world-record of 16.5% of samples in a short time is very beneficial in
(aperture area) for a multi module, this time beating this context.
the recently achieved ECN world record in the
Collaboration with various industrial partners
Netherlands. During 2010, this module efficiency
has also identified various processing problems
was further increased to 17% and again confirmed
previously unknown to the manufacturers. In
by the Fraunhofer Institute.
addition, recent research at UNSW has shown how
the luminescence imaging technique could be used
4.3.2.3.9 Photoluminescent Imaging for for in-line process control, for example to remove
Device and Material Characterisation the influence of shunts. Other potential industrially
With the current growth in the photovoltaic relevant in-line applications for luminescence
industry and the trends towards higher cell imaging that are under investigation include crack
efficiencies, generally achieved on lower detection, spatially resolved series resistance
quality and thinner silicon wafers, there is an monitoring, quality control of raw material and
increasing demand in research laboratories and process control of individual key processing steps
in industrial manufacturing for fast and easy to such as the emitter diffusion. The collaborative
use characterisation tools. Recent research at work with various industry partners and several
UNSW has established photoluminescence (PL) journal and conference publications have resulted
techniques as extremely fast and useful tools for in significant interest worldwide in the technique,
the characterisation of silicon wafers and of silicon specifically in a commercial PL imaging system.
solar cells.
A start-up company BT Imaging Pty Ltd has been
In particular, this work has shown that luminescence formed by Centre researchers Robert Bardos and
imaging techniques give two dimensional high Thorsten Trupke in 2007 with commercial sales of
resolution images of electrical parameters such as the system taking place during 2008. The company
localised series resistance values and the minority has developed commercial off-line and in-line

36
luminescence imaging tools, with healthy sales to screen-printed cells world-wide on standard
research groups and cell manufacturers throughout commercial grade p-type CZ wafers. This work is
ARC
the world during 2010. now to continue as an industry funded project in
conjunction with the ASI, with the industry partners PHOTOVOLTAICS
In summary, photoluminescence techniques have CENTRE OF
being Silex and Suntech.
proven particularly useful for the characterisation EXCELLENCE
of various aspects of the screen printed solar Other important areas of the core research are
2010/11
cell process. Photoluminescence imaging was where problems are tackled and the solutions
ANNUAL REPORT
applied to the study of belt furnace processing, can be used for the benefit of all or most industry
while photoluminescence imaging with current partners. Examples include improvements in the
extraction was applied to the characterisation of light induced plating techniques, reducing the laser
n-type solar cells with printed metallisation to induced defects during the laser doping process,
identify failure mechanisms in the metallisation. The reducing feature sizes during inkjet patterning of
accuracy, high speed and non-destructive nature dielectric layers, improving surface passivation
of PL techniques make it an attractive candidate techniques, adapting processes to suit low cost
for use in high-throughput screen printed solar cell silicon etc. The Centre has a large number of
production lines. very talented PhD students contributing to this
core research, the achievements from which are
4.3.2.4 Pilot Production significantly accelerating the progress for most
industry collaborative research projects. This
Many of the above technologies have been
represents a major benefit to companies that
implemented or are in the process of being
engage in funding photovoltaic research at UNSW.
implemented into pilot production. Table 4.3.2.1
summarises the performance of each of the Some core research however is conducted in
technologies in pilot production using standard conjunction with companies, but where the
commercial wafers with commercial processes companies are not funding the work at UNSW and
and equipment. have no claim over any resulting IP from the work at
UNSW and cannot exercise any restriction regarding
4.3.2.5 Core Research its use. An example of this is the collaboration with
The distinguishing feature of core research is that it Centrotherm to evaluate the compatibility between
is funded by UNSW and the Centre of Excellence or the LDSE technology and the standard Centrotherm
other Government funding scheme. The priorities screen-printing technology. Cells from a standard
for the work are therefore set by the Centre rather Centrotherm screen-printing line for producing
than by industry. Any IP is therefore wholly owned homogeneous emitter solar cells of 18% efficiency
by UNSW, but with any such developments made were fabricated without front silver metal. The
available at no cost to industry partners of the emitter sheet resistivity was the only processing
Centre such as those funding collaborative research variation, being raised to 120 ohms per square.
programs or else licensing Centre technology. Such After laser doping and plating at UNSW, the finished
core research is therefore of great importance in the 5 inch cells were sent to the Fraunhoffer Institute in
initial stages of development of a new technology Germany where they were confirmed to be 19.3%
to take the development to the point where efficiency. This is believed to be the highest ever
industry interest is gained and industry is willing independently confirmed for industrial solar cells
to fund the ongoing development. An example fabricated on standard commercial grade p-type CZ
of this is the Advanced Semiconductor Finger wafers. This record may however be shortly broken
Solar Cell described in Section 4.3.2.3.3 where through collaborative work with Shinsung where
core research has facilitated the development and similar devices fabricated using similar wafers have
demonstration of the concept, to the point where been measured to be 19.6% efficiency and have
test structures can show that efficiencies above been sent to Fraunhofer for confirmation.
19% are achievable, making these the most efficient

Jsc Voc Effic Area Wafer

Technology ARC (mA/cm2) (mV) FF (%) (%) (cm2) type
1 BCSC SiO2 36.8 619 79.0 18.0 148.6 (p-type CZ)
1 BCSC SiN 33.4 604 77.4 15.6 156 (p-type multi)
2 IBBC SiN 37.1 675 79.4 19.8 46 (n-type CZ)
3 Screen-print SiN 37.1 632 78.8 18.5 148.6 (p-type CZ)
3 Screen-print SiN 33.3 614 75.5 16.0 156 (p-type multi)
4 SCF SiN 36.7 635 79.0 18.4 148.6 (p-type CZ)
5 LDSE (Sunrise) SiN 37.8 638 79.0 19.0 225 (p-type CZ)
5 LDSE (UNSW) SiN 37.6 638 79.6 19.0 148.6 (p-type CZ)
5 LDSE (Centrotherm) SiN 38.4 638 78.9 19.3 148.6 (p-type CZ)
5 LDSE (Roth & Rau) SiN 37.8 635 78.4 18.8 225 (p-type CZ) Performance of each of
5. LDSE (Shinsung) SiN 38.6 640 79.3 19.6 148.6 (p-type CZ) the technologies in pilot
production using standard
5 LDSE SiN 35.4 628 77.9 17.3 156 (p-type multi) commercial wafers with
5 LDSE SiN 36.0 645 79.9 18.6 148.6 (n-type CZ) commercial processes and
6 Pluto SiN 38.4 638 79.5 19.5 148.6 (p-type CZ) equipment.

6 Pluto SiN 38.9 638 81.9 20.4 4.0 (p-type CZ) table 4.3.2.1
6 Pluto SiN 35.7 633 78.6 17.7 156 (p-type multi)
7 PERL ZnS/MgF2 41.7 704 82.5 24.2 46 (p-type FZ) 37
 4.4 Second Generation:
Silicon,
ARC Organic and other
“Earth
PHOTOVOLTAICS
CENTRE OF Abundant” Thin-Films
EXCELLENCE
2010/11
ANNUAL REPORT

4.4.1.1 Summary is a thick pigmented resin (P150) with a

Research Team The focus of the silicon thin-film group high load of TiO2, while the best BSR for
research has been on further improvements a textured cell is a highly reflective white
University Staff paint. Secondly, further development in
in the technology of e-beam evaporated
• Dr Sergey Varlamov (group leader) poly-crystalline (poly-Si) thin-film solar cells, plasmonic poly-Si thin-film cells led to
• Prof. Martin Green which is conducted within an ARC Linkage record-breaking Jsc enhancement of 50%.
grant project with CSG Solar. Thirdly, a simple and effective Si film etch-
Research Fellows back texturing process has been introduced
• Dr Thomas Soderstrom (since 6/2010) The UNSW e-beam evaporator, the tool to improve light-trapping in the e-beam
supplying precursor Si films for the poly-Si cells on planar glass, which allowed
Technical Staff project, underwent major refurbishment achieving a record current of 26.6 mA/cm2,
• Dr Patrick Campbell during 2010, allowing greatly improved the highest ever reported for evaporated
• Mark Griffin reliability and up-time of the tool, to poly-Si solar cells.
reduce background contamination level
Postgraduate Research Students and to achieve more accurate control of The first encouraging results have been
• Johnson Wong (PhD) (until 9/2010) Si deposition rate and doping, allowing obtained for transient heating defect
• Zi Ouyang (PhD) (until 9/2010) deposition of high quality Si films. After anneal of poly-Si films. The open-circuit
processing the films into solar cells and voltage improvement of about 58 mV was
• Guangyao Jin (PhD) (until 9/2010)
applying a newly developed etch-back Si demonstrated after poly-Si cell structures
• Jialiang Huang (PhD)
texturing process a new UNSW efficiency were exposed to diode laser or flash lamp
• Yuguo Tao (PhD)
record for evaporated cells of 7.1% (Voc radiation for only a few milliseconds. The
• Peter Gress (PhD)
458 mV, Jsc 26.6 mA/cm2, FF 58%) has been diode laser was also successfully applied to
• Hongtao Cui (PhD)
demonstrated. large area boron diffusion to prepare back
• Do Soon Jung (Masters)
surface field (BSF) layers.
• Bonne Eggleston (PhD) Outstanding progress has been made
• Wei Li (PhD) in the area of light-trapping in poly-Si A large amount of data has been collected
• Qian Wang (PhD, since 3/2010) thin-film cells. Firstly, a comprehensive about effects of different annealing
• Kyung Hun Kim (Masters, since 3/2010) study of various back-surface reflectors treatments on solid phase crystallisation
• Anthony Teal (PhD, since 3/2010) (BSR) of practical significance has been (SPC) of Si films and resulting poly-Si film
• Jae Sung Yun (PhD, since 3/2010) conducted. It allows choosing the best BSR quality and cell performance. A focus was
• Chaowei Xue (PhD, since 9/2010) depending on the cell design and structure. on application of higher temperatures
For example, the best BSR for a planar cell during either the whole SPC process or
Visiting Academic
A. Prof. Soon Kweon (since 1/2010)
38
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Structure of a p-type poly-Si solar cell on planar glass

(layer thicknesses and grain size not to scale).

Figure 4.4.1.1

its stages, such as the incubation, to shorten the remove and passivate defects, and then
crystallisation while maintaining or even improving processed into metallised modules
the cell performance. The result is that, regardless [4.4.1.3]. The highest achieved efficiency
of when a higher temperature is used, it always for 94 cm2 mini-modules produced by
causes poorer poly-Si crystal and electronic quality. CSG technology is 10.5% [4.4.1.4], and full
However, the degree to which the cell performance scale 7-8% efficient modules have been
deteriorates, depends on the Si film preparation manufactured in Thalheim, Germany
conditions. E-beam cells and hybrid cells, i.e. the since 2006.
cells with PECVD emitter and e-beam absorber and
Research in the thin-film group at
BSF, can be crystallised at 640C within only two
UNSW focuses on critical cell fabrication
hours (as compared to 20 hrs at 600C) without a
processes and explores a range of
significant loss in their performance. In the area of
advanced approaches to improve the
the SPC emitter seed-layer approach, the factors
performance and manufacturability
leading to the best seed crystal quality are being
of poly-Si thin-film cells. The group’s
studied.
work over the past years has led to
A comprehensive concept of performance limiting innovative solutions to the key steps of Process sequence of the four
recombination in poly-Si has been developed. the cell fabrication process, including Al induced types of poly-Si thin-film on glass
solar cells under development at
According to the concept it is intragrain defects, glass texturing (AIT) [4.4.1.5] and localised surface
UNSW. All cells are designed for the
such as dislocations that are responsible for limiting plasmons (SP) in metal nanoparticles [4.4.1.6] for superstrate configuration (i.e., the
the minority carrier lifetime in poly-Si material and enhanced light trapping, silicon deposition by sunlight enters the cell through
thus the cell performance. In the practical range of high rate PECVD [4.4.1.7] and e-beam evaporation the glass).
dopant concentrations in the quasi-neutral region [4.4.1.8], defect anneal and passivation, and cell Figure 4.4.1.2
of the cell absorber (> ~5E15 cm-3), the lifetime metallisation and interconnection [4.4.1.9]. The
limiting recombination occurs via shallow levels following sections summarise the thin-film cell
linked to dislocations. At dopant concentrations research projects and their results.
lower than ~5E15 cm-3 the lifetime is limited by
recombination via deep levels introduced by 4.4.1.3 Solar cell fabrication sequence
charged impurities at dislocations and possibly
The thin-film group research in 2010 focused on
grain-boundaries. The dislocation density in poly-Si
two different solar cell types: cells deposited by
thin-film cells was estimated from the TEM images
PECVD and by e-beam evaporation. A schematic
and found to be of an order of 1E10 cm-2, which
representation of a poly-Si thin film cell is shown in
is consistent with experimental voltages of about
Figure 4.4.1.1 and Figure 4.4.1.2 describes the cell
500 mV.
fabrication sequence.

4.4.1.2 Introduction Additionally hybrid cells combining a PECVD emitter

with an e-beam absorber and BSF were produced
The technology of thin-film poly-Si on glass solar
in collaboration with CSG Solar. The substrate
cells is developing steadily and it is expected
used for all cells is 3.3 mm thick borosilicate glass
ultimately to become cost-competitive with
(Schott Borofloat33). Both planar and textured glass
other thin-film technologies [4.4.1.1, 4.4.1.2]. The
substrates are investigated: typically e-beam and
presently best developed poly-Si on glass PV
hybrid cells are fabricated on the planar glass while
technology has been commercialised by CSG Solar
PECVD cells are fabricated on textured glass. The
AG where an approximately 1.5 mm thick PECVD
texture on the Si facing glass surface aims to reduce
amorphous silicon (a-Si) precursor diode is solid-
the reflection losses and to enhance the light
phase-crystallised to form a poly-Si film, which is
trapping in a weakly absorbing poly-Si thin-film. The
annealed and hydrogenated to activate dopants,

39
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE 110
100 680 C 640 C 600 C
2010/11 90

ANNUAL REPORT 80
70
60

Quv [%]
50
40
30
20 E-beam
10 Hybrid
0 PECVD
-10
5 50 500
Annealing Time [min]

Normalised UV crystallinity factor (Quv) of PECVD,

Calculated random scatter absorption limit (black
E-beam and Hybrid films as a function of the annealing
dots) and measured optical absorption in 2 μm thick
time at 600ºC, 640ºC and 680ºC.
poly-Si thin films on planar, AIT, and SAIT glass with air
as the BSR. Figure 4.4.1.5
Figure 4.4.1.4

texture is prepared by the UNSW-developed AIT mini-modules of 36 cm2 area using the proprietary
method [4.4.1.10] described below and consisting CSG Solar metallisation technology described
of an irregular rough array of sub-micron sized elsewhere [4.4.1.4].
dimples.

The next step is the deposition of an

4.4.1.4 Glass Texturing
approximately 75 nm thick SiN film (refractive An advanced glass texturing method was
index ~2.1 at 633 nm) by PECVD at about 400°C developed at UNSW and it is referred to as
or by reactive sputtering at 200°C. The SiN film “aluminium induced texture” (AIT) [4.4.1.5, 4.4.1.10].
acts both as an AR coating and a barrier layer A thin sacrificial Al film is deposited onto planar
for contaminants from the glass. An a-Si n+/p/ glass, followed by annealing at about ~600°C in
p+ precursor diode is then deposited by either an inert atmosphere. The anneal initiates a red-ox
e-beam evaporation or PECVD on planar or AIT reaction whereby Al is oxidised to Al2O3 and SiO2
glass respectively, followed by SPC at 600°C. The from the glass is reduced to silicon, as follows:
resulting as-crystallised poly-Si diodes have Voc
of only about 150-180 mV. A high temperature
treatment is then performed at ~1000°C for 30-60
A surface texture is imparted to the glass according
s to activate the dopants and to anneal defects,
to the nucleation conditions provided during this
which increases Voc up to 220-250 mV. The final
anneal. Subsequent wet-chemical etching removes
step of the material preparation is hydrogenation
the reaction products from the glass surface and
in a remote hydrogen plasma to passivate the
reveals the glass texture (Figure 4.4.1.3a). Further
remaining defects resulting in Voc of 450~500 mV
improvement in the AIT process conditions led to
typical for the finished cells.
development of so-called “sub-micron AIT” (SAIT)
The cells are metallised using a combination glass with smaller and steeper feathers (Figure
of photolithographic patterning with dielectric 4.4.1.3b), which ensure even better light absorption
and metal deposition processes to create an in a poly-Si film prepared on such texture (Figure
interdigitated electrode structure on the rear (Si 4.4.1.4) [4.4.1.16].
side) of the cells. Typically, the BSF electrode of the
Typical atomic force microscope (AFM) images of
PECVD cells covers the whole rear cell surface and
the bare AIT and SAIT glass surfaces with the RMS
the cells can only be illuminated from the glass
roughness of 800 and 150 nm respectively and an
side [4.4.1.11, 4.4.1.12]. The BSF electrode of the
SEM image of the AIT glass coated with a 2 μm thick
e-beam cells consists of thin Al fingers covering only
poly-Si film are shown in Figure 4.4.1.3.
4% of the rear surface, thus allowing bifacial (i.e.
illumination from both glass and Si side) operation As shown in Figure 4.4.1.4, the absorption in
AFM images of bare AIT (a) and SAIT
(b) glass and SEM image of 2 μm of the cells [4.4.1.13, 4.4.1.14, 4.4.1.15]. The details the 2 μm thick poly-Si films on the best AIT
thick poly-Si film on AIT glass (c). of the cell metallisation are described in respective and particularly SAIT glass in the 500~1000 nm
sections below. wavelength interval is very close to the calculated
Figure 4.4.1.3
random scattering absorption limit, based on
The hybrid cells are made on planar glass with the
Monte Carlo ray tracing. It should be noted that
SiN layer and the emitter deposited in the CSG Solar
the significantly higher absorption than the
large area KAI PECVD tool and the absorber and BSF
theoretical limit at 1000 nm and above can be
layers in the UNSW e-beam evaporator. The films
attributed to the enhanced parasitic absorption in
are processed into either individual 2 cm2 large solar
the glass and/or measurements errors as discussed
cells using the UNSW bifacial metalisation or solar
elsewhere [4.4.1.11].

40
 ARC
Films Method Incubation Time [min] Crystallisation Time [min] PHOTOVOLTAICS
600°C 640°C 680°C 600°C 640°C 680°C CENTRE OF
OTM 300 50 16 780 140 29
EXCELLENCE
Raman 360 50 12 720 125 28
2010/11
XRD 270 62 16 1020 120 32
UV-R 270 50 14 900 125 32 ANNUAL REPORT
PECVD Average 300 53 14 855 128 30
OTM 300 35 11 720 90 23
Raman 300 32 10 660 90 24
XRD 270 31 12 1020 105 24
UV-R 270 31 9 900 90 25
E-beam Average 285 32 10 825 94 24
OTM 300 45 14 780 130 28
Raman 360 40 12 690 120 28
Incubation and
XRD 270 60 15 1020 135 30 crystallisation times
UV-R 270 45 12 900 120 30 for Si films (solar cell
Hybrid Average 300 48 13 848 126 29 structures) deposited by
Generic equation [4.4.1.19] 660 108 22 different techniques.

Table 4.4.1.1

4.4.1.5 Solid Phase Crystallisation summary of the incubation and full crystallisation
Annealing Temperature [ C]
Solid phase crystallisation (SPC) is a thermally times based on all film characterisation methods is 680 640 600
given in Table 4.4.1.1. -6
activated process of transformation of a-Si into poly- E-beam = 3.01 eV
Si, which is used for poly-Si thin-film cell fabrication. Hybrid = 2.82 eV
Based on the data from the table it is possible -7 PECVD = 2.72 eV

ln(1/time) [s-1]
The SPC parameters, such as the exact temperature to estimate the activation energies (Ea) for the
profile, as well as the film structure, such as the incubation and crystal growth processes using -8
composition and thickness of the individual the Arrhenius law. The Arrhenius plots are shown
layers, have effects on the resulting crystal and in Figure 4.4.1.6. The Ea calculated from the plots -9
electronic quality of poly-Si film and thus on the cell Incubation
are 3.0 and 3.3 eV (e-beam), and 2.7 and 3.2
performance. A detailed study of the SPC kinetics eV (PECVD), and 2.8 and 3.2 eV (hybrid) for the -10
was conducted to clarify such effects. Several -7 E-beam = 3.31 eV
incubation and crystal growth respectively. The Hybrid = 3.17 eV
techniques – optical transmission microscopy PECVD = 3.18 eV
ln(1/time) [s-1]
Ea for the crystal growth falls within the reported -8
(OTM), Raman, UV reflection (UV-R), and X-ray 3.1-3.4 eV range, but the Ea of less than 3.0 eV for
diffraction (XRD) spectroscopes, scanning electron the incubation is significantly smaller than ~3.4 eV -9
microscopy (SEM) – were used to characterise the previously found [4.4.1.20]. Most likely it is because
SPC kinetics and the crystal quality of the films. -10 Crystal growth
the nucleation rate, to which the incubation time
External Quantum Efficiency (EQE) and Suns-Voc is closely related, is enhanced by particular film -11
measurements were used to characterise the cell deposition conditions and/or the particular film 12.0 12.5 13.0 13.5
performance [4.4.1.17, 4.4.1.18]. 1/kT [eV-1]
structure, such as the presence of the heavily
Representative SPC kinetics for Si films deposited doped layers and interfaces.
Arrhenius plots for the incubation
by different techniques (PECVD, e-beam, and The average grain size in the e-beam poly-Si films and crystal growth processes in
hybrid) at different temperatures based on the UV-R crystallised at different temperatures was estimated PECVD (♦), E-beam (■) and Hybrid
measurements are shown in Figure 4.4.1.5. Similar (●) films with the n+/p-/p+ solar
from the SEM images shown in Figure 4.4.1.7, where
cell structure.
so-called “S-curves” were obtained by the other the grain boundaries were enhanced by Secco
characterisation techniques mentioned above. The Figure 4.4.1.6

Voc Jsc Absorber Doping Diffusion Length [nm]

Films T [°C] [mV] [mA/cm2] p.FF [%] p.Eff [%] [/cm3] Emitter Absorber Performance parameters of
600 474 12.7 76 4.6 5.4×1015 160 2400 the poly-Si thin film solar cells
640 466 11.5 70 3.8 5.8×1015 140 1800 obtained by integration of the
PECVD 680 456 10.8 63 3.1 4.6×1015 135 1500 EQE curves over the AM1.5G
600 474 13.4 76 4.8 9.8×1015 180 2150 solar spectrum, from Suns-Voc
640 462 13.3 75 4.6 8.6×1015 155 2100 measurements, and C-V analysis.
E-beam 680 432 12.9 74 4.1 9.5×1015 120 2000 All cells have an area of 2 cm2.
600 459 14.4 75 4.9 2.9×1015 235 4000 Table 4.4.1.2
600 455 14.1 75 4.8 3.2×1015 210 4000
Hybrid 640 433 13.6 73 4.2 2.5×1015 170 4000

(a) (b) (c)

SEM images of the fully

crystallized and Secco etched
e-beam poly-Si films for different
SPC temperatures.
3 m 3 m 3 m Figure 4.4.1.7

41
 2.7
ARC 2.4 c-Si

Average Grain Size [µm]

E-beam 600 C
PHOTOVOLTAICS 2.1

Normallised intensity
PECVD 640 C
CENTRE OF 1.8 Hybrid
680 C
T
EXCELLENCE 1.5
2010/11 1.2
ANNUAL REPORT
0.9

0.6
560 600 640 680 720
SPC Temperature [℃] 508 512 516 520 524 528 532
Shift [/cm]

a) The average grain size of PECVD (♦), E-beam (■) and

Hybrid (●) films obtained from respective SEM images
as a function of the SPC temperature.
b) Raman spectra of the E-beam films crystallised at
different temperatures (c-Si as a reference). The peak
intensities are normalised.

Figure 4.4.1.8

Intrinsic P 8E18, cm-3 P 1e20, cm-3 P 3E20, cm-3

after SPC [4.4.1.21]. If it is
used as a seed layer followed
Thickness, nm Shift, cm -1
FWHM, cm -1
Shift, cm FWHM, cm Shift, cm
-1 -1
FWHM, cm -1
Shift, cm
-1
FWHM, cm -1 -1

by solid-phase epitaxy (SPE)

50 517.7 7.46 517.6 7.43 517.7 6.86 517.4 7.87
of the absorber and BSF
100 518.4 7.42 517.3 6.92 517.8 6.72 517.0 7.23
layers, it can help to improve
200 518.5 6.75 518 6.90 518.5 6.94 518.5 6.76
the quality of the cell poly-Si
material and to lead to
The Raman peak shift and etching. The grain size decreases from 2.3 µm to 1.5
better cell performance. To
FWHM for emitter seed layers µm and 1.0 µm as the SPC temperature rises from
find preparation conditions that result in better
of different thickness and with 600C to 640C and 680C. Similar trends are observed
different P doping. seed layer properties a series of films was fabricated
for PECVD and hybrid Si films as shown in Figure
with different P concentrations and thickness. The
Table 4.4.1.3 4.1.1.8. The Raman peaks for Si films of all types are
seed layer quality was assessed by Raman (Table
slightly shifted to the left of the c-Si peak, and their
4.4.1.3) and UV-R spectroscopy and the Hall mobility
FWHM are larger for the
(Table 4.4.1.4). The best quality is found for the
higher SPC temperatures
Thickness, nm Carrier conc., cm-3 Hall coeff., cm3/C Mobility, cm2/V*s thickest seed layer of 200 nm and intermediate P
similarly indicating a slightly
concentration of 1e20 cm-3. Using the optimised
50 8.22E19 -9.12E-08 63 poorer crystal quality
emitter seed layers high crystal quality poly-Si cell
100 8.66E19 -8.67E-08 53 of the films crystallised
structures were grown by SPE as confirmed by the
200 8.76E19 -8.56E-08 35 at higher temperatures
98% Quv quality factor (Figure 4.4.1.9).
(Figure 4.4.1.9).
Hall mobilities for emitter seed All performance parameters of the poly-Si thin-film 4.4.1.6 Transient heating defect anneal
layers of different thickness and with
P doping of 3.3E20 cm-3.
solar cells degrade for higher SPC temperatures As was mentioned in section 4.4.1.3, the defect
but to a different extent (Table 4.4.1.2). The PECVD annealing in poly-Si films after SPC is a very
Table 4.4.1.4 cells are most affected, with the pseudo efficiency important process for improving cell Voc. Finding
decreasing from 4.6% to 3.1% for SPC temperatures the best conditions for this process is complicated
of 600C and 680C respectively. The e-beam and due to conflicting trends. On one hand, higher
hybrid cell pseudo-efficiencies annealing temperatures and longer times are
decrease more modestly, from preferred because of better dopant activation
4.9% to 4.2% for the same SPC and more complete defect dissolution. On the
temperatures. There is hardly any other hand, the same conditions cause excessive
loss in the e-beam and hybrid cell dopant smearing and glass distortion, which need
performance after SPC at 640C but it to be avoided. Typically, for any given annealing
allows shortening the SPC time from temperature, there is an optimum time balancing
typical 20 hrs to only 2 hrs. It makes favourable processes with unfavourable ones. For
640C a preferred SPC temperature example, at 950ºC such an optimum time is about
for such cells, particularly for process 3 min, while a similar Voc effect can be achieved at
development purposes when 0.1- 1050ºC for only 50 s. Following this trend, it can be
0.2% loss in the efficiency is not estimated that, at temperatures near the Si melting
important. point (1414 ºC), the optimum annealing times
The research continued in the area can be of order of a few milliseconds, because of
of the SPC emitter seed layer. It was which this type of processes are generally called
UV reflection peaks at 274 nm for previously reported that a heavily “millisecond annealing” [4.4.1.22]. Very importantly,
SPE poly-Si cell structures and c-Si phosphorus doped (>1e20 cm-3) Si layer can form a for such short times the thermal diffusion length
wafer as a reference. in glass is of an order of tens of microns only
high crystal and electronic quality poly-Si material
Figure 4.4.1.9 [4.4.1.23], which means the bulk of the glass stays

42
 ARC
rear metal PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
Si film
ANNUAL REPORT
glass
sunlight front metal

Schematic representation of the interdigitated Schematic cross section of a poly-Si thin-film solar cell
metallisation scheme of PECVD poly-Si thin-film with a point-contacted rear surface (not to scale).
solar cells on glass (not to scale). Figure 4.4.1.11

Figure 4.4.1.10

cold resulting in less distortion and lower thermal an area of 4.0 cm2, and an intended
140
budget for the annealing process. structure of glass/SiN/n+p-p+. The
120
standard processing sequence
Two recently available techniques, which allow

Current [mA]
100
includes SPC at 600oC for ~ 15 hrs;
heating a few micron thick Si film up to its melting 80
rapid thermal defect annealing and Voc 510mV
point within a few millisecond long time are diode 60 Jsc 29.0mA/cm2
dopant activation at ~1000ºC for
lasers and flash lamps [4.4.1.22, 4.4.1.24]. With the 40 FF 62.6%
~ 1 min; remote hydrogen plasma Eff 9.3%
diode lasers in particular, the exposure time can 20
defect passivation at 600ºC for ~30
be easily controlled in the CW mode from about 1 0
min; followed by metallisation using -200 -100 0 100 200 300 400 500 600
ms to 100 ms by the laser beam scanning speed,
the interdigitated contact scheme Voltage [mV]
while the degree of Si film heating is controlled by
shown in Figure 4.4.1.10 and
adjusting the laser power.
described elsewhere [4.4.1.11].
Initial experiments on effects of millisecond Measured current-voltage of a
One of recent advances in improving the PECVD cell 9.3% efficient PECVD solar cell
annealing of poly-Si cell structures have been
performance was the development of an optimised made on AIT textured glass
conducted using either the UNSW diode laser (silicon thickness 2.3 µm, cell area
metallisation scheme with a superior BSR made of
(LIMO450-12x0.3) or the flash lamp system 4.0 cm2, approximated AM1.5G
a combination of a thin silica layer and evaporated
provided by external collaborators (Research Center spectrum, 100 mW/cm2, cell
Al. The rear Al electrode makes a contact to the cell temperature 25°C).
Forschungszentrum Dresden-Rossendorf, Dresden,
BSF layer through small vias photolithographically
FZD). In the very first tests the Si film samples Figure 4.4.1.12
defined in the silica film (Figure 4.4.1.11). When
developed severe cracking and even delamination
the PECVD cells are fabricated on the AIT glass and
after diode laser treatments at room temperature,
metallised using such a point contacted rear-surface
so all following experiments were performed at
structure they possess enhanced light-trapping
elevated glass temperatures from 550C to 700C. A
and achieve both high Jsc of 29 mA/cm2 and record
Voc increase up to about 60 mV has been observed
efficiency of 9.3% [4.4.1.7].
for poly-Si film samples on both planar and textured
BSG as well as SLG for the exposures between 3 and The measured I-V curve of the record cell together
20 ms (Tables 4.4.1.5 and 4.4.1.6). It is only slightly with the other important cell parameters is
lower than the Voc increase of about 75 mV typical illustrated in Figure 4.4.1.12. The measurements
for the standard RTA treatment at 950-1000C for were apertured to 4.0 cm2 area and a NREL
tens of seconds. calibrated cell was used as a reference.

4.4.1.7 PECVD cells

PECVD cell structure and Power density, Exposure, Temperature, V(0.1), Voc, Voc change,
performance kW/cm2 ms oC mV mV mV The Voc results for diode
10.3 3 550 354 433 59 laser treated poly-Si on glass
The poly-Si thin-film cells made with
10.0 3 600 336 418 44 cell structures.
the use of the PECVD are the most
1.1 150 550 261 347 -27
technologically advanced and have Table 4.4.1.5
1.1 150 600 257 344 -30
achieved the highest efficiencies
0.8 360 550 283 364 -10
compared to other cell types. 0.8 360 600 283 365 -9

The cells are made

on 3.3 mm thick Power density, Exposure, Temperature, Voc, Voc gain, Voc results for flash lamp
textured Borofloat Glass kW/cm2 ms Pulse shape oC mV mV treated poly-Si on glass
glass, have about SLG planar 2.4 20 Gaussian 700 385 56 cell structures.
2-2.5 μm thickness, BSG planar 7.8 3 Gaussian 700 407 40
Table 4.4.1.6
BSG textured 1.8 20 Structured 700 430 59
BSG textured 1.5 20 Gaussian 700 425 54 43
 Metallisation on cell level
completed: top view of a 5x5 cm2
ARC sample showing 4 individual cells
PHOTOVOLTAICS plus cross sectional view (SiN and
silica layers not shown) with the
CENTRE OF emitter and air side busbars.
EXCELLENCE
Figure 4.4.1.15
2010/11
ANNUAL REPORT

A schematic air-side top view of a

mini-module using wire-bonded
cell interconnection.

Figure 4.4.4.16

High rate PECVD of a-Si

films for poly-Si cells a-Si:H material confirms an identical crystal quality
The manufacturing of poly-Si thin-film to that of the standard rate poly-Si as shown in
solar cells is a commercially viable process Figure 4.4.1.14.
[4.4.1.1]. However, the deposition of
the precursor a-Si films by currently The poly-Si cell structures were fabricated with
conventional PECVD has been identified the thin emitter and BSF layers deposited at
as having high cost and a low throughput the standard rate and the ~2 µm thick absorber
due to its low deposition rate (25-35 nm/ deposited at the high rate of 250 nm/min in a
min) and the high equipment cost [4.4.1.3]. different PECVD chamber. After the standard
Si evaporation by electron beam (e-beam) material processing sequence described above, the
Deposition rate of a-Si:H as a has been investigated as a higher rate and Voc of the high rate cell was higher than the Voc of
function of power for the SPC- lower cost alternative. A challenge with e-beam the standard rate cell as shown in Table 4.4.1.7.
optimised PECVD process.
evaporation is its directional nature (“line-of-sight” The current research in PECVD poly-Si cells is
Figure 4.4.1.13 deposition), leading to non-conformal coating focusing on developing advanced cells, which
of textured surfaces, which results in combine the best AIT glass for enhanced light
microstructural defects and the poor trapping with the high rate PECVD Si films to
1.2
textured cell performance [4.4.1.25, achieve efficiencies exceeding 10%.
hi gh r at e
1
4.4.1.26]. Another high rate process, which
s t d. r at e
0.8 in contrast to the evaporation can provide PECVD cell metallisation – cell
Si waf er
a conformal coating of textures, is hot-wire
0.6 interconnection
CVD. However, no performance results have
0.4 During fabrication of large size thin-film PV
been reported for the SPC hot-wire CVD
modules, the as-deposited films are divided
0.2
solar cells, and it is also known that the
into smaller area unit cells, which are then
0 hot-wire CVD a-SiH has very high nucleation
505 510 515 520 525 530 535 interconnected in series to reduce the resistive
rates [4.4.1.23], which may lead to poor
ohmic losses. A novel approach allowing a low
crystal and electronic quality of SPC poly-Si.
loss interconnection of individual thin-film
Raman spectra (normalised) of the
poly-Si films obtained by SPC of the A major part of research conducted on PECVD of cells into a minimodule using wire-bonding
standard (pink) and high (blue) rate a-Si:H has been traditionally focused on minimising was recently developed in the UNSW thin film
PECVD a-Si:H. The spectrum of c-Si the defect density for a-Si:H solar cell applications. group [4.4.1.9]. Figure 4.4.1.15 shows a sample
is (black) shown for reference. However, device-grade a-Si:H, which performs well containing four individual PLASMA cells prior to the
Figure 4.4.1.14 for a-Si:H cells, is not necessarily the best precursor interconnection process.
material for making poly-Si cells by
The interconnection method consists of laser
SPC and a higher rate PECVD a-Si:H
Absorber deposition rate, 1-Sun Voc, n=1 Voc, n=2 Voc, scribing an isolation groove along the glass side
may also be suitable for producing
nm/min mV mV mV aluminium busbar to define the cells. Wire-bonds
a good electronic quality poly-Si
35 494.6 ± 8.9 496 666 are then placed over this groove, connecting the
films. Using a conventional 13.56
250 512.4 ± 4.5 513 711 air-side BSF busbar to the glass side emitter busbar
MHz PECVD system and a process
of the adjacent cell as shown in Figure 4.4.4.16. This
Suns-Voc results (the average of five optimised for the SPC application,
method can be expanded by applying a conductive
points) after RTA and hydrogenation much higher deposition rates up to 265 nm/
of SPC poly-Si diodes on AIT material (with a non-conductive adhesive) to the
min have been demonstrated as shown in Figure
glass with PECVD a-Si:H absorber surface of the cell, and wire-bonding the first and
4.4.1.13 [4.4.1.7].
deposited at 35 nm/min (“low rate”) the last cell busbars to the surface of the conductive
250 nm/min (“high rate”). Raman microprobe characterisation of the poly-Si material as also shown in the figure.
Table 4.4.1.7 films obtained after SPC from the high rate PECVD

44
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
I-V curves of the individual cells and 2010/11
the wire-bonded minimodule. ANNUAL REPORT
Figure 4.4.1.17

Schematic the of e-beam Si

evaporation tool with boron
and phosphorus effusion cells
for in-situ doping.

Figure 4.4.1.18

functional 2 cm2 area 6% efficient cells.

Figure 4.4.1.19 illustrates the progress
with the efficiencies of the e-beam
cells and Table 4.4.1.9 summarises the
design features of the cells typically
Table 4.4.1.8 and Figure 4.4.1.17 show the example fabricated at UNSW. The cells are also
electrical results from wire-bonding of two adjacent made with the reverse polarity structure,
PECVD cells. As can be seen, the interconnection i.e. the p-type glass side emitter, and the
scheme improves the Jsc and, thus, the overall n-type absorber and air-side BSF layers.
module efficiency compared to the individual cells,
During previous years the focus of the
which is primarily due to the large reduction in
e-beam cell research was on improving
lateral series resistance along each busbar to the
the poly-Si material quality and optimising Evolution of the PECVD and e-beam
nearest series or module wire-bond connection cell efficiencies at UNSW.
the cell structure and post deposition treatments
point. All I-V measurements were done with
(RTA, hydrogenation) to achieve better voltages. Figure 4.4.1.19
aperture to define the cell area.
Respectable Voc in the range up to 500 mV were
measured by the Suns-Voc technique on the non-
4.4.1.8 E-Beam evaporated Si cells metallised cells. More recently the focus has shifted
to developing a working metallisation scheme,
E-Beam cell structure, fabrication, and
improving the cell currents and fill factors (FF), i.e.
performance producing functional e-beam cells with appreciable
Deposition of Si using an e-beam evaporator can conversion efficiencies.
be performed at a very high rate of up to 1 mm/
min. Other advantages of the method include
absence of toxic gases, good Si source material
usage and compatibility with a continuous in-line
deposition mode. Thus, if performed in a non-UHV AB PECVD thin-film cell
environment (base pressure > 1×10-8 Torr, pressure Cell A Before Cell B Before interconnected (prior to interconnection,
encapsulation) and module
during Si evaporation > 1×10-7 Torr), e-beam Voc (mV) 508.7 507.9 1010.6
(after interconnection,
evaporation is a cost effective Si deposition method Isc (mA) 96.4 97.6 101.3 encapsulation)
for the thin-film PV applications (Figure 4.4.1.18). FF (%) 69.78 68.45 70.9 electrical properties.
Eff (%) 7.8 7.7 8.3
The thin-film group at UNSW has been researching Table 4.4.1.8
e-beam Si thin-film cells since 2004, and significant
progress has been achieved in advancing the
cell technology from a few square millimetre
area mesa devices at the start to recent fully Parameter Details
Glass 3.3 mm planar Borofloat33
AR coating SiN (~75 nm, n ~2.1)
Typical design parameters of
Emitter n+ (~100 nm, P, 5e19~1e20 cm-3, 400~500 W/sq)
the UNSW E-beam Si cells.
Base p (~2 µm, B, 1e16~1e17 cm-3)
Table 4.4.1.9
BSF p+ (~150 nm, B, up to 1~5x1019 cm-3, ~1000 Ω/sq)
RTA 10~20 s at 1000°C or ~4 min at 900°C
Hydrogenation 15~20 min at ~600°C, remote plasma
Metal 0.5-1.5 μm thick interdigitated Al fingers on front & rear
Back reflector Diffuse while paint

45
 Voc (mV) Jsc (mA/cm2) FF (%) Eff (%) Rs (Ωcm2)
ARC 479 20.1 66 6.3 4
PHOTOVOLTAICS
CENTRE OF Measured 1 sun J-V results of
EXCELLENCE 1.8 µm thick planar e-beam
2010/11 cells with aligned bifacial
metallisation ([4.4.1.13,
ANNUAL REPORT 4.4.1.14]). The shunt resistance
of the cells is > 2000 Ωcm2
and has a negligible effect on
the performance.

Table 4.4.1.10

Schematic (left) and actual (right) views of a metallised e-beam cell (white
paint BSR not shown).

Figure 4.4.1.20

Doping (cm-3) Voc (mV) FF (%) Jsc (mA/cm2) Rs (Ωcm2) Eff (%) pEff (%) The dopant concentration was determined using
1.7×1015 413 64.1 19.2 2.0 5.09 5.5 the Z-analysis technique described elsewhere
5.7×1015 418 61.9 17.6 3.4 4.67 5.3 [4.4.1.27]. This technique has an advantage of
1.0×1017 449 62.5 6.5 7.9 2.16 2.5 yielding the electrically active dopant density,
8.3×1017 403 54.4 4.0 19 1.04 1.3 which is most relevant to the cell performance. The
cells were characterised by Suns-Voc, EQE, light
Performance parameters of e-beam E-Beam cell metallisation J-V measurements and the results are shown in
cells with different absorber
The e-beam cell metallisation scheme, illustrated Table 4.4.1.11. To account for the spectral mismatch
doping densities.
in Figure 4.4.1.20, was developed in 2008 and allow between the light I-V tester and the AM1.5G solar
Table 4.4.1.11 bifacial illumination and versatile back surface test spectrum, which causes measurement artifacts,
reflector (BSR) study. It features interdigitated Al line a conservative approach was used where the short-
contacts for both heavily doped layers, the emitter circuit current density during the J-V measurements
and BSF. The BSF electrodes only cover about 4% of was adjusted to match those determined from
the area. The emitter electrode is formed in a way EQE measurements.
that avoids contact with the absorber layer entirely.
The major observed effect of the doping density
Instead of plasma etching the grooves
is a sharp reduction of the cell current with
through the whole Si thickness, a timed etch
increased doping. It dominates a weaker effect
stop is used such that a thinned emitter layer
on the cell voltage, where Voc rises slightly up to
still remains at the bottom of the grooves.
dopant density of about 1e17/cm3 before falling
Then, the glass-side comb-like electrode is
sharply at the higher densities. The best cell
centred in the grooves leaving significant
efficiencies are obtained at the lowest active dopant
space (>10 μm) between the edges of the
concentrations in the range between 1e15/cm3 and
electrodes and the groove sidewalls. After
5e15/cm3. Figure 4.4.1.21 shows the effect of the
metallisation the entire cell rear surface is
doping densities on the key cell parameters.
then coated with a layer of a white paint as
a diffuse BSR. The typical cell performance
parameters obtained with the described
Role of back surface filed
metallisation are listed in Table 4.4.1.10. The BSF is very important for the poly-Si solar cell.
Besides transporting the holes to the metallic
E-Beam cell absorber contacts, it also repels the electrons from the rear
doping density cell surface thus reducing recombination at this
interface. In fact, this effect is crucial for the poly-Si
The electronic properties of the absorber
thin-film device as shown in Figure 4.4.1.22. Without
layer, mainly the effective minority carrier
the BSF the Voc of the cell is reduced from 435 mV to
diffusion length and life-time, are expected
only 270 mV. Increasing the BSF thickness from 30
to depend strongly on the layer doping level,
to 90 nm decreases the sheet resistance by a factor
which thus can have a large effect on the cell
3 (as expected) but does not impact the Voc of the
performance. After introducing a working
Measured values for Voc, pFF, device as presented in Figure 4.4.1.22. Increasing
metalisation scheme, an absorber doping
Jsc(EQE), and pEff as a function the deposition rate of the BSF has a positive effect
of the absorber doping. Jsc is optimisation experiment was conducted in order
on the Voc. The Voc increases by 40 mV when the
calculated from EQE results. to estimate this effect. All cells in the experiment
deposition rate increases from 5 A/s to 10 A/s.
Jsc(EQE) and pEff were obtained were intended to be nominally identical but with
under two different back surface the different doping densities in the absorber The best efficiency reported previously for the
reflector (BSR) schemes: air (black
(about 2 μm thick, made on the SiN coated planar poly-Si cell with the identical metallization was
closed squares) and white paint (red
open squares). glass, and with the metallisation and the white- 5.2 % with a voltage of 435 mV [4.4.1.13]. The
paint BSR as described in the previous section). development of the high rate BSF clearly improved
Figure 4.4.1.21

46
 ARC
PHOTOVOLTAICS
CENTRE OF
2 m thick film
480 EXCELLENCE
20.0
465
BSF deposition rate 2010/11
5 [A/s] 17.5
10 [A/s] +40 mV ANNUAL REPORT
450 2.5 [A/s] 15.0
12.5

Jsc[mA/cm2]
Voc [mV]

435
10.0
420
7.5
+165 mV
405 5.0
270 2.5 No BSR 482 mV 68% 14.8mA/cm2 4.9%
White paint 479mV 66% 20.1mA/cm2 6.3%
255 0.0
0 30 60 90 0 50 100 150 200 250 300 350 400 450 500
BSF thickness [nm] Voc [mV]

Voc of 2 µm thick evaporated poly-Si solar cells with I-V curves of 2 µm thick planar poly-Si thin film solar
various BSF layers thickness and deposition rate. cells with and without white paint as back reflector.

Figure 4.4.1.22 Figure 4.4.1.23

the voltage. The I-V curves data of the solar cells recently reported [4.4.1.26].
with and without a BSR are presented in Figure Such a defective structural
4.4.1.23. The poly-Si solar cells with a white paint morphology is formed due to
BSR has a conversion efficiency of 6.3% with two inherent characteristics
Voc=479 mV, FF=66% and Jsc=20.1 mA/cm2. The of the evaporation, high
BSR increases the Jsc by 5.3 mA/cm2 and decreases directionality of the arriving
the FF by 2% due to the increase of Jsc and (increase atomic flux and the adatom low
of ohmic losses in the metallization). surface mobility. The deposition
only takes place on the surface
Light- trapping in e-beam poly-Si areas in direct line-of-sight of
thin‑film cells the evaporation source resulting in discontinuities FIB cross section of an evaporated Si
in the areas shaded from the source either by the film on textured glass substrate. (a)
Poly-Si thin-film solar cells need effective light-
surface topographical features or by the previously is a grey-scale image, and (b) is the
trapping to compensate for the moderate same image but set to a black and
deposited atoms themselves. The earlier arrived
absorption. This is typically achieved by glass white mode.
adatoms cast “shadows” for subsequently arriving
substrate texturing for PECVD cells. However,
adatoms, thus creating a network of parallel high Figure 4.4.1.24
evaporated poly-Si cells are not compatible with
density columns surrounded by lower density
textured glass due to low density material grown on
material, as it is schematically shown in Figure
textured substrates as shown in Figure 4.4.1.24. A
4.4.1.25. Besides, in contrast to PECVD deposition,
cross-sectional FIB investigation of the evaporated
the evaporated adatoms have a sticking coefficient
Si films on the textured glass reveals noticeable
close to unity; they do not have or receive the
morphological differences from the films on the
excessive energy to move in any way from
planar glass. On relatively smoother textures the Si
the place of their initial landing to find a more
film has apparently vague areas of lower material
thermodynamically favorable position, thus leading
density, or high defect density, extending from the
to morphological and possibly electronic defects.
inflexions in the texture features. Such defective
areas further develop into microcracks or voids Possible approaches to reducing an extent of the
either after the thermal treatment (SPC, RTA) or defect formation during deposition by evaporation
on the rougher textures. When the glass texture is using smoother glass texturing and/or to texture
has vertical or overhanging features, extended the Si film surface instead of, or in addition to Schematic representation of
discontinuities in the evaporated Si film are found. formation of low density material
smooth glass textures or a completely different
and voids during evaporation,
approach, the localized surface plasmon (LSP) where already adsorbed adatoms
Most described defects are believed to be related
enhanced light-trapping. The different structures cast shadows for newly arriving
to the columnar microstructure, which is typical for
are presented in Figure 4.4.1.26. adatoms thus forming parallel
evaporated films. Clear experimental evidence for channels filled with no or low
the columnar microstructure and the low density density material.
void network present in the SPC poly-Si films was
Figure 4.4.1.25

Schematic representation
different light trapping
scheme. A) planar, B)
textured glass, C) back
texture, D) nanoparticles.

Figure 4.4.1.26

47
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Light-scattering by surface
plasmons (top-left); SEM image
of Ag nanoparticles formed
from 16 nm thin precursor film
(top-right); AFM micrograph of
Si cell surface coated with Ag
nanoparticles (left). AFM micrograph of etch-back
textured poly Si film surface.
Figure 4.4.1.27
Figure 4.4.1.28

The plasmonic light-trapping relies on scattering collects transmitted light, which otherwise would
by Ag nanoparticles as shown in Figure 4.4.1.27 be lost, and redirects it back to the cell thus
[4.4.1.28]. Such nanoparticles can be relatively increasing the cell optical thickness at least twice.
simply formed by depositing a 5-30 nm thin Ag Additionally, most practically used BSRs are diffuse
film on already metallised and characterised and thus scatter the light, contributing to better
poly-Si e-beam cells followed by annealing at light- trapping. Depending on particular cell
about 200ºC leading to formation of 50-200 nm design, different BSRs are used: UNSW PECVD cells
wide irregularly shaped particles. The AFM profile use a combination of silicon dioxide layer with Al;
and SEM image of the Ag nanoparticles on a UNSW e-beam cells use a diffuse white paint or Ag
poly-Si film are presented in Figure 4.4.1.27. nanoparticles (or both); CSG Solar cells (both PECVD
and e-beam) use a combination of the diffuse
The silicon back texture can be simply made by
pigmented resist coated with Al [4.4.1.1]. However,
etching the silicon film in a KOH solution. After
until recently there was no systematic study of BSRs
a few second, pyramids formation appears as
to know which one is best suited particular poly-Si
presented in Figure 4.4.1.28. The roughness is
cell design.
typically around 100 nm whereas the roughness
of the nanoparticles is around 15nm. This A comprehensive comparative study has been
demonstrates the different mechanism of light conducted during 2010, which characterised and
Schematic structure of plasmonic
poly-Si thin-film solar cell with 44%
scattering involved into the two methods. compared the performance of different BSRs on
Jsc enhancement. poly-Si solar cells. The cell material prepared by
The optimisation of the two approaches is currently
PECVD was provided by CSG Solar and processed
Figure 4.4.1.29 ongoing at UNSW. The best developed process for
into metallised cell using the UNSW bifacial
plasmonic light-trapping uses a Ag nanoparticle
metallisation described in the previous section
array formed from 16 nm thick precursor Ag film
(Figure 4.4.1.30). To ensure fair comparison, all
directly on the poly-Si cell coated with a
BSRs were tested on the same cell, which was
350 nm thick MgF2 layer, with a white paint
characterised prior to and after application of each
100 BSR on the top, a spacing layer between the
BSR. The following BSRs were examined: acrylic
90
nanoparticles and the back reflector, typically
80 white paint (WP); P150 resist loaded with titania
350 nm of MgF2 (Figure 4.4.1.29) [4.4.1.6]. The
70
particles (P150W); P150W coated with ~ 150 nm
Jsc enhancement due to the plasmonic light-
EQE, 1-R [a.u.]

60
Al (P150W-Al); P150 with a double load of titania
50 trapping is 44%.
40 (P150DW); twice thicker P150DW (DP150DW); Ag
Jsc [mA/cm2]
30
Ag nano 23.5 The Si-back texture can be made by wet nanoparticles. The results are summarised in Table
20
white paint 21.9 chemical etching, plasma etching, or a 4.4.1.12 and also plotted in Figure 4.4.1.31.
10
Si back text 24.2
combination of both. The current gain
0
300 400 500 600 700 800 900 1000 1100 For the planar cells the Ag nanoparticles provide
wavelength [nm] achieved due to the Si-back texture is about
the best Jsc enhancement of about 48% on average
66%, significantly higher than the gain due to
(Jsc ~21 mA/cm2) (the best cell enhancement is
the plasmonic light-trapping (Figure 4.4.1.30).
EQEs of a cell with planar interfaces 50%) followed by the double P150DW and the
However, both approaches have further potential
(black), a plasmonic cell, and a cell white paint. For the textured cell one has to note a
with Si-back texture. for improvement, currently being evaluated.
lot higher Jsc (~24 mA/cm2) compared to the planar
Figure 4.4.1.30 cells even without any BSR due to excellent light-
4.4.1.9 Back-Surface Reflector study trapping providing by the random glass texture.
Most thin-film solar cells, and poly-Si cells in The enhancement due to BSRs are very modest
particular, are not thick enough to fully absorb with the white paint performing the best. The Ag
incident light even if the cells are textured to nanoparticle BSR is not tested for the textured cells
improve light-trapping. A back-surface reflector because nanoparticles cannot be formed on the
(BSR) is a necessary part of cell structure which textured surface.

48
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Current enhancement in planar (left) and textured (right)

poly-Si solar cells dur to different back-surface reflectors.
Note that the reference Jsc is 14.25 and 24. 18 mA/cm2
for the planar and textured cell respectively with no BSR.

Figure 4.4.1.31

4.4.1.10 Recombination Planar Textured

Jsc enhancement by different
back-surface reflectors on
processes in thin-film Si on BSR Jsc, mA/cm 2
Gain, % Jsc, mA/cm 2
Gain, % planar and textured poly-Si
glass solar cells No BSR 14.25 24.18 (80%) thin-film solar cells.
The current output of poly-Si thin-film WP 18.86 32.4 26.19 8.3
Table 4.4.1.12
solar cells is approaching the so- P150W 17.16 20.4 24.37 0.8

called Lambertian limit, which serves P150W/Al 17.75 24.6 24.58 1.7
P150DW 18.56 30.2 25.35 4.9
as a gauge for quantum efficiencies.
DP150DW 19.09 34.0 25.44 5.2
However, there is little understanding
Ag nanoparticles 21.51 50.9
about how high the Voc could be
– the lowest foreseeable practical
and absorber dopant concentrations, a concept
limit is the Voc of multicrystalline Si wafer based
consistent with experimental observations, which
solar cells. Thus, the voltage of poly-Si cells as an
describes the dominant recombination in poly-Si
academic topic is still a research frontier where the
has emerged [4.4.1.31, 4.4.1.32, 4.4.1.33, 4.4.1.34]. qΦ
fundamental limits are unknown and speculations Ec
abound. In previous years many researchers have According to the concept, the carrier recombination EF
attributed the lower lifetimes of poly-Si solar cells to in poly-Si on glass solar cells can be modelled as 1
their small grain size, a view which holds merit if the a superposition of two processes. The first one
2
grain boundaries are not well passivated [4.4.1.29], involves the electronic transition between shallow
but in more recent years it is becoming clear that states which are 0.05-0.07 eV below the conduction
intragrain defects are also potent enough to limit band and 0.06-0.09 eV above the valence band, Ev
the lifetime to the nanosecond range [4.4.1.30]. respectively, which are consistent with
While the grain boundary recombination can be the shallow bands at silicon dislocations
successfully mitigated by enlarging the grain size or (Figure 4.4.1.32a). The second process
optimising the passivation of dangling bonds within occurs via deep levels at charged defects.
the grain boundaries, so far there has not been an The shallow band recombination
effective method to circumvent the detrimental dictates the solar cell properties over
effects of intragrain defects. The goal of research at the entire practical absorber doping
UNSW is to elucidate the dominating recombination range of 5E15~1E17 cm-3; the deep level
pathway in the poly-Si thin-film solar cells so that recombination originates from either
the attainment of higher Voc becomes a more charged dislocations or grain boundaries
tractable problem. and only becomes influential at dopant
concentration lower than ~5E15 cm-3
As a result of a comprehensive study of the poly-Si
(Figure 4.4.1.32b).
on glass solar cell characteristics (Suns-Voc, EQE)
and film properties over a range of temperatures
(a) The dominant recombination
pathways in poly-Si solar cells
involving shallow defect levels: 1)
direct transition between shallow
bands; 2)transition between
deep level and shallow band; (b)
Simulation of the lifetime at Voc vs
dopant concentration. The short
dashed line represents shallow
bands recombination, and the long
dashed line represents deep level
recombination at grain boundaries.

Figure 4.4.1.32

49
 1.E-08
1.E-08
ARC PECVD 2.2e15 PECVD 3e15
PECVD 1e16
PHOTOVOLTAICS PECVD 5.1e15 PECVD 2.4e16
e-beam 6.4e15 PECVD 1e16
CENTRE OF 1.E-09 e-beam 1.56e17
ALICE 2e16 1.E-09 e-beam 7.2e15
EXCELLENCE ALICE 4.5e16 e-beam 6.9e16
e-beam 2.5e16
2010/11 ALICE 5e16

lifetime (s)

lifetime (s)
ANNUAL REPORT 1.E-10
1.E-10

1.E-11 1.E-11

1.E-12 1.E-12
30 40 50 60 70 80 30 40 50 60 70 80
q/kT q/kT

Lifetime vs q/kT plots for a) p-type, and b) n-type

poly-Si thin-film cells.

Figure 4.4.1.35

Arrhenius plots for the diode In the ideal sc-Si the diode saturation current J01 Further confirmation of the shallow-level related
saturation current J01/(JL/I) (a) and follows the Arrhenius law with the activation energy recombination in poly-Si solar cells comes from
for short-circuit current Jsc (c) for
Ea equal to the Si bandgap when extrapolated to the study of the bulk lifetimes as a function
poly-Si and c-Si solar cells; b) Voc
plots extrapolated to 0K for poly-Si 0K. The short-circuit current density, Jsc, in sc-Si is of temperature which allows a more detailed
and c-Si solar cells. only very weakly dependent on the temperature understanding of the behaviour of these shallow
(due to the effect of band-gap variation). However, levels in the device quasi-neutral, bulk regions.
Figure 4.4.1.33
poly-Si solar cells exhibit different characteristics Minority carrier diffusion lengths in the absorber
[4.4.1.31]. The Arrhenius plots for J01 of poly-Si layers of various poly-Si thin-film cells were
solar cells have distinctively a smaller slope, which extracted by fitting the device EQE simultaneously
reflects a lower Ea associated with J01 by 0.15- with its optical reflectance, a procedure which
0.18 eV. As a consequence, the Voc plots versus T has become a routine for poly-Si thin-film cells
extrapolated to 0K have a lower intercept whose short lifetimes precludes quantification
in the range 1.0~1.1 eV as compared by the standard techniques such as the quasi
to 1.27 eV for sc-Si solar cells (Figure steady state photoconductance (QSSPC). The EQE
4.4.1.33a, b). The Arrhenius plots for Jsc measurement and fitting procedure were repeated
of poly-Si cells whose absorber diffusion for each sample at different temperatures ranging
length is shorter than the absorber from 120K to 320K. Figure 4.4.1.34 superposes the
thickness have steeper slopes indicating front-side EQE curves of the same cell at different
stronger temperature dependence of the temperatures (note that the heights of the curves
poly-Si cell current (Figure 4.4.1.33c). are reduced slightly due to additional reflectance
by a quartz window on the cryostat in which the
Although the smaller Ea and the
sample is placed). Clearly the quantum efficiency of
intercept could in principle be due to the
the cell varies strongly with temperature.
narrower band gap in poly-Si material as
compared to c-Si, it cannot explain the Figures 4.4.1.35 a) and b) plot the extracted lifetime
temperature sensitivity of Jsc. On the τ against q/kT for p- and n-type cells respectively.
other hand, both effects are consistent Although the actual lifetime values at a given
Typical sequence of EQE curves
with existence of shallow sub-bandgap temperature differ greatly from cell to cell, all the
at different temperatures ranging
from 120K to 320K states acting as either minority carrier traps or lifetime curves exhibit Arrhenius law with the
recombination centres. activation energy (Ea) of 0.17-0.21 eV near room
Figure 4.4.1.34 temperature [4.4.1.32]. These activation energies

50
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Typical TEM DFWB image of a poly-

Si film on glass.

0.5 μm Figure 4.4.1.37

are much larger than those reported for plastically greater role in determining the Voc of AIC
1.E-09
deformed Si and multicrystalline Si [4.4.1.35], cells in fact, of the two rightmost triangle lifetime at 232K (s) Series1
ALICE (n-type)
AIC
where the dominant recombination is thought data points representing AIC cells at e-beam (n- type)
to be via transitions between the shallow and about 5E16 cm-3 doping in Figure 4.1.1.36, PECVD (n-
e-beam (p type)
deep levels. Assuming direct transitions between the higher lifetime point originates from PECVD (p-type)
shallow levels, the Ea is expected to be (Ec-Ede)+(Edh- (100) preferentially oriented material and Power (Series1)
1.E-10
Ev)+2.5kT=180~120 meV (where Ede and Edh are the lower lifetime point comes from (111)
-0.9619
shallow band energies near the conduction and the preferentially oriented material. y = 462576x
2
R = 0.8161
valence bands respectively), which is in remarkable
All the above evidence points at
agreement with the experimental values.
intragrain defects, as a likely limit
Furthermore, as a consequence of electron on poly-Si thin-film cell lifetimes. In 1.E-11
1.00E+15 1.00E+16 1.00E+17 1.00E+18
transitions between the shallow bands, the particular, dislocations are the most likely
dopant concentration (cm-
recombination rate becomes proportional to the candidates because they provide the
concentrations of both minority and majority strain fields necessary to create split-off
carriers, as in the case of radiative recombination, states from the band edges that can act Lifetimes at 232K for various poly-Si
in contrast to recombination via deep levels, as shallow levels [4.4.1.36]. To estimate dislocation thin-film cells plotted against
dopant concentration.
where the rate is proportional to the minority density in poly-Si solar cells a TEM study was
carrier concentration only. It leads to minority conducted using the Weak-Beam-Dark-Field (DFWB) Figure 4.4.1.36
carrier lifetime inversely proportional to the technique. An example image is shown in Figure
majority carrier concentration, or roughly to the 4.4.1.37 and the dislocation density deduced by the
dopant concentration [4.4.1.32]. The experimental image analysis is in the order of 1E10 cm-2. There
confirmation of this effect is shown in Figure is very rough correlation between the dislocation
4.4.1.36 which plots the minority carrier lifetimes density found in poly-Si solar cells and the cell Voc.
in a number of poly-Si solar cells versus the dopant The best cells with Voc > 500 mV have dislocation
density. There is a clear inverse relationship density of 6~8x109 cm-2, while for the cell with very
between the lifetime and dopant density. poor Voc < 450 mV the dislocation density is about
1.5x1010 cm-2 [4.4.1.37].
Amongst the n-type cells in Figure 4.4.1.35b, three
are made from small grained (~1 μm) poly-Si (circle As the dopant density decreases, the shallow
or square data points), and three are made from band recombination gradually becomes less and
large grained (~5 μm) poly-Si epitaxially grown on less significant and eventually, at sufficiently low
an Al induced crystallised (AIC) seed layer (triangle dopant densities, less than about 5x1015 cm-3 the
data points). The similarity in the temperature recombination involving deep levels starts to
behaviour of the lifetime, regardless of the grain dominate (process 2 in Figure 4.4.1.32a). Such deep
size, indicates that large and small grain materials levels are associated with defects concentrating in
have a common lifetime limiting mechanism. Also, dislocations and grain boundaries. The defects are
in Figure 4.4.1.36 one readily sees that the AIC cells usually charged to the same polarity as the majority
(triangles) can take on the lifetime values either carriers thus creating potential barriers which
above or below the power law trend line, indicating attract the minority carriers. The lifetime in this case
that they do not have consistently superior lifetimes depends on the potential barrier heights, which
just by virtue of their larger grain size. It was found become greater with lowering dopant density.
that the predominant grain orientation plays a

51
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

The effect of recombination via deep levels 4.4.1.7 P. I. Widenborg, G. Jin, P. J. Gress, and S. Varlamov,
at charged defects is to place an upper “High rate 13.56MHz PECVD a-Si:H depositions for
SPC poly-Si thin film solar cells”, Proceedings of the
limit on the minority carrier lifetime (thus
24th EUPVSEC, Hamburg, 2009.
the cell Voc). Figure 4.4.1.38 shows Voc
4.4.1.8 Z. Ouyang, O. Kunz, A. B. Sproul, S. Varlamov,
simulation results of poly-Si solar cells “Influence of the absorber doping for p-type
where two recombination processes in evaporated polycrystalline silicon thin-film solar
the absorber, both via shallow bands and cells on glass, J. Appl. Phys., 2011, 109, 060104.
via deep levels, taking place in parallel. 4.4.1.9 P.J. Gress, P.I. Widenborg, S. Varlamov, A.G. Aberle,
Above 1x1016 cm-3 doping, the Voc is fairly Wire bonding as a cell interconnection technique
for polycrystalline silicon thin-film solar cells on
constant and approaches the 485 mV
glass”, Progress in Photovoltaics: Research and
limit corresponding to the case where Applications, 2010, v. 18, pp. 221-228.
only shallow band recombination occurs. 4.4.1.10 P.I. Widenborg, N. Chuangsuwanich, A.G. Aberle,
Below 1x1016 cm-3 doping, deep level Glass texturing, Patent AU2004228064A1.
Simulated Voc of a poly-Si solar recombination becomes influential to the 4.4.1.11 P.I. Widenborg, A.G. Aberle, “Polycrystalline
cell (2.2 µm thick n-type absorber, device performance eroding the Voc by about 21 Silicon Thin-Film Solar Cells on AIT-Textured Glass
Jsc 27 mA/cm2) in which both Superstrates”, Advances in Optoelectronics, 2007,
mV when the doping reaches 1x1015 cm-3. Thus, in
shallow bands recombination and article ID 24584.
deep level recombination at grain the practically significant dopant concentration
boundaries (with electrostatic range of 5x1015~5x1016 cm-3, an immediate 4.4.1.12 P.I. Widenborg, S.V. Chan, T. Walsh, and A.G. Aberle,
fluctuations) occur. “Thin-Film Poly-Si Solar Cells on AIT-textured Glass
challenge to consistently raising the Voc above 500
– Importance of the Rear Reflector”, Proceedings of
Figure 4.4.1.38
mV is to reduce the dislocation density to below the 33rd IEEE PVSC, San-Diego, 2008.
the 1x109 cm-2 level. If and when this is achieved 4.4.1.13 O. Kunz, Z. Ouyang, S. Varlamov, and A.G. Aberle,
the next step to even higher Voc is minimisation of “5% efficient evaporated solid-phase crystallised
charged defects likely associated with impurities. polycrystalline silicon thin-film solar cells”, Progress
in Photovoltaics: Research and Applications, v. 17,
p. 567, 2009.
4.4.1.11 References
4.4.1.14 O. Kunz, Z. Ouyang, J. Wong, and A.G. Aberle,
4.4.1.1 M.A. Green, “Polycrystalline silicon on glass “Advances in evaporated solid-phase-
for thin-film solar cells”, Applied Physics A: 96, crystallized poly-Si thin-film solar cells on glass
p. 153, 2009. (EVA)”, Advances in Optoelectronics, 2008,
4.4.1.2 A. G. Aberle, “Fabrication and characterisation of article ID 532351.
crystalline silicon thin-film materials for solar cells”, 4.4.1.15 O. Kunz, J. Wong, J. Janssens, J. Bauer, O.
Thin Solid Films, 26, p. 511, 2006. Breitenstein, A.G. Aberle, “Shunting Problems Due
4.4.1.3 P. A. Basore, Proceedings of the 21st EUPVSEC, p. to Sub-Micron Pinholes in Evaporated Solid-Phase
544, Dresden, Germany, 2006. Crystallised Poly-Si Thin-Film Solar Cells on Glass”,
4.4.1.4 M. Keevers, T.L. Young, U. Schubert, R. Evans, Progress in Photovoltaics, 2009, v. 17, pp.35-46.
R.J. Egan and M.A. Green, “10% Efficient Csg 4.4.1.16 H. Cui, G. Jin, M. Wolf, P. Campbell, M. Geen,
minimodules:, Proceedings of the 22nd EUPVSEC, “Enhanced Absorption of SPC Poly-Si Thin Films
p. 1783, Milan, Italy, 2007. on Aluminium Induced Textured (AIT) Glass with
4.4.1.5 G. Jin, P. I. Widenborg, P. Campbell, S. Varlamov, Submicron Texture Feature Size”, Proceedings of
“Lambertian matched absorption enhancement the 25th EUPVSEC, Valencia, Spain, 2010.
in pecvd poly-Si thin film on aluminium 4.4.1.17 S. Varlamov, Y. Tao, J. Wong, O. Kunz, R. Egan,
induced textured glass superstrates for solar cell “Crystallisation kinetics of e-beam evaporated Si
applications”, Progress in Photovoltaics: Research films and its effects on poly-Si film and solar cell
and Application, 2010, v. 18, pp. 482-589. properties”, 25th EUPVSEC, Valencia, Spain, 2010.
4.4.1.6 Z. Ouyang, S. Pillai, F. Beck, O. Kunz, S. Varlamov, K. 4.4.1.18 Y. Tao, S. Varlamov, J. Wong, O. Kunz, R. Egan,
R. Catchpole, P. Campbell, M. A. Green, “Effective “Effects of SPC temperature on properties of
Light Trapping in Polycrystalline Silicon Thin-film evaporated poly-si thin films and soalr cells”, 35th
Solar Cells by Means of Rear Localised Surface IEEE PVSC, Hawaii, USA, 2010.
Plasmons”, Appl. Phys. Lett., 2010, v. 96, 261109.

52
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

4.4.1.19 A. Blum, C.J. Feldmann, J. Non-crystall. Solids, 1972, and applications to solar cells”, J. Appl. Phys., 91,
v. 11, 242. p. 2002.
4.4.1.20 C. Spinella, F. Priolo, “Crystal grain nucleation in 4.4.1.30 L. Carnel, I. Gordon, D. Van Gestel, G. Beaucarne,
amorphous silicon”, Appl. Phys. Rev., 1998, v.84, J. Poortmans, A. Stesmans, “High open-circuit
pp.5383-5414. voltage values on fine-grained thin-film polysilicon
4.4.1.21 T.Matsuyama, M. Tanaka, S. Tsuda, S. Nakano, Y. solar cells”, J. Appl. Phys., 100, p. 063702, 2006.
Kuwano, “Improvement of n-type poly-Si film 4.4.1.31 J. Wong, J. L. Huang, O. Kunz, Z. Ouyang, S. He,
properties by solid-phase crystallisation method”, P. I. Widenborg, A. G. Aberle, M. Keevers, M. A.
Jap. J. Appl. Phys., 1993, v. 32, pp. 3720-3728. Green, “Anomalous temperature dependence of
4.4.1.22 P. Timans, J. Gelpey, S. McCoy, W. Lerch, S. Paul, diode saturation currents in polycrystalline silicon
“Millisecond Annealing: Past, Present and Future”, thin-film solar cells on glass”, J. Appl. Phys., 105, p.
Mat. Res. Soc. Symp. Proc., 2006, v. 912, 103705, 2009.

4.4.1.23 K. Ohdaira, T. Fujiwara, Y. Endo, S. Nishizaki, H. 4.4.1.32 J. Wong, J. Huang. B. Eggleston, M. Green, O.
Matsumura, “Formation of Several-Micrometer- Kunz, M. Keevers, R. Egan, “Lifetime limiting
Thick Polycrystalline Silicon Films on Soda Lime recombination pathway in thin-film polycrystalline
Glass by Flash Lamp Annealing”, Jap. J. Appl. Phys., silicon on glass solar cells”, J. Appl. Phys. V. 107,
47, p. 8239, 2008. 123705.

4.4.1.24 K. Ohdaira, T. Fujiwara, Y. Endo, S. Nishizaki, 4.4.1.33 J. Wong, J. Huang, M. Keevers, M. Green, “Voc-
H. Matsumura, “Explosive crystallisation of Limiting recombination mechanism in thin film
amorphous Si films by flash lamp annealing” J. silicon on glass solar cells”, Proc. 35th IEEE PVSC,
Appl. Phys. 106, 044907, 2009. Hawaii, USA, 2010.

4.4.1.25 Z. Ouyang, O. Kunz, M. Wolf, P. Widenborg, G. 4.4.1.34 J. Wong, J. Huang, S. Varlamov, M. Green. R. Evans,
Jin and S. Varlamov, “Challenges of Evaporated M. Keevers, R. Egan, “Structural inhomogeneities
Solid-Phase-Crystallised Poly-Si Thin-Film Solar in polycrystalline silicon on glass solar cells and
Cells on Textured Glass”, Proceedings of the their effects on device characteristics”, Progress in
18th International Photovoltaic Science and Photovoltaics, 2011, DOI: 10.1002/pip.1089.
Engineering Conference & Exhibition, Kolkata, 4.4.1.35 M. Kittler, W. Seifert, K.W. Schroeder, “Temperature
India, 2009. dependent EBIC diffusion length measurements
4.4.1.26 M. Werner, U. Schubert, C. Hagendorf, J. Schneider, in silicon”, Phys. Stat. Sol. (a), 2986, v. 93,
M. Keevers, R. Egan, “Thin film morphology, growth pp.K191-K104.
and defect structure of e-beam deposited silicon 4.4.1.36 V. Kveder, M. Kittler, W. Schroter, “Recombination
on glass”, Proceedings of the 24th EUPVSEC, activity of contaminated dislocations in silicon:
Hamburg, Germany, 2009. a model describing electron-beam induced
4.4.1.27 A. Straub, R. Gebs, H. Habenicht, S. Trunk, R. A. current contrast behaviour”, Phys. Rev. B, 63,
Bardos, A. B. Sproul, and A. G. Aberle, “Impedance p. 115208 2001.
analysis: A powerful method for the determination 4.4.1.37 J. Huang, J. Wong, M. Keevers, M. Green,
of the doping concentration and built-in potential “Cell performance limitation investigation of
of nonideal semiconductor p-n diodes”, J. Appl, polycrystalline silicon thin-film solar cells on glass
Phys. 97, p. 083703, 2005. by transmission electron microscopy”, Proc. 25th
4.4.1.28 K.A. Catchpole, A. Polman, “Plasmonic solar cells”, EUPVSEC, Valencia, Spain, 2010, pp.3565-3567.
Optics Express, 2008, v. 16, pp. 21794-21800. 4.4.1.38 J. Wong, J. Huang, S. Varlamov, M. Green, “The roles
4.4.1.29 P. Altermatt and G. Heiser, “Development of a of shallow and deep levels in the recombination
three-dimensional numerical model of grain behaviours of polycrystalline silicon solar cells”,
boundaries in highly doped polycrystalline silicon Progress in Photovoltaics, 2011, in press.

53
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Recent improvements in independently confirmed

efficiency for small area (≤ 1cm2) organic solar cells
(extracted from Green et al. [4.4.2.2]).

Figure 4.4.2.1

4.4.2.1 Introduction phases. The scale of the phase separation

University Staff A new organic photovoltaic (OPV) research depends on the solvent, solubility of the
• Dr Ashraf Uddin lab has been established at Level-10 in materials and parameters of the spin-
• Prof. Martin Green the UNSW Chemical Science Building coating process such as the spin speed
• A/Prof. Gavin Conibeer to accommodate work on organic and temperature. If the morphology
semiconductor materials and OPV devices. could be controlled on a molecular
Senior Fellow scale the efficiency of charge separation
Today this lab is equipped with basic
Dr Dirk König facilities organised with support from an and transport could be expected to be
ARC Discovery Grant, UNSW equipment substantially higher. Now we are working
Research Fellow for the morphology optimization of
grant, a family grant from the Green family,
Dr Shujuan Huang a Faculty grant and with the support of organic films for OPV devices in respect to
the ARC Photovoltaic Centre of Excellence. fundamental issues such as light trapping,
Research Assistant control of electronic structure at films
Currently, the research team consists of
Xiaohan Yang interfaces, exciton dissociation and carrier
three academics, two research fellows, one
research assistant, 6 PhD, one M.Eng. and 8 transport for photovoltaic operation to
Postgraduate Research Students
honours students. achieve our goals.
• Lara Treiber
• Xiaohan Yang We have already produced some meaningful The detailed physical understanding of OPV
• Miga Jung data from our research work. Two journal devices still lags behind their application,
• Xiaolei Liu papers, three conference proceeding on account of fundamental differences in
papers and one patent are already in the optoelectronic properties of organic
• Kim Taekyun
preparation. The nanoscale texture or thin materials compared to conventional
• Fred Oi
film morphology of the donor/acceptor semiconductors. This lack of understanding
• Matthew Wright
blends used in most OPV devices is a limits the scope of material and device
Undergraduate Thesis Students critical variable that can dominate both design. Among those features that
the performance of new materials being distinguish OPV devices from inorganic
• Thoo Chee Hong
optimized in the lab and efforts to move semiconductor based devices are: (i) the
• Matthew Wright
from laboratory-scale to factory-scale charges and excited states are localized
• Xinda Yang
production. Although OPV conversion on individual molecules or molecular
• Jizhong Yao
efficiencies (up to 8.3%) have improved segments, with the result that charge and
• Adam Sen Ann Ong
significantly in recent years, progress in energy transport processes are relatively
• Lihui Song
morphology optimization still occurs largely slow; (ii) the dielectric permittivity is low,
• Buyi Yan
by trial and error, in part because much of leading to stronger space charge effects;
• Viren Nathoo
our basic understanding of how nanoscale (iii) the organic semiconductor materials
are electronically disordered, dispersing
Taste of Research Student morphology affects the optoelectronic
properties of these heterogeneous organic the rates of charge transfer and transport
Solomon Freer
semiconductor films has to be inferred processes; (iv) the active layers are often
indirectly from macroscopic measurements. heterogeneous, either as multi-component
The morphology of the active film is a films or because of nonuniform molecular
very important factor in producing high ordering; and (v) the organic semiconductor
efficiency devices. is usually not doped, thus precluding the
conditions that allow charge dynamics to
Films spin-coated from blend solutions be linearized in the description of device
undergo separation of the donor/acceptor physics. The need for appropriate device

54
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT
(a) The schematic band energy
structure of bulk-heterojunction
organic cell used to produce the
results of Fig. 4.4.2.1; (b) (left)
a schematic showing random
distribution of donor (P3HT) and
acceptor (PCBM) regions in the
blended bulk-heterojunction
organic solar cell, (right) a
schematic diagram of systematic
alignment of donor and
acceptor layers.

Figure 4.4.2.2

physics for organic semiconductor materials “bulk heterojunction” cell structure shown in Fig.
is of particular relevance to the organic bulk 4.4.2.2(a). From this baseline device structure, the
heterojunction solar cell. Centre believes it has several strengths that will
allow it to contribute to the ongoing cell efficiency
4.4.2.2 OPV Conversion Efficiency evolution documented in Fig. 4.4.2.1 and to the
development of more stable and durable devices.
Interest in OPV with increasing conversion efficiency
has grown exponentially over recent years. The The challenge in the OPV is that absorbing light in
key development in the OPV device is the bulk an organic donor material produces coulombically
heterojunction cell [4.4.2.1] obtained by blending bound excitons that require dissociation at the
donor and acceptor layers, a device structure well donor/acceptor interface. The energy-level offset
suited to the associated short excitonic diffusion of the heterojunction (donor/acceptor interface)
lengths. The workhorse for OPV research has is also believed to play an important role for the
become polymer-fullerene bulk heterojunction dissociation of bound excitons in OPV cells. An
devices fabricated by coating a thin layer of efficient exciton separation into free charge carriers
PEDOT:PSS on the top of cleaned ITO substrates at the interface of an acceptor and donor materials
followed by a blend of PCBM/P3HT (or other donor will increase the photocurrent of the solar cell
polymer), then by vacuum evaporation of Al and [4.4.2.4]. The transport of excitons to the interface
annealing. Both packaging and cell efficiency had also limits the conversion efficiency of OPV. The
improved by 2006 to the stage where independent details of the exciton transport mechanism in OPV
measurement of cell efficiency was both feasible are still not well understood. There is a need for more
and warranted, with subsequent efficiency systematic study to gain in-depth understanding
improvements well documented, after review of the mechanisms of light absorption, exciton
by a team that includes one of the present team dissociation and charge transport, particularly their
members [4.4.2.2]. Using the above structure, relationships with molecular and morphological
Konarka, USA established 4.8% efficiency for a tiny structures of the materials as well as the nano-scale
0.14 cm2 cell in July 2005. Konarka increased this architectural design of the devices. Figure 4.4.2.2(b)
to 5.15% in December 2006 with 5.24% posted shows the two examples of donor/acceptor blending
for a 0.7 cm2 cell in July 2007. Companies such as design (left) and systematic alignment of donor
Plextronics have reported roughly comparable and acceptor layers for bulk-heterjunction organic
results using similar materials with 5.4% efficiency solar cells.
confirmed in July 2007 for a 0.1 cm2 device, 6.0% in
August 2008 for an even smaller 0.04 cm2 cell and We believe that the conversion efficiency of OPV can
2.0% as recently as 28 January 2009 for a 224 cm2 be increased to over 10% for possible commercial
module. OPV device developer Solarmer Energy applications through in-depth understanding of
achieved 7.9% efficiency in December 2009 [4.4.2.3]. the light harvesting behaviour of organic materials
In November 2010, Konarka achieved their highest and by selecting suitable materials and new device
OPV efficiency 8.3%. The history of OPV conversion architectures.
efficiency is shown for small area (≤ 1cm2) cells in
Figure 4.4.2.1. 4.4.2.4 Active Layer Annealing
Thermal annealing is an effective method that
4.4.2.3 Standard OPV Structures increases the polymer crystallinity and improves
The fullerene based acceptor material (PCBM) is the phase segregation of donor and acceptor
interspersed with a donor polymer, commonly domains. The surface morphology is also affected
P3HT, with line-bond diagrams for these materials by annealing as shown in Fig. 4.4.2.3. Usually,
and the energy band diagram of the standard as-cast P3HT:PCBM films have a featureless and

55
 ARC
PHOTOVOLTAICS
CENTRE OF
(a) Annealing effects on the
EXCELLENCE absorbance of P3HT/PCBM
2010/11 blended layers on glass substrate
(temperatures from 125°C to 145°C
ANNUAL REPORT
for 5 to 15 minutes duration). (b)
Average absorbance compared for
different annealing temperatures
and duration. The optimum
annealing conditions were found
as 125°C for 15 minutes for the best
quality layer.

Figure 4.4.2.4

homogeneous surface, remain the same under all annealing temperatures

while annealing produces a and times. However, the intensity of the spectra
coarser textured surface with changes. As mentioned, the optimum annealing
obvious phase segregation. temperature was found as 125°C for 15 minutes for
The majority of device studies the best quality layer.
on P3HT:PCBM blends utilize
We have started to fabricate our OPV cells with
annealing to increase device
average Voc ≈ 0.6 V, FF ≈ 70 % and Jsc ≈ 10 mA/cm2.
efficiency. This is believed to
The average conversion efficiency is around 3 - 4%.
be due to changes in both
The measured Voc of our OPV devices as a function
the morphology of the active
of light intensity is shown in Figure 4.4.2.5.
layer whereby P3HT and PCBM
self-organize into nanoscale We have used different types of P3HT to fabricate
domains as well as the better our OPV cells. Their corresponding Voc and FF are
intermolecular alignment of recorded in Table 4.4.2.1.
the P3HT giving higher hole
mobilities. 4.4.2.5 Light Trapping in OPV
In our studies, we have Given the high absorption coefficient of organic
found that annealing the materials, 100 nm thick organic layers are typically
device after the application sufficient to absorb most of sunlight. The diffusion
of the cathode improves length of excitons in the organic materials is
the efficiency of the device typically in the range of 5–10 nm. A partial
much more significantly that solution to the problem of low charge carrier and
annealing before the cathode exciton mobility is the reduction of the active
deposition. The reason for layer thickness, that also implies the loss of a large
AFM images of surface morphology
of P3HT layer on glass substrate (a) this is believed to be reduced mobility of the PCBM fraction of the impinging light. The use of a light
as spin coated; and (b) annealed molecules under the increased confinement of the trapping systems has been actively investigated to
at 140°C for 10 minutes. After cathode. In our studies we have explored a range increase the total path length of light into the active
annealing the layer structure material without needing to increase its physical
of anneal temperatures and find that the optimum
become more clearly visible. The
surface roughness is also reduced annealing temperature depends upon the blend thickness. The concept is to induce sunlight to travel
by annealing. ratio, and in general lies in the range of 120°C to a longer distance in the active layer by a multi-pass
150°C. The optimum anneal temperature for our path in order to exploit most of the radiation.
Figure 4.4.2.3
blend appears to be ~125°C (for 15 minutes) – a
Many different light trapping schemes have been
value subsequently used in the remainder
proposed to enhance the quantity of light absorbed
of the optimization experiments detailed
in OPV cells such as metal gratings [4.4.2.6], buried
below. An improved transport property
nanoelectrodes [4.4.2.7] and multireflection
of the active layer is also observed which
structures [4.4.2.8]. However, the requirement that
leads to a reduced series resistance and in
the feature sizes of the scattering structures be
turn a better device performance. Recently,
comparable to the film thickness of the OPV cells,
microwave annealing has been proposed
i.e., 50–300 nm, and at the same time larger than the
as an alternative thermal treatment
wavelength of light to effectively alter the photon
method [4.4.2.5].
propagation direction is intrinsically contradictory.
Voc and fill-factor (FF) of our OPV Figure 4.4.2.4 shows the absorbance The electrical properties can be negatively affected
cells, fabricated with different spectra of the blended active layer of P3HT/ by the introduction of light scattering elements in
donor type materials of P3HT. PCBM, annealed at different temperature and for direct contact with such thin active layers: defects
different durations. The absorbance spectra almost and shorts may easily occur. Alternate approaches
Table 4.4.2.1

56
 100.00
Data Fit
Voc V_0.1 ARC
10.00 MPP Rsh PHOTOVOLTAICS
V_n1 fit V_n2 fit
CENTRE OF
light intensity (Suns)

V_n1 V_n2
1.00 Vx EXCELLENCE
2010/11
0.10 ANNUAL REPORT

0.01

0.00
0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80
Voc (V)

Open circuit voltage (Voc) of OPV devices as a function

of light intensity. The Voc increases logarithmically
with light intensity as for conventional silicon solar
cells.

Figure 4.4.2.5

in improving absorption efficiency have exploited film devices, as well as the use of plasmonics
the properties of surface plasmons (SPs). SPs are for light-trapping in thin, plane-parallel
electromagnetic surface waves confined to a metal- photovoltaic structures [4.4.2.10].
dielectric interface by coupling to the free electron
The Centre is also particularly interested in
plasma in metals [4.4.2.9]. The evanescent nature of
hybrid organic/inorganic systems as a way of
SPs permits the manipulation and enhancement of
improving both performance and stability.
optical fields below the diffraction limit, allowing
Some work in this area has already commenced
the use of thin OPV layers without sacrificing their
as reported below. Hybrid solar cells are a mixer
absorption potential. The integration of plasmonic
of nanostructures of both organic and inorganic
structures with OPVs has been previously examined
materials. They combine the unique properties
in the context of metallic nanoparticles, which can
of inorganic semiconductor nanoparticles with
be used to increase optical absorption. We have also
properties of organic/polymeric materials. Absorbance spectra of conventional
started working on the plasmonic light trapping
Inorganic semiconductor nanoparticles or quantum OPV cell (ITO/PEDOT:PSS/
system to increase the efficiency of our OPV cells.
dots may have high absorption coefficients and P3HT:PCBM/Al without TiOx spacer
particle size induced tunability of the optical band- layer and with TiOx layer.
4.4.2.6 Simulation Work gap. Band-gap tuning in inorganic nanoparticles Figure 4.4.2.6
The Centre has purchased an upgrade of the with different nanoparticle sizes can be used for
Gaussian Density Functional Theory (DFT) program realization of device architectures, such as tandem
package to Gaussian09. We can model larger solar cells in which the different bandgaps can
systems in the so-called ONIOM model which be obtained by modifying only one chemical
consists of up to three model shells of different compound. Thus, the organic/inorganic hybrid
accuracy. The highest accuracy is limited to the concept for photovoltaic solar cells is becoming
molecule of interest, while the lower accuracy attractively interesting in recent years. The
regions are bigger and cover the electronic solubility of the n-type and p-type components
environment of the species under investigation. is an important parameter in the construction of
This can include neighbouring molecules for charge hybrid solar cells processed from solutions.
transfer as well as electrodes. An improved tool for
calculating infrared- and Raman-spectra is included 4.4.2.7 Ordered Nanoparticle Arrays
as well, providing an important link to experimental For Hybrid Organic/Inorganic
observations at the organic species. The present
Solar Cells
strengths include a dedicated ab-initio molecular
simulation capability (Fig. 4.4.2.7) presently rated at Experimental work on fabrication of highly
1080 gigaflops but with upgrading to 1.6 teraflops ordered arrays of nanoparticles as absorber
planned for the near-term. To date, this facility materials originally for hot carrier solar cells Dedicated Linux 64 bit cluster for
has been used for ab-initio modelling of silicon has been initiated as a potential means to realise ab-initio molecular simulations
ordered superlattice structures. In this work, the aim presently rated at 1080 gigaflops.
quantum dots in a dielectric environment. The
extended capability for excited state modelling is to establish a fabrication system for depositing Figure 4.4.2.7
should allow the screening of new candidate OPV sequential monolayers of nanoparticles with
materials for both performance and durability. uniform shells.

Other strengths include past experience with a We have installed a Langmuir-Blodgett (LB) system
number of engineered vertical junction structures for fabrication of highly ordered nanoparticle
and with light-trapping, not yet fully exploited monolayers as shown in Fig. 4.4.2.8. The LB
in OPV devices. The Centre has pioneered light- technique leads to the development of ordered
trapping in both wafer-based and silicon thin- monolayers at an air-water interface while
exploiting the self-organization mechanism of

57
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Schematic of the LB apparatus used to fabricate a

monolayer of encapsulated nanoparticles. Si nanoparticle encapsulated with organosilanes.

Figure 4.4.2.8 Figure 4.4.2.9

is very attractive for organic and inorganic hybrid

device applications. Organic/inorganic hetero-
junctions have generated a great interest for their
potential in developing hybrid devices as well as
to understand organic semiconductors and their
interfacial properties. We are also investigating
the possibility of poly-Si/organic hybrid solar
cells. Figure 4.4.2.10 shows the absorption spectra
of poly-Si/PEDOT:PSS/P3HT on glass substrate
compared with the absorption spectra of c-Si/
PEDOT:PSS/P3HT structure.

4.4.2.9 References
4.4.2.1 G. Yu, K. Pakbaz and A. J. Heeger, Appl. Phys. Lett.
Absorption spectra of poly-Si/ colloidal dispersion. Compression of the monolayer 64 (1994), p3422.
PEDOT:PSS/P3HT on glass substrate is monitored via measurements of surface pressure 4.4.2.2 M.A. Green, K. Emery, Y. Hishikawa and W. Warta,
compared with absorption spectra and then controlled by a feedback loop. This unique “Solar Cell Efficiency Tables”, Versions 26-34,
of c-Si/PEDOT:PSS/P3HT structure.
technique allows transfer of this ordered monolayer Progress in Photovoltaics, 2006-2009.
Figure 4.4.2.10 onto a wide range of solid substrates such as 4.4.2.3 http://www.solarmer.com
glass or Si wafers. By controlling the interspacing 4.4.2.4 A. Holzhey, C. Uhrich, E. Brier, E. Reinhild, P. Bauerle,
between adjacent particles, i.e. the shell thickness, K. Leo and M. Hoffmann, J. Appl. Phys. 104 (2008),
by varying the molecular weight of capping species, p064510.

we can control the periodicity of the film - leading 4.4.2.5 K. Lee, J. Y. Kim and A. J. Heeger, SPIE Organic
Photonic Materials and Devices VIII, San Jose, CA,
to new optical and electrical properties.
January 23-26, 2006.
Silicon (Si) nanoparticles are being used as core 4.4.2.6 L.S. Roman, O. Inganäs, T. Granlund, T. Nyberg,
materials. In order to control the interspacing M. Svensson, M.R. Andersson, J.C. Hummelen,
Advanced Materials 12 (3) (2000), p189–195.
between the particles, the termination of Si
nanopaticles is carried out using organosilanes of 4.4.2.7 M. Niggemann, M. Glatthaar, A. Gombert, A.
Hinsch, V. Wittwer, Thin Solid Films 451–452 (2004),
varying alkyl chain lengths, as shown in Fig. 4.4.2.9. p619–623.
Progress to date is reported elsewhere [4.4.2.11]. 4.4.2.8 K. Tvingstedt, V. Andersson, F. Zhang, O. Inganas,
Once the assembly approach is mastered, it Applied Physics Letters 91 (2007) p123514.

should be possible to build up device structures 4.4.2.9 N.C. Lindquist, A. Lesuffleur, and S. Oh, Applied
Physics Letters 91 (2007), p253105.
incorporating layers of quantum dots with the
doping in each layer individually controlled. 4.4.2.10 S. Pillai, K.R. Catchpole, T. Trupke, and M.A. Green,
“Surface Plasmon Enhanced Silicon Solar Cells”,
Journal of Applied Physics 101 (2007), p093105.
4.4.2.8 Poly-Si/Organic Hetero- 4.4.2.11 L. Treiber, C. Bumby, S. Huang, G. Conibeer, 23rd
Junction Cell European Photovoltaic Solar Energy Conference,
Valencia, Spain (2008).
Recently organic/inorganic hetero-junction thin
4.4.2.12 R. A. Halton, et al., Organic Electronics 10
films solar cells have gained importance for their
(2009), p388.
cost-effective potential applications [4.4.2.12]. The
combination of hydrogenated amorphous (a-Si:H)
and/or poly-silicon (poly-Si:H) and organic materials

58
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Synthetic silicon. Portion of the Periodic

Table showing pathways to engineering
semiconductors with 4 valence electrons/atom.

Figure 4.4.3.1

4.4.3 Earth-Abundant Photovoltaic international work in this area. The starting point is
Devices (CZTS) a viable high throughput manufacturing process.
Efforts are to develop technology suitable for the
University Staff desired process rather than persevering with solution
growth processes that have given the best laboratory
• Dr Xiaojing Hao
results to date. Materials selection would be
• Prof. Martin Green
guided by stable compositions formed in naturally
Postgraduate Research Student occurring minerals such as kesterite and stannites.
Xiaolei Liu Another key area that is to be investigated in detail
is epitaxial relationships to silicon to investigate the
suitability for tandem cell stacks, including stacks
involving silicon as the lowermost layer.
4.4.3.1 Strand Outline
All successfully commercialised non-concentrating 4.4.3.2 Simplified fabrication of CZTS
photovoltaic technologies to date are based on thin film cells
silicon or the chalcogenides (semiconductors Though various approaches have been applied
containing Group VI elements, specifically Te, Se to fabricate CIGS solar cell, CIGS with the highest
and S). As indicated by Figure 4.4.3.1, the successful efficiency (up to 20.3% so far) have been achieved
chalcogenide materials, CdTe and CuInSe2, can be by the physical vapour deposition (PVD) approach,
regarded as “synthetic silicon” where the balance with the first 10% efficient cell demonstrated in
between atoms in these materials provides the 1980, the first 15% cell in 1993 and the first 20% cell
same average number of valence band electrons as in 2008. As a substitute for CIGS thin film solar cells,
in silicon. CZTS has many similarities to CIGS as mentioned
Unfortunately Cd and Se are toxic “heavy metals” above. As such, it is logical that the full potential
while Te and In are amongst the 12 most rare of high efficiency CZTS thin film solar cells will be
elements in the Earth’s crust, factors that would extracted using the PVD approach.
seem to limit the long-term potential of the Though a major milestone was reached at the
established chalcogenide technologies. However, end of 2009 with the demonstration of 9.7% CZTS
as indicated in Figure 4.4.3.1, by investigating more efficiency (some Se also in device) by a research
deeply into the Periodic Table, the compound team at IBM by a “hydrazine-based solution”
Cu2ZnSnS4 is uncovered with the same number of approach [4.4.3.1], solution growth is believed
valence band electrons on average but involving inherently incapable of giving the material control
earth-abundant, non-toxic elements. required for best-possible performance. Evidence
Although the potential of this material is relatively for this arises in the similar case of CIGS where the
unexplored for photovoltaics, initial results have same IBM team holds the record of 12.2% efficiency
been promising with a group at IBM reporting 9.7% for solution-deposited CIGS (again hydrazine–
efficiency for small cells based on related materials based) [4.4.3.2], while in contrast the PVD approach
(alloy of the above sulphide and the corresponding gives 20.3% efficiency.
selenide). The appearance of the cells involved Evaporation and sputtering are the two most
(Fig. 4.4.3.2) is quite close to the appearance of important PVD techniques. The disadvantage of
cells made using CIGS technology (CuInSe2 plus evaporation is that it is very difficult to control the
CuGaSe2 alloy). Cu evaporation source; and the lack of commercially
The Centre’s work in this new strand of activity available equipment for large area thermal
takes a different direction from most of the present evaporation. In contrast, magnetron sputtering is

59
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

SEM cross-section of CZTS-type cells (Todorov et al., Adv. Mater. 22,

E156, 2010).

Figure 4.4.3.2

a more attractive process because large area/high germanium thin film by magnetron sputtering
rate sputtering is a better developed technique results in the significant improvement of the
than evaporation. In addition, sputtering has performance of germanium thin film, appears
commercially available equipment that gives good highly relevant.
uniformity over large areas. UNSW has considerable
As such, the idea of one-step reactive sputtering
experience in PVD techniques, such as the
by in-situ introducing H2S gas into the deposited
achievement of the first silicon quantum dot solar
CZT (Cu2ZnSn) film is a logical step to attempt a
cell on quartz [4.4.3.3], a recent patent relating to
better CZTS film quality. However, there have been
crystal Ge on Si wafer by a sputtering technique and
few attempts to date at such a simplified one-step
the polycrystalline silicon thin film solar cell on glass
process by introducing the sulphide into the CZTS
by an evaporation technique [4.4.3.4].
film during in-situ deposition. In addition, this
Almost all of the methods that have been process is not well understood and significantly
attempted for CZTS fabrication involve two steps; more effort is required in order to understand
deposition of a metal/metal-sulfide precursor growth mechanisms involved in the reactive
and a subsequent anneal at high temperature sputtering process. Comparison of reactively
in a hydrogen sulphide or elemental sulphur sputtered films with 2-step sulphidisation processes
environment to incorporate more sulfur into the will provide the early emphasis of the Centre’s work.
film and evolve the correct phase [4.4.3.5-4.4.3.7].
It is reported that the H2S/sulphur anneal is a 4.4.3.3 References
destructive process that can potentially introduce 4.4.3.1 T.K. Todorov, K.B. Reuter, D.B. Mitzi, Advanced
defects into the film due to the out-diffusion of Materials 2010, 22, E156.
metals. In addition, for large scale production of 4.4.3.2 W. Liu, D.B. Mitzi, M. Yuan, A.J. Kellock, S.J. Chey,
solar cells, a direct one-step, large area deposition and O. Gunawan, Chem.Mater., 2010, 22(3), 1010-
process is preferred. 1014.
4.4.3.3 I. Perez-Wurfl, X.J. Hao, A .Gentle, D-H. Kim, G.
Fortunately, reactive sputtering is a well suited Conibeer, M.A. Green, Applied Physics Letters 95
technique because of the above-addressed feature (2009), 153506.
of precise control on the stoichiometry of deposited 4.4.3.4 Z. Ouyang, S. Pillai, F. Beck, O. Kunz, S. Varlamov, et
films over a large area with high uniformity at a al., Applied Physics Letters, 2010, 96 (26), 261109.
relatively low cost. The experience documented in 4.4.3.5 J.J. Scragg et al., Thin Solid Films, 517 (2009), pp.
our patent of “A Method of Forming A Germanium 2481-2484.

Layer On A Silicon Substrate And A Photovoltaic 4.3.3.6 R. Schuee et al., Thin Solid Films, 517 (2009) pp.
2465-2468.
Device Including A Germanium Layer”, where the
introduction of H2 gas into the in-situ deposited 4.4.3.7 Hironori Katagiri et al., Thin Solid Films, 480-481
(2005), pp. 426-432.

60
 4.5 Third Generation Strand –
Advanced Concepts ARC
PHOTOVOLTAICS
University Staff: • Bo Zhang CENTRE OF
• A/Prof. Gavin Conibeer (group leader) • Andy Hsieh EXCELLENCE
• Dr Richard Corkish • Dawei Di 2010/11
• Prof. Martin Green • Zhenyu “Wayne” Wan
ANNUAL REPORT
• Craig Johnson
Senior Research Fellows: • Sammy Lee
• Dr. Dirk König • Haixiang Zhang

Research Fellows: Casual Staff:

• Dr Shujuan Huang James Rudd (part time)
• Dr. Ivan Perez-Wurfl
Visiting Researchers:
Lecturers: • Dr. Yukiko Kamikawa (AIST, Tsukuba,
• Dr Santosh Shrestha Japan, June 2009 – Dec 2012)
• Prof. Bingqing Zhou (Inner Mongolia
Postdoctoral Fellows: Normal University, China, Sept 2009 –
• Dr Supriya Pillai (part time) Aug 2010)
• Dr Xiaojing “Jeana“ Hao • Dr. Mallar Ray (Bengal Eng. and Sci.
• Dr Sangwook Park (to Apr 2010) University, May 2010)

Professional officers: Undergraduate Students:

• Dr Tom Puzzer (part time)
4th Year Thesis Projects:
• Dr Patrick Campbell (shared with Thin
Film) • Yu Feng
• Mark Griffin (shared with Thin Film) • Tao Zhan
• Chen Pan
Research Associates: • Xi Li
• Dr Didier Debuf (adjunct fellow) • Yao Yao
• Yidan Huang • Linzhi Ma
• Lei “Adrian” Shi (from Sept 2009) • Lin Dong

Higher Degree Students: Masters Project:

• Lara Treiber Chandraprasad Ramachandran
• Pasquale Aliberti
Visiting Practicum students:
• Yong-Heng So
• Robert Patterson • Stephan Michard (RWTH Aachen,
Germany, Aug 2009 – Feb 2010)
• Binesh Puthen Veettil
• Dolf Timmerman (Amsterdam
• Chao-Yang “Jeff” Tsao
University, Oct – Dec 2010)

Abstract and performance in the photovoltaic devices. Hot Carrier cells have seen very significant
There has been approximately equal work on Improved models for the understanding improvement in demonstrated resonance
the two major projects in Third Generation in of doping effects in these materials have in energy selective contacts using Si
2010. These are the Group IV nanostructure also been established. Work on alternate nanostructure layers, as well as a development
tandem cells project - the “all-Si” tandem cell matrices for Si quantum dots, in both silicon of 2 and 3D modelling of transport in these
- and the Hot Carrier solar cell project, with nitride and carbide, has seen development of structures. Modelling of Hot Carrier efficiencies
its continuing funding from GCEP (Global composite structures which have improved continues to get more sophisticated with
Climate and Energy Project). There has also transport in the growth direction, whilst application to real material systems such as
been further work on Up-conversion and on maintaining quantum confinement in the III-nitrides and inclusion of Auger processes
Plasmonics. plane, but which also increase the uniformity which become significant at high carrier
of QD sizes. Work on Ge nanostructures has concentrations. Work on absorbers has
The Si nanostructure work has seen increased also improved with high electrical p-type allowed modelling of the phononic properties
understanding of the mechanisms for conductivity established for Ge quantum of a range of bulk materials, in conjunction
transport and quantum confinement. More dots and excellent pseudo single crystalline with time resolved photoluminescence
sophisticated modelling of both has been growth quality for Ge quantum wells in a measurement of carrier cooling in some of
tied more directly to improved interpretation nitride matrix. Heterojunction photovoltaic these materials. Also modelling of coherent
of experimental results. This has led to devices combining the advantages of two of nanoparticle nanostructures, which emulate
establishment of a predictive ‘equivalent circuit these different material types are now being the phononic properties required, has
modeller’ which will allow optimisation of Si investigated. developed into direct application to structures
QD device parameters to maximise transport grown directly. These structures include the

61
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Loss processes in a standard solar

cell: (1) non-absorption of below
band gap photons; (2) lattice
thermalisation loss; (3) and (4)
junction and contact voltage losses;
(5) recombination loss.

Figure 4.5.1

colloidal dispersion of Si nanoparticles, which cells, light trapping by these non-texturing methods
has seen coherent arrays successfully grown, is very important for third generation devices, which
and growth of III-V quantum dot structures with consist of very thin layers of material.
collaborators, with modelling and characterisation
This progress in all the main Third Generation
indicating their usefulness as absorbers. The
project areas is improving understanding and
parameters needed for a real hot carrier solar cell
allowing optimisation of modelling, structures
are now better defined, and candidates for its
and devices. Developments to come in the
structure are now being modelled in more detail.
next year will see significant advancement in
This will feed into fabrication and characterisation
these areas, with excellent prospects for good
of some candidate structures in the near future.
demonstration devices.
The up-conversion project in 2009 reported the
approach of use of porous-Si as a host material for 4.5.1 Third Generation Photovoltaics
an Erbium upconverter. This has the advantage
The “Third Generation” photovoltaic approach is
of much greater versatility in fabrication and
to achieve high efficiency whilst still using “thin
electrochemical doping with Er, compared to
film” second generation deposition methods. The
other host phosphors. Work in 2010 has further
concept is to do this with only a small increase
demonstrated up-conversion of below band gap
in areal costs and to use abundant and non-toxic
photons. The ease of control of porous-Si growth at
materials and hence reduce the cost per Watt peak
different refractive indices has also been exploited
[4.5.1]. Thus these “third generation” technologies
to create fully integrated multiple Distributed Bragg
will be compatible with large scale implementation
Reflectors. These allow a tuning of the forbidden
of photovoltaics. The aim is to decrease costs to well
photonic band such that regions just outside
below US$0.50/W, towards US$0.20/W or better, by
the bandgap (which have enhanced density of
dramatically increasing efficiencies but maintaining
photonic states) lie at the 1500nm absorption
the economic and environmental cost advantages
window of Er. This effectively concentrates light
of thin film deposition techniques (see Fig. 4.1.3
into this absorbance window and boosts up-
showing the three PV generations) [4.5.1, 4.5.2]. To
conversion quadratically. Enhancements of up
achieve such efficiency improvements such devices
to 80 are possible with multiple levels, with a
aim to circumvent the Shockley-Queisser limit for
consequent narrowing of the wavelength range
single band gap devices that limits efficiencies to
enhanced. Enhancements of 40 are very practical.
the “Present limit” indicated in Fig. 4.1.3 of either
This represents a big potential improvement for up-
31% or 41% (depending on concentration ratio).
conversion efficiency.
This requires multiple energy threshold devices
Plasmonics has previously been applied to both such as the tandem or multi-colour solar cell. The
first generation and second generation Si cells with Third Generation Strand is investigating several
very significant enhancements in absorbance of approaches to achieve such multiple energy
near band gap photons. Work in 2010 has applied threshold device [4.5.1, 4.5.3].
plasmonic silver nanoparticles to nanostructured
The two most important power loss mechanisms
Si QD layers. Again significant enhancements of
in single-band gap cells are the inability to absorb
photoluminescence (reciprocal with absorption)
photons with energy less than the band gap (1 in
have been absorbed, up to 16x for wavelengths
Fig. 4.5.1), and thermalisation of photon energy
just shorter than the effective band gap. Also
exceeding the band gap, (2 in Fig. 4.5.1). These
application to the rear of the layer is seen to be
two mechanisms alone amount to the loss of
better than the front because the reflection of the
about half of the incident solar energy in solar cell
silver particles themselves is beneficial rather than
conversion to electricity. Multiple energy threshold
parasitic. Just as with thin film second generation

62
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

A tandem photovoltaic cell using quantum confined QDs or QWs

to engineer the band gap of the top cell and potentially also the
lower cells. Short wavelength light is absorbed in the top cell and
longer wavelengths in lower cells, thus boosting the overall voltage
generated and hence efficiency.

Figure 4.5.2

approaches can utilise some of this lost energy. Previously we have demonstrated the ability to
Such approaches do not in fact disprove the validity fabricate materials which exhibit a blue shift in the
of the Shockley-Queisser limit, rather they avoid it effective band gap as the QD or QW size is reduced,
by the exploitation of more than one energy level using photouminescence [4.5.4] and absorption
for which the limit does not apply. The limit which [4.5.5] data [4.5.6]. A thin film deposition of a self-
does apply is the thermodynamic limit shown in organised QD nanostructure is achieved through a
Fig. 4.1.3, of 68.2% or 86.8% (again depending sputtered multi-layer of alternating Si rich material
on concentration). and stoichiometric dielectric [4.5.4]. On annealing
the excess Si precipitates into small nanocrystals
In the Third Generation Strand, we aim to introduce
which are limited in size by the layer thickness,
multiple energy levels by fabricating a tandem cell
thus giving reasonable size uniformity, as first
based on silicon and its oxides, nitrides and carbides
demonstrated by Zacharias [4.5.7]. Demonstration
using reduced dimension silicon nanostructures to
of doping of these layers with both phosphorus
engineer the band gap of an upper cell material.
and boron to create a rectifying p-n junction has
We are aiming to collect photo-generated carriers
resulted in devices with a photovoltaic open circuit
before they thermalise in the “Hot Carrier” solar cell.
voltage of 490mV [4.5.6, 4.5.7, 4.5.8].
Also we are investigating absorption of two below
bandgap photons to produce an electron-hole Formation of Si (or Ge or Sn) QDs through layered
pair in the cell by up-conversion in a layer behind thin film deposition of Si rich material which
the Si cell using erbium doped host materials. crystallises into uniform sized QDs on annealing.
In order to optimise the requisite properties,
A cell based entirely of Si, or other group IV
all these structures are likely to be thin hence
elements, and their dielectric compounds with
maximising absorption of light in thin structures
other abundant elements (i.e. silicon oxide, nitride
through light trapping is very important. Hence we
or carbide) fabricated with thin film techniques, is
are also investigating localised surface plasmon
advantageous in terms of potential for large scale
enhanced coupling of light into these Third
manufacturability and in long term availability of
Generation devices.
its constituents. Such thin film implementation
implies low temperature deposition without
4.5.2 Si nanostructure solar cells melt processing, it hence also involves imperfect
crystallisation with high defect densities. Hence
4.5.2.1 The ‘‘all-Si’’ Tandem cell
devices must be thin to limit recombination due to
their short diffusion lengths, which in turn means
Researchers:
they must have high absorption coefficients.
Shujuan Huang, Ivan Perez-Wurfl, Dirk König,
Tom Puzzer, Xiaojing Hao, Sangwook Park, Bo For photovoltaic applications, nanocrystal materials
Zhang, Dawei Di, Yong-Heng So, Zhenyu Wan, may allow the fabrication of higher band gap solar
Sammy Lee, Yidan Huang, Gavin Conibeer, cells that can be used as tandem cell elements on
Martin Green top of normal Si cells [4.5.11, 4.5.12]. For an AM1.5
solar spectrum the optimal band gap of the top
cell required to maximize conversion efficiency is
We are developing a material based on Si (or other
~1.7 to 1.8eV for a 2-cell tandem with a Si bottom
group IV) quantum dot (QD) or quantum well
cell [4.5.13]. To date, considerable work has been
(QW) nanostructures, from which we can engineer
completed on the growth and characterization of Si
a wider band gap material to be used in tandem
nanocrystals embedded in oxide [4.5.7, 4.5.14] and
photovoltaic cell element(s) positioned above a thin
nitride [4.5.15, 4.5.16] dielectric matrices. However,
film bulk Si cell, see Fig. 4.5.2.
little has been reported on the experimental

63
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

FMG scheme employed

for finding electronic
states and the
confined energy.

Figure 4.5.5

properties of Si nanocrystals embedded For Si QDs in SiO2 the precipitation occurs according
in SiC matrix [4,5,17]. These are of to the following:
Si rich dielectric
2-5nm particular interest for application in
photovoltaic devices because of an
Stoichiometric expected significant increase in carrier
dielectric, 2-5nm
transport due to a decrease in the barrier
Precipitation of excess Si from Si rich dielectrics in
height between adjacent nanocrystals
SiNx and SiC follows a similar crystallisation reaction
[4.5.18]. As a result, sufficient carrier
Quartz or glass substrate as Si precipitates from the amorphous matrix. The
mobility can be obtained to satisfy device
Substrate techniques has also been applied to growth of Sn
fabrication requirements.
and Ge QDs in SiNx in SiO2. Ge quantum dots can be
precipitated at substantially lower temperature, as
  4.5.2.2 Fabrication of Si QD
discussed below.
Multilayer deposition of alternating nanostructures
Si rich dielectric and stoichimetric Thin film techniques are used for nanostructure
dielectric in layers of a few nm. On 4.5.2.2.1 Carrier tunnelling transport in Si
annealing the Si precipitates out
fabrication. These include sputtering and plasma QD superlattices
to form small nanocrystals of a size enhanced chemical vapour deposition (PECVD).
Transport properties are expected to depend on
determined by the layer thickness. The deposition is a variation of the multi-layer
Nanocrystal or quantum dot size is the matrix in which the silicon quantum dots are
alternating ‘stoichiometric dielectric / Si rich
therefore uniform. embedded. As shown in Fig. 4.5.4 different matrices
dielectric’ process, shown in Fig. 4.5.3, followed
produce different transport barriers between the
Figure 4.5.3 by an anneal during which Si nanocrystals
Si dot and the matrix, with tunnelling probability
precipitate limited in size by the Si rich
heavily dependent on the height of this barrier.
layer thickness [4.5.12, 4.5.7]. The most
Si3N4 and SiC give lower barriers than SiO2 allowing
SiO2 successful and hence most commonly
larger dot spacing for a given tunnelling current.
used technique is sputtering, because
Si3N 4
SiC
3.2 eV of its large amount of control over
1.9 eV deposition material, deposition rate
0.5 eV
and abruptness of layers. A multi-target
c-Si 1.1 eV c-Si 1.1 eV c-Si 1.1 eV remote plasma sputtering machine with
two independent RF power supplies as The results suggest that dots in a SiO2 matrix would
0.9 eV
2.3 eV well as dditional DC power supplies is have to be separated by no more than 1-2 nm of
4.7 eV
used in this work. matrix, while they could be separated by more than
RF magnetron sputtering is used to 4 nm of SiC. Fluctuations in spacing and size of the
deposit alternating layers of SiO2 and dots can be investigated using similar calculations.
SRO of thicknesses down to 2nm. [SRO It is also found that the calculated Bloch mobilities
Bulk band alignments between
silicon and its carbide, nitride refers to Si rich oxide, formed by co-sputtering Si do not depend strongly on variations in the dot
and oxide. Tunnelling probability and SiO2 .] Deposition of multi-layers, consisting spacing but do depend strongly on dot size within
between QDs separated by d the QD material [4.5.18]. Hence, transport between
typically of 20 to 50 bi-layers, is followed by an
depends exponentially on the
anneal in N2 from 1050 to 1150°C. During the anneal dots can be significantly increased by using
square root of the barrier height
(ΔE1/2) multiplied by d (e.g. the excess silicon in the SRO layer precipitates to alternative matrices with a lower barrier height, ∆E.
[4.5.19] p244). form Si nanocrystals between the stoichiometric For the same tunnelling current the spacing of QDs
oxide layers. can increase for oxide to nitride to carbide matrix.
Figure 4.5.4

64
4.5.2.2.2 Modelling of quantum In order to demonstrate the conformity of results
confinement and tunnelling transport from this model with experimental results,
ARC
we compared it with experimental results
for Silicon QDs in SiO2 dielectric grown using PHOTOVOLTAICS
Researchers: sputtering methods and characterized using CENTRE OF
Binesh Puthen-Veettil, Robert Patterson, photoluminescence (PL) [4.5.20]. Figure 4.5.6 shows EXCELLENCE
Dirk König the dependency of confined energy on the size 2010/11
of QDs for Si/SiO2 QD structures. Confined energy ANNUAL REPORT
increases exponentially as QD size decreases.
Band gap engineering for these Si nanostructure Comparison shows that for larger QDs experimental
materials requires a large number of almost and theoretical values are similar, but for smaller
identical quantum dots (QDs), evenly distributed dots the experimental results for the confined
in a dielectric layer. As the proximity of dots energy shows a lower degree of confinement
increases, confined wavefunctions in the QDs than the theoretical result. This difference can be
interact with one another to form mini-bands. It attributed to interface defects between Si and SiO2
is also seen that higher Si content in the Si rich and to defects within the SiO2 matrix but near the
layer gives rise to bigger QDs, due to the tendency interface. For QDs below a diameter of about 4 nm
of QDs to merge together and form bigger dots these interface effects have a larger influence on the
during annealing [4.5.12, 4.5.20]. If the dielectric confined energy level. This is because the ground
layer is a good diffusion barrier for Si (such as the state electron density, which is concentrated at
nitride layer in Si/Si3N4 or Si3N4 interlayer samples, the centre of the QD, is closer to these interface
see sections 4.5.2.3.1 and 4.5.2.3.2), this effect is defects for the smaller QDs and hence more
strongly reduced. For relatively thick layers in which strongly influenced.
diffusion is not constrained, the shapes of these
QDs can differ greatly from an ideal spherical shape. As an extreme example of irregularly
Calculation of confined energy in QD structures is of shaped QD, we have investigated
interest since their electronic and optical properties electronic states in aribitrarily
can be determined by determining the confined shaped QDs. Here we have used the
energy levels in the structure. “horseshoe” morphology [4.5.24],
formed by lateral growth of Si QD
Since these calculations are extremely difficult to inside SiO2 dielectric shown in
perform using ab-initio methods due to the large Figure 4.5.7 (left). These states are
memory and computation time requirements, found to be very different from the
we have developed a model to quantitatively expected electronic energy levels
analyse electronic states and their interactions in that are associated with spherical
a QD array, in the framework of the effective mass QDs as shown in Fig. 4.5.7 (right).
approximation (EMA) calibrated by ab-initio DFT The degeneracy in spherical QDs (3
calculations. This model is realized by solving the in first excited level and 5 in second Simulation results – multigrid
three dimensional time independent discretized method and Density Functional
excited level) is lost due to the asymmetric shape.
Theory (DFT) calculations [4.5.23]
Schrödinger’s equation in the EMA, employing a The deviation from the expected confinement level compared with the experimental
modified Full Multi Grid method (FMG) [4.5.21] to becomes significant at higher confined levels. results for Silicon QDs in
overcome the difficulty of computational intensity. SiO2 dielectric.
By using exact solution methods to solve the Eigen This thus indicates the importance of knowing
the regime of QD growth relevant for the material Figure 4.5.6
equation for the coarsest grid, the non convergence
problems [4.5.22] arising from oscillatory conditions. As shown in the TEM tomography of
eigenvalues are eliminated. Since initial guesses Section 4.5.2.2.3, under appropriate conditions of
of the solution are very close to the approximate Si density and annealing, a spherical morphology
solution, this method proves to be much faster than can be maintained and QD merging avoided,
other iterative methods. The flow diagram for this thus giving greater control over the confined
process is shown in Fig. 4.5.5. energy levels.

First three modes of electronic

wave-functions in a “horseshoe”
shaped QD with high surface to
volume ratio (far left). The ground
state is located near the centre of
the dot and is approximately in a
spherical shape with zero nodes.
The number of nodes increases
with the increase in the energy
level. Energy levels associated with
a horseshoe shaped QD compared
with that of a 6nm diameter
spherical QD (left). The two
morphologies give energies which
are most similar for their ground
states, with the horseshoe shape
giving significantly lower energies
than spherical for higher confined
levels. The degeneracy in energy
levels is lost in the horseshoe
shaped dot.

FigurE 4.5.7

65
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Energy filtered TEM tomography images of Si QDs

in a SiO2 matrix formed as multilayers (left), and
top view of one Si QD layer (right).

Figure 4.5.8

The spectra fittings of the R and T data. The inset

shows the corresponding employed optical model in
the fitting of a 20 bilayer structure composted of 5 nm
SRN / 3 nm Si3N4.

Figure 4.5.9

4.5.2.2.3 3D EF-TEM tomography on Si

QDs embedded in oxide matrix formed the ability of electron tomography to reconstruct
as multilayers (Collaboration with the morphology and internal structure of a Si QD
Cornell University) nanostructure. Fig. 4.5.8 (left) shows the aligned
“on-axis” reconstructed EFTEM tomography, and
Fig. 4.5.8 (right) shows a top-view of the Si QD size
Researchers: and shape distribution in one Si QD layer. With such
Xiaojing Hao, Shujuan Huang, Lena Fitting information, we can confirm that inside the Si QD
Kourkoutis (Cornell University), Ivan layer, most of the Si QDs are reasonably spherical
Perez‑Wurfl, Tom Puzzer and well separated from each other. This indicates
that individual QDs have not merged to form larger
particles under these conditions. This is important
A significant challenge to the design of nano-scale in designing the optimum concentrations to
materials and devices arises from the difficulty give minimum QD spacing whilst maintaining
of characterising complex three-dimensional individual QDs.
structures on small length scales, which is
nonetheless critical in understanding the material 4.5.2.3. Different materials for
performance. Whilst a
QD nanostructures
number of techniques, such
as conventional transmission As described in section 4.5.2.2.1, different matrices
electron microscopy, are useful to modify both the tunnelling probability
provide sufficient two- between adjacent quantum dots and the energy
dimensional (2-D) resolution, levels in the quantum dots themselves. This
they have insufficient approach has been carried out for both Si QDs in
depth sensitivity to obtain SiNx and in SiC. It is also possible to have asymmetric
internal three-dimensional superstructures in which quantum confinement
(3-D) structure. Electron and hence enhanced band gap is maintained in
tomography can be applied the plane, but in which tunnelling probability is
to obtain such information maximised in the direction normal to the plane such
by using high-angle dark- that transport of carriers to contacts is maximised.
field scanning transmission Interlayers use for such asymmetric structures
electron microscopic can also act as diffusion barriers enhancing size
(STEM) tomography and uniformity of QDs.
energy filtered transmission Alternative group IV materials such as Ge offer the
electron microscopy (EFTEM) possibility of lower temperature precipitation of
tomography. We have applied nanostructures and the potential for band gaps
this EFTEM tomography to lower than that of silicon should these be required
investigate Si QDs embedded for tandem cell elements under a silicon cell. We
in an oxide matrix formed have investigated both QDs and QWs of Ge in either
as multilayers, in order to SiO2 or SiNx matrices.
better understand the dot
Real (ε1) and imaginary (ε2) parts size, shape and distribution for further optimization
of the complex dielectric function of Si QD device performance. Fig. 4.5.8 illustrates
of Si QDs samples with various sizes
obtained from spectral fittings.

Figure 4.5.10

66
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

(a) Normalized PL spectra of Si QDs samples (S3-S5)

with various sizes. (b) Comparison of band gap
expansion (ΔEg) of Si QD as a function of QD size.

Figure 4.5.11

4.5.2.3.1 Silicon QDs embedded in silicon

nitride matrix

Researchers:
Yong-Heng So, Shujuan Huang

(i) Size-dependent optical properties of Si

QDs in Si-rich nitride/Si3N4 superlattice
An approach similar to spectroscopic ellipsometry
analysis has been proposed to determine the size
dependent optical properties of silicon quantum
dots (Si QDs) in Si-rich nitride/silicon nitride
superlattice structure (SRN/Si3N4-SL) [4.5.26]. The
optical properties of Si QDs are modelled using
the Tauc-Lorentz (TL) model and Bruggeman
effective medium approximation that can yield the
energy bandgap of the Si QDs based on spectral
(ii) n-type conductivity of nanostructured XPS spectra of (a) Si 2p and (b) Sb
fitting of the reflection (R) and transmission (T) of
thin film composed of antimony-doped Si 3d3/2 for the undoped and 0.54
characterized samples, as shown in Fig. 4.5.9. A four- at.% Sb-doped Si-NC. (c) Fitting
phase optical model as shown in the inset of Fig.
nanocrystals in silicon nitride matrix
of Sb 3d3/2 spectrum by two peaks
4.5.9 was employed for the spectral fitting. Highly conductive thin films composed of antimony separated by ~0.8 eV in energy (δ1
(Sb)-doped Si nanocrystals (Si-NCs) embedded and δ2).
Figure 4.5.10 shows that the dielectric functions
in Si3N4 matrix were prepared by a co-sputtering Figure 4.5.12
of Si QDs are strongly size dependent. Sample S3,
technique. Results from structural characterizations
S4 and S5 correspond to SRN/Si3N4 thicknesses
suggest that doping with Sb concentration of 0.54
of 3/3 nm, 4/3 nm and 5/3 nm, respectively. The
at. % has negligible affect on the crystallization
suppressed imaginary dielectric function of Si QDs
properties of Si-rich nitride (SRN) films. The X-ray
exhibit a broad peak centred between transition
Photoelectron (XPS) data of Fig. 4.5.12(a) and (b)
energies E1 and E2 of bulk crystalline Si and which
show the Si 2p and Sb 3d3/2 spectra of the undoped
blue shift towards E2 as the QD size reduces.
and Sb-doped SRN films. The slight asymmetric Sb
Figure 4.5.11(b) shows that the band gap 3d3/2 peak was analyzed using a curve-fitting routine
expansion indicated by the TL model when the and could be decomposed into two components,
size of Si QD reduces is in good agreement with δ1 (537.0 eV) and δ2 (537.8 eV) as shown in Fig.
PL measurements (Fig. 4.5.11(a)). The bandgap 4.5.12(c). The two components δ1 and δ2 (~0.8 eV
expansion with the reduction of Si QD size is well apart) can be assigned to neutral Sb and positively
supported by the ab-initio calculations of confined charged donors, respectively.
energy levels from [4.5.25].
Thus, the appearance of δ2 suggests the presence
of Sb-Si bonds which implies that Sb stoms were
either incorporated within the Si-NCs or located at
the interface between the NCs and Si3N4 matrix.

67
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

(a) Conductivities measured

throughout the entire voltage
range. (b) ln(σ) vs 1/T1/2 plot
from 120 to 220 K, for the 0.54
at.% Sb-doped sample. The
solid line is least square fit
to the data. (c) Temperature-
dependent conductivities for
the undoped and 0.54 at.%
Sb-doped samples.

Figure 4.5.13

The temperature dependent conductivity data 4.5.2.3.2 Si QDs in SiO2/Si3N4

of Fig. 4.5.13(a) indicate that the conductivity of hybrid matrix
0.54 at.% Sb-doped SRN film (~2.8×10-2 S/cm)
is six orders of magnitude higher than that for Reserachers:
undoped material (~7.3×10-8 S/cm), which could be
Dawei Di, Ivan Perez-Wurfl, Gavin Conibeer
attributed to an increase in carrier concentration.
Furthermore, n-type electrical behaviour with
carrier concentration of 2.4×1016 cm-3 and mobility To improve the current transport properties in the
of 2.94 cm2/V s in the doped films as observed from vertical direction and to obtain better size control of
Hall measurements was attributed to free carrier Si quantum dots, we proposed a new design based
generation due to the effective Sb doping. on Si QDs embedded in a SiO2/Si3N4 hybrid matrix
[4.5.27, 4.5.28]. By replacing the SiO2 tunnel barriers
The temperature-dependent conductivities of with the Si3N4 layers, the new material manages
both undoped and Sb-doped samples are shown to constrain the growth of doped Si quantum
in Fig. 4.5.13(b). As can be seen, doping of the dots effectively and enhances the apparent band
Si-NCs strongly influences the electronic transport gap. Also electrical characterisation on Si QD/c-
properties of the films. Arrhenius-like temperature Si hetero-interface test structures indicates the
dependence is observed in the T range between new material possesses improved vertical carrier
220 and 320 K, attributable to thermally activated transport properties.
conduction. From the slope of the Arrhenius plot,
we found that the activation energy EA decreases to Doped and undoped samples with different
0.182 eV for the Sb-doped Si-NCs film, suggesting barrier dielectrics (SiO2 and Si3N4) are compared.
effective n-type doping of the Si-NCs. Nevertheless, Samples were fabricated using the co-sputtering
the extracted EA is much larger than that of bulk Si multilayer technique with layers containing excess
with the same dopant. One possible explanation for Si consisting of SRO for both materials, with an SRO
the large EA observed would be the deeper donor thickness of 4 nm. Samples were either undoped or
level expected in Si-NCs. Also, it is possible that the co-sputtered with either P2O5 or B for n- and p-type
number of free carriers available for conduction doping respectively. After annealing at 1100⁰C
may be limited due to a trap density distributed the crystalline properties were characterised by
within the bandgap that is comparable to the X-ray diffraction (XRD), as shown in Fig. 4.5.14. For
doping density. samples with Si3N4 barriers, the measured Si NC
sizes are 4.3 nm when undoped, 3.5 nm when B
Interestingly, the conductivity for the Sb-doped doped and 5.0 nm when P2O5 doped. For samples
sample at T < 220 K deviates from Arrhenius with SiO2 barriers, grain sizes are 7.7 nm when
behavior and is best described by the expression, σ undoped, 7.2 nm when B doped, and 15 nm when
= σ0 exp [-(T0/T) 1/2] as illustrated in Fig. 4.5.13(c). The P2O5 doped. The Si NC sizes vary significantly
charge transport mechanism can be explained well although all samples have the same silicon-to-
by a percolation-hopping model. oxygen ratio in the SRO layers and the same as-
deposited layer thickness.

It can be observed from the experiment that

samples with Si3N4 barriers contain generally
smaller Si QDs than samples with SiO2 barriers.
This is because Si3N4, a denser and stiffer material
than SiO2, acts as a better diffusion barrier than
SiO2 [4.5.27]. It could also be thermodynamically
related to the interface free energy between

68
 10000 Samples with SiO2 barriers

9000 (111)
8000

7000
(220)
ARC
Intensity (a.u.)

6000 (311)
P2O5 doped
5000
PHOTOVOLTAICS
4000

3000 undoped
CENTRE OF
2000 EXCELLENCE
1000 B doped 2010/11
0
15 20 25 30 35 40 45 50 55 60 65 ANNUAL REPORT
2 Theta (degrees)

10000 Samples with Si3N4 barriers

9000

8000

7000 (111)
Intensity (a.u.)

6000

5000 (220) (311) P2O5 doped

4000

3000 undoped
XRD patterns of (a) samples with SiO2 barriers, (b)
2000
samples with Si3N4 barriers.
1000 B doped
0
Figure 4.5.14
15 20 25 30 35 40 45 50 55 60 65
2 Theta (degrees)

1.0E+02 1.0E+01
B doped samples P2O5 doped samples

nitride barrier nitride barrier

1.0E+01 oxide barrier oxide barrier
Current density (A/cm )
Current density (A/cm )

2
2

1.0E+00

1.0E+00

1.0E-01
1.0E-01

1.0E-02 1.0E-02
0.0 0.2 0.4 0.6 0.8 1.0 1.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2
Bias voltage (V) Bias voltage (V)

the Si NC and the matrix dielectrics during the not related to carrier recombination. It can be Vertical current density
growth of the Si NCs. As a result, Si3N4 is able to clearly seen from the measurement that for B measurements on (a) B doped
samples, (b) P2O5 doped samples.
constrain the growth of Si NCs very effectively, doped materials, Si3N4 barriers result in a current
Insets show how barrier type and
within ±1 nm of the intended diameter (i.e. 4 nm, enhancement of approximately an order of NC dimension influence the vertical
the deposition thickness of SRO). In contrast, Si magnitude over those with SiO2 barriers (Fig. 4.5.15 current transport process (showing
NCs with SiO2 barriers are significantly larger than (a)). This can be qualitatively explained by [4.5.19]: dominant effect only in each case,
not to scale).
the as-deposited thickness of SRO - over 180%
larger for undoped and B doped and 375% for P2O5 Figure 4.5.15
doped. Hence, it can be noted that samples with
phosphorus (P2O5) dopants tend to form larger sized
Si NCs than undoped samples, while introduction
of boron doping tends to suppress this size increase (4.5.1)

and slightly reduces NC size. This phenomenon is where Te is the tunnelling probability between
more pronounced for samples with SiO2 barriers quantum dots, d is the barrier thickness or the
(Fig. 4.5.14 (a)) than those with Si3N4 barriers (Fig. separation between dots, m* is the effective mass of
4.5.14 (b)). electron, ħ is the reduced Plank constant, ΔE is the
Si3N4 barriers have lower band gaps (~5.3 eV) energy difference between the conduction band
than SiO2 (~9 eV) therefore they should be more edge of the barrier material and the conduction
transparent to charge carrier transport because band edge of the confined QDs.
of the larger tunnelling probability for electrons Although the difference in Si NC sizes would play an
and holes. To verify this hypothesis in practice, important role in the tunnelling event, the variation
we compared the current transport of SiO2 of ΔE is still the dominant factor for B doping. On
barrier structures with Si3N4 barrier structures the other hand, for the P2O5 doping, samples with
via the measurement of vertical currents under a Si3N4 and SiO2 barriers have similar currents (Fig.
bias voltage. 4.5.15 (b)). These apparently contradictory results
The current-voltage characteristics of these test can be explained by referring back to the XRD
structures are shown in Fig. 4.5.15. It should be results. The P2O5 doped sample with SiO2 barriers
noted that these samples are not PV devices. contains Si NCs with an average size of 15 nm, this
Therefore, the dark currents measured are primarily suggests that a single Si NC is physically penetrating
affected by the materials’ conductivities and are three thin layers on average (inset of Fig. 4.5.15 (b)).

69
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

(a) Raman spectra of SRC samples after RTA and (b) the
same samples with additional furnace annealing. (Bulk
Si wafer peak at 520.5 cm-1 for reference.)

Figure 4.5.16

This greatly reduces the number of barrier layers (i) Study of RTA induced stress
that the carriers need to tunnel through. This size Two different annealing processes have been
effect becomes so significant that it effectively applied on all SRC amorphous samples: furnace
competes with and largely cancels the increased annealing at 1100oC/1hr and RTA at 1100oC/30
conductivity due to lower ΔE for the Si3N4 barrier sec. After annealing, Si and SiC nano-crystals were
material. clearly observed in TEM and XRD. The peak positions
This result is very promising for both QD size of Raman spectra are down-shifted to lower
control and increased vertical currnt transport in wavenumbers for RTA samples (Δ=5.4cm-1 from
these nitide barrier interlayer Si QD materials. bulk Si) as compared to furnace annealed samples
(Δ=2cm-1 from bulk Si), indicating greater tensile
4.5.2.3.3 Silicon QD nanocrystals stress in the Si nano-crystals, as shown in Fig. 4.5.16
(a). This is because of a different thermal expansion
HRTEM image of annealed 2.0nm embedded in silicon carbide matrix
coefficient for Si and SiC crystals and the fast
UT-Si3N4 sample. The maintenance
temperature ramping rate in RTA. In order to release
of a layered structure and the Researchers:
presence of Si-QDs can be seen. the stress in RTA samples an additional annealing
Zhenyu Wan, Shujuan Huang, Gavin Conibeer process was carried out. Raman analysis indicates
Figure 4.5.17
that after an additional furnace anneal at 1100°C for
Silicon carbide (SiC) has a lower barrier height 30 min, all samples with different Si concentrations
than either Si3N4 or SiO2. Therefore it seems likely could release most of their residual stress, as shown
that it should offer a higher tunnelling probability in Fig. 4.5.16 (b). However, an additional RTA could
between QDs and hence a higher conductivity not release residual stress in low Si concentration
matrix than either. In previous work, we have samples due to insufficient duration [4.5.31].
demonstrated that both Si and SiC have been
crystallised by high temperature annealing of a (ii) Si QDs embedded in SiC matrix with
single thick Si-rich SiC (SRC) layer or of a Si1-xCx/ Si3N4 barrier layers
SiC multilayer structure [4.5.28, 4.5.29]. We
Si3N4 is also considered preferable to SiO2 in term
believe that the formation of β-SiC nanocrystals
of carrier transport. It has been proven as a good
may hinder the formation of Si QDs. Also they
diffusion barrier to suppress inter-diffusion between
may cause current leakage via the SiC grain
silicon rich layers in multilayer structures during
boundary traps to increase the shunt current in
annealing [4.5.27]. In 2010, we have successfully
the solar cell. Therefore, the work in 2009 focused
introduced ultra-thin Si3N4 (UT-SiN) barriers (0.2nm-
on studying the mechanism of the crystallisation
2.0nm) into a Si-NC in SiC matrix structure using
of SiC and optimising the materials by comparing
sputtering followed by RTA. Crystallisation of the
the annealing methods of rapid thermal
SiC matrix has been greatly suppressed and a clear
annealing (RTA) and conventional furnace
layered superlattice structure can be observed as
annealing. We found that RTA annealed samples
shown in the TEM of Fig. 4.5.17. The nanocrystal
revealed a better degree of crystallisation
size of all samples was calculated using XRD peak
XRD analysis and corresponding of Si nanocrystals with a smaller residue of
analysis to quantitatively investigate the optimum
grain sizes from the Scherrer amorphous Si [4.5.30]. The work in 2010 has been to
formula for all the Si3N4 barrier thickness of the Si3N4 barrier layer required to
further improve the structure by reducing the stress
thicknesses. Insert: Si (111) and SiC produce confined crystalline Si-QDs, as shown in
(111) peaks of each sample. caused by RTA and by replacing SiC barrier layers
Fig. 4.5.18. It is seen that as the Si3N4 barrier layer
with thin silicon nitride layers in order to suppress
Figure 4.5.18 thickness increases, the β-SiC peaks decrease, such
the growth of SiC crystals.
that β-SiC-NCs almost disappear when the Si3N4
barrier thickness is over 0.8nm.

70
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Absorption edge fitting for UT-SiN2.0 and UT-SiN0.2 Temperature-depentent conductitity of different Si3N4
samples respectively, (insert: band gap energy for barrier thickness.
different Si3N4 thicknesses).
Figure 4.5.20
Figure 4.5.19

Optical characterization has been performed 4.5.2.3.4.1 Quantum Dots in SiO2

to investigate quantum confinement effects for
different nanocrystal sizes. Optical analysis reveals a Researchers:
blue shift in the strong absorption edge, consistent Santosh Shrestha, Bo Zhang, Pasquale Aliberti,
with quantum confinement effects in small Si NCs Gavin Conibeer
matching qualitatively calculated results using
effective mass theory from other groups, as shown
in Fig. 4.5.19.
(i) Fabrication of Germanium
Temperature dependent I-V measurements were quantum dots
carried out to investigate the carrier transport The current work builds on earlier work on Ge
mechanism. The conductivity is best described nanocrystals carried out in our group [4.5.32]. The
by the expression, σ = σ0 exp [-(T0/T) 1/2] when new work uses a different approach to annealing of
the Si3N4 barrier thickness is greater than 0.5 nm, the samples which gives a wider range of control Cross-section HRTEM image of a
as illustrated in Fig. 4.5.20. The charge transport of Ge NC formation. Ge NC samples have been
typical sample containing GeRO
mechanism can be explained well by the layers between GeO2/SiO2 layers
fabricated by RF magnetron sputtering using a following annealing.
percolation-hopping model, similar to the result combination of a combination target comprising
found in Section 4.5.2.3.1 for Sb doped SiQDs in a fused quartz disc and high purity Ge strips. Figure 4.5.21
Si3N4. Finally, we conclude that in future, a 0.5-0.8 Sputtering of the Ge-rich oxide (GeRO) layer is
nm thick layer of Si3N4 barrier layer would be achieved by sputtering with just an Ar background,
optimum to achieve both good Si-NC confinement
and minimal film resistivity for a candidate material
for photovoltaic applications.

4.5.2.3.4 Germanium Nanostructures

Although the main focus has been on Si
nanocrystals during the last decade, other group IV
nanocrystals have also been studied. Since Ge has
smaller electron and hole effective masses and a
larger dielectric constant than Si, the excitonic Bohr
radius of bulk Ge is larger than that of Si. This leads
to a more prominent quantum confinement effect
in Ge NCs. Furthermore the lower melting point of
Ge at 938.3oC implies that Ge NCs should be able
to form at lower temperatures than Si NCs. This is
indeed a significant advantage both for processing (a) Raman spectra and (b) GIXRD
compatibility and for processing costs, although whereas sputtering of the oxide layers, GeO2/ image of the as-deposited and
annealed films.
set against this are the greater cost and lower SiO2, is performed by reactive sputtering with O2.
abundance of Ge as compared to Si. These layers are alternately deposited to obtain Figure 4.5.22
the desired multilayered structure. A thick oxide
capping layer is deposited to prevent oxidation of
the GeRO layers during subsequent annealing. Post
deposition annealing of the samples was performed
under a low pressure (in-situ in the growth
chamber) during which Ge NCs are formed. Results

71
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Raman spectra of the samples annealed for Raman spectra of the samples annealed at
different durations. different temperatures.

Figure 4.5.23 Figure 4.5.24

from structural characterization using TEM, shows three sharp peaks at 27.12°, 45.13° o and
Raman, XRD and PL have been obtained. 53.21°, corresponding to the groups of planes {111},
{220} and {311} of crystalline Ge, respectively. This
Figure 4.5.21 shows a high resolution TEM
observation confirms good crystallinity of the Ge
image of a section of a typical sample
phase in the film and agrees well with our Raman
containing GeRO layers between GeO2/
results. The average size of the Ge NCs, calculated
SiO2. The sample was annealed at 650°C.
from the {111} peak broadening using the Scherer
Ge nanocrystals of about 5nm diameter
equation, is about 4.5 nm. This value is slightly
are evident. Although there is a slight
smaller than that estimated from the HRTEM image.
variation, the NCs appear to be spherical
The difference in Ge NCs sizes obtained from these
in shape. The clear lattice fringes observed
two methods may possibly be due to spatial non-
in the TEM image give direct evidence
uniformity of the Ge NCs size; with TEM probing
of the formation of the Ge NCs. It should
a much smaller sample region compared to XRD
be noted that the distance between the
Room temperature PL of a measurement. In addition, the penetration depth
lattice fringes is 3.3 Å, which is consistent
multilayer film containing Ge NCs in of the incident X-rays is larger than the thickness
GeO2/SiO2. with the lattice spacing of the {111} planes of the Ge
of our film, thus information obtained from XRD is
diamond structure.
Figure 4.5.25 averaged throughout the whole film.
Fig. 4.5.22 (a) shows Raman spectra of the as-
The crystallisation of Ge NCs with annealing
deposited and the annealed samples, including
duration has been investigated with Raman
those for bulk Ge as a reference. For these samples,
spectroscopy. Fig. 4.5.23 shows Raman spectra
the sputtering times for each GeRO layer and
for identical multilayer samples with each GeRO
GeO2/SiO2 layer were 8 minutes and 6 minutes,
and GeO2/SiO2 layer deposited for 6 minutes. The
respectively, and the post deposition
samples were annealed at 685°C for different
annealing was performed at 650°C for 40
durations as indicated in the diagram. The
minutes. A broad hump at around 270
crystallization of Ge is found to take place within
cm-1 is observed in the spectra of the
Growth Temp. the first few minutes of annealing. However, the
Intensity (arb. units)

445 oC as-deposited film which is attributed to

III noticeable broad hump, which is attributed to small
420 oC the non-crystalline Ge phase. But in the
NCs, suggests an early stage of the crystallization
400 oC annealed film, it is replaced by a sharp
process at this annealing duration. As annealing
II 380 oC peak at 300.5 cm-1, which is very close to
duration increases, Raman peaks become sharper
350 oC the Ge-Ge optical phonon mode for bulk
320 oC and narrower indicating an increase in Ge
I RT Ge (300.2 cm-1), indicating the formation
crystallinity. It is also observed that the peaks show
of Ge NCs with good crystallinity. Peak
150 200 250 300 350 400 negligible difference for annealing durations longer
R aman Shift (cm-1) broadening and an asymmetric shoulder
than 10 minutes.
on the lower frequency side can be
interpreted by the model of the optical The average size of NCs, as calculated from XRD
Raman spectra from Ge-NCs in
single layer samples grown at phonon confinement effect in nanocrystals data, increases with annealing duration. This is
different substrate temperatures. [4.5.33]. However, it seems that in our Ge system the consistent with the Raman results and agrees
high frequency side of the Raman peak does not well with the growth dynamics by diffusion of
Figure 4.5.26
show a shift to lower phonon frequencies which is neighbouring Ge atoms. Most importantly, the
usually the case for NCs. increase in the NCs size begins to level off after 15
minutes of annealing in this particular case. This is
Figure 4.5.22 (b) shows GIXRD patterns for this
similar to the size confinement effect observed for
sample. In the as-deposited film there are no
the growth of Si NCs in SiO2 matrix [4.5.7, 4.5.34].
obvious peaks but only two broad bands at around
In our superlattice structure, Ge NCs are confined
2θ = 26° and 2θ = 49°. After annealing, the sample
within the GeRO layers between GeO2/SiO2 layers,

72
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

A High resolution TEM image of

multilayered Ge NCs in SiO2 matrix.
The sample was grown at 380°C.

Figure 4.5.27

which work as barrier layers to the crystal growth. growth and no post annealing process was used
Hence, Ge tends to precipitate with a diameter [4.5.36].
approximately equal to the thickness of the GeRO
Figure 4.5.26 shows Raman spectra for a single layer
layers. This control mechanism is more significant
GeRO film deposited at different temperatures as
for thin layers with thickness of a few nanometers,
indicated in the diagram. For clarity, the graph has
within which there is 2D rather than a 3D growth.
been divided into three different regimes. In regime
Figure 4.5.24 shows the Raman spectra of the I, the samples grown below 350°C are shown.
samples annealed at different temperatures for 40 In this case, the spectra show no Raman peaks
minutes, which is enough to complete the growth related to nanocrystalline Ge, instead broad bands
process as discussed above. As expected, the small centred around 280 cm-1 are observed. The bands
nanocrystalline asymmetric hump is reduced with are shifted toward higher frequency by about 10
the increase in annealing temperature within the cm-1 compared with that of amorphous Ge (α-Ge)
range of temperatures studied here, just as in located at around 270 cm-1. This indicates that α-Ge
the case of annealing durations. The results also coexists with very small particles (1~2 nm) [4.5.37].
illustrate the formation of Ge NCs in SiO2 matrix at a
For the sample grown at Tg = 320°C, the observation
temperature as low as 620 °C.
of a hump at ~ 280 cm-1 as well as a shoulder at ~
A sample with the identical process sequence as 298 cm-1 implies an increase in the number and
those for structural characterizations was used for size of small particles. The rapid growth of Ge NCs
photoluminescence measurements, except that is found to occur in Regime II, in which samples
the film was deposited on quartz. Fig. 4.5.25 shows grown at temperatures between 350°C and 400°C
results of room-temperature PL measurement. are shown. The Raman peak corresponding to the
The PL spectrum consists of a single broad band TO phonon mode of the crystalline Ge (c-Ge), near
centred at 1.77eV (corresponding to a wavelength 300.4 cm-1, appears at Tg = 350°C and becomes
of 700 nm) which can be fitted with three Gaussian sharper with increase in the growth temperature.
distributions. Measurement on GeO2/SiO2 film This range of temperature is very close to the onset
deposited on quartz under similar conditions temperature for Ge crystallisation reported in
did not show any observable PL in the range of [4.5.38]. However, when the growth temperature Tg
wavelengths concerned. Thus the PL signal from the reaches 420°C (Regime III), a drastic degradation of
multilayer sample can be attributed to Ge NCs in the Ge-Ge peak intensity and shape is observed. These
GeRO layers. Furthermore, a blue shift of PL energy phenomena indicate the absence of Ge crystallinity
with NCs size has been shown in previous work in this temperature range.
[4.5.32]. We tentatively consider that this is due to
Based on the Raman results discussed above, it is
the band gap increase induced by the quantum
expected that the growth temperature window is
confinement effect in Ge NCs [4.5.35].
between 350oC and 420oC, in which Ge atoms or
clusters can accumulate and eventually grow into
(ii) Low temperature growth of highly crystallised nanocrystals. These temperatures
Germanium quantum dots are much lower than the usual post-deposition
In this section, a low temperature growth of Ge annealing temperatures used for Ge NCs fabrication.
nanocrystals is discussed. This is advantageous
A HRTEM image of a multilayered sample consisting
because of the potential for reduced processing
of alternate layers of GeRO and GeO2/SiO2 films is
cost and suitability for cheaper substrates such
shown in Fig. 4.5.27. In this case the sample was
as soda-lime glass. Samples were grown with RF
grown at Tg = 380oC without further post-deposition
magnetron sputtering as described earlier, except
thermal treatment. The TEM image shows close to
in this case the substrate was heated during the film
spherical Ge NCs with fairly uniform size, separated

73
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Schematic diagram of a Ge-NC/n:c-Si HJ diode.

Figure 4.5.28

by the barrier layers. The Ge NC sizes are in moderate amounts of Ga and Sb dopants was also
the range of 4.8 - 5.8 nm which is consistent investigated. The doped films exhibit similar
with the average size (~ 4.8 nm) determined conduction properties to the intrinsic films, which
from XRD. The image in the inset shows means the films were still dominated by surface
clear lattice fringes inside one of the Ge- state induced hole conduction and that the dopants
NCs (indicated by the arrow). The fringes were not effectively activated. This is not surprising
are consistent with {111} planes of the Ge if one realizes the screening of shallow dopants in
diamond structure. NCs due to the increase of binding energy and
ionization energy. This effect together with the
(iii) Electrical properties of Ge-NC inherent hole generation effect can make it very
thin films challenging to produce n-type Ge-NC thin films.

Ge-NC thin films about 250 nm ~ 300 nm

were deposited. The coverage of Ge was ~ (iv) Ge-nc/c-Si Heterojunction Devices
20% of the area of the composite target. Heterojunction (HJ) devices employing Ge-NC thin
XPS has shown that this resulted in a Ge films on lightly doped n-type crystalline silicon
atomic concentration of ~ 35% in the films. substrates with impurity concentration of ~ 1 x
Undoped, Ga-doped and Sb-doped Ge 1015 cm-3 (n:c-Si) were fabricated to evaluate the
targets were used. During the sputtering compatibility of the nanostructured thin film and
process the substrates were intentionally investigate the design parameters required for its
heated up to ~ 380 oC for in-situ growth of application in photovoltaic devices. The schematic
(a) The semi-logarithmic plot of Ge-NCs. Finally, RTA treatments at different diagram of the HJ device is illustrated in Fig. 4.5.28.
dark I-V curves of a typical i:Ge-NC/
temperatures, including 650oC, 700oC, 750oC, 800oC The total area of the device is 1 cm2. The thickness
n:c-Si HJ device in both polarities
at room temperature. (b) The were carried out in nitrogen [4.5.39]. of the Ge-NC layer is about 250 nm. The Ge-NC
measured suns-Voc characteristics thin films were post-annealed by RTA at 800oC in
The current conduction characteristics of i:Ge-NC
of the HJ device. The red solid circle order to achieve lower film resistivity and higher
indicates the 1-sun point. thin films were measured using HP4140B pA Meter/
carrier concentration. The Al front fingers and rear
DC voltage source with tri-axial wires to eliminate
Figure 4.5.29 contact were deposited by thermal evaporation. No
noise. The relatively high conductivity and carrier
passivation or sintering process was performed on
concentration in the i:Ge-NC thin films is quite
the devices.
surprising, taking into consideration that they were
not intentionally doped. Temperature dependent Figure 4.5.29 shows the dark I-V curves and suns
measurement revealed a relationship, VOC of a typical i:Ge-NC/n:c-Si HJ device at room
suggesting a thermally activated nearest hopping temperature. The device shows good current
conduction mechanism in these films. The carrier rectification of three orders of magnitude at ± 1.5
transport was considered to occur at the surface V. Since both front and rear electrodes were ohmic
state energy level and a theoretical calculation contacts, the rectification effect can be attributed
predicts that the density of surface states of Ge can to the junction. The apparent photovoltage was
provide sufficient free holes to explain the observed detected from the illuminated device and the 1-sun
conductivity. RTA treatments further increase the Voc was found to be ~ 314 mV. The best fitting to the
film conductivity without changing much the experimental data using a two diode model predicts
structural properties of nanocrystals. This an effective ideality factor of ~1.01 throughout
improvement was tentatively attributed to the the entire injection range, which is indicative of a
modification of surface structure of NCs and dominating bulk and surface recombination in the
reduction of oxygen-deficiency-related defects in heterojunction device. Whilst this HJ device does
the SiO2 matrix. The effect of incorporating not demonstrate photovoltaic behaviour of the Ge
NC material it does show good rectification and is

74
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

XRD spectra of Ge QDs of various Ge content in Ge-rich nitride layer with a fixed substrate temperature of 600 ºC (left) and
TEM image of 40 vol% Ge self-assembled at 600 ºC (right).

Figure 4.5.30

encouraging as a starting point for using Ge-NC in with sizes in agreement with the NC size observed Ge vol% ∆(2Ө) of (111) NC Size (nm)
true photovoltaic applications. in TEM images. 30 5.62 NA
40 3.85 2.22
Figure 4.5.31 summarises the band gap engineering
4.5.2.3.4.2 Germanium QW and QD trend with the size of Ge QDs. The absorption edge
50 2.41 3.54
60 2.38 3.59
nanostructures embedded in silicon shows a tendency to shift to high photon energy 70 1.92 4.45
nitride matrix when Ge nanocrystal size decreases. The absorption
coefficient was calculated by a simplified equation; Ge NC size estimation from XRD.
Researchers:
Table 4.5.1
Sammy Lee, Jian Chen, Shujuan Huang,
Martin Green
where T and R are the transmission and reflection
and d is the thickness of the film. With
As discussed above in section 4.5.2.3.1, a silicon
decrease in Ge content in the GRN layer,
nitride matrix (Si3N4) should increase tunneling
and hence in the diameter of the QDs, the
probability and hence increase conductivity of Ge
absorption edge appears to increase by up
nanostructure materials. In addition a nitride will
to ~1.0 eV, as estimated with the Tauc-plot
strongly suppress the possibility of oxidation of Ge.
method. However, these data should be
Ge QDs and QWs were fabricated by co-sputtering
considered as indicative of the trend only
followed by post-annealing and self-organisation by
as there is error in quantitative values for
substrate heating.
the shift in absorption edge due to
uncertainty in extrapolating back to the
(i) Ge QDs embedded in silicon energy axis.
nitride matrix
Using a very similar approach to our previous work (ii) Ge QWs with silicon nitride
on Ge QDs in oxide and Sn QDs in nitride matrices, barriers Absorption edge observed from
alternating layers of germanium-rich nitride (GRN) transmission and reflectance
With a similar approach to the Ge QD fabrication,
layers and Si3N4 layer were alternately co-sputtered measurements.
Ge QWs were fabricated by co-sputtering of
on a heated substrate of either a silicon wafer Figure 4.5.31.
alternating GRN and Si3N4 layers and followed by
or quartz slide. The Ge QDs crystallise due to
furnace annealing or RTA. The Ge content in GRN
minimisation of surface energy and the Si3N4 layer
was higher in this case in order to give continuous
deposited on top of the GRN layer truncates crystal
layers. It was varied between 60vol% to 100vol%
growth. The temperature of the substrate and the
and the annealing temperature in N2 was also varied
Ge content in GRN layers were the dominant factors
from 600ºC to 900ºC for 1 hour. The structural and
in controlling the size of the NCs. The structure of
optical properties were studied by GIXRD, Raman
the NCs was studied by GIXRD, Raman spectroscopy
spectroscopy, TEM and transmission and reflectance
and TEM, and the optical properties were studied
measurements.
with transmittance and reflectance measurements
analysed using a Tauc-plot. Fig. 4.5.32 shows the continuous Ge layer with
its QW-like-structure between amorphous Si3N4
Figure 4.5.30 shows the XRD spectra of samples of
barriers. Highly ordered lattice planes can clearly
various Ge content in GRN. The Bragg peaks of {111},
be seen in these Ge QW, indicating formation
{220} and {311} planes sharpen with increasing Ge
of large single crystal type structures. Based on
content, indicating the increase of the nanocrystal TEM image of the sample
Raman and XRD characterisation, the crystallisation
sizes. Using the Scherrer equation the size of the containing 60 vol% Ge and furnace
temperature shows a significant dependence annealed at 900ºC.
nanocrystals was estimated as shown in Table 1,
on the thickness of either Ge QW or Si3N4 barrier
Figure 4.5.32

75
 ARC +
W
PHOTOVOLTAICS V m(Temp)
I -
CENTRE OF SRO 4nm Al
EXCELLENCE contacts
B doped B doped B doped
2010/11
SRO SRO SRO
ANNUAL REPORT SiO2, 2nm
bilayers bilayers bilayers

Wa
P doped SRO bilayers with
sheet resistance Rsheet
Non-uniform current
injection
Quartz substrate

Schematic diagram of the of the fabricated inter-

digitated Si QD in SiO2 devices, with B and P doping
to create the p-n junctions and illustrating the effect
of current crowding due to high lateral resitnce to the
back contact.

Figure 4.5.33

layers, varying from 700ºC to 900ºC. The absorption 4.5.2.4.1 Equivalent circuit model of
coefficient estimated from the transmittance and nanocrystal devices on quartz substrates
reflectance measurements also shows an apparent
In order to better understand the limitation
increase in band gap by up to 1.1eV when the
imposed by the device on the solar cell
thickness of the Ge QWs is 1.5nm.
performance, it is necessary to find a good
Both Ge QDs and QWs nanostructures in Si3N4 equivalent circuit model specific to our devices.
show evidence of band gap enhancement. Further The high resistivity of the base layer makes it
work will investigate electrical properties and the imperative to consider the two dimensional effects
possibility of device fabrication. of current flow [4.5.41]. In order to analyse the I-V
characteristics of these diodes we first generalised
4.5.2.4 Silicon nanocrystal devices on the model to include any number of diodes in
quartz substrates series. The series connection of diodes is used
to explain the ideality factors higher than two
normally observed in our structures. We believe this
Researchers:
is a reasonable model as the ideality factor observed
Ivan Perez-Wurfl, Xiaojing Hao, Dawei Dai, is almost independent of the diode current. This
Adrian Shi type of behaviour has also been observed in
multi quantum well laser diodes [4.5.42], where
it has been proven to arise from an unintended
As previously reported in the 2008 and 2009 series combination of diodes. Based on this series
annual reports devices have been fabricated combination of diodes, it is possible to obtain an
using the SiQD in SiO2 materials, with p-n junction expression relating temperature dependent I-V
formation using B or P of multilayers, respectively measurements to the band gap [4.5.9]. It is further
[4.5.9, 4.5.10, 4.5.40]. The fabricated p-n diodes possible to linearise the expression around an
consisted of sputtered alternating layers of SiO2 average measuring temperature, Tavg, as follows:
and SRO onto quartz substrates with in-situ
boron and phosphorus doping. The top B doped
bi-layers were selectively etched to create isolated
p-type mesas and to access the buried P doped
bi-layers. Aluminium contacts were deposited by
evaporation, patterned and sintered to create ohmic
contacts on both p and n-type layers. The fabricated
interdigitated solar cells have an effective area of (4.5.2)
up to 0.12cm2. The devices exhibit rectification and
where k is Boltzmann’s constant, q is the electron
a photovoltaic response with VOC up to 493 mV, but
charge, Tnom is the temperature at which the
with as yet very small currents and bad fill factors.
saturation current, Ioi , is defined, ni is the diode
These are partly due to the very high resistance
ideality factor, and αi is the saturation current
of the material and in particular to the relatively
temperature exponent. Notice that this equation
long lateral paths to contacts at the back contact,
shows that the I-V characteristics are related to a
necessitated by groth on an insulating quartz
sum of band gaps.
substrate. The device structure with appropriate
contacting is shown schematically in Fig. 4.5.33. As the current flows from the base contact to
This also shows the current crowding effect which the emitter, a linear voltage drop along the base
results from the high lateral resistance. and under the diode isolation mesa causes an
exponential change in the diode current. This
crowding of the current at the edge of the diode

76
mesa, depicted in Fig. 4.5.33, can be modelled photocurrent flows only through this fractional
adding a current dependence on the series area, the series resistance is inversely proportional
ARC
resistance. This series resistance, RS, can be to this fraction: R1=Rtot/fraction. Since the
expressed as the sum of a current independent, Rext photocurrent, Iph, flows through R1, the illuminated PHOTOVOLTAICS
and a current dependent series resistance Rint arising I-V characteristics are limited by a larger resistance CENTRE OF
from current crowding: than that observed in the dark condition, as long as EXCELLENCE
the fraction of the diode area is smaller than one 2010/11
half. The simulations depicted in Fig. 4.5.35 show ANNUAL REPORT
the reduction of ISC as the fraction, f, is varied from
(4.5.3) 99% to 1% of the total diode area.

The value of this resistance can be found by With the circuit model described, it will be possible
numerically solving the following transcendental to extract complementary information from dark
equation: and illuminated I-V measurements.

In view of these simulations, it is clear that the

electrical characterisation of our devices needs
(4.5.4)
to take into consideration previously overlooked
Within this mathematical framework it is then limitations. For example, great care should be
possible to extract the value of the series taken when interpreting the Quantum Efficiency
resistance, remove its effect from the measured extracted from a spectral response measurement
I-V characteristics to extract the actual ideality of as the assumed condition of short circuit current
the diode current. We normally observe an ideality may be incorrect even if the device is externally
factor of 3. Based on the proposed model of at least short circuited (internally, the diode
two diodes in series, without any assumptions, it may be forward biased). Moreover,
is only possible to establish that the value of the as the current is proportional to the
band gap extracted from temperature dependent photon flux, and the flux is generally
I-V measurements corresponds to a sum of the band different at each wavelength tested,
gaps (or activation energies) of these diodes. the internal bias of the device can be
different at each point of the spectrum
An interesting behaviour observed in these investigated.
devices was the apparent lack of correspondence
between the dark and light I-V characteristics. The
4.5.2.4.2 Demonstrator program
series resistance extracted from the diode dark I-V
for Si nanostructure devices
characteristic is too small to explain the limited
short circuit current as well as the low fill factor Using the approach described in
measured under a simulated 1-sun condition. A Section 4.5.2.4.1, a device simulator Equivalent Spice circuit
more complete circuit model is necessary to explain has been developed in an MS Excel spreadsheet. representation of the
The Demonstrator is a 2D simulator of the diodes fabricated devices.
this discrepancy. We have proposed a model where
the observed behaviour is due to two distinct fabricated from Si QD nanostructures. It is based on Figure 4.5.34
areas in the fabricated devices. The photocurrent is a simple diode model with a few fitting parameters
produced only in a small area of the device, this area and uses the method in section 0 and in [4.5.9].
being proportional to a fraction of the normalized There are three areas which the user can modify:
diode area. In a Spice circuit model this area is given
a value smaller than 1, that we denote as fraction. 1. Measured parameters: these come from
This will be a fitting parameter to reproduce the measured properties of the fabricated diodes.
The base resistivity and its temperature
measured dark and illuminated I-V characteristics.
dependence are measured. The
Figure 4.5.34 shows the circuit proposed.
device geometry is known. The
A relatively large percentage of the device area default values are representative
is responsible for the measured characteristics in ones for the devices.
the dark. The current in this area is caused by the 2. J0, ideality factor, band gap and XTO
can be varied to give flexibility in
diffusion of minority carriers caused by the applied
fitting to a specific diode properties. f=0.01
voltage (V1 in Fig. 4.5.34) from the p or n side to
3. The variation parameters are Current f=0.03
the opposite region. This current is expected to
and Temperature.
be large due to the low lifetime of the minority f=0.1
The outputs are:
carriers (mostly recombination current in the
depletion region). The observed series resistance 1. The I-V curves for the range of
will be proportional to the total series resistance, temperatures chosen, in both log-lin f=0.31
Rtot, and inversely proportional to the area, R2=Rtot/ and lin-lin output.
(1-fraction), where the diffusion current occurs. The 2. The variation of the ideality
f=0.99
dark I-V behaviour is modelled by the top branch of factor with current for
the circuit depicted in Fig. 4.5.34. D_QD1 and D_Sch various temperatures.
represent the series connection of diodes whose 3. The series resistance
band gaps are extracted using Eq. (4.5.2). The series with temperature.
Spice simulations of 1-sun I-V
resistance, R2, has the temperature and current 4. The extracted band gap, using either a
characteristics based on the circuit
calculation incorporating a series resistance
dependence detailed in Eqs. (4.5.3) and (4.5.4). depicted in Fig. 4.5.34. The fraction,
correction or with both a series resistance and a f, represents the normalized area of
Only a small part of the diode area may have a large current modification term. the diode where the photocurrent
enough lifetime to produce a photocurrent. As this is produced.

Figure 4.5.35

77
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11 Dark I-V curves.
ANNUAL REPORT
Figure 4.5.36

Local ideality factor.

Figure 4.5.37

Some example outputs, using the default parameter doping by incorporation of B and P during
settings, follow: sputtering growth [4.5.9, 4.5.10]. Also, formation of
p- and n-type materials is clearly demonstrated in
Figure 4.5.36 shows that a reasonable diode
the Si QD nanostructure materials with Si nitride
characteristic is shown, with good turn-on and
interlayers doped with either P or B discussed in
rectification.
Section 4.5.2.3.2; in the Sb doped n-type Si QDs in
Figure 4.5.37 shows the increase in the ideality Si3N4 in Section 4.5.2.3.1; in the p-type Ge QD in
factor with current is as expected with a SiO2 material in section 4.5.2.3.4; and in MOS type
diode transiting from domination by radiative devices doped with either P or B by a diffusion
recombination (n=1) to being dominated by anneal discussed in [4.5.43].
Shockley Read Hall recombination (n=2). But the
However, the doping mechanisms taking place
large regime in which n=3 and then region where
in these structures are not well understood.
it goes to much higher values, are indicative of
Theoretical work has shown that direct doping of
a non-physical aspect to the model. The most
the QDs is prevented by segregation of impurity
likely explanation is that rather than two diodes
atoms from the perfectly crystalline QDs [4.5.44,
in series there are actually several, probably a
4.5.23]. Experimentally this is supported by data
variable number in different parts of the device, the
on the free electron density in Si nanocrystals
combination of which lead to a composite ideality
using Electron Paramagnetic Resonance (EPR),
factor.
which show that 95% of P atoms are segregated
The demonstrator is useful as an iterative tool used to the surface of the nanocrystals and that their
to simulate the real measured data from a device. contribution to doping is at least an order of
The parameters used for such a fit then define the magnitude lower than the atomic concentration
values of the circuit elements in the EC model. [4.5.45]. Further evidence that doping causes
Attachment of physical meaning to these elements electronic changes comes from the quenching of
will then require further development and interface luminescence on the incorporation of P (or Au) in
with the EMA/quantum mechanical model. Si nanostructures [4.5.46] and also from a small
enhancement of luminescence observed at low
4.5.2.5 Doping in Si QD nanostructures P doping levels followed by quenching at higher
levels [4.5.47]. This latter being explained by the
Fabrication of a PV device from the Si QD materials
passivation of QD surface states by low levels of P
requires a control of work function such as to allow
increasing luminescence and then the saturation of
separation of photogenerated electron-hole pairs.
this mechanism by excess P. This also explains the
Methods by which this can be achieved include
non-monotonic behaviour of activation energy seen
fabrication of a grown or diffused p–n junction
in P doped nanostructures [4.5.10].
or p–i–n junction with the Si QD multilayers as
the i-region. These require careful control of If direct doping of the QDs is not occurring,
the work functions of the p and n-regions. P-n then another possibility is modulation doping
junction devices have been fabricated and doping of the matrix SiO2, as is commonly used in III-V
demonstrated. Progress on the theory explaining nanostructures; but this is ruled out because of
these doping effects using conventional dopants the very high ionisation energies of about 5eV that
and the possibility of other modulation doping would be required for the resultant deep defects
approaches has been made. in a dielectric matrix [4.5.48]. These clearly will
not occur at room temperature, or even at higher
As discussed in Section 4.5.2.4, rectifying p-n and
temperatures.
p-i-n structures can be fabricated with ‘conventional’

78
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Stoichiometric
dielectr ic, 2-5 nm a) b)

Two models for doping of Si QD nanostructures. (a)

An extended sub-oxide region surrounding the Si QDs
allows doping by P or B at relatively shallow levels. These
then can provide carriers to be captured by the QDs
through a modulation doping mechanism. (b) QD size
enhancement (suppression) by P (B) doping results in a
type I heterojunction between P and B doped material.
This facilitates carrier separation due to the much greater
mobility of electrons – the Dember effect [4.5.49].

Figure 4.5.40

4.5.2.5.1 Doping mechanisms in

Si QD nanostructures

Researchers:
Dawei Di, Xiaoming Hao, Ivan Perez‑Wurfl,
Gavin Conibeer

As there is clear evidence for doping by P and B,

a third possibility is that it is the sub-oxide QD/
matrix interface region, with its shallower doping
defect levels, that is doped by the P or B atoms, thus
providing free carriers to be captured by the QDs
[4.5.49]. As shown in Fig. 4.5.38, ab-initio calculation
of the levels associated with B atoms associated due to an amorphous region around the Si QDs. An Ab-initio calculation using
explanation for the band gap of this being larger GAUSSIAN: (a) the density of states
directly with the interface region on the surface of for a Si83/84OH62 oxide terminated QD
a Si QD indicate that there are no shallow or deep than bulk Si is if this region consists of sub-oxide
with and without a B atom at the
levels introduced by the B within the effective band material. If this region were doped with B or P, it interface; (b) the optimised Si QD,
gap (HOMO-LUMO gap). The reason is that the could provide free carriers to be captured by the consisting of Si83B-OH62, with the B
QDs in a modulation doping mechanism, thus atom at top centre (in pink).
strong affinity of B for O results in very strong B-O
bonds, effectively splitting any available OMOs and giving rise to the p- or n-type behaviour observed. Figure 4.5.38
UMOs a long way apart and well inside the already (This is illustrated schematically in Fig. 4.5.40 (a).)
existing density of states. An alternative explanation for the rectifying
Wavelength (nm)
However, this approach considers the interface character of the junctions obtained can 300 400 500 600 700 800
25
between Si QD and SiO2 matrix to be very abrupt. be based on the modification to the Si QD 4.8
4.4

crystalisation discussed in Section 4.5.2.3.2, 4.0

Energy (eV)

In fact there is highly likely to be an extended 20

3.6

transition region of a sub-oxide around the QD. If for P and B doping of Si QDs in oxide with 2.4
Absorption (106cm-1)

2.2

this region is a sufficiently extended silicon sub- nitride interlayers [4.5.49]. The fact that P (B) 15
2.0
1.8
m=2
m=1/2

oxide (SiOx) it could provide flat Bloch bands able doped material produces larger (smaller) 0.8 0.9 1.0 1.1
O/Si ratio
1.2 1.3

to be doped in a pseudo bulk-like regime. In order QDs means that its confined energy levels 10
SiO0.86/SiO2
that B and P doping levels are shallow enough to and hence effective band gap will be smaller
SiO1.00/SiO2
be ionised in such a region, x would need to be less (larger) than undoped material under 5 SiO1.30/SiO2
than about 0.5. Such a region would be amorphous otherwise similar conditions. Thus a junction
and hence not able to be modelled with ab-initio between a P and a B doped region will 0
5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5
methods. But the presence of such regions can be actually be a type-I heterojunction between Energy (eV)
determined from absorption measurements. small and large band gap materials,
respectively, as illustrated in Fig. 4.5.40 (b).
There is some support for this theory in the Room temperature absorption
absorption data for Si QDs in SiO2 [4.5.5] shown in It is also very likely that the mobility of electrons co-efficient of annealed SiOx/
in these materials, whilst not big, will still be much SiO2 multilayer films with various
Fig. 4.5.39. This shows a region of strong absorption x (1100oC, 1hour). The inset is
with an absorption edge which blue shifts as greater than that of holes. Hence photogenerated
an estimate of the approximate
the O to Si ratio increases, which translates to an electrons near the junction would experience a optical band gap (filled squares:
increase in QD size. But there is also an additional drift field sweeping them into the P doped material strong absorption; empty circles:
with its lower conduction band edge, whilst holes absorption tail) [4.5.5].
weakly absorbing tail which extends well into the
effective band gap region. This tail is likely to be with their very limited mobility would be immobile. Figure 4.5.39
This would result in electrons accumulating in the P

79
 Electronic density of states (DOS) of Si10 QD in 2 ML
ARC SiO2 (dotted black), fully oxidised nanocrystal (grey)
PHOTOVOLTAICS and Si10 QD in 2 ML SiO2 with Sc replacing Si atom in
CENTRE OF the SiO2 shell at the outermost position (top). DOS of
Sc-doped approximants split into α (spin up, red) and
EXCELLENCE β (spin down, blue) DOS because of non-even number
2010/11 of electrons within the approximant due to acceptor
ANNUAL REPORT state of Sc. The b-highest occupied molecular orbital
(HOMO) 1193 indicated in magenta can take up 0.07
electrons from the Si10 QD over two 2 ML (6 Å) of perfect
SiO2. The β-lowest unoccupied MO (LUMO) 1194 shown
in green can take up a much bigger proportion of an
electron if the Si QD is slightly bigger (e.g. Si35, dQD =
11 Å). The Sc-doped nanocrystals are shown with the
iso-density plots of 1.35×10−3 e/Å3 for β-HOMO 1193
(lower left) and β-LUMO 1194 (lower right). The atoms
of the Si10 QD and the Sc atom are highlighted in cyan.
Si atoms in SiO2 are grey, O atoms are red, H-atoms
terminating the outermost O-bonds are white. All DFT
computations carried out with a Hartree-Fock 3-21G(d)
optimisation and B3LYP/6-31G(d) electronic structure
calculation [4.5.23].

Figure 4.5.42

material and being collected in the external circuit atom, and they must have one single oxidation
as if from an n-type material, with electrons injected number. The transition metal scandium (Sc) fulfils
back into the device through the B doped material the last two requirements. It also has a very suitable
where they would recombine with the immobile valence state configuration which consists of a
holes, thus making the B material appear p-type. full (doubly occupied) 4s shell and one electron
This qualitatively mirrors the behaviour observed, in the 3d1 state located within the 4s shell. On the
but will require corroboration. As it relies on the energy scale, the 4s electrons have a higher binding
difference in electron and hole mobilities, it is energy as compared to the 3d1 and 3d2 states. This
essentially the same as the Dember effect, which provides the 3d1 and 3d2 states with an electrostatic
is known to produce a photovoltaic effect through screening which results in a higher binding energy
carrier separation, albeit not very efficiently. of these states. Thereby an unoccupied 3d2 state
exists at a high binding energy which is able to take
It is also possible that both the sub-oxide doping
up an electron from a nearby source such as a Si QD,
effect and the mismatched QD size effects operate
see Fig. 4.5.41.
in the device. The fact that both mechanisms act
in the same direction is fortuitous as both could Ab-initio Density Functional Theory calculations
therefore contribute to the observed rectifying on a Si10 QD in 2 mono layers (MLs) of SiO2 have
photovoltaic behaviour. Differentiation between been carried out, with a Si atom at the outermost
these or a determination of the relative strengths position with full O termination replaced by a Sc
Spatial distribution of 4s and 3d
electronic states of Sc (upper) and
will require further study on the effects of doping, atom. The Si10 atom cluster is the smallest size which
energetic arrangement of 4s and 3d interlayers and band alignments. These will also still behaves as a Si QD [4.5.23]. Such a nanoparticle
states (lower). further optimisation of the devices and lead to allows us to directly observe the electronic structure
better photovoltaic performance from these Si QD of Sc as a modulation acceptor in SiO2 in the
Figure 4.5.41
nanostructure devices. proximity of a Si QD, see Fig. 4.5.42.

Figure 4.5.42 shows that 0.07 electrons stemming

4.5.2.5.2 Ab-initio modelling of from the Si10 QD (dQD = 7.3Å) are localized at
modulation doping possibilities in the Sc atom over two 2 ML (6 Å) of perfect SiO2.
Si QDs nanostructures Technological relevant QD sizes start at dQD ≈ 20Å,
with a much lesser degree of quantum confinement
Researchers: as compared to a 7.3Å QD. It is thus reasonable to
Dirk König, James Rudd, Daniel Hiller (IMTEK, assume that a Si QD with dQD ≥ 20Å will be positively
University of Freiburg, Germany) ionized to a much greater extent, ensuring that
holes are majority carriers in a Si QD super lattice
(SL) embedded in SiO2. Experimental verification
As discussed in Section 4.5.2.5, direct doping of of this doping mechanism is being carried out. The
QDs by dopants is thermodynamically very unlikely. difference to conventional acceptor modulation
The alternative of modulation doping of the SiO2 doping as used in III-V electronic devices is the
matrix is also not feasible due to the strong anionic way the Si or Ge QDs are ionised by Sc in SiO2. As
nature of O leading to very deep ionisation energies Sc induces a deep acceptor level in SiO2 it cannot
[4.5.48]. We have concentrated recently on possible be thermally ionized from the SiO2 valence band.
candidates for more efficienct acceptor doping Instead, the Sc acceptor state takes up the electron
of Si QDs. Modulation acceptors must have an directly from the initially occupied state at a lower
energetic position below the confined hole level, binding energy presented by a nearby QD, see Fig.
they must not introduce defect levels in the QD 4.5.43. This requires the Sc acceptor to be within
band gap if located at the interface as an active 50Å of the Si QD in order to exploit field emission
dopant and as a completely saturated foreign

80
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT
Acceptor modulation doping:
Conventional example with GaS
QD in AlAs matrix and electron
thermally emitted into acceptor
(Mg), such that holes are captured
by the QD (left). In SiO2, charge
transfer occurs directly as the
Sc state is too deep within the
band gap to be thermally ionized.
Instead, the acceptor state is
ionised directly from the QD by
field emission of an electron into
the Sc acceptor state.

Figure 4.5.43

which is a consequence of the energy difference of P and B dopants, and modelling of possible
between the QD HOMO and the Sc acceptor state. doping mechanisms based on modified modulation
doping and these size effects. Alternative
Acceptor modulation doping of SiO2 can be
modulation dopants are being modelled with
accomplished by incorporation of a small amount
scandium as an acceptor looking like a strong
of Sc, say one monolayer of Sc, into the SiO2 barriers
candidate. Further work will build on this increased
between the SRO QD array layers. During the
understanding of anisoptropic structures, growth of
segregation anneal, Sc is incorporated into the
different materials and understanding of doping to
SiO2 matrix and thereby activated as an acceptor.
improve materials and devices further.
Ab-initio calculation indicates that Sc attached
to the Si QD does not create any levels within
the HOMO-LUMO gap of a Si84(OH)64 QD. We can
4.5.3 Hot Carrier cells
therefore assume that a Sc atoms bonded onto a
Si QD does not result in a detrimental change in Researchers:
its electronic structure. This concept is now being
Shujuan Huang, Santosh Shreshtha, Dirk
investigated experimentally.
König, Robert Patterson, Pasquale Aliberti,
Binesh Puthen Veettil, Lara Treiber, Ivan
4.5.2.6 Summary of Group IV Perez-Wurfl, Andy Hsieh, Yu Feng, James
nanostructures for tandem cell elements Rudd, Stephan Michard, Martin Green,
In 2010, significant progress has been made on Gavin Conibeer
understanding the transfer of growth parameters
of Si QDs to other matrices. The use of anisotropic
structures both to provide diffusion barriers Hot carrier solar cells offer the possibility of very
to excessive QD growth and to provide higher high efficiencies (limiting efficiency above 65% for
conductivity paths in the transport direction, has unconcentrated illumination) but with a structure
shown the validity of these approaches both for that could be conceptually simple compared to
Si3N4 and for SiC matrices. Knowledge of Ge QD other very high efficiency PV devices – such as
nanostructures has improved significantly with multi-junction monolithic tandem cells. For this
clear p-type behaviour of Ge QD s in SiO2 and very reason, the approach lends itself to ‘thin film’
successful growth of Ge QWs in Si3N4 matrix. The deposition techniques, with their attendant low
increase in sophistication of modelling of confined costs in materials and energy usage and facility to
energy in these structures now matches reasonably use abundant, non-toxic elements.
well with experimental data. An ideal Hot Carrier cell would absorb a wide range
Improved modelling of confined levels in of photon energies and extract a large fraction
irregular shaped particles also matches well of the energy to give very high efficiencies by
with experimental data of other groups. Higher extracting ‘hot’ carriers before they thermalise to
level device modelling is now able to describe the band edges. Hence an important property of a
the anomalous effects due to high lateral series hot carrier cell is to slow the rate of carrier cooling
resistance in devices and consequent current to allow hot carriers to be collected whilst they are
crowding. This is now being used to optimise still at elevated energies (“hot”), and thus allowing
nanostructure growth and device design, so as to higher voltages to be achieved from the cell and
push beyond the current open circuit voltages of hence higher efficiency. A Hot Carrier cell must also
490mV. Part of this is an understanding of doping. only allow extraction of carriers from the device
This has progressed with greater knowledge and through contacts which accept only a very narrow
control over the effects on QSD size of introduction range of energies (energy selective contacts or

81
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Band diagram of the Hot Carrier cell. The device has

four stringent requirements:

Figure 4.5.44

nanocrystals superlattice arrays has

bee applied to real material systems.
The growth of such systems in both
III-V QD superlattices with collaborators
and with colloidal Langmuir-Blodgett
dispersion of Si nanocrsystals has
produced structures which are
now being characterised for their
modulation of phononic properties.
Also meausurement of carrier cooling
rates has been extended to other large
phononic gap bulk materials including
InN, demonstrating the importance of
material quality. Design of structures
(a) Schematic representation ESCs). This is necessary in order to prevent cold for hot carrier cells which should be
of energy and particle fluxes carriers in the contact from cooling the hot carriers, practical and realisable has developed, with the
interactions used in the model
(Particle fluxes - full line arrow,
i.e. the increase in entropy on carrier extraction is device properties more carefully specified and plans
energy fluxes- dotted line arrow). minimized [4.5.50]. The limiting efficiency for the for fabricating such structures in real devices.
(b) HCSC efficiency as a function hot carrier cell is over 65% at 1 sun and 85% at
of carrier extraction energy level. maximum concentration – very close to the limits
Parameters used are: thermalisation
4.5.3.1 Modelling of Hot Carrier Solar Cell
for an infinite number of energy levels [4.5.1, 4.5.51, Efficiency
time = 100 ps, concentration =
1000, lattice temperature = 300K 4.5.52]. Fig. 4.5.44 is a schematic band diagram of a
and absorber layer thickness = 50 Hot Carrier cell illustrating these two requirements.
nm. [J, F and E are current density Researchers:
and particle and energy fluxes a) To absorb a wide range of photon energies;
Pasquale Aliberti, Yu Feng, Santosh Shrestha, Gavin
as denoted by subscripts for A b) To slow the rate of photogenerated carrier
Conibeer, Martin Green
absorption, E emission and IA cooling in the absorber;
Auger processes.] c) To extract these ‘hot carriers’ over a narrow range Collaboration with:
Figure 4.5.45 of energies, such that excess carrier energy is not Yasuhiko Takeda (Toyota Central Research
lost to the cold contacts; Laboratories, Nagoya)
d) To allow efficient renormalisation of carrier
energy via carrier-carrier scattering.
Previous work was focused on developing a new
In 2010 modelling of Hot Carrier efficiencies has model to calculate limiting efficiency of a real HCSC
progressed with implementation of real based on an Indium Nitride (InN) absorber layer. InN
material properties to give more realistic has been chosen as a potential material because of
efficiencies for InN which include Auger its narrow electronic band gap for absorption of a
processes and more realistic contact wide range of photon energies, whilst also having a
structures. Significant progress has been wide phonon band gap. This is good for suppression
made on demonstrating resonance in of phonon decay, and hence for slowing carrier
double barrier selective energy structures. cooling [4.5.53, 4.5.54].
Further work on triple barrier double Si
QW structures has been carried out for Calculation of limiting efficiency was performed
Carriers transmission probability rectifying ESCs. This is complemented by taking into account real optical and electronic
versus energy for (a) ideal ESC, (b) improvements in 2/3D modelling of transport in properties of InN, removing most of the ideality
non-ideal ESC. these ESC structures. For absorbers, modelling of assumptions used in other models [4.5.51, 4.5.52,
Figure 4.5.46 4.5.55, 4.5.56]. The detailed band structure

82
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

(a) HCSC efficiency as a function of extraction width

of ESCs for different extraction energies ΔE. (b) HCSC
efficiency as a function of extraction energy ΔE for
different ESCs energy width. Thermalisation time is
100 ps, lattice temperature is 300 K. Absorber layer
thickness is 50 nm.

Figure 4.5.47

of wurtzite bulk InN has been considered in The maximum efficiency has been found for a ΔE
performing computation of carrier densities, between 1.15 eV and 1.2 eV with a transmission
pseudo-Fermi potentials and II-AR time constants energy window δE of 0.02 eV. The value of limiting
[4.5.57]. Results have been calculated considering efficiency is 39.6% compared to 43.6% calculated
ideal energy selective contacts for the HCSC, which in the previous section using ideal ESCs. The drop
means that contacts have a very high conductivity in efficiency is mostly due to the decrease of open
and a discrete energy transmission level. circuit voltage related to the decreased extraction
Recently the limiting efficiency for the hot carrier level, equation (4.5.7). This is partially compensated
InN solar cell has been calculated considering by an increase in extracted current due to increased
non-ideal ESCs. In this case the carriers are not II rate.
extracted on a single energy level, but in a finite Figure 4.5.47 (a) shows calculated efficiency as a
energy window. Calculations have been performed function of for several values of extraction energy. In
taking into account contact resistance and entropy all the curves two different trends can be identified.
generation effects. If the value of δE is too close to zero, the efficiency
The flux of current travelling through the ESCs is very low due to low carrier extraction, thus a very
towards the cold metal electrodes can be described small value of short circuit current. The conductivity
using the following relation. of the contact in this case is indefinitely large.
Enlarging δE, the number of carriers available for
extraction increases, improving JSC, and so the
maximum efficiency. In general the efficiency peak
has been found for values of δE from 0.02 eV to 0.1
eV depending on the extraction energy ΔE. For the
(4.5.5) configurations which show higher efficiencies, ΔE
< 1.35 eV, the optimum value of δE goes from 0.02
The current density in this case is proportional to eV to 0.05 eV. This optimum value for δE of between
the occupation probability at the two sides of the is very close to kTRT. This represents the variation
ESC. Equation (4.5.5) has been derived assuming no in energy in the contacts such that approximately
correlation of energy of electrons in three different the kTRT will inevitably be lost anyway by carriers
directions as shown in (4.5.6). This assumption is thermalising within the contacts. Thus it sets a
acceptable if there is a parabolic dispersion relation lower limit on a reasonable δE. Therefore this result
at minimum energy point along the three different indicates that the transmission energy range has
directions. to be very small and confirms once again the high
selectivity requirements of ESCs for HCSC [4.5.58].

(4.5.6) In Fig. 4.5.47 (b) the value of maximum efficiency as

a function of ΔE is reported for different values of
Based on the energy and carrier conservation, Δμ δE. It can be observed that for small transmission
and TC at steady state are calculated. energy window the extraction energy which allows
maximum efficiency is lower compared to the one
calculated using ideal ESCs. This effect is related
(4.5.7) to the higher occupancy at lower energies, which
increases the value of JSC for contacts with a small
[quantities as defined in Fig. 4.5.45 (a).] transmission window.

83
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Peak at 1.205eV of the conductance vs. energy plot for

different values of σ for configurational disorder (left).
Conductance vs. energy plot with peak at 1.573 eV for
different values of σ for morphological disorder (right).
Conductance given in units of 2e2/h.

Figure 4.5.48

During fabrication of the quantum dots in a

dielectric matrix for selective energy contacts,
different kinds of irregularities can be present
in the structure, the major disorders being
configurational (disorders in the position of the
dots) and morphological (disorders in the size of
the dots). The extent to which configurational and
morphological disorders determine the electrical
properties of the overall structure is investigated
using simulation runs of the model. The disorders
are assumed to follow a normal distribution from
the mean position and size. The results show the
outcome of an average of 1000 simulation runs with
Planar view of the double barrier 4.5.3.2 Energy Selective Contacts different standard deviation (σ) values.
structure formed by Si QDs in SiO2
matrix, with SiC barriers of 2nm The requirement for a narrow range of contact Figure 4.5.48 shows the simulation results for
width (left). Mean diameter of QDs energies can be met by an energy selective resonance in 2.6nm Si dot in SiO2 matrix under
is 1.8nm. I-V characteristics for the contact (ESC) based on double barrier resonant different orders of configurational disorders. As
double barrier structure with SiO2
tunnelling. Tunnelling to the confined energy levels the disorders increases from σ =0 to σ =1 the
and SiC barriers at temperature 10K
and 300K (right). in a quantum dot layer embedded between two conductance decreases by 53%, but the resonant
dielectric barrier layers, can give a conductance energy remains the same at 1.205eV. This shows
Figure 4.5.49 sharply peaked at the line up of the Fermi level on the confined energy in the QDs does not change
the ‘hot’ absorber side of the contact with the QD as their size is fixed but the effective filtering
confined energy level. Conductance both below reduces dramatically. Figure 4.5.49 shows the
this energy and above it should be very significantly simulation results for a 2.2 nm Si dot in a SiO2
lower. This is the basis of the current work on matrix under different orders of morphological
double barrier resonant tunnelling ESCs. disorders. As the disorder increases from σ =0 to
σ = 1, the conductance decreases by 60% and the
4.5.3.2.1 Modelling of QD structures resonant peak remains the same at 1.573eV. But
the morphological disorders cause major impact
Researchers compared to configurational disorders because of
Binesh Puthen-Veettil, Dirk König, the widening of the energy selection window. This
Gavin Conibeer is due to the distribution in size of the QDs, since
QDs with different sizes have different resonant
energies which are slightly different from the mean
We developed a robust 2 dimensional model resonant energy, the average of them all increase
for describing the transport properties through the spread of the resonant peak thus reducing the
quantum dot structures and have used this model efficiency of the double barrier structure as energy
to understand the filtering characteristics of Energy selective contacts.
Selective Contacts (ESCs). In this way we are able
A planar representation of the double barrier
to compute the effective filtering in 2 dimensions
structure is shown as in Fig. 4.5.49 (left). The barriers
by running numerical simulations. The model is
are usually high band gap dielectric materials like
developed from a discretized Schrödinger equation
SiO2, Si3N4 or SiC. SiC barriers have advantages
by considering the sample volume as a collection
over SiO2 barriers that SiC barriers in double barrier
of discrete points and using an effective mass
structure makes a very good diffusion barrier
approximation method over the entire volume.
for silicon during processing, which can yield a

84
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
250 – 500 nm 500 – 550 nm 550 – 600 nm
2010/11
ANNUAL REPORT

600 – 650 nm 650 – 700 nm 700 – 750 nm

Principle of oa-IV, shown for electron injection into the sample

(left). If free carriers obtain additional energy by a massive
photon flux significantly above the band gap, a smaller bias field
is required to reach a resonant condition. Sample stage of oa-IV
(right); wavelength range indicated. Every 50 nm spectral range
has a photon flux of approximately 50 Suns.

Figure 4.5.50

structure with uniform QD size in the transport free carriers with additional energy
direction thus ensuring sharp resonances at the by optical means, see Fig. 4.5.50. The
resonant energies of the structure. Since the resonant level of the ESC can then
conductance of the structure is more in the case of be reached by carriers at a lower bias
SiC barriers because of the lower barrier height of field, shifting the tunnelling resonance
SiC, thicker barriers can be used to give the same of the ESC – here a QD array – to lower
conductance as SiO2, for which the very thin barriers bias voltages Vbias.
are difficult to fabricate.
Samples based on Si were provided
Figure 4.5.49 (right) shows the I-V characteristics for by the University of Freiburg, see
the double barrier QD structure with SiO2 dielectric Fig. 4.5.51. Si is very far from being
and SiC barriers at temperature 10K and 300K. an ideal Hot Carrier Absorber due to
As can be seen, the structures with SiC potential the complete lack of a gap between
barriers have higher current density at both low optical and acoustic phonon modes
temperature and at room temperature. As the and hence no propensity to slow
temperature increases, the current density as well as carrier cooling. Furthermore, Si
the width of the resonant peak increases. Also, the has an indirect band gap in the
negative differential resistances appear at slightly spectral range of optical excitation
lower voltage in structures with SiC barrier than which renders the interpretation
that in structures with SiO2 barrier. This is due to the of measured data rather complex
increased leakage and thus the lower confinement due to the absorption coefficient aSi Sample structure for oa-IV (upper).
Contact area on QD array is 1
energy for the SiC. changing over two orders of magnitude. However,
mm2, with ≈ 1010 QDs. Optical
for initial proof of concept it is adequate providing
generation rate Gopt within
4.5.3.2.2 Characterization of Energy the excitation of hot electron-hole pairs in the
ballistic transport range
Selective Contacts using ‘Optically material is sufficiently large and collection can be
to the ESC and its relative
assisted I-V’ made from very close to the ESC.
change for photon wave
A Xenon (Xe) arc lamp is a continuous light source length intervals used in
Researchers: with one of the most constant photon fluxes Fhn in oa-IV (middle). For these, the
Dirk König, Stephan Michard (RWTH Aachen, the spectral range 1000 to 500 nm. Nevertheless, absorption coefficient of c-Si
Germany), Binesh Puthen-Veettil there are significant deviations in particular in the aSi and the photon flux of the
long wavelength region (> 800 nm). We therefore Xe arc lamp Fhn were used
Collaboration with: compare oa-IV curves which have the same Vbias (lower). Full green rectangles
for the current transition from forward to reverse show range of measurements
Daniel Hiller (IMTEK, University of
direction, Vbias(j → 0), indicating a constant carrier which can be compared.
Freiburg, Germany)
product (constant quasi-Fermi level positions) at Dashed green rectangles show
the ESC. Under these conditions, the current density measurements which cannot
Optically assisted IV (oa-IV) is a new characterisation as a function of quasi-Fermi level position can be be compared, but approach
method which investigates an energy selective assumed to be identical for these curves and this the high optical injection limit.
contact (ESC) fabricated on a Hot Carrier Absorber current density can be treated as a constant offset Please note breaks on left
material with optical excitations in a pre-defined for each Vbias value. ordinate in both graphs.
wave length range. If a very large optical generation
For liquid nitrogen (LN2, T = 77 K) measurements, Figure 4.5.51
rate Gopt exists adjacent to the ESC, it can probe this
resonant tunnelling behaviour (negative differential
HC population immediately next to the contact.
resistance – NDR) was detected for all spectral
Alternatively, if the HC population within the
ranges. With the method mentioned above, only
absorber is known, the ESC can be tested for its
the oa-IV curves from 800 – 750 nm and 650 –
energy selectivity. The basic principle is to provide

85
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Schematic representation
of time resolved
photoluminescence
arrangement.

Figure 4.5.53

600 nm with Fhn ≈ 50 Suns for For all oa-IV curves, a tunnelling feature for electron
each interval could be directly extraction can be seen at 760 mV forward bias.
compared to each other. In However, the background current density is much
addition, we evaluated the curves higher than the tunnelling portion of the current
with optical excitations in the density in this forward bias range, so no NDR can
range of 1000 – 500 nm (Fhn ≈ 600 be seen.
Suns) and 550 – 250 nm (Fhn ≈
At room temperature (T = 300 K), NDR was detected
140 Suns). The former is near the
only for λ ranges ≤ 600 nm, while the impact of
practical high concentration limit
resonant electron injection is still visible for curves
for solar cells of 1000 Suns.
in the 800 – 600 nm range. The background current
As shown in Fig. 4.5.52, for the density in the bias range around the tunnelling
curves in the 800 to 600 nm event increases by about an order of magnitude
region at LN2 temperature, the when going from LN2 to room temperature.
potential difference between Vbias(j This is due to thermal activation of trap states
→ 0) and the average voltage of and increased thermal scattering during carrier
the resonance (VRes,avg) shows a transport through the QD array. It has the same
decrease of 48 mV with increasing effect on NDR suppression as explained above for
photon energy (decreasing electron injection under forward bias.
λ range). As carrier densities
The QD arrays were processed by segregation
(quasi-Fermi level positions) are
anneals, comprising some 1010 QDs as ESCs
identical for these curves, this
under the contact pad. It is very encouraging to
potential drop corresponds to an
see resonant tunnelling transport even at room
increase in free hole energy of
temperature in a material system which was not
48 meV. As this energy difference
grown epitaxially. At LN2 temperature (77 K), an
is equal to the effective kT, it can
increase in carrier temperature of DT = 560 K was
oa-IV results of a typical sample be interpreted as an increase
detected when increasing the photon energy from
with optimum barrier thicknesses in carrier temperature of DT = 560 K when going
1.6 ± 0.05 eV to 1.985 ± 0.075 eV under a constant
and QD size at LN2 temperature from 800 – 750 nm (1.55 – 1.65 eV) to 650 – 600 nm
(upper) and at room temperature photon flux of Fhn ≈ 50 Suns. More suitable Hot
(1.91 – 2.06 eV). Even at Fhn ≈ 50 Suns, this effect
(lower). Curves in the wave length Carrier Absorber materials will be investigated.
is small. This is due to Si being a very unsuitable
intervals from 800 to 600 nm can
be compared due to their identical absorber material, further corroborated by the ratio
offset of Vbias to absolute zero Vbias., of peak to valley current densities (quality factor of 4.5.3.3 Hot Carrier Absorbers: slowing of
showing the free carrier product to NDR – QFNDR) dropping from 2.2 to 1.5. Si does not carrier cooling
be constant. Curves for 550 – 250
have a phononic band gap so that optical phonons
nm and 1000 to 500 nm show Researchers:
maximum optical injection limit. undergo a rapid decay which tremendously
Both curve groups printed in bold. accelerates carrier cooling. Robert Patterson, Gavin Conibeer, Santosh
Shreshtha, Dirk König, Pasquale Aliberti,
Figure 4.5.52 For the 1000 – 500 nm and 550 – 250 nm ranges,
Shujuan Huang, Yukio Kamikawa, Lara Treiber,
the QFNDR increases significantly to 4.1 and 3.2,
Martin Green
respectively. The potential difference Vbias(j → 0) –
VRes,avg is shifted by 365 and 200 mV, respectively;
a quantitative comparison of both curves to each Carrier cooling in a semiconductor proceeds
other or to the curves in the spectral range from predominantly by carriers scattering their energy
800 – 600 nm is thus not feasible. However, it shows with optical phonons. This builds up a non-
that the energy selectivity may increase under high equilibrium ‘hot’ population of optical phonons
concentration Fhn. which, if it remains hot, will drive a reverse reaction

86
to re-heat the carrier population, thus slowing was found that the limiting efficiency is strongly
further carrier cooling. Therefore the critical related to hot carriers relaxation velocity in the
ARC
factor is the mechanism by which these optical absorber [4.5.57].
phonons decay into acoustic phonons, or heat in PHOTOVOLTAICS
A comparison of femtosecond time resolved CENTRE OF
the lattice. The principal mechanism by which this
photoluminescence (tr-PL) spectroscopy between EXCELLENCE
can occur is the Klemens mechanism, in which the
InP and GaAs was reported in last year’s annual
optical phonon decays into two acoustic phonons 2010/11
report and is also now published [4.5.64]. This
of half its energy and of equal and opposite ANNUAL REPORT
showed a distinctly longer carrier cooling time
momenta [4.5.59].
constant for the wide phononic gap InP as
The build up of emitted optical phonons is compared to almost zero phononic gap of GaAs.
strongly peaked at zone centre both for compound It also showed further evidence for excitation
semiconductor due to the Frölich interaction into higher side valleys for both GaAs and InP for
and for elemental semiconducotrs due to the appropriate excitation wavelengths. In 2010 the
deformation potential interaction. The strong hot carrier cooling in InN has been investigated
coupling of the Frölich interaction also means that using tr-PL. The wide gap between optical and
high energy optical phonons are also constrained acoustic branches in the InN phononic dispersion
to near zone centre even if parabolicty of the bands relation (wider than that for InP) prevents the
is no longer valid [4.5.60]. This zone centre optical Klemens decay of optical phonon into acoustic
phonon population determines that the dominant phonons. This can lead to slower carrier cooling due
optical phonon decay mechanism is this pure to “Hot Phonon Effect” [4.5.65]. The decay of hot
Klemens decay. carriers for different excitation wavelengths InN has
been investigated.
4.5.3.3.1 Suppression of phonon decay in Tr-PL experiments have been performed
bulk materials on InN samples using the measurement
In some bulk semiconductors, with a large configuration shown in Fig. 4.5.53. In this
difference in their anion and cation masses, there technique a laser pulse acts as a switching
can be a large gap between the highest acoustic gate relating the photoluminescence
phonon energy and the lowest optical phonon signal to the time domain. The PL signal
energy, possibly large enough to block operation is collected from the sample, after a
of this Klemens mechanism, which can be termed femtosecond laser excitation, and focused
a ‘phononic band gap’. Work previously presented in a non linear crystal. The gate signal
[4.5.61] using a simple 1D force constant model and is generated from the same laser and is
complemented by high accuracy DFT computation focused on the same crystal after passing
[4.5.62] indicated that GaN, InN and InP all have through an optical delay stage. The signal
large phononic band gaps, which are close to those is detected using a monochromator and
found experimentally [4.5.63]. We are using time a PMT. Our system configuration provides
resolved photoluminescence (tr-PL) to investigate 150 fs pulses with tunable wavelength
3D representation of InN time
the carrier cooling rates in these materials. over a range of 256 nm (4.84 eV) to 2.6 µm (0.48 eV). resolved PL data.
Figure 4.5.54 is a three dimensional representation Figure 4.5.54
4.5.3.3.2 Time resolved
of PL as a function of time for all the probed
photoluminescence measurements of wavelengths. It can be observed that the PL sharply
bulk phononic band gap materials rises when the carriers are photo-excited by the
laser pulse. The fast decay of the PL shows the
Researchers thermalisation of carriers towards respective band
Pasquale Aliberti, Dirk König, Santosh edges. The decay is faster for highly energetic
Shrestha, Gavin Conibeer, Martin Green carriers compared to carriers closer to the bandgap.
Thus the carrier population quickly degenerates
Collaboration with: towards the band edges during the thermalisation
process. In InN the thermalisation is most probably
Raphael Clady, Murad Tayebjee, Tim Schmidt
due to interaction of highly energetic electrons and
(University of Sydney), Nicholas Ekins-Daukes
holes with LO phonons [4.5.66].
(Imperial College)
To investigate the rate of the carrier cooling
process, the effective temperature of the carrier
The potential efficiency boost, which can be
population has been calculated fitting the high
achieved by Hot Carrier solar cells, is directly
energy tail of the PL spectrum for every single
related to the possibility of extracting high
time during the cooling transient. The PL has been
energy carriers from the absorber layer before
fitted assuming that carriers form a Boltzmann-like
thermalisation, increasing the voltage and hence
distribution in a femtosecond time scale using the
the conversion efficiency. The poor conversion
following equation.
efficiency of photons with energies above the band
gap of the absorber is the main loss mechanism
in conventional single junction solar cells. The
investigation of thermalisation time constants of
hot carriers is a crucial step towards the engineering
of Hot Carrier cells. The efficiency of an InN based L(ε) represents the measured PL intensity at
hot carrier solar cell has been calculated using a energy ε, α(ε) is the measured sample absorption
complex theoretical model, see Section 4.5.3.1. I coefficient, EG is the InN energy gap, 0.7 eV in this

87
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Some images of the structure as grown and modelled.

(Left) the near-perfect stacking of the structure is
shown. (Centre) in-plane InAs QD arrangement.
(Right) simple hexagonal superlattice structure
showing stacking. The QDs are similar to flattened
disks with in plane dimensions of 4050 nm and
heights of 7 nm.

Figure 4.5.57

case, and kB is the Boltzmann constant. in the superlattice dispersion open up [4.5.61]. There
TC is the fitted parameter and represents is a close analogy with photonic structures in which
the hot carrier temperature. Figure 4.5.55 modulation of the refractive index in a periodic
shows the carrier temperature transient, system opens up gaps of disallowed photon
which seems to follow an exponential energies. Here modulation of the ease with which
behaviour quite well (blue line – fit). phonons are transmitted (the acoustic impedance)
opens up gaps of disallowed phonon energies.
The fitting of the calculated temperature
data has been performed using a
single exponential.
4.5.3.3.3.1 Force constant modelling of
III-V QD materials by SK growth

Researchers:
R. Patterson, Y. Kamikawa, G. Conibeer
Carrier relaxation curves for InN.
The photoexcited carrier density
is 1.5 x 1019 cm-3. The blue dashed
Here τTH represents the carriers thermalisation Collaboration with:
curve is a single exponential fit. time constant, whereas C and K are two constant Yoshitaka Okada – University of Tokyo
parameters. The fitted value for τTH is 7 ps. This
Figure 4.5.55
long cooling constant can be attributed to the hot
phonon effect due to the long lifetime of the A1(LO) 3D force constant modelling, using the reasonable
phonon [4.5.64,4.5.66]. This hot carrier relaxation assumption of simple harmonic motion of atoms
velocity is still too high to achieve a considerable in a matrix around their rest or lowest energy
efficiency gain in an InN hot carrier solar cell. position, reveals such phononic gaps [4.5.68].
However, it has been demonstrated that, for InN, The model calculates longitudinal and transverse
the carrier cooling velocity is strictly related to the modes and can be used to calculate dispersions in
quality of the material and slower carrier cooling a variety of symmetry directions and for different
constants compared to the ones calculated in this combinations of QD sub-lattice structure and super-
chapter have been reported in the literature [4.5.67]. lattice structure.

III-V Stranski-Krastinov grown QD arrays of InAs in

4.5.3.3.3 Nanostructures for absorbers InGaAs and InGaAlAs matrices are fabricated at the
Nanostructures offer the possibility of modification University of Tokyo using MBE. We are investigating
of the phonon dispersion of a composite material. these for evidence of phonon dispersion
III-V compounds or indeed most of the cubic and modulation and potential slowed carrier cooling.
hexagonal compounds can be considered as very In order to understand the expected phonon
fine nanostructures consisting of ‘quantum dots’ dispersions these are being modeeled using the 3D
of only one atom (say In) in a matrix (say N) with force constant technique.
only one atom separating each ‘QD’ and arranged
Lattice matched and strain compensated material
in two interpenetrating fcc lattices. Modelling of
pairs that may produce large phonon bandgaps are
the 1D phonon dispersion in this way gives a close
of interest. Previous iterations in the design of these
agreement with the phonon dispersion for zinc-
Schematic of the simple hexagonal structures indicated the importance of separating
blende InN extracted from real measured data for
superlattice of InAs QD disks, “light” and “heavy” atoms to different parts of the
wurtzite material.
showing their relative dimensions. nanostructure. Initially, the lightest atom in the
The red square indicates Similar ‘phonon band gaps’ should appear in system, As, was present in both the QD and the
orthorhombic repeat unit for the
good quality nanostructure superlattices, through matrix. This meant that the reduced mass of both
simple hexagonal system.
coherent Bragg reflection of modes such that gaps regions (proportional to the sum of the inverses
Figure 4.5.56

88
 Dispersion curves for InAs QDs in
a simple hexagonal SL. The matrix
is In0.5Ga0.5-xAlxAs with (x = 0.4).
There is a large phonon bandgap
present in the system due to the
presence of Al. The gap seems
very tolerant to Al location. (Far
left): <100>, (Centre): <110>, (Left):
<111> directions.

Figure 4.5.58

Densities of states (DOS) for the

dispersion curve immediately to the
left. Note small changes in the DOS
could be due to pseudo-random
locations of the Ga and Al particles
or to differences in crystalline
direction. (Far left): <100>,
(Centre): <110>, (Left): <111>
directions respectively.

Figure 4.5.59

of each atomic mass) was very similar as the light 4.5.3.3.3.2 Fabrication and
element dominates in this case. characterisation of highly ordered
On this iteration structures with an In0.5Ga0.5-xAlxAs nanoparticle arrays for Hot
matrix (with x=0.4) and InAs QDs were grown. Carrier absorber
Significant Al content was introduced into these
structures with the expectation that this light Researchers:
element, segregated to the matrix material, Lara Treiber, Shujuan Huang,
might produce appreciable phonon bandgaps. A Gavin Conibeer
schematic of the structure is shown in Fig. 4.5.56.
Some images derived from characterisation of
the structure are presented in Fig. 4.5.57. The As demonstrated by the work on
superlattice of QDs has a simple hexagonal modelling phonon modulation,
structure. Extraordinary periodic out-of-plane periodic core-shell QD arrays offer
stacking is achievable and largely defect free a way to significantly change the
structures can be grown on the order of microns. phonon modulation in a superlattice
Langmuir-Blodgett set-up for
because the core and shell can be of materials of
Force constant modelling of this structure predicts film fabrication with surface
very different force constant, directly leading to a functionalized Si NPs.
an appreciable phonon bandgap, as shown in
strong phonon confinement. Deposition methods
Fig. 4.5.58 and Fig. 4.5.59. This bandgap is due Figure 4.5.60
are being investigated to fabricate such highly
almost entirely to the presence of the Al in the
ordered QD arrays.
system. A small bandgap is present due to the
mass difference between In, Ga and As, but it is The Langmuir-Blodgett (LB) technique fabricates
less than a quarter of the size shown in Fig. 4.5.58 thin films from colloidal dispersions of quantum
and Fig. 4.5.59. Due to computational constraints dots (QDs). The LB technique allows for controllable,
the size of the QDs is quite small, only about a uniform film formation and subsequent transfer
nanometre in diameter. While actual sizes for these onto a solid substrate, such as quartz [4.5.69].
structures are too computationally intensive to Through surface passivation of the QDs,
model without extreme effort, recent modelling interparticle spacing can also be controlled. Figure
with gradually increasing size suggests that the 4.5.60 illustrates the set-up of the LB apparatus and
dispersion relations scale linearly with size. That the functionalized QDs.
is, once the discrete distances (bond lengths) are  
small relative to the superlattice unit cell dimension, In 2010 work has focused on fabrication of periodic
the dispersion should look exactly the same thin films of Si nanoparticles (NPs) provided by
when scaled such that the relative dimensions are the University of Minnesota. The Si NPs under
preserved. This will be investigated in greater detail investigation are surface passivated with dodecane
in further work. (C12H26), a long chain hydrocarbon rendering the
surface hydrophobic. The particles were dispersed
in chloroform and investigated using TEM, UV-vis
spectroscopy and photoluminescence (PL). Using
TEM imaging the average diameter was found to be
3.0 ±0.5 nm. Fig. 4.5.61 shows a droplet of solution
(dispersed in chloroform) and the inset highlights
the lattice fringes confirming the presence of silicon (Upper) TEM images of Si NPs dried
on a grid. (Lower) TEM image of
- a 3.14 Å lattice spacing corresponding to the
Si NP monolayer deposited at 50
{111} planes. mN/m of SP.

Figure 4.5.61

89
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11 40mN/m
60mN/m
ANNUAL REPORT

absorption (a.u)
65mN/m
collapse

0.5 1.5 2.5 3.5 4.5

Energy (eV)

(Left) Optical absorption of Si monolayers deposited at

various surface pressures (SP). (Right) PL of Si NP films
at varying SP, as compared to the Si NP solution.

Figure 4.5.62

4.5.3.3.3.3 Implementation of
  nanostructures for Hot Carrier cells
External contact
In order to use a nanoparticle array as a Hot Carrier
  Selective energy contact:
absorber, and hence utilise slowed carrier cooling
double (or triple) barrier QD resonant tunnelling
  Slowed carrier cooling absorber layer: due to modified phonon dispersion, then a way
§ Periodic QD array tuned to block optical phonon
decay by Klemens mechanism; to fabricate a complete structure with ESCs must
§ Core shell QD for ease of tuning;
§ Must also allow renormalization of carrier energy by be established. A structure in which nanoparticles
e-e & h-h scattering.
Proof of concept with MOVPE/MBE III-V QD array.
are arranged in a uniform 3D array should give the
Transfer to thin film or self-assembled QD array. required phononic band gap. The degree to which
  Selective energy contact:
selecting opposite carrier type this needs to be ordered is still under investigation.
External contact Certainly a large difference between masses is
beneficial between the nanoparticles and the
matrix and this can be best achieved in an array of
Conceptual Integrated Device Monolayer structures of Si NPs have been ‘core shell QDs’ in which the core and shell have very
Figure 4.5.63
successfully deposited using the LB method and different acoustic impedance in order to promote
characterisation has been on-going using TEM, UV- coherent reflection and hence confinement of
vis, PL and Raman spectroscopy. Figure 4.5.61 (right) phonons [4.5.61]. The matrix in which such an array
illustrates TEM image of the Si films fabricated at 50 is embedded needs to allow transport of carriers to
mN/m surface pressures (SP) of the water trough. contacts and also electron-electron and preferably
As noted, high contrast is difficult to achieve due to hole-hole scattering to renormalise carrier energies.
the small density difference of Si and the carbon- Such a structure should also have a narrow
passivating chain. As a result, to further quantify electronic band gap so as to absorb a wide range
the coverage and regularity, computer recognition of photon energies. This combination of properties
software is required; on which work is on-going. It is is challenging but not mutually exclusive because
difficult to see for this reason, but Fig. 4.5.61 (right) phononic properties are largely independent of
appears to show a dense packing of Si NPs. electronic properties. Energy selective contacts
UV-Vis spectroscopy was used to study the optical are also required for such a structure. These would
absorption of the films. As illustrated in Fig. most likely be arranged at the top and bottom of
4.5.62 (left) for single layers of Si NPs at varying the absorber. (A hole contact might not need to be
compression surface pressures (SP), the absorption selective because the hole population only contains
clearly increases when SP increases, corresponding a small fraction of the hot carrier energy and hence
to higher packing densities of Si NPs. The PL spectra thermalisation of holes is less important.) These
of the above films show narrow and symmetric ESCs would be QD or QW double barrier structures.
peaks at 1.6 ~1.7 eV, as shown in Fig. 4.5.62 (right), (Addition of an extra layer to give double QD/QW
which are in good agreement with the estimated triple barrier structures should also give rectification
bandgap using the ‘effective mass approximation’. at the contact.) Such a QD array / double barrier QD
The red shift in emission energy from Si NP films ESC structure is shown schematically in Fig. 4.5.63.
as compared to the solution corresponds to the
electronic energy transfer from small particles to
larger ones when they are closely packed [4.5.70].

Investigation of the electronic and phononic

properties of the monolayer and multilayer
structures are on-going using low-wavenumber
Raman and time resolve PL spectroscopy.

90
Physical picture of the ND HC Solar
Cell (upper) with a carrier diffusion
ARC
field (core/shell thickness ratio PHOTOVOLTAICS
not to scale), embedded into a CENTRE OF
wide band gap solar cell [4.5.71].
Iso-valent impurities provide an
EXCELLENCE
energy selective level for electrons. 2010/11
Band diagram of the ND HC Solar ANNUAL REPORT
Cell matched to the physical
structure (lower). Direct generation
of hot exciton (1), energy selective
extraction of hot electron from Hot
Carrier absorber (2), thermionic
emission/diffusion of holes over
the barrier (3), electron-electron-
scattering with renormalization
of carrier energies and possible
energy loss by phonon emission
(4), free carrier re-absorption (5),
impact ionization (6) and carrier
generation in the wide band gap
matrix (7).

Figure 4.5.64

Nanoparticle cell design incorporating 4.5.3.3.4 Hot Carrier Absorber:

ESCs and absorber Choice of Materials

Researcher: Researchers:
Dirk König G. Conibeer, R. Patterson, P. Aliberti, S. Huang,
Y. Kamikawa, D. König, Binesh Puthen-Veettil,
S.Shrestha, M.A. Green
A full Hot Carrier cell requires control of optical,
electronic and phononic properties in the same
structure. One possible approach is the ‘nano-dot The properties required for a good hot carrier
(ND) Hot Carrier cell’ embedded into the i-region of absorber material are listed below in order
a wide band gap solar cell as shown in Fig. 4.5.64. of priority.

Hot Carrier solar cells must have a continuous 1. Large phononic band gap (EO(min) - ELA) - in
electronic density of states (DOS) to maximise order to suppress Klemens decay of optical
maximum photon absorption and elastic electron- phonons this must be at least as large as the
maximum acoustic phonon energy. Hence a
electron scattering. The latter refills the depleted
large mass difference (or large force constant
electronic energy levels from which electrons are
difference) between constituent elements is
extracted through the ESC. On the other hand, if required [4.5.61].
they do not have a phononic band gap themselves
2. Narrow optical phonon energy dispersion (ELO -
the NDs must not be much bigger than the ballistic EO(min)) – in order to minimise the loss of energy
mean free path in order to prevent inelastic to TO phonons by Ridley decay. This requires
scattering by which carriers cool down rapidly. a high symmetry atomic or nano-structure,
However, optical phonons can be confined in the preferably cubic with degenerate optical
nanocrystals as mentioned in Section 4.5.3.3.3 phonon energies at zone centre.
by arranging for a mismatch between optical 3. Small electronic band gap – to allow a wide
phonon energies in the two materials. Core shell range of photon absorption. This should be less
nanocrystals can be used to realise this [4.5.61, than 1eV. For 1 sun concentration the optimum
4.5.62]. In addition the shell can be doped with is 0.7eV - as a band gap below this energy
gives no advantage in the balance between
iso-valent impurities which form split-off subbands
absorption and emission. As the concentration
within the electronic band gap of the material,
ratio increases this optimum band gap decreases
such as boron replacing gallium in gallium nitride to zero at maximum concentration [4.5.1].
(GaN) [4.5.71]. Such an ESC would only work for
4. A small LO optical phonon energy (ELO). This
one carrier type as co-doping of a thin nano-shell reduces the amount of energy lost per LO
appears not to be technologically feasible. For some phonon emission, requiring a greater number
III-V compounds with a very high effective mass to be emitted for a given energy loss [4.5.62].
ratio between holes and electrons such that most of However, it is difficult to have both a small ELO
the excess energy during optical generation is taken and the large phononic gap required in 1. This
up by the electron, thermalisation of holes does not condition requires further investigation, but
represent a large loss in energy. Hence thermionic does still argue for a small ELO.
emission of holes over a low barrier can be used 5. An absolute small maximum acoustic phonon
to transport holes into the bulk of the device to be energy (ELA). This maximises the phononic gap if
ELO is also small. A small ELA requires a large mass
collected by the junction with little loss in efficiency.
for the heavy atom and/or small force constant
for its bonds.
6. Good renormalisation rates in the material,
i.e. good e-e and h-h scattering (e=electron,

91
 IIA IIIA IB IIB IIIB IVB VB VIB
ARC
PHOTOVOLTAICS
Be B C N O
CENTRE OF
Mg Al Si P S
EXCELLENCE
2010/11
Ca Sc Cu Zn Ga Ge As Se
ANNUAL REPORT
Sr Y Ag Cd In Sn Sb Te

Ba La Au Hg Tl Pb Bi Po

Er

Use of the periodic table to analyse possible analogue compounds of InN

based on atomic mass combination and electro-negativity.

Figure 4.5.65

h=hole). This requires a reasonable DOS at II-IV compounds is analogous and is now quite 1D models. However being all group IVs they
all energies above Eg, or at least only very widely being investigated in the Cu2ZnSnS4 only form weak compounds. Unfortunately
narrow gaps between energy levels, ≤ analogue to CuInS2 [4.5.72]. SiC, whether 3C, 4H or 6H, has too narrow
kT300K. This in turn requires a good overlap a phononic gap. Nonetheless GeC does
of wavefunction for carriers through the ZnGeN can be fabricated [4.5.73] and is most
form a compound and is of significant
material, i.e. poor electronic confinement directly analogous with Si and GaAs. However,
interest [4.5.76].
or a reasonable conductor. This condition is its band gap is large at 1.9eV. It also has a small
met in all inorganic semiconducotrs quite calculated phononic band gap [4.5.74]. ZnSnN There are also several other inherent
easily, with e-e scattering rates of less than has a smaller electronic gap (1 eV) and larger advantages of group IV compounds/alloys all
100fs for reasonable carrier concentrations. calculated phononic gap [4.5.74]. It is however of which are associated with the four valence
It may not be met in organic
difficult to fabricate, and also its phononic gap electrons of the group IVs which result in
semiconductors or in nanostructures with
is not as large as the acoustic phonon energy predominantly covalent bonding:
large barrier heights.
making it difficult to block Klemens decay
7. Good carrier transport in order to allow a) The elements form completely covalently
transport of hot carriers to the contacts. completely. HgSnN or HgPbN should both bonded crystals primarily in a diamond
This is similar to 6, except that it only have smaller Eg and larger phononic gaps. structure (tetragonal is also possible as in
need be in the direction of the contacts, These materials have not yet been fabricated βSn). However for group IV compounds
probably the z growth direction. A [4.5.75]. the decreasing electronegativity down the
reasonably low resistance is probably good. group results in partially ionic bonding.
8. It should be possible to make good quality, Large mass cation: This is not strong in SiC and whilst it tends
highly ordered, low defect material. to give co-ordination numbers of 4, can
The Bi and Sb compounds have large
Preferably it should be easy and cheap to nonetheless result in several allotropes of
predicted phononic gaps and Bi is a relatively decreasing symmetry: 4c, 4h, 6h. However,
do this.
abundant material, with only low toxicity as the difference in period increases for the
9. Earth abundant and readily available and
[4.5.75]. BiB has the largest phononic gap as yet theoretical group IV compounds, so
processable constituent elements and
processes. but AlBi, Bi2S5, Bi2O3 (bismuthine) are also too does the difference in electronegativity
attractive. Similarly SbB has a large predicted and hence also the bond ionicity and the
10. No or low toxicity of elements, compounds
phononic gap. That for AlSb is the same size degree of order. For a hot carrier absorber
and processes.
as the acoustic phonon energy and its band this is ideal because it is just such a large
[(ELO , EO(min) & ELA are the maximum optical,
gap is 1.5eV, making it marginal as an absorber difference in the period which is needed
minimum optical and maximum acoustic
to give the large mass difference and
phonon energies respectively.] material and similar to InP.
hence large phononic gaps. All of GeC,
SnSi, SnC (and the Pb compounds) have
InN has most of these properties, except 4, 8 & Group IIIA III-Vs computed phononic gaps large enough
9, and is therefore a good model material for a LaN and YN both have large phononic gaps to block Klemens decay, and should also
hot carrier cell absorber. whilst that for ScN is too small. tend to form ordered diamond structure
compounds.
The Lanthanides can also form III-Vs. ErN and b) Because of their covalent bonding, the
Analogues of InN
other RE nitrides can be grown by MBE. The group IV elements have relatively small
As InN is a model material, but has the phononic band gaps of the Er compounds are electronic band gaps as compared to their
problems of abundance and bad material predicted to be large, because of the heavy more ionic III-V and much more ionic II-VI
quality, another approach is to use analogues Er cation, but its discrete energy levels make analogues in the same period: e.g. Sn
of InN to attempt to emulate its near ideal it not useful as an absorber, although the 0.15eV, InSb 0.4eV, CdTe 1.5eV. In fact to
properties. These analogues can be II-IV-nitride combination of properties in a nanostructure achieve approximately the same electronic
compounds, large mass anion III-Vs, group IV could be advantageous. band gap one must go down one period
compounds/alloys or nanostructures. from group IV to III-V and down another
period from III-V to II-VI: e.g. Si 1.1eV, GaAs
II-IV-Vs: ZnSnN; ZnPbN; HgSnN; HgPbN Group IV alloy/compounds: 1.45eV, CdTe 1.5eV. This means that for
With reference to Fig 4.5.65, it can be seen that All of the combinations Si/Sn, Ge/C or Sn/C group IV compounds there is greater scope
replacement of In on the III sub-lattice with look attractive with large gaps predicted in for large mass difference compounds

92
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Cross-sectional electron micrograph of a 30-bilayer

porous silicon distributed Bragg reflector.

Figure 4.5.66

whilst still maintaining small electronic 4.5.3.3.5 Summary of Hot Carrier 4.5.4 Up-conversion
band gaps. A small band gap of course solar cell research
being important for broadband absorption Researchers:
in an absorber - property 3 in the desirable 2010 has seen significant development in
properties for hot carrier absorbers listed most areas of Hot Carrier solar cell work. Craig Johnson, Gavin Conibeer
above. The modelling of efficiencies not only now
c) The smaller Eg would tend to be for the includes real material parameters for highly Collaboration with:
larger mass compounds of Pb or Sn. Which, promising absorber materials such as InN, Peter Reece (Physics, UNSW)
to give large mass difference, would be but also is now extended to Energy Selective
compounded with Si or Ge. This trend contacts of finite width. The transport across
towards the lower periods of group IV also such contacts is further modelled for a range
means that the maximum optical phonon Up-conversion in novel silicon-based
of QD matrix combinations. Work on ESCs has
and maximum acoustic phonon energies seen further more detailed demonstration
materials
will be smaller for a given mass ratio - the Up-conversion (UC) in erbium-doped
of the necessary resonance in double barrier
desirable properties 4 and 5.
resonant tunnelling structures, with additional phosphor compounds (particularly NaYF4:Er)
d) Furthermore, unlike most groups, the has been shown to be a promising means
evidence for hot carrier populations, albeit
group IV elements remain abundant for the
very small, from illuminated I-V measurements of enhancing the sub-band-gap spectral
higher mass number elements – desirable
with the ‘optically assisted I-V’ technique. response of conventional Si solar cells without
property 9. Property 10 is also satisfied
because the group IVs have low toxicity. Measurement of InN with time-resolved PL modification of the electrical properties of the
has indicated some evidence for slowed cell [4.5.77]. In this scheme, a layer containing
Nanostructures: carrier cooling, further corroborating the the phosphor is applied to the rear of a high-
As discussed in section 4.5.3.3.3 QD importance of a large phonon band gap efficiency bifacial cell. After absorbing two
nanostructures can be viewed in the same to block optical phonon decay, but also long-wavelength (~1500nm) photons - which
way as compounds. Their phononic properties highlights the importance of material are transmitted by the cell - the excited Er
can be estimated from consideration of their quality. Modelling of nanostructures for ions can relax by emitting a photon with an
combination force constants. Hence it is absorber materials has focussed on real III-V energy greater than the Si band gap, thereby
possible to ‘engineer’ phononic properties in QD structures, showing phonon bandgaps increasing the current that can be extracted
a wider range of nanostructure combinations. which will soon be measured in phonon from the cell.
Of the materials discussed above the Group IVs dispersion measurements. Progress on the
While phosphors have demonstrated
lend themselves most readily to formation of Langmuir-Blodgett approach to ordered
high-efficiency UC behaviour, their use
nanostructures instead of compounds due to nanoparticle arrays has seen development of
presents particular challenges with regard
their predominantly covalent bonding, which ordered single layer arrays of Si nanoparticles.
to fabrication and cost. Our work in the last
allows variation in the coordination number. The potential application of nanostructures
year has focused on the development of
Therefore the nanostructure approaches of to fully integrated devices has started to
Er-doped porous Si (PSi:Er) as an alternative
section 4.5.3.3.3 are consistent with a similar be investigated conceptually, with various
UC material. PSi is unique in that its porosity -
description as analogues of InN, whether it be designs considered. Similarly the possibility
and hence the material refractive index - can
III-V QDs, colloidally dispersed QDs or for core of absorber materials which are analogous
be varied as a function of depth, allowing for
shell QDs. to InN is also being investigated. These many
the elaboration of high-quality monolithic Si
aspects of Hot Carrier cells will see further
optical structures such as distributed Bragg
development and consolidation in 2011
reflectors (DBRs). A cross-sectional electron
with recent success in significant additional
micrograph of such a structure is shown in Fig.
funding.
4.5.66. Its porous substructure also allows for
deep infusion of dopant atoms via techniques
such as electroplating.

93
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Simulated reflectivity spectra of

PSi DBRs with varying numbers
of bilayers; layer thicknesses
were adjusted to fix the short-
wavelength stop band edge
at 1550nm (marked with red
dashed vertical line).

Figure 4.5.68

changes abruptly between transmission and

reflection of incident photons as determined by
the superposition of Bloch waves. Insofar as the
group velocity vg of an incident wave can be said
to transition between “positive” and “negative”
values - implying energy propagation into and out
of the structure, respectively, in bands of strong
transmission and strong reflection - there exists an
inversion point at which vg is zero, corresponding to
a standing wave established by the superposition of
scattered Bloch components. As can be seen in the
right panel of Fig. 4.5.67, group velocity is drastically
suppressed across a broad region near the zero-
point. For steady-state equilibrium conditions,
Band structure in the neighborhood We have previously reported room-temperature UC
this slowing of energy propagation requires a
of the fundamental band gap of luminescence from Er-doped nanoporous Si when
an infinite 1D photonic crystal proportional increase in energy density, that is,
excited at very high power densities with a 1550nm
(left) and group velocity in the an enhancement of the electromagnetic field
material (right). laser. In the past year we have focused on increasing
strength within the structure. It can be shown that
the efficiency of this UC luminescence by modifying
Figure 4.5.67 the efficiency increase for a two-step UC process is
the PSi structure as described above.
proportional to (c/nvg – 1)I0, where c is the speed of
The dependence of the efficiency of the UC process light in vacuum, n is the average refractive index of
on irradiance is non-linear in the range of “sun- the multilayer material, and I0 is the incident field
like” incident power. That is, the concentration intensity [4.5.78]. In this way the slow-light mode
of the incident radiation into a small Er-doped acts to augment the interaction that converts
region results in greater efficiency than for non- energy from the field into atomic potential energy,
concentrated radiation over a proportionally large- resulting in a boost in the efficiency of the process
area device. Typically, concentration is achieved until saturation is reached.
using lenses or non-imaging optics, but we have
examined an analogous electromagnetic field Simulation of field enhancement in 1D
enhancement in the vicinity of optically-active Er PC structures
ions by the excitation of slow-light modes in these Using a 1D transfer matrix calculation, we simulated
PSi:Er DBRs. the reflectivity characteristics of a series of PSi DBRs
with 10, 20, 30 and 40 high-/low-porosity bilayers.
Field enhancement in 1D PSi:Er As shown in Fig. 4.5.68, the reflectivity characteristic
photonic crystals of a DBR contains a wide photonic stop-band.
Though their use preceded the advent of the The structural parameters were tuned slightly in
photonic crystal concept, DBRs can be described each case as the number of bilayers increased to
as one-dimensional photonic crystals. The regular produce a short-wavelength band edge minimum
variation of the optical thickness of multiple at 1550nm, the excitation wavelength of interest
alternating layers results in wavelength-dependent (marked with a dashed red line).
interference behaviour manifesting as discrete
Maximum transmission of ~1550nm light into the
bands of allowed and disallowed electromagnetic
structure is clearly required for efficient coupling
modes, as shown in the left panel of Fig. 4.5.67. The
into the Er ions. From Fig. 4.5.68 it is evident that
“edge” of a photonic band is a spectral region in
an increased number of layers “compresses” the
which the effect of coherent scattering processes
reflectivity characteristic, resulting in a steeper

94
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
Intensity-valued mode profile in DBRs
with varying numbers of high/low-
ANNUAL REPORT
porosity bilayers for a leading band edge
fixed at ~1550nm. All intensities are
normalized to the incident field intensity
and each plot is independently color-
scaled to show maximum detail. The air
and substrate interfaces with the DBR are
indicated by the dashed black (upper)
and red (lower) lines, respectively.

Figure 4.5.69

band edge and a narrower transmission window 4.5.5 Plasmonics for 3rd
between the stop band and the neighboring Generation structures
interference fringe.

Figure 4.5.69 shows the calculated field intensity

Researcher:
profile for these DBRs in the spectral region Supriya Pillai
between 1500 and 1600nm. The results have
been normalized to the incident field intensity
Plasmonics is the study of the interaction of light
and each plot is independently colour-scaled to
on a thin metal dielectric interface resulting in
show maximum detail. They clearly demonstrate
the collective oscillation of the free electrons in
that considerable field enhancement is possible
the metal. This gives rise to exciting properties
throughout the depth of such a multilayer structure,
that are very different from the bulk metal. It is
with peak relative intensities of more than 11 for
an emerging area of technology for photovoltaic
the 30-bilayer structure and more than 18 for the
applications with potential for light management
40-bilayer structure. However, it is also clear that the
at the nanoscale. Ever since our proof-of
regions of enhancement are increasingly narrowly-
concept results were published in 2007 [4.5.79],
concentrated as the number of bilayers increases.
there has been a huge increase in interest in
This is due to the additional interference fringes
incorporating nanoparticles on solar cells due
in the reflectivity characteristic of the DBR that
to their demonstrated light trapping properties.
arise with additional layers, as is apparent from Fig.
The large optical polarisation associated with the
4.5.69. Calculations for more than 40 layers show
nanoparticles results in a strong enhancement of
that enhancements of up to 80 are possible for a
the electric field around the vicinity of the particles
large number of bilayers though the peak narrows
along with strong scattering. This optical property is
to less than 1nm wide for 100 bilayers.
being utilised for improving light trapping in thin-
film solar cells.
Conclusion
The scattering or near field enhancement is
These simulations show that drastic field intensity
strongest at the surface plasmon resonance (SPR)
enhancement is in principle achievable in Er-doped
frequency and this can be tuned by changing the
PSi DBRs. Our current work is focused on the
size, shape, dielectric medium around the particles.
fabrication of suitable samples by anodic chemical
SPR tunability offers control in manipulating light
etching, electroplating, and high-temperature
in the regime where the absorption of the solar
annealing. Angular-dependent room temperature
cells are weak. Our recent work (see section 4.4
photoluminescence studies are underway to
Thin Film) has shown the benefits of having the
probe the effect of the photonic band edge on the
particles on the rear as opposed to previous work
efficiency of UC in the optically-active Er ions within
with front located particles [4.5.80]. By having the
the structures. This Si-based approach to efficient
particles on the back of the cell any potential loss
up-conversion is conceptually promising and by
due to absorption in the metal or losses due to
optimising our fabrication techniques to improve
sub-resonance photocurrent suppression resulting
the quality of our optical structures we expect it will
from destructive interference of the scattered and
prove to be valuable work.
transmitted light into the semiconductor layer
can be avoided [4.5.81]. An additional advantage
is that because the scattering plasmonics layer
is electrically decoupled from the cell, the
antireflection layer on the front and the plasmons

95
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE Increasing order of wavelength
2010/11
ANNUAL REPORT

Decreasing order of bandgap energy

Increasing order of particle size

of the semiconductor
Schematic representation showing
an example of surface plasmon
resonance tunability by varying the
of size of the metal nanoparticles
for wavelength dependent light
trapping in a tandem solar cell.

Figure 4.5.70

can be optimised independently of each other. Our particularly noted for particles corresponding to
recent work has also made use of an additional thicker Ag layers as can be seen from the PL plots in
reflecting layer (paint) along with the nanoparticle Fig. 4.5.71. The red-shift of the enhancement peaks
layer on the rear side of the cell to maximise the with larger particles are also clearly evident. More
randomisation of light and to ensure no light is details of this work has been reported in [4.5.83].
lost [4.5.80].
Rear located nanoparticles perform better than
In this study we focus our interest on the use of front located nanoparticles, consistent with the
plasmonics for low-dimensional structures like results from thin-film cells. We believe this is
quantum dot structures for possible applications because of the change in scattering cross-section of
in tandem solar cells. PL measurements were the nanoparticles due to the changes in the driving
used to characterise the samples. The analysis field and also due to the change in the mode of
is based on the reciprocity of light that a good excitation (from air for front located nanoparticles
absorber can be a good emitter and the proof- and from Si for rear located nanoparticles) for
of concept established with EL measurements the two different locations on the samples. More
in an earlier study [4.5.82]. Hence the PL work need to be done to better understand the
enhancement would be synonymous to an mechanism. Further work will also concentrate
increase in photocurrent for an optimised QD cell. on using alternate approaches for tuning the
The tunability of plasmon resonance allows the resonance position close to the band-edge
possibility of increasing the photocurrent of each (emission wavelength for PL) to increase the
cell in a tandem cell configuration. Optimising scattering properties and hence absorption in the
the nanoparticle parameters can help achieve weakly absorbing range.
wavelength dependent light trapping which will
be an encouraging step towards increasing the 4.5.6 Concluding remarks for the Third
performance of a tandem solar cell. Figure 4.5.70 Generation section
shows one way of achieving the SPR tunability by
In 2010 work has proceeded significantly in all the
varying the size of the nanoparticles in a tandem
areas of Third Generation research, with improved
cell configuration.
fabrication and characterisation of materials and
The sample under study is a 4nm single layer silicon complexity of modelling which together give an
QD structure in an oxide matrix with a 6nm capping overall better understanding and optimisation
oxide layer with an emission wavelength of 930nm of devices.
(~1.3eV). Mass thicknesses of 10, 14, 18 and 22 nm
Group IV based nanostructure materials have seen
silver were deposited and the particles annealed
significant improvement in device design. With
at 200°C in nitrogen for an hour. A 532nm, 10mW
interlayers, in both SiNx and SiC matrices, used to
Nd:YAG laser was used to illuminate the samples
effectively control both the uniformity of size of Si
for both the front (incident on QD layer) and rear
QDs and carrier transport in the growth direction.
side (incident on the quartz slide) locations of the
Modelling now allows the confined energy levels
silver nanoparticles and the PL plots studied. PL
of complex QD shapes to be calculated and their
measurements were carried out on these samples
overlap to form mini-bands estimated in more
and the effect of different size particles were
detail. At the same tiome improved characterisation
investigated both before and after the deposition of
is giving a much clearer idea on the shape and
the nanoparticles.
distributions of Si QDs and is now allowing choice
The self assembly techniques of nanoparticle of growth parameters to give uniform size spherical
fabrication give a varied size and shape distribution QDs to be grown. Ge nanostructure materials have
very suitable for a broadband response. This is seen improved control of p-type conductivity in

96
 7
no MNP (a) Front 1.0
10nm Ag
6 14nm Ag 0.9

5
18nm Ag 0.8 ARC

PL enhancement
22nm Ag
0.7
PHOTOVOLTAICS

PL (a.u)
4 0.6

3
0.5 CENTRE OF
0.4
2 0.3
EXCELLENCE
1
0.2 2010/11
0.1
0 0.0
ANNUAL REPORT
600 700 800 900 1000
Wavelength (nm)
3.2
20 no MNP PL plots from a 4nm single
(b) Rear
18 Ag reflector 2.8
10nm Ag layer QD structure in an oxide
16 14nmAg
2.4 matrix with a 6nm capping
18nm Ag
PL Enhancement

14 22nm Ag oxide layer (right axis) along

2.0 with the corresponding
12

PL (a.u)
1.6 enhancement plots (left axis)
10
with silver nanoparticles
8 1.2 corresponding to 10, 14, 18
6 and 22 nm mass thickness of
0.8
4 silver for (a) silver nanoparticles
0.4 on the front and (b) silver
2
0 0.0 nanoparticles on the rear.
600 700 800 900 1000
Wavelength (nm) Figure 4.5.71

Ge QDs and very highly crystalline growth of Ge thus tackling one of the key problems of rare earth
QWs. Homojunction Si nanostructure devices up-conversion that it is inherently non-linear and
have demonstrated improved conduction, there dependent on concentration.
has been continued demonstration of rectifying
Plasmonics, whic has previously been shown to
properties for heterojunctions with c-Si and
enhance emission and absorption from both 1st
heterojunctions between two nanostructured
and 2nd generation cells, has now been applied
materials are now starting to exploit the advantages
to the Si annostructured layers of 3rd generatrion
of both.
materials. This has shown significant enhancement
Hot carrier cells have seen further development in luminescence demonstrating the strong potential
of efficiency modelling which now not only to achieve local concentration with a global 1 sun
includes material parameters such as Auger illumination. This is particularly useful for several of
coefficients for real promising absorber materials the non-linear 3rd generation approaches.
such as InN, but also includes the effects of real
The development of all the 3rd generation projects
non-ideal energy selective contacts. There has
in 2010 now allows much greater understanding
been significant improvement in measurement
of the materials and devices. Work in 2011 will see
of negative differential resistance in Si QD based
consolidation of this into improved devices. Several
energy selective contacts is now allowing and this
new areas of funding will contribute to this and also
overlaps well with continuing improvement in
allow development of new project areas.
modelling the transport through such structures
in various matrices. Carrier cooling measurements
on the candidate absorber material InN have References:
shown evidence for phonon bandgaps building on 4.5.1 M. A. Green, “Third Generation Photovoltaics:
Ultra-High Efficiency at Low Cost” (Springer-
the slowed carrier cooling seen in InP previously. Verlag, 2003).
Modelling of analogues of these materials in
4.5.2 J. Nelson, “The Physics of Solar Cells”, Imperial
nanostructures now predicts the phonon properties College press, 2003.
of rea lIII-V QD arrays grown by collaborators. In 4.5.3 G. Conibeer, “Third Generation Photovoltaics”,
addition colloidally dispersed Si nanoparticles Materials Today, 10 (11) (2007) 42-50.
have now been fabricated as coherent single layer 4.5.4 E-C. Cho, S. Park, X. Hao, D. Song, G. Conibeer,
nanoparticle arrays and are progressing towards S-C. Park, M.A. Green, “Silicon quantum dot/
multi-layer structures. Theoretical work on the crystalline silicon solar cells”, Nanotechnology, 19
(2008) 245201
potential materials for absorbers with controlled
phon dispersion and hence slowed carrier cooling 4.5.5 Hao X J, Podhorodecki A, Shen Y S, Zatryb G,
Misiewicz J, Green M A, “Effects of Si-rich oxide
has progressed on approapratie analogues for
layer stoichiometry on the structural and
InN and devices involving various nanostructure optical properties of Si QDs/SiO2 multilayer film”,
configurations are being analysed conceptually. Nanotechnology 20, 485703 (2009).
4.5.6 M.A. Green, G. Conibeer, D. König, I. Perez-Wurfl,
Up-conversion has seen a further significant
A. Gentle, S. Huang, D. Song, X.J. Hao, F. Gao, S.
development of the use of porous-Si as a host for Er Park, C. Flynn, Y. So, B. Zhang, D. Di, P. Campbell,
up converting species. The porosity of porous-Si can Y. Huang and T. Puzzer, “Progress with All-Silicon
be controlled to give layers of alternating refractive Tandem Cells On Glass Using Silicon Quantum
Dots in Silicon-Based Dielectric Matrices”, 24th
index which can be used as distributed Bragg
European Photovoltaic Solar Energy Conference,
reflectors to modifiy the photonic density of states Hamburg, Germany, 21-25 September 2009.
such that there is strong enhancement of photonic
4.5.7 M. Zacharias, J. Heitmann, R. Scholz, U. Kahler, M.
modes at the critical Er absorption at 1500nm. Schmidt, Appl. Phys. Lett. 80(2002) 661.
This has been demonstrated and can lead directly 4.5.8 X.J. Hao, I. Perez-Wurfl, G. Conibeer, M.A. Green,
to large increase in the up-conversion efficiency, 19th PVSEC, Korea, Nov 2009.

97
 4.5.9 I. Perez-Wurfl, X. Hao, A. Gentle, D. Kim, G. nanocrystal precipitation in SiC matrix”, Thin
Conibeer, M.A. Green, “Si nanocrystal p-i-n Solid Films 516 (2008) 3824–3830, doi:10.1016/j.
ARC diodes fabricated on quartz substrates for third tsf.2007.06.150
PHOTOVOLTAICS generation solar cell applications,”Applied Physics 4.5.29 D. Song, E-C. Cho, G. Conibeer, Y. Huang, C. Flynn,
Letters, 95 (2009) 153506. MA. Green, “Structural characterization of annealed
CENTRE OF
4.5.10 X.J. Hao, E-C. Cho, G. Scardera, Y.S. Shen, E. Si1-xCx/SiC multilayers targeting formation of Si
EXCELLENCE Bellet-Amalric, D. Bellet, G. Conibeer, M.A. Green, nanocrystals in a SiC matrix”, Journal of Applied
2010/11 “Phosphorus-doped silicon quantum dots for Physics, 103 (2008) 083544.
ANNUAL REPORT all-silicon quantum dot tandem solar cells”, Solar 4.5.30 Z. Wan, S. Huang, M.A. Green, G. Conibeer, Physica
Energy Materials and Solar Cells, 93 (2009) 1524. Status Solidi C, “Residual stress study of silicon
4.5.11 G. Conibeer, M.A. Green, R. Corkish, Y.-H. Cho, E.-C. quantum dot in silicon carbide matrix by Raman
Cho, C.-W. Jiang, T. Fangsuwannarak, E. Pink, Y. measurement”, EXCON ‘10 Proceedings special
Huang, T. Puzzer, T. Trupke, B. Richards, A. Shalav, issue, accepted September 2010.
K.-L. Lin, Thin Solid Films 511/512 (2006) 654. 4.5.31 Zhenyu Wan, Shujuan Huang, Martin Green
4.5.12 E.-C. Cho, Y.H. Cho, T. Trupke, R. Corkish, G. and Gavin Conibeer, Rapid thermal annealing
Conibeer and M.A. Green, Proc. 19th European and crystallization mechanisms study of silicon
Photovoltaic Solar Energy Conference, Paris, (June nanocrystal in silicon carbide matrix, Nanoscale
2004), p. 235. Research Letters, 7541289134788882, accepted
4.5.13 F. Meillaud, A. Shah, C. Droz, E. Vallat-Sauvain, January 2011.
C. Miazza, Sol. Energy Mater. Sol. Cells 90, 4.5.32 F. Gao, M.A. Green, G. Conibeer, E-C. Cho, Y. Huang,
29522 (2006). I. Perez-Wurfl, C. Flynn, “Fabrication of multilayered
4.5.14 T. Arguirov, T. Mchedlidze, M. Kittler, R. Rölver, B. Ge nanocrystals by magnetron sputtering and
Berghoff, M. Först, B. Spangenberg, Appl. Phys. annealing”, Nanotechnology, 19 (2008) 455611.
Lett. 89 (2006) 053111. 4.5.33 Richter et al., Solid State Commun., 39 (1981), 625;
4.5.15 T.-W. Kim, C.-H. Cho, B.-H. Kim, S.-J. Park, Appl. Phys. I.H. Campbell, P.M. Fauchet, Solid State Commun.
Lett. 88 (2006) 123102. 58 (1986) 739; Y. Sasaki, C. Horie, Phys. Rev. B 47
(1993) 3811.
4.5.16 Y.-H. Cho, E.-C. Cho, Y. Huang, C.-W. Jiang, G.
Conibeer, M.A. Green, Proceeding of 20th 4.5.34 G. Conibeer, M. Green, E-C. Cho, D. König, Y-H. Cho,
European Photovoltaic Solar Energy Conference, T. Fangsuwannarak, G. Scardera, E. Pink, Y. Huang, T.
Barcelona, Spain, June 6-10, 2005, p.47. Puzzer, S. Huang, D. Song, C. Flynn, S. Park, X. Hao,
D. Mansfield, “Silicon quantum dot nanostructures
4.5.17 Y. Kurokawa, S. Miyajima, A. Yamada and M. for tandem photovoltaic cells”, Thin Solid Films, 516
Konagai, Jpn. J. Appl. Phys. 45, L1064 (2006). (July 2008) 6748.
4.5.18 C.-W. Jiang, M.A. Green, J. Appl. Phys. 99 4.5.35 B. Zhang, S.K. Shrestha, P. Aliberti, M.A. Green, G.J.
(2006) 114902. Conibeer, Characterisation of size-controlled and
4.5.19 K. Boer, “Survey of Semiconductor Physics”, van red luminescent Ge nanocrystals in multilayered
Nostrand Reinhold, 1990. superlattice structure, Thin Solid Films 518 (2010)
4.5.20 Aliberti, P., et al., “Study of silicon quantum dots 5483-5487.
in a SiO2 matrix for energy selective contacts 4.5.36 B. Zhang, S.K. Shrestha, M.A. Green, G.J. Conibeer,
applications”, Solar Energy Materials and Solar “Size controlled synthesis of Ge nanocrystals in
Cells. 94(11): p. 1936-1941 (2010). SiO2 at temperatures below 400 degree C using
4.5.21 E.A.B.Cole, “Mathematical and Numerical magnetron sputtering”, Applied Physics Letters, 96,
Modelling of Heterostructure Semiconductor 26901 (2010).
Devices: From Theory to Programming”. 2009: 4.5.37 Y. Sasaki, C. Horie, Phys. Rev. B 47 (1993) 3811;
Springer. 406. Kartopu et al., J. Appl. Phys. 103 (2008) 113518.
4.5.22 Grinstein, F.F., H. Rabitz, and A. Askar, “The 4.5.38 Karmous et al., Phys. Rev. B 73 (2006) 075323.
Multigrid Method for Accelerated Solution of 4.5.39 B. Zhang, S. Shrestha , M.A. Green , G. Conibeer,
the Discretized Schrodinger Equation”. Journal of Surface states induced high p-type conductivity
Computational Physics, 1983. 51: p. 423-443. in nanostructured thin film composed of Ge
4.5.23 D. König, J. Rudd, M.A. Green, G. Conibeer, “Role of nanocrystals in SiO2 matrix, Applied Physics
the interface for the electronic structure of silicon Letters, 97, 132109 (2010).
quantum dots”, Physical Review B, 2008. 78(3): 4.5.40 X.J. Hao, E-C. Cho, C. Flynn, Y.S. Shen, S.C. Park,
p. 035339. G. Conibeer, M.A. Green, Solar Energy Materials
4.5.24 Yurtsever, A., M. Weyland, and D.A. Muller, & Solar Cells, “Synthesis and characterization of
“Three-dimensional imaging of nonspherical boron-doped Si quantum dots for all-Si quantum
silicon nanoparticles embedded in silicon oxide dot tandem solar cells”, 93 (2009) 273–279.
by plasmon tomography”, Applied Physics Letters, 4.5.41 I. Perez-Wurfl, A. Gentle, X. Hao, M.A. Green, G.
2006. 89(15): p. 151920-151920-3. Conibeer and D-H. Kim, Proceedings of 24th
4.5.25 L.-W. Wang and A. Zunger, J. Phys. Chem. 98, EPVSEC, Hamburg, Germany, (2009).
2158 (1994). 4.5.42 J.M. Shah, Y.-L. Li, Th. Gessmann, E. F. Schubert.,
4.5.26 Y-H. So, M.A. Green, G. Conibeer, S. Huang, A. Journal of applied physics. 94 (2003) 2627.
Gentle, “Size dependent optical properties of Si 4.5.43 L. Ma, D. Lin, G. Conibeer, I. Perez-Wurfl,
quantum dots in Si-rich nitride/Si3N4 superlattice “Introducing dopants by diffusion to improve the
synthesized by magnetron sputtering”, Applied conductivity of silicon quantum dot materials
Physics Letters, accepted Feb 2011. in 3rd generation photovoltaic devices”, Physica
4.5.27 D. Di, I. Perez-Wurfl, G. Conibeer, M.A. Green, Solar Status Solidi, accepted Sept10
Energy Materials & Solar Cells 94 (2010) 2238-2243. 4.5.44 G. Cantele, E. Degoli, E. Luppi, R. Magri, D. Ninno,
4.5.28 G. Conibeer, M.A. Green, D. Konig, I. Perez-Wurfl, G. Iadonisi, S, Ossicini S, “First-principles study of n-
S. Huang, X. Hao, D. Di, L. Shi, S. Shrestha, B. and p-doped silicon nanoclusters”, Physical Review
Puthen-Veettil, Y. So, B. Zhang and Z. Wan, “Silicon B, 72 (2005) 113303; G.M. Dalpian, J.R. Chelikowsky,
quantum dot based solar cells: addressing the “Self-Purification in Semiconductor Nanocrystals”,
issues of doping, voltage and current transport”, Physical Review Letters, 96 (2006) 226802.
Progress in Photovoltaics: Res. Appl. 18 (2010) In 4.5.45 A.R. Stegner, R.N. Pereira, R. Lechner, K. Klein,
press, Published online in Wiley Online Library H. Wiggers, M. Stutzmann, Brandt MS, “Doping
(wileyonlinelibrary.com). DOI: 10.1002/pip.1045, efficiency in freestanding silicon nanocrystals from
accepted Sept 2010 the gas phase: Phosphorus incorporation and
4.5.28 D. Song, E. Cho, Y. Cho, G. Conibeer, Y. Huang, defect-induced compensation”, Physical Review B,
S. Huang, MA. Green, “Evolution of Si (and SiC) 80 (2009) 165326.

98
4.5.46 A.L. Tchebotareva, M.J.A. de Dood, J.S. Biteen, H.A. 4.5.63 H. Bilz, W. Kress, “Phonon dispersion relations in
Atwater, A. Polmana, “Quenching of Si nanocrystal Insulators”, Springer-Verlag (1979).
photoluminescence by doping with gold or 4.5.64 R. Clady, M.J.Y. Tayebjee, P. Aliberti, D. Konig, N.J. ARC
phosphorous”, Journal of Luminescence, 114 Ekins-Daukes, G.J. Conibeer, T.W. Schmidt, M.A. PHOTOVOLTAICS
(2005) 137-144. Green, “Interplay between Hot phonon effect and CENTRE OF
4.5.47 M. Fujii, A. Mimura, S. Hayashi, K. Yamamoto, C. Intervalley scattering on the cooling rate of hot
Urakawa, H. Ohta, Journal of Appl. Phys., 87 (2000) carriers in GaAs and InP”, Progress in Photovoltaics, EXCELLENCE
1855-1857. accepted March 2011. 2010/11
4.5.48 D. König, S. Shrestha, J. Rudd, G. Conibeer, 4.5.65 K.T. Tsen, J.G. Kiang, D.K. Ferry, H. Lu, W.J. Schaff, ANNUAL REPORT
M. A. Green, “Doping of Si-Based Dielectrics H.W. Lin, S. Gwo, “Direct measurements of the
for Providing Majorities to Si Quantum Dots: lifetimes of longitudinal optical phonon modes
Acceptors in SiO2”, Proc. 24th European PVSEC, and their dynamics in InN”, Applied Physics Letters,
(Hamburg 2009). 90 (2007) 3.
4.5.49 G. Conibeer, M.A. Green, D. Konig, I. Perez-Wurfl, 4.5.66 Y.C. Wen, Y.C., C.Y. Chen, C.H. Shen, S. Gwo, C.K. Sun,
S. Huang, X. Hao, D. Di, L. Shi, S. Shrestha, B. “Ultrafast carrier thermalization in InN”, Applied
Puthen-Veettil, Y. So, B. Zhang and Z. Wan, “Silicon Physics Letters, 89 (2006) 3; M.D. Yang, et al.,
quantum dot based solar cells: addressing the “Hot carrier photoluminescence in InN epilayers”,
issues of doping, voltage and current transport”, Applied Physics A - Materials Science & Processing,
Progress in Photovoltaics: Res. Appl. 18 (2010) In 90 (2008) 123-127.
press, Published online in Wiley Online Library 4.5.67 F. Chen, A.N. Cartwright, H. Lu, W.J. Schaff, “Ultrafast
(wileyonlinelibrary.com). DOI: 10.1002/pip.1045, carrier dynamics in InN epilayers”, Journal of
accepted Sept 2010. Crystal Growth, 269 (2004) 10-14.
4.5.50 Santosh K. Shrestha, Pasquale Aliberti, Gavin J. 4.5.68 R. Patterson, M. Kirkengen, B. Puthen Veettil, D.
Conibeer, Energy selective contacts for hot carrier Konig, M.A. Green, G. Conibeer, Sol. Ener. Mats. And
solar cells, Solar Energy Materials and Solar Cells, Sol. Cells, 94 (2010) 1931-1935.
94 (Sept 2010) 1546-1550.
4.5.69 S. Huang, K. Minami, H. Sakaue, S. Shingubara, T.
4.5.51 P. Würfel, Sol. Energy Mats. and Sol. Cells. 46 Takahagi, “Effects of the surface pressure on the
(1997) 43. formation of Langmuir-Blodgett monolayer of
4.5.52 R. Ross, A.J. Nozik, J. Appl Phys (1982) 53, 3318. nanoparticles”, Langmuir, 20 (2004), 2274.
4.5.53 V.Y. Davydov, et al., “Absorption and emission of 4.5.70 C. R. Kagan, C. B. Murray, M. Nirmal, M. G. Bawendi,
hexagonal InN: Evidence of narrow fundamental Phys. Rev. Lett. “Electronic Energy Transfer in CdSe
band gap”, Physica Status Solidi (B), 229 (2002) p. Quantum Dot Solids”, 76 (1996) 1517.
r1-r3. 4.5.71 D. König, “Photovoltaic Device Physics at the
4.5.54 D. Fritsch, H. Schmidt, M. Grundmann, “Band Nanoscale”, Chapter 3 (pp. 73 - 146) in L. Tsakalakos
dispersion relations of zinc-blende and wurtzite (Ed.), “Nanotechnology for Photovoltaics”, Taylor &
InN”, Physical Review B, 69 (2004) 165204. Francis, April 2010; ISBN-13: 9781420076745
4.5.55 Y. Takeda, T. Ito, T. Motohiro, D. König, S.K. Shrestha, 4.5.72 T.K. Todorov, K.B. Reuter, D.B. Mitzi, Advanced
G. Conibeer, “Hot carrier solar cells operating under Materials, 22 (21010) 1–4.
practical conditions”, Journal of Applied Physics, 4.5.73 W.R. L. Lambrecht, “Structure and phonons of
105 (2009) 074905-10. ZnGeN2”, Phys. Rev. B 72 (2005) 155202
4.5.56 Wurfel, P., A.S. Brown, T.E. Humphrey, and M.A. 4.5.74 T.R. Paudel, W.R.L. Lambrecht, Phys. Rev. B, 79
Green, Particle conservation in the hot-carrier solar (2009) 245205.
cell”, Progress in Photovoltaics, 13 (2005) 277-285.
4.5.75 S.C. Erwin, I. Zutic, Nature Materials, 3 (2004)
4.5.57 P. Aliberti, Y. Feng, Y. Takeda, S.K. Shrestha, M.A. 410‑414.
Green, G.J. Conibeer, “Investigation of theoretical
efficiency limit of hot carrier solar cells with bulk 4.5.76 Z.T. Liu, J.Z. Zhu, N.K. Xu, X.L. Zheng, Jap. J. Appl.
InN absorber”, Journal of Applied Physics, 108 Phys., 36 (1997) 3625.
(2010) 094507-10. 4.5.77 A. Shalav, B.S. Richards, M.A. Green, “Luminescent
4.5.58 G. Conibeer, C.W. Jiang, D. König, S.K. Shrestha, T. layers for enhanced silicon solar cell performance:
Walsh, M.A. Green, “Selective energy contacts for Up-conversion”, Sol. Energy Mats. and Solar Cells,
hot carrier solar cells”, Thin Solid Films, 216 (2008) 91 (2007) 829-842.
6968-6973. 4.5.78 C.M. Johnson, P.J. Reece, G.J. Conibeer, “Up-
4.5.59 P.G. Klemens, Phys. Rev. 148 (1966) 845. conversion luminescence enhancement in
erbium-doped porous silicon photonic crystals for
4.5.60 G. Conibeer, R. Patterson, P. Aliberti, L. Huang, J-F. photovoltaics”, Photonics West Conference, San
Guillemoles, D. König, S. Shrestha, R. Clady, M. Francisco, Jan 2011.
Tayebjee, T. Schmidt, M.A. Green “Hot Carrier solar
cell Absorbers”, 24th European PVSEC, Hamburg 4.5.79 S. Pillai, K. R. Catchpole, T. Trupke, and M. A. Green,
(Sept. 2009). J. Appl. Phys., 101 (2007) 093105.

4.5.61 G. Conibeer, N.J. Ekins-Daukes, J-F. Guillemoles, D. 4.5.80 Z. Ouyang, S. Pillai, F. Beck, O. Kunz, S. Varlamov, K.
König, E-C. Cho, C-W. Jiang, S. Shrestha and M.A. R. Catchpole, P. Campbell, and M. A. Green, App.
Green, “Progress on Hot Carrier Cells”, Solar Energy Phys. Lett., 96 (2010) 261109.
Materials and Solar Cells, 2009, 93, 713. 4.5.81 S. H. Lim, W. Mar, P. Matheu, D. Derkacs, and E. T. Yu,
4.5.62 D. König, K. Casalenuovo, Y. Takeda, G. Conibeer, J.F. J. Appl. Phys., 101 (2007) 104309.
Guillemoles, R. Patterson, L.M. Huang, M.A. Green, 4.5.82 S. Pillai, K. R. Catchpole, T. Trupke, G. Zhang, J. Zhao,
“Hot Carrier Solar Cells: Principles, Materials and and M. A. Green, Appl. Phys. Lett. 88 (2006) 161102.
Design”, Physica E, 42 (2010) 2862-2866. 4.5.83 S. Pillai, I. Perez-Wurfl, G. J. Conibeer, and M. A.
Green, physica status solidi (c) 8 (2011) 181.

99
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

4.6 Photonics and device

characterisation
include series resistance imaging and
4.6.1 Photoluminescence based 4.6.1.1 Background diffusion length imaging on fully processed
characterisation of silicon In photoluminescence (PL) imaging an intense cells. The research of the PL group at UNSW
light source is used to illuminate large area continues to contribute significantly to these
samples such as silicon bricks, wafers or developments. This year the research by our
University Staff
complete solar cells homogeneously with group was focussed on the application of the
A/Prof. Thorsten Trupke typically one-sun equivalent illumination photoluminescence intensity ratio method
intensity. A CCD camera captures a high for measurements of the bulk lifetime on
Project Scientists and resolution picture of the luminescent light silicon bricks (Section 4.6.1.2), proof of concept
Technicians that the sample emits. Figure 4.6.1.1 shows a studies on an application of luminescence
Allen Yee schematic diagram of a luminescence imaging imaging for measuring the emitter sheet
system. The power supply that is shown on resistance on diffused wafers (4.6.1.4) and on
Undergraduate Students the bottom left can be used as an excitation analysing local shunt or defect currents from
source in electroluminescence (EL) imaging PL images on metallised and non-metallised
Matthias Juhl
experiments on finished cells [4.6.1.3] and also samples in a quantitative fashion (4.6.1.5).
for applications in which a solar cell is held
Postgraduate Students The commercialisation of PL imaging by
at specific operating points under external
• Yael Augarten UNSW spin-off company BT Imaging Pty
illumination, such as the series resistance
• Henner Kampwerth Ltd has also made significant progress, with
imaging methods that were introduced by our
• Bernhard Mitchell a growing number of leading wafer and
group [4.6.1.4-4.6.1.6].
cell manufacturers and research institutes
• Li Hua
High resolution (typical one megapixel) worldwide now using the BTi-R1 R&D tool
External Collaborators luminescence images can be captured that was introduced into the market in 2008.
• Ron Sinton, Sinton Consulting within seconds or even fractions of a second, In addition, as-cut wafer inspection with PL
allowing very fast and detailed studies of imaging (see Section 4.6.1.3) is increasingly
• BT Imaging Pty Ltd
specific material and device properties. recognised as an ideal tool for incoming
• Otwin Breitenstein, Jan Bauer, Max
Planck Institute for Microstructure Over the last five years PL imaging has seen wafer inspection in solar cell production
Physics, Halle, Germany the development of a range of specific or for outgoing quality control in wafer
applications and also rapid adoption by the manufacturing [4.6.1.2, 4.6.1.7-4.6.1.8].
PV community. Specific applications that New in-line PL imaging tools that are fast
are applicable across the PV value chain enough to keep up with typical production
are now available and the number of these line throughput at full one megapixel image
applications is growing rapidly. Examples

100
 ARC
PHOTOVOLTAICS
CENTRE OF
Schematic diagram of a
EXCELLENCE
luminescence imaging system. In 2010/11
photoluminescence (PL) imaging an ANNUAL REPORT
intense laser is used to illuminate
the entire sample and a sensitive
IR camera then captures the
luminescent emission from the
sample. In PL imaging the sample
can be a brick prior to wafering,
an as-cut or partially processed
wafer or a fully processed solar
cell (as shown). A power supply
is used in electroluminescence
imaging on finished solar cells
and in combination with the
light source in series resistance
imaging techniques.

Figure 4.6.1.1

resolution were introduced by BT Imaging in 2010 measurements on silicon bricks, which avoids
and are currently entering the market. the need for a separate resistivity measurement.
The method is based on analysing the
4.6.1.2 PL imaging on Si bricks photoluminescence intensity ratio (PLIR) between
two PL images taken with different spectral filters
Photoluminescence imaging is an ideal tool for
mounted in front of the camera [4.6.1.10-4.6.1.11].
fast characterisation of silicon bricks, giving instant
It is equivalent to the diffusion length imaging
information about the position of low lifetime
method proposed and first demonstrated at
regions that are commonly observed near the
UNSW by Peter Würfel on finished solar cells using
bottom and top of a brick and also exposing areas
electroluminescence imaging [4.6.1.12]. In fact, the
of high structural defect density. This information
elimination of voltage variations from a diffusion
can be used as valuable feedback during the
length image that is achieved in the PLIR method
production of silicon wafers. Initial work on
on cells is in perfect analogy to the elimination
luminescence imaging on bricks showed excellent
of doping density variations when applying this
qualitative correlation between the as measured PL
method to PL images on bricks. Silicon bricks
intensity and for example μ-PCD effective minority
represent an ideal test case for the PLIR method,
carrier lifetime maps [4.6.1.9]. More recently we
since uncertainties associated with variations in
also demonstrated a quantitative interpretation
surface texture and with the optical and electronic
of the PL signal in terms of the bulk minority
properties of the rear surface are almost completely
carrier lifetime mb. The bulk lifetime is a more useful
avoided. In cells and wafers these uncertainties can Bulk lifetime images obtained from
parameter than the effective lifetime, especially in
lead to large experimental errors. a single doping density normalised
unpassivated samples, where the measurement PL image (upper) and from the PLIR
result is strongly influenced by surface Figure 4.6.1.2 shows a comparison of two (lower), see text for details.
recombination. PL imaging thus has advantages bulk lifetime images obtained from a doping
Figure 4.6.1.2
compared to other measurement techniques such normalised PL image and from the PL intensity
as μ-PCD, which have commonly been used to date ratio, respectively. The latter image was obtained
for the characterisation of bricks in production, and using a transfer function that was calculated
which report only effective lifetime values. taking into account sample temperature
and the exact spectral properties of the
In our previous approach to measure bulk lifetime
measurement system.
from PL images we converted the measured PL
intensity form a single PL image into bulk lifetime Deviations are observed near the low
using a pre-determined tool specific transfer lifetime edges at the bottom and top,
function. While that approach proved to be a where the PLIR overestimates the bulk
reliable approach for obtaining bulk lifetime on lifetime (see also Fig.4.6.1.3, showing a
a number of samples and without the need for comparison of cross sections through
additional calibration, one disadvantage is the the two bulk lifetime images). Defect
fact that it requires a separate doping density luminescence contributions, which become
measurement. The PL intensity in low injection more noticeable in the low lifetime regions
conditions is given as PL~Dn∙ND, i.e. it is proportional with reduced band to band luminescence
to both the effective lifetime and the background signal is currently believed to cause those
doping density ND. Normalising the measurement deviations. Spectral PL measurements
Cross sections through the bulk
for doping variations is required in order to obtain currently under way will shed more light on this lifetime images from Fig. 4.6.1.2,
information on Dn and thereby the lifetime. issue. Several approaches to mitigate the impact showing good agreement in the
of the above measurement errors in the PLIR centre and artefacts near the
Recent work in the Photoluminescence Group bottom and top regions.
near the bottom and top regions are currently
focussed on a modified approach for bulk lifetime
under investigation. Figure 4.6.1.3

101
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

PL images of two multicrystalline

silicon wafers. (a) Centre wafer with
low dislocation density, (b) Centre
wafer with high dislocation density,
(c) bottom wafer with low lifetime
across the entire wafer area, (d)
corner wafer with low dislocation
density [4.6.1.2].

Figure 4.6.1.4

Very good quantitative agreement The dominant spatial features that are commonly
is observed between the two observed in mc-Si as-cut wafers are dislocation
measurements in the central part, networks, recombination active grain boundaries
i.e. across the majority of the brick. and impurity rich regions in edge/corner and top/
This region is the section of the brick bottom wafers. Figure 4.6.1.4 shows four typical
that is supposed to be used for wafer examples, (a) a wafer from a center brick with low
manufacturing (the very low lifetime dislocation density, (b) a wafer from a center brick
top and bottom regions are supposed with high dislocation density, (c) a wafer from the
to be removed from the ingot). impurity rich area at the bottom and (d) a wafer
from a corner brick with low dislocation density
4.6.1.3 Raw wafer inspection [4.6.1.2]. All spatial features that appear dark in
these images are indicative of highly recombination
Wafer quality in industrial
active regions within the wafers. Automated image
manufacturing of silicon solar cells
processing methods allow determining wafer
varies substantially and can have a
quality metrics that are related to the area fraction
strong impact on the efficiency of
of those features and finally correlation with cell
solar cells, particularly in the case
efficiency parameters such as Isc and Voc [4.6.1.1,
of multicrystalline silicon. As-cut
4.6.1.7, 4.6.1.13]. An example for such correlation
wafer inspection of incoming quality
is shown in Fig. 4.6.1.5, showing the correlation
control (IQC) in solar cell production
between the cell efficiency, Voc and (given in
is therefore indispensable in order
relative numbers, normalised to unity) and the
to maintain high efficiencies at high
dislocation density as measured on the as cut
yield in production. To date IQC in
Correlation of cell performance wafers for a number of cells. A strong correlation
production typically includes optical
data (normalised) with the is observed.
defect density obtained from inspection for cracks and large chips, thickness
the PL images on the as-cut and total thickness variation, surface morphology, The observed correlation of these metrics with
wafers [4.6.1.1]. resistivity measurements and in some cases cell performance data, which will depend on the
μ-PCD lifetime scanning. A problem with these detailed processing conditions, can then be used
Figure 4.6.1.5
techniques is that the electrical cell performance for a range of actions, all aiming at higher yield,
does not correlate sufficiently strongly with the and higher average efficiency in production: (i)
data resulting from these measurements. One main wafer rejection, (ii) wafer quality based pricing, (iii)
reason is surface recombination dominating total wafer quality specific processing, or (iv) assignment
recombination (and thus effective lifetime) in as-cut of specific types of wafers to specially tuned
wafers. Only wafers with very low bulk lifetime production lines. Separating specific wafers for
(<10 ms) or specific areas exhibiting very low bulk high efficiency processing lines such as a selective
lifetime show up sufficiently strongly in average or emitter line, which requires higher quality wafers
spatially resolved lifetime. While this complicates to realise the full efficiency potential and may not
the analysis of spatially averaged lifetime values be as robust against specific types of defects as the
and makes correlation with final cell efficiency less common screen printed process, is one specific
reliable, particularly for wafers with overall high example for that latter application.
lifetime, it is beneficial since such areas are generally
expected to also have the strongest impact on The PL intensity emitted from as-cut silicon wafers
final cell efficiency. In other words, features that is extremely low, a result of the strong dominance
are visible with significant contrast in effective of surface recombination, resulting in luminescence
lifetime images are very recombination active and quantum efficiencies on the order of 10-8 (i.e. one
have bulk lifetime of only a few ms or below one ms luminescence photon is emitted for 100 million
incident photons). This low photoluminescence

102
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

BT-Imaging W1 PL imaging system,

with integrated automated image
processing algorithms and a current
throughput of up to 2400 wafers
per hour with full one megapixel
image resolution.

Figure 4.6.1.6

response of as-cut wafers makes inline applications The luminescence from the non-illuminated area
with a throughput of up to 3600 wafers per hour, is generated by lateral flow of excess carriers from
a significant experimental challenge. BT Imaging the illuminated into the non-illuminated parts of
recently presented the W1 (see Fig. 4.6.1.6), the the wafer, in other words, electroluminescence
first inline PL imaging system capable of taking is generated in a contactless fashion from that
PL images on as-cut wafers with a throughput of region, as reported previously elsewhere [4.6.1.14].
2400 wafers per hour and using automated image The illumination intensity and the emitter sheet
processing to identify a range of relevant specific resistance have a strong impact on the carrier
defect metrics. Using these tools, application in redistribution via lateral carrier flows and thereby
production of the above concepts, which so far has on the steady state excess carrier concentration,
only been proven off-line on relatively small sample the latter reflected in the luminescence intensity.
sets, will demonstrate the value of using PL imaging Combined with theoretical modelling this allows
for IQC. interpretation of the luminescence intensity
distribution in terms of the emitter sheet resistance.
4.6.1.4 Emitter sheet resistance imaging An example of a PL image taken on a diffused PL image of a diffused silicon
In solar cell production, the value of the emitter wafer with 60 Ω/sq is shown in Figure 4.6.1.7. The wafer taken with a 1-inch circular
resistance typically needs to be maintained within circular filter is visible in the centre of the wafer. long pass filter located in close
proximity to the wafer surface. The
tight specifications in order to avoid subsequent The luminescence signal from underneath the luminescence signal emitted from
problems with contact formation such as junction filter represents contactless electroluminescence underneath the filter is caused
shunting or high contact resistance. Previous work as described above. Figure 4.6.1.8 shows three by lateral current injection from
in the PL group focussed on luminescence based PL intensity profiles from the centre of the filter the illuminated wafer area via
emitter sheet resistance imaging measurements the emitter.
to the outside region taken on three different
on fully processed metallised solar cells. In a new samples with emitter sheet resistivities ranging Figure 4.6.1.7
approach, the aim is to measure the emitter sheet between 30 Ω/sq and 100 Ω/sq. With increasing
resistance in a contactless fashion from PL images emitter resistance the PL intensity decays with a
on non-metallised samples, which would allow steeper slope towards the centre of
application of that method immediately after the the non-illuminated region, which is
emitter diffusion in production. Since the emitter a result of the limited ability of the
sheet resistance is not expected to exhibit lateral emitter to transport current laterally.
variations over short distances, a method with An analytical model for that relative
comparatively low spatial resolution is sufficient. PL profile was developed and used
Here we investigated the principle of using a PL to fit the measured data within the
image for a coarse measurement of the emitter non-illuminated part of the image.
sheet resistance, providing a single average data Resulting emitter sheet resistance
point for a large area. values were found to agree with eddy
current measurements of the emitter
The measurement principle is to take a PL image
sheet resistance.
with non-uniform illumination and obtain the
emitter sheet resistance information from the The analytical model so far describes
resulting PL intensity variation, in combination only the PL profile in the non-
with theoretical modelling. A 1-inch circular long Cross section of the normalised
illuminated filter region. Further work
PL intensity from the centre of the
pass filter is placed in close proximity to the sample will aim to extend that model to the illuminated filter (x=0) to the outside region
during the PL imaging measurement. The filter part of the wafer, resulting in increased accuracy. through PL images as shown in Fig
blocks effectively all (>99.9 %) incident light form The impact of the wafer background doping density 4.6.1.7 for different samples with
the area underneath the filter and transmits >98% variable emitter sheet resistance.
on the measurement results is another topic for
of the luminescence that is emitted from that area. further work. Figure 4.6.1.8

103
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

0.5 cm
(a) (b)

PL images taken on a test structure. An external Extracted current as calculated from PL images
resistor is connected in parallel to the structure using a using a new analysis methodology as a function
contact probe (visible in the bottom right). The images of the measured current extracted over a resistor,
show the impact of a large resistance (mild shunt, demonstrating that PL imaging allows a quantitative
left) and a small resistance (strong shunt, right) on the analysis of the extracted current from localised shunts
PL image. or defects.

Figure 4.6.1.9 Figure 4.6.1.10

4.6.1.5 Luminescence based shunt terms of the diode voltage near a shunt site allows
imaging quantification of individual local shunts.
Qualitative shunt detection from luminescence Applications of this method to finished cells
images has been demonstrated in the past by the are affected by current extraction via the metal
UNSW PL group (amongst others) [4.6.1.15-4.6.1.17]. grid, which causes experimental errors. The
The vicinity of shunted areas typically appears as a impact of this effect was assessed experimentally
blurred region of reduced luminescence intensity using specific test structures, where the relative
in both PL and EL images. This blurring is caused by position with respect to the nearest grid finger of
voltage drops associated with lateral current flow artificially introduced shunts could be varied. These
through the emitter and the front surface grid. experiments indicate that the above experimental
errors in fully processed cells can at least partially
A luminescence imaging based method for
be corrected using empirically determined look
determining quantitative shunt values in silicon
up tables.
solar cells from open circuit photoluminescence
images has been developed and demonstrated When applied to partially processed cells prior to
in collaboration with the Max Planck Institute for metallisation the method can be used to quantify
Microstructure Physics, Halle, Germany. The method local shunts, to determine the total recombination
is based on interpretation of the luminescence current in local and distributed defects and also
intensity around a local shunt in terms of the to quantify recombination channels that are
extracted current density [4.6.1.18]. Under the introduced by specific processing steps such as
assumption of a unity ideality factor the local laser processing.
reduction in PL count rate in the vicinity of a shunt
or local defect is proportional to the current being 4.6.1.6 Summary
extracted. The theoretical framework for quantifying
An exceptional variety of material and solar cell
the total current extracted by a local defect by
parameters can be measured on silicon bricks,
integrating the reduction of the PL signal in the
silicon wafers and silicon solar cells with high
surrounding of the shunt or defect and by applying
lateral spatial resolution and short measurement
suitable calibration procedures was developed.
time using luminescence imaging techniques.
Experimental verification of the method was The range of applications continues to grow, with
achieved using a specifically prepared test structure, new applications for PL imaging being developed
i.e. a solar cell structure with single point contact at UNSW, and increasingly also in other research
on the front surface. That structure allowed institutes and by R&D groups in PV companies.
generating local shunts with variable and well PL imaging, introduced at UNSW only five years
defined external shunt resistance. Figure 4.6.9 ago, has now been broadly adopted as a standard
shows two PL images of that structure with a high characterisation method, with PL tools now in
(left) and low (right) Ohmic resistor connected in use at virtually all leading research institutes
parallel with that structure. Figure 4.6.1.10 shows worldwide and also by leading wafer and solar cell
the extracted current, as obtained from the above manufacturers. A range of opportunities for in-line
methodology as a function of the current flowing quality control, process monitoring and process
across the resistor, as measured with a multimeter. control are enabled by the high resolution and
Excellent agreement is observed. Combining the speed of PL imaging. Demonstration by UNSW
above current measurement technique with an spin-off company BT Imaging of those applications
interpretation of the local luminescence signal in in beta trials at several company sites is currently
in progress.

104
References 4.6.2 General Characterisation
4.6.1.1 McMillan, W. et al., In-line monitoring of electrical
ARC
wafer quality using photoluminescence imaging. University Staff
Proceedings of 25th EPVSC, Valencia, Spain, PHOTOVOLTAICS
Dr Henner Kampwerth
September, 2010. CENTRE OF
4.6.1.2 Trupke, T., Nyhus, J. and Haunschild, J., EXCELLENCE
Luminescence imaging for inline characterisation in Project Scientists and Technicians
silicon photovoltaics. Phys.Stat.Solidi RRL, 2011: p.
2010/11
Alan Yee ANNUAL REPORT
published online February 2011.
4.6.1.3 Fuyuki, T. et al., One shot mapping of minority
carrier diffusion lenth in polycrystalline Si solar cells The range of research undertaken by the Centre
using electroluminescence. 31st IEEE Photovoltaic
Specialists Conference, Orlando, USA, 2005.
involves a variety of measurement techniques
for the characterisation of photovoltaic materials
4.6.1.4 Trupke, T. et al., Fast photoluminescence imaging of
silicon wafers. WCPEC-4, Waikoloa, USA, 2006. and devices. Understanding and utilising these
4.6.1.5 Trupke, T., et al., Spatially resolved series resistance
techniques appropriately, and improving them
of silicon solar cells obtained from luminescence where possible, is crucial in order to maintain the
imaging. Appl. Phys. Lett., 2007. 90: p. 093506. world-class reputation of our research output.
4.6.1.6 Kampwerth, H. et al., Advanced luminescence based
effective series resistance imaging of silicon solar
Improvements to measurement techniques with the
cells. Appl. Phys. Lett., 2008. 93: p. 202102. goal of producing more accurate and meaningful
4.6.1.7 Haunschild, J. et al., Quality control of as- results will have a direct impact both on research
cut multicrystalline silicon wafers using activities and on the optimization of procedures
photoluminescene imaging for solar cell production. used in manufacturing. The time and number
Solar Energy Materials and Solar Cells, 2010. of experiments needed to understand a certain
94(12): p. 2007-2012.
phenomena can be greatly reduced. The need to
4.6.1.8 You, D. et al., The dislocation distribution
measure parameters of interest more precisely,
characterisitics of multi-crystalline silicon ingot
and its impact on cell efficiency. 35th IEEE PVSC, without the possibility of misinterpretation, is
Honululu, USA, June, 2010. therefore high. The General Characterisation group
4.6.1.9 Trupke, T., Bardos, R.A. and Nyhus, J., was formed in July 2010 to address this need.
Photoluminescence characterisation of silicon wafers
and solar cells. 18th workshop on Crystalline Silicon The group primarily focuses on optical
Solar cells & Modules, 2008. characterization methods such as
4.6.1.10 Trupke, T. et al., Bulk minority carrier lifeitme photoluminescence spectroscopy and absorption
from luminescence intensity ratios measured spectroscopy. In recent years, it has been shown
on silicon bricks. 25th EPVSC, Valencia, Spain, that these optical measurements can be used to
September 2010.
extract a large number of material and device
4.6.1.11 Mitchell, B. et al., Bulk minority carrier lifetimes and
properties. In addition the ability to perform
doping of silicon bricks from photoluminescence
intensity ratios. J. Appl.Phys., 2011: p. accepted contactless non-destructive probing of small areas
for publication. down to a few ten micrometers in diameter makes
4.6.1.12 Würfel, P. et al., Diffusion lengths of silicon solar optical measurement techniques very attractive for
cells obtained from luminescence images. Journal of samples that cannot be measured otherwise.
Applied Physics, 2007. 101: p. 123110.
In addition, the group will review some
4.6.1.13 Demant, M. et al., Analysis of luminescence
images applying pattern recoginition techniques. commonly used measurement methods with
Proceedings of 25th EPVSC, Valencia, Spain, the aim of automating parameter extraction
September, 2010. from measurement data. It is common that less
4.6.1.14 Sinton, R.A., Contactless electroluminescence for experienced researchers misinterpret data and draw
shunt-value measurement in solar cells. Proceedings incorrect conclusions. Computer programs that
of the 23rd EPVSC, Valencia, September, 2008:
assist in various levels of interpretation therefore
p. 1157.
can prove their value. Some computer programs
4.6.1.15 Trupke, T. et al., Luminescence imaging for fast
shunt localisation in silicon solar cells and silicon written in the course of this work will be made
wafers. International Workshop on Science and available for public use.
Technology of Crystalline Silicon Solar Cells,
Sendai, Japan, 2-3 October, 2006.
4.6.1.16 Kasemann, M. et al., Shunt detection capabilities of
luminescence imaging on silicon solar cells. 22nd
European Photovoltaic Solar Energy Conference,
Milan, Italy, 2007.
4.6.1.17 Breitenstein, O. et al., On the detection of shunts in
silicon solar cells by photo- and electroluminescence
imaging progress in Photovoltaics, 2008. 16: p. 325.
4.6.1.18 Augarten, Y. et al., Luminescence shunt imaging:
qualitative and quantitative shunt images using
photoluminesccence imaging. 24th European
Photovoltaic Solar Energy Conference, Hamburg,
September, 2009.

105
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

5. EDUCATION are appropriately qualified to enter the field upon completion of their
studies. The undergraduate programs have experienced continuing
Summary strong demand growth since 2007. Since 2003, the School has seen a
total of 160 students graduate from its undergraduate programs, and
The ARC Photovoltaics Centre of Excellence incorporates the activities
126 students from its postgraduate degrees. These graduates are now
of the former Key Centre for Photovoltaic Engineering. The former
taking advantage of a range of opportunities being created in the
Key Centre started in 1999, after the award of special funding from
booming photovoltaics and renewable energy industries.
the Australian Government to promote teaching and research in the
area of photovoltaics. A major initiative of this Key Centre was the The higher media profile of global warming, greenhouse and energy
establishment of the world’s first undergraduate degree in Photovoltaics issues in the general Australian community is helping to raise the
and Solar Energy. The Centre was one of only eight such Key Centres School’s profile locally and the industry growth in Asia raised the
awarded Australia-wide across all disciplines, demonstrating the School’s profile internationally
Government’s understanding of the importance of the field of
The granting of a large number of sponsorships was finalised in August
renewable energy. In 2003 the Key Centre and its activities were
2007 for up to $5.2m support from the Australian Government for
incorporated into this Centre of Excellence, awarded to the same team
study at the School, as part on the Asia-Pacific Partnership on Clean
at the University of New South Wales. On 1 January 2006 UNSW officially
Development and Climate. In 2010, these scholarships attracted 21 new
formed a new School within the Faculty of Engineering, the School of
undergraduate and 48 (29 in Semester 1 and 19 in Semester 2) new
Photovoltaic and Renewable Energy Engineering, which includes the
postgraduate coursework students to the School. The number of formal
ARC Photovoltaics Centre of Excellence.
agreements with Chinese universities established as a result of the
The School offers undergraduate, postgraduate and research programs Partnership grew to seven during 2010.
encompassing a range of aspects relating to the photovoltaic and other
At the end of 2010, there were 437 undergraduate students, 88
renewable energy industries. These programs have been developed in
postgraduate coursework students and 63 research students (11
consultation with representatives from industry to ensure graduates
Masters by research and 52 PhD) enrolled in the School’s programs.

106
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

700

600

500
Total No. of students

400

300

200

100

0
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Year

Overall enrolment numbers in

second semester, by year.

5.1 Undergraduate
Educational Programs
The School offers two undergraduate coursework engineering expected to graduate in March 2011 after completing their studies in
programs, each of which can be taken in combination with a Bachelor 2010. At the end of 2010, a total of 310 students were enrolled.
of Arts (5 years), Bachelor of Science (5 years), Bachelor of Commerce
The second, broader degree program, in Renewable Energy Engineering,
(5.5 years) or a Bachelor of Laws (6 years). These degrees aim to develop
was introduced in 2003. In addition to photovoltaic devices, students
well-educated graduates with the basic skills, attributes and knowledge
in this program study solar architectural technologies, wind energy,
required to practise as professional engineers in the booming
biomass, solar thermal and renewable energy policy as part of their core
renewable energy industries.
curriculum. At the end of 2009, a total of 127 students were enrolled.
The Photovoltaics and Solar Energy Engineering program draws on 2010 saw 15 new graduates of this program, bringing the total to 32,
the long history of expertise of Centre staff in this field and was the and 5 more are expected to graduate in March 2011. At the end of 2010,
first of its kind internationally. Two new programs were established in 24% of enrolled undergraduate students were women.
2009 in China, at Nanchang University in Jianxi Province and at North
Graduates of these programs are gaining employment within a variety
China Electric Power University in Beijing. The UNSW program includes
of organisations, including in the areas of energy efficiency and
education in technology development, manufacturing, quality control,
sustainable design. Each year a number of graduates start on a research
reliability, life cycle assessment, system design, maintenance and fault
program at the Centre, elsewhere at UNSW or at other institutions
diagnosis, marketing, policy development, energy efficiency and other
while others are employed in industry by solar cell and equipment
renewable energy technologies. It has two unique features: a group
manufacturers, system design and integration companies, electricity
project in the second year of the program, and a ‘strand’, which is a
utilities, energy efficiency and environmental design companies and
minor area of specialisation. Since 2003, a total of 128 students have
major end users of products.
graduated from the program, 24 graduated in 2010 and another 34 are

107
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Measurement of light
distribution in a horizontal
plane. A lantern is shown in
the inset.

5.1.1 The Strand 5.1.2.1 Developing Countries

The strand is unique within UNSW to the The School has been involved with projects in the
Photovoltaics and Solar Energy Engineering developing countries for the last eight years. The
program, and is essentially a minor area of application of photovoltaics and other renewable
specialisation which complements a student’s energy technologies can make a great difference to
study of photovoltaics. Students can take strands people’s lives and living standards in these places.
covering a wide variety of areas including Students involved with this project in the past have
computing, electronics, mathematics, physics, installed and maintained photovoltaic lighting
mechanical engineering, civil engineering, and systems in rural locations in Nicaragua, Vanuatu
architecture. The aim of the strand is to provide and Nepal and photovoltaics powered water
students with broader engineering backgrounds purification in Sri Lanka. Skills that the students had
important for the cross-disciplinary nature of developed throughout the year were put to good
Top side of Glowstar PCB showing
photovoltaic applications. use in the field as invaluable lessons were gained
several features such as fuses
and microprocessor.
regarding project management of activities in
In 2011 strands will be offered to Renewable
developing countries. Students faced and overcame
Energy Engineering students as well. These will be
many technical and non-technical issues in project
less formally structured than in the Photovoltaic
implementation. With the rapidly growing student
Engineering program and take the form of sets
population the School is no longer involved in
of electives associated with various renewable
directly supervising international journeys for
energy technologies.
groups of second year students but, instead,
suggests interested students involve themselves in
5.1.2 Undergraduate Second-Year the UNSW Chapter of Engineers Without Borders.
Student Group Projects
In the second year of the Photovoltaics and Solar Solar Lanterns Project
Energy program, students undertake a year-long
Energy is fundamental to alleviate poverty in
group project. The main emphasis of the second
health, education, communication, economic
year group project course is hands-on project
development and social equity. Electricity impacts
engineering. The course has a lecture component
the lifestyles and living conditions, primarily
covering project engineering, report writing and
through electric lighting. However, approximately
presentation skills and each project has a research
1.5 billion people in developing countries have
component, a planning and design component, a
no access to electricity. The majority of these
hands-on component and a presentation/reporting
people use kerosene lamps for lighting. Kerosene
component. This course helps to prepare students
lamps are expensive to run, have health impacts,
for their fourth year thesis, which is undertaken by
are potentially dangerous and are a source of
all students enrolled in both the Photovoltaics and
greenhouse gases.
Solar Energy and Renewable Energy Engineering
undergraduate programs. Solar lanterns have potential to improve the living
conditions of people in developing countries. It
In 2010, second-year project groups worked on the
can provide much needed study hours for school
following projects: silicon and cuprous oxide thin
children as well as support after-hour income
(Upper) Student-built current-to- film solar cells, photoluminescence analysis of solar
generating activities. One study has found that solar
voltage converter. (Lower) Student- cells, solar thermal water heating, Sunswift solar race
built phototransistor circuit.
lanterns have significantly benefited school-aged
car, solar water pumping, solar lanterns, cookers and
children in India. The extended study hours made
a “solar suitcase” for developing countries and ice
possible with solar lanterns shown to have a very
air conditioning.
positive impact on their performance.

108
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Thermal image of experiments determining the Image of one groups insulated pot
efficiency and heat loss to the environment of various cooking food.
cooking devices.

There are many solar lanterns are available in the 5.1.2.2 Silicon Thin-Film Solar
market. The performance of these lanterns varies Processing Equipment Project
widely making it difficult for consumers to select
This project aims to give students a broad
a product.In this project students have studies
experience in enhancing and maintaining
a range of solar lanterns commonly available in
a world-class thin-film solar cell research
developing contries. Using a set of tools they
laboratory. This is achieved through a two-
compared performance of solar panel, ballast
stage process. Firstly, students develop a
electronics, charging-discharging circuits, light
fundamental understanding of equipment
output and efficiency.
and systems within the Centre’s thin-film
laboratory, including vacuum systems, gas
Solar Cookers Project delivery systems, leak testing equipment,
Stand alone PV systems (PV/diesel/battery) supply and plasma processing machines. The
power to many households in remote parts of second stage involves the students
the world. Remote households in Australia and in integrating what they have learnt, by applying it to Cake successfully cooked!

particular Aboriginal communities utilise such PV a specific project in the thin-film laboratory.
systems. Cooking is an energy load that needs a
In 2010, six second-year students undertook this
substantial amount of energy. In many instances
project. A group of three students worked on
designers for stand-alone PV systems advocate use
upgrading a vacuum test chamber to redesign the
of gas or microwave cookers for cooking. However,
vacuum control interface and build it. A second
often gas bottles are a very expensive way to get
group of three students worked on halogen lamp
energy for cooking.
vacuum annealing stage, with the aim of being
Is it possible that as the price of PV falls that some able to anneal samples at up to 1000°C. This
form of electric cooking could be the best option? Is group successfully designed and built a halogen
microwave cooking the most efficient technology? lamp assembly that was both functional and easy
To answer this question we asked students: What to maintain.
is the most efficient way to cook? This project
Both groups benefited greatly from the course
explored low energy approaches to cooking and PV
as they learnt to integrate project management
system design to supply the energy to power such
and engineering problem solving to successfully
systems. One approach investigated “slow cookers”.
complete their projects.
Typically they are cooking systems that utilise low
energy but for long periods of time. Typically they
are uninsulated, so at present not very efficient! 5.1.2.3 Cuprous Oxide Thin-Film Solar
Cells Project
Students investigated various approaches to
Students this year revived a decades-old
insulate their cooking devices and used PV panels
photovoltaics technology that was once the
to charge a battery and then use the energy to heat
globally leading technology, before silicon came to
a pot of food. Energy flows (current and voltage)
the fore in the 1950s. The manufacturing process for
were monitored, as well as the temperature of the
cuprous oxide cells is relatively simple and students
food. Students investigated, designed, built and
made and tested their own cells from scratch.
monitored whether small PV systems are capable of
supplying such cooking systems.
5.1.2.4 Water Pumping Project
A group of students was charged with the task
of demonstrating the pumping of water with
photovoltaic power from ground level to the

109
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Seventy SPREE students illustrated their final year thesis work on 22 October
2010 in the largest yet of the school’s graduating classes.

roof of the UNSW Electrical Engineering building. Array

The students worked on realistic conceptual The solar array is essentially an off-grid solar
applications such as toilet flushing in the building system, with the added complications of very
with excess stormwater, water supply for a rural fragile modules, and the requirement that it must
village or stock watering on a cattle station. For withstand the conditions of driving 3000km of
their demonstration project, they were required dusty, bumpy road.
to provide nominal flow rate and storage to
demonstrate the concept. The skills learnt are The students designed the module and array sizing
directly applicable to solar water pumping for to conform to the race regulations, and selected
a wide range of applications, including in rural the appropriate cabling and bypass diodes. They
villages around the world. also performed array calculations and simulations
in order to calculate the optimum wiring scheme
5.1.2.5 Solar Thermal Water Heating that allowed for maximum output power, the
most efficient battery charging, and maximum
Project
robustness against shading.
In our daily life, we are using lots of hot water for
different purposes, needing a considerable amount Two types of cells were used for maximum
of electrical or other energy sources. In this project efficiency, standard industrial cells, and UNSW-
the students learned how to use solar thermal produced Topcell cells. The team characterised
energy to heat cold water. In this project student the Topcells, initially by open-circuit voltage
studied the basic principle of solar heating system measurements, and later by full IV sweeps. The cells
and designed a new water heating system. They were then encapsulated, with assistance from CSG
have also learned what materials should be used Solar, and assembled into modules.
in the systems and investigated the possibilities
Finally, the array team and other Sunswift members
to improve the design for better conversion
assembled the array, mounting and soldering the
efficiency. They built their designed systems using
fragile modules onto the solar car.
their selected materials, as shown in the following
figures. They tested their systems and collected and
Composites
analysed data to find conversion efficiency, cost
evaluation, etc. Carbon fibre composites are extremely strong,
lightweight materials, which are currently used,
e.g. in aeroplanes and wind farms, to increase the
5.1.2.6 UNSW Sunswift Project
energy conversion efficiency of the systems.
Over the past year, second year SPREE students
have helped to design and build several major The composites team had the substantial job of
components of the UNSW solar car, which placed building the carbon fibre composite components
the first in the silicon class in the 2009 Global Green for a car which was almost entirely carbon fibre.
Challenge (GGC), the premier international solar car These included the top shell, which housed the
race. The result was the best result achieved by the solar array, the bottom shell, which formed the
team in its 14 year history. structural chassis to safely house the driver, and
the structural components which held the braking,
Students were involved in two solar car related
steering and suspension systems. The team also
projects – design and construction of the
successfully constructed the first new carbon fibre
photovoltaic array that powers the car, and
wheels for the car in 8 years. The wheels are the
construction of the lightweight composite
most complicated composite part to construct,
components that reduce the weight, and hence the
require a great deal more care and attention, and
energy usage of the car.
must be thoroughly tested. The new wheels were

110
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Upper: Copper pipes mounted on a copper sheet for water heating.

Lower: A complete system with glass covered on top.

Student-built thermal water heating system for domestic use in their 2nd year project.

manufactured extremely well, and performed efficiency dye sensitised solar cells, optimization of
beautifully over the 3000km of rough terrain. high temperature ITO, time of flight measurement
setup, sustainable transport, building integrated
The construction of all of the components took over
PV system design and testing, evaluation of pv/
six months, and was done at the Boeing Aerospace
diesel mini grid applications in the NT, solar
facilities in Bankstown. Boeing sponsored the
power for a weeding robot, silicon quantum dots,
team the carbon fibre and other materials, and
design of a domestic solar water heating system,
the students learned how to work with composite
performance assessment and evaluation of the
materials from the experts.
UNSW Quad building’s rooftop PV System, feasibility
The carbon work on the car by these students of electric cars in Sydney, management strategies
was easily some of the highest quality work in a for building sustainability, and deploying energy
field of 40 international solar cars, including semi efficiency policy to offset cost of meeting MRET in
professional teams, and also suffered no issues the residential sector, energy efficiency in buildings
during the race. & Greenstar.

The Poster Prize selection in November 2010,

5.1.3 Fourth Year Thesis for students completing at end of Semester 2,
The thesis project is usually completed in the last resulted in a wins for Hua Fan for his poster about
two sessions of an engineering undergraduate “Solar Electrolysis System for Remote Production
student’s studies. Students may start their project in of Disinfectant Solutions” (Photovoltaic and
either semester. They undertake directed laboratory Solar Energy) and Kah Howe Chan for his poster
and/or other research work on an approved on “Wind and other renewable energies at the
subject under guidance of the School’s academic UNSW Wellington property” (Renewable Energy
staff, sometimes with industry participation or Engineering). The other topics on which theses
co-supervision. Typically, the thesis involves the were completed included: study of carrier transport
design and construction of experimental apparatus mechanism in organic semiconductor for different
together with practical tests. Each student is electrode metals for organic solar cells, laser
required to present a seminar in their first semester doping of thin film silicon solar cells, wind energy
and submit a written report and present an Open forecasting, impact of electric vehicles on LV
Day poster in their second semester. electrical network, optical fibre/nanowire hybrid
structures for dye-sensitized solar cells, design of
Some students finished and presented their
metamorphic tandem solar cells, the 46% efficient
projects at the end of Semester 1, 2010. The
cell project: cell modelling & testing, comparison
Photovoltaic and Solar Energy Thesis Poster Prize
of properties of Si QD film treated by rapid thermal
was won by Lei Zhang (Laser Doping Technique
annealing and converntional furnace annealing,
Investigation and Optimisation for High Efficiency
feasibility study of biogas/electric hybrid cars,
Laser Doped Solar Cells Fabrication). The Renewable
benefits of NSW wind power for the NEM, electrical
Energy Engineering Thesis Poster Prize was won by
characterisation of Ge nanocrystals embedded
Nicholas Boerema (Renewable Energy Integration
in amorphous silica matrix, 46% efficient split
into the National Electricity Market). Other projects
spectrum cells, biogas production on a Wisconsin
coming to completion at that time were concerned
farm, Investigation of n/p-type doping of Ge
with wind energy and its grid integration, annealing
Ncs in SiO2 towards tandem photovoltaic cells,
effects on silicon thin film solar cells, determination
optimisation of thermal annealing parameters on
of carrier diffusion length and saturation current
planarisation and electrical properties of thin-film
density in silicon solar cells, laser doped emitters
poly-crystalline silicon solar cells on glass, home
and back surface fields, metalization effects on the
charging for electric vehicles, reduction in office
optical properties of nanostructure silicon, high

111
 tenant energy, doping concentration effects on 5.1.4 Dean’s Awards
metal ohmic contact formation to silicon wafer, Each year the Dean of Engineering presents awards
ARC
energy efficiency on UNSW campus, optical for outstanding students in each non-final year of
PHOTOVOLTAICS characterisation of silicon quantum dots in SiOx/ study in each Engineering School. Prizes awarded
CENTRE OF Si3N4 for optical detection of defects, design and during 2010 for academic performance in the
EXCELLENCE construction of PV solar electrolysis system for previous year went to Dongchen Lan (stage 1),
2010/11 medical aid kit, zero energy buildings, fabrication Benjamin Hughes (stage 1), John Durrant (stage
ANNUAL REPORT and characterisation of Si DQs nanostructures 2), Vincent Allen (stage 2), Xue Bai (stage 3), Dong
grown by PECVD, adapting direct etching method Lin (stage 3), Yu Feng (stage 3) and Shen-Long Ooi
for texturing of silicon, design of light trapping (stage 4).
structure for organic solar cells, low energy water
pump systems, PV, pumping, micro-hydro or wind
5.1.5 N&M Radiant Design Prize
at UNSW’s Wellington property, passive PV cooling
using stack effect, investigation of phosphorous The N&M prize is awarded to one student group
doping of Si-QD materials by furnace diffusion, undertaking the School’s project in the first-
sustainable energy in Willoughby city council year Faculty of Engineering course, ENGG1000,
buildings, design of light trapping structure for Engineering Design. The project is to design, build
organic solar cell, boron and phosphorus doping and demonstrate a solar powered “space elevator”
of silicon rich oxide layers by diffusion of dopants, to climb a suspended tether in the final weeks of
solar water pump testing, design of high efficiency the semester. The Prize is awarded to the fastest
organic solar cells, investigation of emitter sheet team. The winners in 2010 were “Captain Planet”,
resistance using photolumescence imaging, team members being Mingyang Jiang, Daniel
aerosol jet etching for silicon solar cells, zero energy Moctezuma-Baker, Syed Syed Ihsan, Edric Verbeek-
buildings, study of temperature and doping effects Martin, Jack Wong and Heng Wu.
on silicon wet-etching, low contact area laser
doping, inverters on unstable grids, on-demand N&M Suntastic Project Prize
street light, electric motorcycle, investigations into This prize, donated by the same sponsors as the
over-floating of laser-doped cells, energy selective N&M Prize, was offered again in 2010. It recognises
contact formed by thermal oxidation, growth of Si best performance in SOLA2052 second year project
QD at low temperature, determination of carrier course. It was won by Jarred Rudman.
diffusion length and saturation current density in
silicon solar cells (I and II generation solar cells), Photovoltaics Prize
zero energy building for China, optics of the
This prize honours best performance in SOLA3540,
heliostat field and how this impacts uniformity
Applied Photovoltaics, and was awarded in 2010 to
at the receiver, novel approach to plating for low
Yan, Xia and Yang, Chuanxi.
cost solar cells, size controlled growth of Ge QD
in SiO2, solar access shading survery of UNSW
Steve Robinson Memorial Prize
Kensington campus, optimisation of rear localised
surface plasmon enhanced light trapping in This prize is in memory of a former PhD student,
polycrystalline silicon thin-film solar cells, hybrid Steve Robinson, and recognises the best
bulk heterojunction solar cells, carrier transport postgraduate performance in SOLA9002, Solar
mechanism in organic semiconductor for different Cells and Systems. It was awarded in 2010 to Pei-
electrode metals for organic solar cells, aerosol jet Chieh Hsiao.
etching for silicon solar cells, PV partial shading and
minimisation of its impact, novel techniques for 5.1.6 Engineers Australia Student of the
low cost high efficiency solar cells, temperature and Year (Sydney Division)
doping effects on silicon wet etching, innovative SPREE student Chris McGrath, won the Student
rear point contact scheme for silicon solar cells, Engineer of the Year Award at the Sydney Division’s
PV and RE at Wellington campus, efficiency Engineering Excellence Awards in September 2010.
prediction on advanced photovoltaic converter Chris finished his studies in 2009 with First Class
concepts, modelling of recombination activities in Honours with a thesis on designing and installing a
polycrystalline silicon thin film solar cells, properties nano- hydro electric power system for the regional
of Sb-doped polycrystalline germanium thin film by primary school in a remote village in Vanuatu. He
a magnetron sputtering method, and boron doping continues to support this work with a 2010 UNSW
of Si-QD materials by diffusion. final year undergraduate student and to promote
The $500 Photovoltaics Thesis Prize for 2010 was further sustainable energy development projects
won by Xiang Zhao and Eleanor Wood was awarded in Vanuatu.
the Renewable Energy Thesis Prize. These prizes
are awarded for the highest marks in the final year 5.1.7 University Medals
thesis subjects. Three students from the School are to be awarded
the University Medal when they graduate in early
2011 following completion of their studies in the
School’s programs in 2010. The students are Xue Bai
and Yu Feng (Photovoltaics and Solar Energy) and
Eleanor Wood (Renewable Energy).

112
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Chris McGrath and Imaki

village workers marking the
completion of the penstock
inlet for the hydro turbine
(Image: D. McGrath).

Engineers Australia
Student of the Year
(Sydney Division)

5.2 Postgraduate Programs

The School offers three postgraduate coursework programs, a Graduate At the end of 2010, the School had 63 students enrolled in postgraduate
Certificate of Engineering Science, a Graduate Diploma of Engineering research degrees, 52 in a Doctor of Philosophy and 11 in Masters
Science and a Master of Engineering Science, and two research by Research. Research topics available to these students cover the
programs, Masters by Research and a Doctor of Philosophy. These entire photovoltaic sector and building energy efficiency, but with
degrees are intended to provide students with an exceptional education greatest emphasis on device theory, design and production, module
in advanced concepts and research in the photovoltaics or other design, balance of system components, and photovoltaic systems
renewable energy areas. and applications. Research students play a pivotal role throughout all
the Centre’s activities. In 2010 the Centre produced six PhDs, and two
The 1.5-year Master of Engineering Science coursework program was
new PhD and one Masters by research are expected to graduate in
developed to build on the prior education of engineers from other
early 2011.
engineering disciplines who are attracted to the photovoltaics and
renewable energy industries. Students study courses chosen from the All the Faculty’s postgraduate coursework programs include an
areas of photovoltaic devices, photovoltaic systems and applications, emphasis on engineering management and all students will be required
energy efficiency and renewable energy technologies. Enrolments in to take some faculty-based “Engineering and Technical Management”
the program have grown significantly from the previous small numbers, courses of general importance. The School introduced its Managing
with total enrolments of 89 at the end of 2010. There were 37 graduates Energy Efficiency (GSOE9017) course into this set of generally available
in 2010 and 21 more are expected in early 2011. courses in 2010 and it proved to be very popular.

5.3 Scholarships
The undergraduate programs at the Centre attract very bright supports study at UNSW by international PhD students (full fees),
students from across the world. The UAI cut off for local students MEngSc coursework students (50% fees), and third and fourth year
for these programs was lifted to 91.7 for the 2010 intake. The Co-Op undergraduate students (100% fees). The undergraduate arrangement
Scholarship Program and the Faculty of Engineering’s Rural Scholarship of shared education requires detailed agreements between UNSW
Program have helped to attract these high-achieving students, while and the selected source universities, which are restricted to China.
the Taste of Research Summer Scholarship Program and occasional These scholarships will address a key need for additional expertise and
casual employment opportunities provide experience in the School’s knowledge in the rapid expansion of the photovoltaics industry in the
laboratories with research leaders and encourage them to pursue region and their award acknowledges UNSW’s importance in the field.
research careers. The Centre’s very high international profile also
In Semester 1 of 2008, these scholarships attracted five undergraduate,
allows it to attract high quality research students, evidenced by the
one postgraduate coursework and three PhD research students to the
number doctoral students undertaking research programs at the
School. In Semester 2 of 2008, these scholarships brought in a further
Centre who have been awarded either of an APA (local students) or IPRS
eight undergraduate, eleven postgraduate coursework and two PhD
(international students) scholarships.
research students. Enrolments for 2009 were much stronger and the APP
students comprised a significant proportion of the year’s intake. Twenty-
5.3.1 Asia-Pacific Partnership on one undergraduate and forty-eight MEngSc APP students commenced
Clean Development and Climate their study in 2010.
The School applied during 2006 for sponsorship funding from the
Agreements, known as “2+2” agreements, have been signed with seven
Australian Government under the Asia-Pacific Partnership on Clean
leading Chinese universities: Sun-Yat-Sen University and South China
Development and Climate (APP) programme to bring students from
University of Science and Technology in Guangzhou; Nankai University
the Asia-Pacific region, specifically China, India and South Korea,
and Tianjin Universities in Tianjin; Zhejiang University in Hangzhou;
to study photovoltaics engineering at UNSW. In January 2007, APP
Nanchang University in Nanchang, Beijing Jiaotong University
Programme investment of $5.2m was approved by the APP Ministers
in Beijingu.
and the first students were able to come in 2008. This significant grant

113
 Six MEngSc students completed the APP program A new sponsor, Infigen, awarded its first Co-Op
at end of Semester 1 2009 and a further ten scholarship and the School’s first for the Renewable
ARC
MEngSc and five undergraduate completed the Energy Engineering Program in 2010.
PHOTOVOLTAICS APP program at end of Semester 2 2009. In 2010,
CENTRE OF 10 MEngSc students completed the APP program 5.3.3 Rural Scholarship Program
EXCELLENCE at end of Semester 1 and 12 at the end of Semester
The Faculty of Engineering established the Rural
2010/11 2 and 8 and 26 BEng students finished their studies
Scholarship Program in 2001 to encourage high-
ANNUAL REPORT in Semesters 1 and 2, respectively. Several of the
achieving students living in rural and isolated areas
graduating students have intended to enter the
to study engineering. The scholarships are valued at
School’s PhD program but the numbers able to
approximately $9,500 per annum for four years of
do so has been restricted by the availability of
full-time study which eases the financial hardship
scholarships.
of relocating to and living in Sydney. Two new rural
scholars began their studies in the School in 2010.
5.3.2 Co-Op Scholarship Program
The Co-Op Program is an industry-linked 5.3.4 Thyne Reid Foundation
scholarship program where students obtain Scholarships
extended work experience with industry sponsors
During 2008 two new undergraduate four-year
as part of their undergraduate studies, through
scholarships were awarded to the School by The
two summer placements, and two six month
Thyne Reid Foundation, one for Photovoltaics and
placements (prior to final year of coursework),
Solar Energy Engineering and one for Renewable
with sponsor companies. In addition to their
Energy Engineering. The two students began
outstanding academic achievements, students
studies in Semester 1 of 2009 and will proceed into
are selected based on their involvement in school
their third year in 2011.
and community activities, their demonstrated
leadership and teamwork skills and their ability
to communicate. Participation in this program
5.3.5 Taste of Research Summer
enables students to apply the knowledge they Scholarship Program
have gained during their studies in real industry The Taste of Research Summer Scholarship Program
projects, as well as opportunities for networking, is primarily for high achieving 3rd year students
recruitment and professional development through and, in exceptional cases, 2nd year students may be
considered. As part of the program, engineering
schools offer 10 week projects for students to
complete during their summer break. These
projects provide students with scholarship support
to gain experience working as part of a research
team, for example in the laboratories of the
Centre. In addition to providing an opportunity
for checking whether research is their ideal career
path, participation in these projects helps students
further develop their technical skills and their
written and oral communication through written
reports and poster presentation.

From December 2009 to February 2010, the

Centre hosted 12 students, taking advantage of
the growing interest in the it’s research programs
among outstanding undergraduates, particularly
from those in this School. Again, in the 2010/2011
summer the Centre was keen to retain for research
training as many as possible of the outstanding
senior undergraduate students and took on 10
Some of the students attracted students through the Taste of Research program.
leadership camps. This program is also highly
to the School by the sponsorship
beneficial to industry sponsors, who have access to
offered by the Australian 5.3.6 Research Scholarships
Commonwealth Department of excellent students, can observe these students in
the Environment, Water, Heritage their workplace, and have the first opportunity of Sowbaranigha Chinnusamy Jayanthi, a student
and the Arts, through the Asia recruiting the best new graduates. from India, was awarded an e8 Sustainable Energy
Pacific Partnership for Clean Development Masters Scholarships for one
Development and Climate. Also Co-Op sponsors currently include, BP Solar, a global academic year so that she may study for a Master of
shown in the group photo are Mr company that has been supplying PV systems to
Demin Zheng and Ms Feng Junying, Engineering Scienceion in Photovoltaics and Solar
Australian markets for 30 years, CSG Solar, a new Energy, at the School. The e8 is an organization
Consuls, Education Office of the
Consulate - General of the Peoples technology startup with an innovative crystalline of leading electricity companies from the global
Republic of China in Sydney, Mr silicon on glass PV product, Suntech Power Co. Ltd., electricity sector that wish to help in protecting
Matthew Hopkins, Department of the leading Chinese photovoltaic manufacturer
Environment, Water, Heritage and
the global environment, and to promote the
and Renewable Energy Corp. (REC), a Norwegian efficient generation and use of electricity. The
Arts and UNSW staff.
based silicon and photovoltaics company that e8 has established a scholarship program to
is establishing photovoltaics manufacturing support outstanding students from developing
in Singapore. countries who plan to study at the Masters level

114
in areas directly related to sustainable energy are much more competitive. Suntech Power also
developmeant. sponsors a PhD scholarship in the Centre. Interest
ARC
in and applications for postgraduate research in
Through thesis projects and scholarships, including PHOTOVOLTAICS
the School boomed during 2009 and the School
the Taste of Research Scholarship program, the CENTRE OF
expects to have four new IPRS recipients begin
Centre encourages some of the best undergraduate EXCELLENCE
their research programs in 2011. The Centre also
students, particularly those from the School, into
has five PhD scholars sponsored by the Australian 2010/11
research. Being an internationally recognised
Government through the Asia-Pacific Partnership on ANNUAL REPORT
research organisation, the Centre attracts very
Clean Development and Climate scheme.
high quality students to its research programs
where students have the opportunity to work
with leaders in this field. The fact that 21 of our
5.3.7 Rhodes Scholarships
doctoral students are currently in receipt of the very The Rhodes Scholarship, arguably the most
competitive Australian Postgraduate Award (APA) prestigious scholarship in the world, is for
for local students or the Australian Government’s postgraduate study at Oxford University. Late
International Postgraduate Research Scholarship in 2010 Alice Lang, a combined undergraduate
(IPRS) for international students is testament to the Photovoltaic Engineering/Arts degrees student
quality of these students. To be awarded an APA, of SPREE with significant research experience in
the student must be a local student with first class the Centre’s laboratories, learned that she would
honours graduate at Bachelor degree level, while become a Rhodes Scholar. Alice will complete her
the IPRS is available to international students and Arts courses in Semester 1 of 2011 and start her
studies at Oxford in Semester 2.

5.4 Educational Resoruces

5.4.1 New Book Translation translation of Silicon Solar Cells: Advanced
A new book in Japanese language, “Fundamentals Principles and Practice” will be published by
and Applications of Solar Cells”, was published Shanghai Jiao Tong University Press in early 2011,
in 2010, including five chapters authored by the bringing to three the number of the Centre’s books
Centre’s Prof. Martin Green. A simplified Chinese available in that language.

5.5 PVSOC
One characteristic of the Centre which has led to House Day (organised by ANZSES and held in
the success of the educational programs is the September) and the PVSOC Annual Dinner. Students
friendly atmosphere that is engendered by being and staff enjoyed this party at Coogee in September
a small (but strongly growing!) school with highly 2010 (held earlier in the year than previously to Alice Lang, Rhodes Scholar for 2011.
motivated students and academic and general avoid distraction from final week assignments and
staff. Students appreciate being able to form exam preparations) where final year students were
friendships and support networks with fellows, farewelled, staff thanked, and the 2011 PVSOC
as well as feeling comfortable and familiar with committee was elected.
academic, technical and administrative staff.
A new aspect of PVSOC’s activities began in
PVSOC (Photovoltaics Society) is a social committee
2010 with the advent of regular consultations by
established by the students which fosters this
the PVSOC President, as a representative of the
atmosphere with organised social events and
student body, with the Head of School and the
activities to encourage student interaction. In 2010,
Undergraduate Coordinator to work with the staff
the committee organised several events, including
to further improve the delivery of education to
barbeques, student participation in Sustainable
the students.

(Left) New Chinese book, “Silicon

Solar Cells: Advanced Principles
and Practice” by M. Green will be
released in early 2011.
(Right) New Japanese book,
“Fundamentals and Applications
of Solar Cells” by M. Yamaguchi, M.
Green, Y. Ohshita and N. Kojima.

115
 5.6 Promotional Activities
ARC
The Centre regularly participates in promotional ceremony with Federal Minister for Climate Change
PHOTOVOLTAICS
activities organised by the Faculty of Engineering and Water Penny Wong and Federal Environment
CENTRE OF and UNSW Student Recruitment. These events are Minister Peter Garrett, Mrs Vivienne Shi, UNSW
EXCELLENCE important for increasing awareness and interest in Chancellor David Gonski, Scientia Professor Martin
2010/11 the Centre’s educational and research programs. Green and Dr Richard Corkish.
ANNUAL REPORT
5.6.1 The Shi’s Family Charitable 5.6.2 Print and Broadcast Media
Foundation and The Sydney Student recruitment advertisements were
Theatre Company also placed in the HSC Survival Magazine, the
The School has been a major participant in the HSC Change of Preference Guide, and the HSC
Sydney Theatre Company’s high-profile “Greening Universities Advisory and Information Day Guide.
the Wharf” project that is improving the water Media mentions of the School and the Centre
efficiency and greenhouse gas impact of the Wharf became more frequent in 2010 as community
Theatre, at Walsh Bay. UNSW, in consultation with interest in climate change and renewable energy
the Shi’s Family Charitable Foundation has agreed issues grew.
that the STC’s “Greening the Wharf” project is a
project suitable to showcase the Pluto technology 5.6.3 UNSW Information Day
as an example of UNSW and Suntech’s research Local undergraduate students must apply for
excellence. The School contributed $2m, donated admission to UNSW programs through the
by the Shi’s Family Charitable Foundation, towards Universities Admissions Centre (UAC), and the first
the establishment of a 380kWp grid-connected week of January is the final opportunity for students
photovoltaic array on the Wharf’s roof, completed to change their preferences for entry to university
and opened in December 2010. The balance of programs. Therefore, the university hosted an
funding for the $4 million project was raised by information day in January 2010 to assist students
Sydney Theatre Company from The Commonwealth obtain information to finalise their preferences.
School administrative and academic staff attended
this event and talked to many prospective students
who were unsure of their career direction. Students
received information on our programs and had
the opportunity to ask questions of staff. A final
round of interviews for the Faculty of Engineering
Admissions Scheme (FEAS) was carried out on the
same day.

5.6.4 UNSW Courses and Careers Day

UNSW Courses and Careers Day is the annual
information day, held in the first week of
September, for prospective students to obtain
information about programs and student life at
the university. As part of this day students and
academic and administrative staff from the School
provided advice and information to prospective
students from information desks in the Roundhouse
and in a marquee in the UNSW Quad, nearby the
Sunsprint model solar car race finals. This year,
the weather was, unfortunately, far from ideal for
the 2010 event. During the day Dr Alistair Sproul
Pictured at the launch of the Wharf presented two public lectures about the School’s
Theatre’s rooftop PV array are, Government’s Green Precincts Fund, NSW
and Centre’s programs as part of the lecture series
(left-right):Dr. Richard Corkish, Government’s Public Facilities Program, corporate
Head of School; Mr George organised centrally by the university.
sponsor Energy Australia and philanthropists.
Maltabarow, Managing Director,
EnergyAustralia; The Hon. Frank The project grew from two sources. Firstly, Dr Shi 5.6.5 Faculty of Engineering Information
Sartor, NSW Minister for Climate Zhengrong, CEO of Suntech Power, expressed
Change and the Environment; Day for High School Students
a wish to make a gift to UNSW and to Sydney
The Hon. Tony Burke MP, Federal Each year the Faculty of Engineering organises
Minister for Sustainability, for helping him start his successful career in
an information day to give high school students
Environment Water, Population photovoltaics and to use it as a promotional tool.
and Communities; Senator The
and their teachers an opportunity to learn about
This led to a study of several prominent sites in
Hon Don Farrell, Parliamentary engineering and the programs offered at UNSW.
Sydney and a pre-feasibility study, carried out as
Secretary for Sustainability and As part of the day students visit three engineering
Urban Water; Cate Blanchett,
a final year thesis project by Samantha Wong. In
schools of their choice and engage in interactive
Artistic co-Director, Sydney Theatre parallel, Cate Blanchett and Andrew Upton, the
activities aimed at demonstrating the relevant
Company; Dr Zhengrong Shi, CEO Artistic Directors of the Sydney Theatre Company,
of Suntech Power Holdings; Mrs engineering area. Once again in 2010 the demand
were seeking opportunities to generate a significant
Vivienne Shi, Director, Shi’s Family to attend this event was so great that two High
part of the Company’s energy from renewable
Charity; Andrew Upton, Artistic co- Schools Days were required, with the School
Director, Sydney Theatre Company. sources. Bringing the two ideas together led to
running several tours.
(Image produced by and used with the official launch in July 2009 of the project at a
permission of AAP).

116
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Some of the students who participated in the UNSW

SunSprint Model Solar Car Challenge.

5.6.6 Honeywell Engineering

Summer School
The Honeywell Engineering Summer School is
an event held in December and is conducted by
Engineers Australia. As part of the summer school
high school students from across NSW and the ACT
about to enter their final year take part in a week of
activities which involves industry visits and lectures
and demonstrations at a number of universities.
In 2010 approximately 30 students visited the
Centre to participate in hands-on activities and
information provision.

5.6.7 Model Solar Vehicle Challenges

SunSprint Model Solar Car Challenge

The SunSprint Model Solar Car Challenge is a
photovoltaics project-based learning event
designed for high school students. In the event,
solar cars are designed and built according The UNSW Faculty of Engineering initiated the
to specifications provided by the Australian- project in 1996 with the Photovoltaics Centre
International Model Solar Challenge (AIMSC), a of Excellence taking over its operation in 1999.
national organization overseeing and linking all of In 2005 the UNSW Faculty of Science joined
the state events. These specifications provide design the Faculty of Engineering and the Centre in
criteria that, when followed, allow students all over jointly sponsoring the event.
Australia to build model solar cars that are uniformly
The 2010 SunSprint Model Solar Car Challenge
competitive with respect to cost and use of
was run on September 3rd and 4th, 2010 under
materials. The students contribute ideas, inspiration,
truly horrible conditions with extraordinary
time and money. The average vehicle costs over
amounts of rain and virtually no sunlight. The
$600 and takes the team over four months to design
winning cars were able to actively compete
and build. Many schools throughout NSW have
using only 10% of the normal amount
allowed final year design and technology students
of sunshine.
to adopt SunSprint as their year-long major project.
High school physics and science teachers have
also used SunSprint as a class-based teaching
project as SunSprint addresses several aspects of
the curriculum.

Only 10% Sunshine and lots of

rain at the UNSW Model Solar
Vehicle Challenges.

117
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

The MiniSprint Competition.

MiniSprint and Model Solar Boats be built in an evening. The kit requires soldering,
The MiniSprint Model Solar Car Challenge gluing and taping.
and the Model Solar Boat Challenge The Model Solar Boat Competition requires much
are offshoot events allowing primary more student design than in the MiniSprint. The
school students to design and build solar boats can be made of any material and can use
vehicles and then experience the thrill of almost any size or shape of solar cells and motors.
competition. The MiniSprint competitors Getting the boat to float right side up and to
start with a kit containing all of the competitively move through the water is quite
important parts of the solar car and can a feat.

Preparations for the Model Solar

Boat Competition. 5.7 Educational
Collaborations
A range of educational collaborations have similar agreements were signed with additional
been established between the Centre two Chinese universities. The two new partner
and other educational institutions and universities are Tianjin University (two agreements,
organisations. These collaborations involve each with a different school) and Zhejiang
the development and implementation of University, Hangzhou.
educational programs and courses, the
A new agreement has now been made in 2009
provision of support for student projects and
with Nanchang University, in Nanchang, which has
theses, and the exchange of students.
formed a new School of Photovoltaic Engineering
and an additional “2+2” agreement was signed in
5.7.1 “2+2” Agreements early 2010 with South China University of Science
The Model Solar Boat Competition. In 2010 the School continued collaborations and Technology, in Guangzhou.
with four Chinese Universities for third
and fourth year students from those 5.7.2 Thyne Reid Foundation
universities to study Photovoltaics and
The Thyne Reid Foundation concluded in 2008
Solar Energy Engineering at UNSW and
an agreement to sponsor the studies of two
for Masters by coursework study with Asia
undergraduate students for four years of their
Pacific Partnership sponsorship. “2+2”
studies, one in Photovoltaics and Solar Energy
agreements were signed in 2007 with two
Engineering and one in Renewable Energy
prestigious Chinese universities, Nankai
Engineering. The students began their studies in the
University in Tianjin and Sun Yat-Sen
first semester of 2009 and finished their second year
(Zhongshan) University in Guangzhou. The
in 2010.
first students under this scheme started at
UNSW in Semester1 of 2008. During 2008

Trophies for primary school

participants.

118
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

The project team and Imaki

village volunteers preparing
the long trench for the power
cables from the turbine to the
village. (Image produced by
and used with permission of
E. Tehan).

5.7.3 Cundall, Infigen, Norcoast industrial training or in “Taste of Research” projects

Refrigeration and Others at Suntech and postgraduate students often travel
there to use vital research equipment. Additionally,
Cundall, a sustainable design consulting company,
Suntech sponsored an undergraduate Cooperative
and the Darcy and Patricia Wentworth Fund joined
Scholarship and two PhD scholarships and has
forces to financially support a final year thesis
made significant financial and in-kind contributions
project to bring hydro power to a remote village
to UNSW to support photovoltaics research.
school in Vanuatu, mentioned in Sec. 5.1.The
student undertaking this project, Eden Tehan, was
strongly supported by an alumnus, Chris McGrath, 5.7.5 Singapore Polytechnic, Ngee Ann
who started the project in 2009. In 2010 the project Polytechnic and Temasek Polytechnic
was extended to include the secondary (boarding) The Polytechnics in Singapore offer three-year
school, medical clinic and church in an integrated diploma programs, including in engineering.
power system supplied by two hydro turbines During 2007 the School arranged for future
and including a new low-power electric vaccine articulation of graduates from the new clean energy
refigerator. Donors for the project extension in 2010 diploma of Clean Energy program at Singapore
included Cundall, Infigen, Norcoast Refrigeration, Polytechnic into either of the School’s Bachelor of
and several private donors: John Tehan, Susan Engineering programs. In 2008 the School set up
McParland, Sean Tehan, Don and Marg McGrath, a similar articulation arrangement with Ngee Ann
Tyler Troy, Helena Colnaric, Leah Greengarten, Polytechnic and Temasek Polytechnic was included
Athron McCann, Peter Baker, Noam Field and David in 2009. Significant numbers of students may flow
Gazal. Eden and Chris also raised $4,500 for the in future years from these institutions, once the first
project at a Vanuatu Shakedown concert, thanks groups of students make their way through the
to donated performances by “The Bakery”, “The new programs, fuelling Singapore’s drive to become
Liberators” and Emma Davis. David Thompson a major photovoltaics manufacturer. Richard
of Gwynneville Electrical Services was the team’s Corkish has accepted in 2010 the role of External
electrician during the first trip, paying his own Examiner for the Diploma in Clean Energy program
way and donating a week of work as well as many Singapore Polytechnic.
electrical components. Dave Selke of All-round
Supplies, Sydney, was a very helpful tradesman who 5.7.6 Darcy Wenworth
also donated some vital electrical components.
Darcy Wentworth donated his services through
Samantha Paver, Sasha Siljanovic and Sean Tehan all
2010 as a visiting academic, supervising
contributed their time to the project.
undergraduate student projects and coordinating
two subjects.
5.7.4 Suntech Power Corporation
Suntech, a partly Australian-owned company 5.7.7 Kema Consulting
operating in China, has been actively involved in
Dr Hanzheng Duo, a practicing Engineer at Kema
assisting UNSW with the development, testing
Consulting, in his role as visiting academic, donated
and evaluation of educational material in the
his time and industrial project experience to
PV area such as in the development, testing and
supervise and assess an undergraduate thesis
evaluation of a teaching software package called
project through 2010.
the Virtual Production Line. Significant numbers
of UNSW undergraduate students participate in

119
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

6. ORGANISATIONAL
STRUCTURE
When the Key Centre for Photovoltaic Engineering account the geographical separation of its members
was awarded in 1999, it was established as (Australia, USA, China, Germany, Italy, Spain),
an autonomous Centre within the Faculty of although more frequent correspondence takes
Engineering of UNSW, becoming independent place when necessary with individual committee
from the School of Electrical Engineering where members either by email or telephone. The
UNSW photovoltaic activities had been previously membership of the Advisory Committee during
located. At that time, the Key Centre was given 2010 has included:
the same operational independence and rights as
• Professor Stuart Wenham, Centre Director
the other Engineering schools, with the Director
• Professor Martin Green, Centre Executive
Stuart Wenham being given the same status and
Research Director
authority as other Heads of Schools. With the
• Associate Professor Gavin Conibeer, Deputy
awarding of the Centre of Excellence in 2003,
Director (3rd Generation PV Research)
the University committed to the Key Centre for
• Associate Professor Thorsten Trupke, Deputy
Photovoltaic Engineering becoming the School Director (Silicon Photonics)
of Photovoltaic Engineering, within which the • Dr Sergey Varlamov, Deputy Director (2nd
ARC Photovoltaics Centre of Excellence would be Generation PV research)
located. This transition occurred in January 2006. • Professor Allen Barnett: world leading academic
The corresponding organisational relationships are in the field; recipient of many prestigious
shown in the upper part of Fig. 6.1. international prizes such as the William Cherry
Award; former President of AstroPower, one of the
Within the Centre of Excellence, Deputy Directors
world’s largest solar cell manufacturers, before
have been appointed for the major strands of being purchased by GE.
research as shown. The Management Committee
• Prof. Andres Cuevas, ANU;
of the Centre comprises the six Directors and
• Dr Francesca Ferrazza, Chief Scientific Officer,
Deputy Directors, along with the Head of School Enitechnologie;
and the Business and Operations Manager. This • David Hogg: COO of Suntech Power; until recently
committee meets fortnightly on the 1st and 3rd CEO of CSG Solar AC, Germany; former CEO of
Fridays of each month, with each Deputy Director Pacific Solar, Australia;
giving a report on the activities in his or her area • Mr. David Jordan, until recently, Manager of New
over the preceding fortnight. This committee Technology, BP Solar International;
takes responsibility for decision making within • Dr Larry Kazmerski, until recently Head,
the Centre that affects the Centre as a whole, Photovoltaic Division, US National Renewable
while the individual Deputy Directors receive their Energy Laboratory;
own annual budgets to allow them to make and • Prof. Antonio Luque, Head, ESTI, Polytechnical
implement decisions that impact only their own University of Madrid; Leading European academic
laboratory areas and research activities. in the photovoltaic field; founder of Isofoton, one
of Europe’s most established manufacturers;
The Advisory Committee for the Centre comprises • Dr Zhengrong Shi, CEO of Suntech-Power, the
the Centre Directors, the Head of School for world’s largest silicon cell manufacturer; recipient
Photovoltaic Engineering, leading academics from of various industry prizes; former Deputy Research
other institutions, industry leaders such as CEO’s Director of Pacific Solar; major collaborator of
of various companies involved in the field, and the Centre;.
research leaders. This committee provides high level • Prof. Peter Würfel, Institut fur Angewandte Physik
advice, feedback and recommendations in relation der Universität Karlsruhe.
to the Centre’s activities and their relevance. The
Advisory committee meets annually, taking into

120
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Centre organisational structure.

Figure 6.1

Other distinguished researchers and industrialists purchasing of appropriate new infrastructure

have been co-opted to the Committee as required. and equipment.

Also within the Management structure and a The Director works closely with the Centre’s
member of the Centre’s Management Committee is Executive Research Director to assist in setting
the Centre’s Business, Technology and Operations high level research direction and priorities for
Manager Mark Silver, given the extensive the Centre’s programs. In the teaching area, with
involvement of Centre staff with industry, large the Key Centre for Photovoltaic Engineering
number of collaborative research projects, and the and its activities being incorporated into the
high level of success of the Centre in generating, Centre of Excellence, the Director also takes
marketing and commercialising technology. The significant responsibility in ensuring the successful
Head of Administrative Support for the Centre development and implementation of the Centre’s
during 2010 was Joyce Ho. Not shown in the educational programs, both at undergraduate and
Management Structure is the Financial Officer postgraduate levels. In this area, the Director works
Julie Kwan who looks after the Centre Accounts, closely with Dr Richard Corkish, the Head of School.
interfaces with the University Financial System,
advises the Centre Directors and Deputy Directors
on budgeting and financial matters, and generally
assists with sourcing of funding, purchasing,
reporting, writing grant applications, and advising
on business opportunities.

A primary responsibility of the Director in the

Centre is to ensure the Management structure as
shown functions efficiently and effectively, with
appropriate support and resources to facilitate
the Centre’s achievement of its milestones and
performance targets. An important aspect of this is
managing the Centre’s finances to ensure suitable
levels of funding are made available for supporting
the major research strands, while simultaneously
ensuring adequate funding is available for repair
and maintenance of existing facilities and the

121
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

7. FINANCIAL
SUMMARY Electric, Schlumberger and ExxonMobil
for the Global Climate & Energy Project,
administered by the Stanford University,
The total income for 2010 for the Australian who contributed $0.61 million. The next
Research Council (ARC) Photovoltaics Centre of three largest cash contributions were from CSG
Excellence was $6.8 million. This does not include Solar from Australia, Shinsung Holdings from Korea
over $2 million from the Host Institution based on and Guodian Solar from China, each contributing
EFTSU and related income. This income has been about $0.3 million. Other cash and in-kind
generated by the Centre through its educational contributions have been made by a large number
activities that have formed the basis of a School of industry collaborators and contributors of whom
within the Faculty of Engineering. This income has the most significant have been Hyundi Heavy
been used for the development of new courses and Industries, the European Commission, Renewable
teaching materials and to fund the salaries of most Energy Corporation (REC), BP Solar, CEEG Nanjing
of the academic staff associated with the Centre. PV Tech, Toyota Motor Group, E-ton Solar, Silex, Roth
The largest two components of Centre income in & Rau, Global Sunrise Energy, Infigen Energy, the
terms of cash contributions were from the ARC Australian Academy of Science and Advent Solar. In
with $2.1 million for the Centre Grant and industry addition, many other companies contribute to the
with $2.0 million. The former was complemented Centre through the expertise and experience they
by a further $0.49 million in ARC funding for other offer in collaborative research areas and the in-kind
grants. However, for the latter, in-kind contributions support they provide through access to equipment,
for collaborative research add a further almost facilities and personnel not available at UNSW.
$1 million, effectively making industry funds the The third largest cash component of income was
largest income source for the Centre for the fourth Host Institution support. UNSW contributed $1.0
consecutive year. With the booming photovoltaic million not including additional Host Institution
industry, the success of Centre PV technology and support in the form of EFTSU and related income.
the high demand for companies wishing to work
with the Centre, this trend is expected to continue. The next largest cash component of income was
Figure 7.2 shows a more detailed breakdown of $0.70 million from the State Government, primarily
income derived from industry and related sources. through the Department of State and Regional
Development (DSRD). This funding has been
The largest industry contributor was again Suntech- particularly important for the purchase of new
Power, but with the majority of the $0.63 million equipment and the development of facilities.
being in-kind contributions resulting from most
of the collaborative research being carried out at Other Centre contracts, consulting work and
the premises of Suntech in China. The largest cash technology transfers are conducted through
contributor was the consortium of Toyota, General NewSouth Innovations, the commercial arm of
the university. These are handled on behalf of

122
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

Sources of Centre income.

Figure 7.1

the Centre of Excellence and its staff through

the NewSouth Innovations accounts and not
included in this financial report.

The Centre also earns income through the

sale of educational CDs, books and computer
software and running short courses.

Centre 2010 Expenditure of ARC Grant was $2.3

million as shown in Figure 7.3, approximately
$200k more than the corresponding income.
By far the largest component of expenditure
was for salaries and scholarships with $1.4
million, including for research appointments,
technical support, administrative staff, financial
management, specialised consultants and
facility cleaning. Appointments include full-
time, part-time and casual.

Equipment expenditure, mainly for small items 2010 industry income to the ARC
or small contributions leveraged to purchase Photovoltaics Centre of Excellence.
large items, was $101k from the ARC Grant. With
Figure 7.2
the large amount of industry funded research,
travel expenditure was quite significant at
$145k, with several staff and students making
many trips to industry partners for industry
collaborative research and technology transfers.
Other travel included attendance and paper
presentation at international conferences.

Another major component of Centre

expenditure from the ARC grant is on
consumables and maintenance with $633K.
These costs are strongly dominated by
laboratory consumables to support the device
research such as high purity gases, chemicals
and general laboratory supplies.

Expenditure of the ARC grant by

expenditure category.

Figure 7.3

123
 ARC
PHOTOVOLTAICS
CENTRE OF
EXCELLENCE
2010/11
ANNUAL REPORT

8. PUBLICATIONS
8.1 BOOK
Y. Yamaguchi, Martin A Green, Y. Ohshita and N. Kojima, “Fundamentals and Applications of Solar Cells”, ISBN 978-4-621-08211-9, 231pp (Maruzen, Japan, 2010)

8.2 BOOK CHAPTERS

G. Conibeer, “Up and Down Conversion for R. Corkish, “Solar Energy”, in Mohd. Nordin Hassan D. König, “Photovoltaic Device Physics at the
Photovoltaics” in “Physics of Solar cells” Ed. Badescu, V and Sukanta Roy (eds.), “Sustainable Energy in Asia Nanoscale”, Chapter 3 (pp. 73 - 146) in L. Tsakalakos
& Paulescu, M, accepted after review April 2009 (Nova and the Pacific”, Akademi Sains Malaysia, 2010 ISBN: (Ed.), Nanotechnology for Photovoltaics, Taylor &
Science Publishers, July 2010). 978.983-9445-43-5 (invited). Francis, April 2010, ISBN-13: 9781420076745 [Amazon
Books bestsellers rank 680,938] (invited).
G. Conibeer, “Applications of Si Nanocrystals in
Photovoltaic Solar Cells” in “Silicon Nanocrystals:
Physics and Applications”, Ed. Rasit Turan and Lorenzo
Pavesi (Wiley VCH, 2010).

8.3 patent applications

I. Brazil and A. Lennon, “Method for Texturing D. König, “Acceptor Modulation Doping of L. Mai, A. Wenham and S.R. Wenham, “Solar Cell
Surfaces”, International Patent Application PCT/ Nanocrystals and Quantum Structures, International Methods and Structures”, International Patent
AU2010/001208, filed on 16 September 2010. Patent Application PCT/AU2010/001043, filed on 13 Application PCT/AU2010/000036, filed on 15 January
August 2010. 2010.
G. Conibeer, S. Shrestha, D. Konig and M.A. Green,
“Hot Carrier Energy Conversion Structure and D. König, “Donor Modulation Doping of Quantum L. Mai, A. Wenham and S.R. Wenham, “Photoplating
Method of Fabricating the Same”, International Structures”, International Patent Application PCT/ of Metal Electrodes for Solar Cells”, International
Patent Application PCT/AU2010/000848, filed on 2 AU2010/001042, filed on 13 August 2010. Patent Application PCT/AU2010/001460, filed on 3
July 2010. November 2010.
A. Lennon, D. Lu and Y. Chen, “Metal Contact Scheme
R. Corkish, B. Jaya, Y. Augarten, C. Rizos and L. Ge, for Solar Cell”, Australian Provisional Application C-Y. Tsao, X.J. Hao and M.A. Green, “A Method
“Method to Detect Faults in Photovoltaic Arrays 2010902694, filed on 2 July 2010. of Forming a Germanium Layer on a Silicon
by Observation of Electroluminescence Signal”, IP Substrate and a Photovoltaic Device Including a
Notification to New South Innovations, 1 July 2010. A. Lennon, S.R. Wenham and A. Ho-Baillie, “Method Germanium Layer”, Australian Provisional Application
of Patterned Etching of Selected Material”, National 2010905488, filed on 15 December 2010.
M.A. Green,“Improved Scheme for Light-Trapping Phase Entry PCT/AU2009/000098 filed on 1
in Organic Photovoltaic and Related Thin-Film August 2010. S.R. Wenham, B. Tjahjono and L. Mai, “Photovoltaic
Optoelectronic Devices”, Australian Provisional Device Structure and Method”, International Patent
Application 2010900576, filed on 12 February 2010. A. Lennon and S.R. Wenham, “A Method for the Application PCT/AU2010/000145, filed on 11
Selective Delivery of Material to a Substrate”, February 2010.
M.A. Green, “Si Cells with III-V Multijunction Cells”, International Patent Application PCT/
Australian Provisional Application 2010900574, filed AU2010/001001, filed on 6 August 2010. S.R. Wenham, B. Tjahjono, A. Lennon and N. Kuepper,
on 12 February 2010. “Improved Metallization Method for Silicon Solar
A. Lennon, Y. Yao, K. Chin and D. Coyle, “Method and Cells”, International Patent Application PCT/
M.A. Green, A. Wenham and S.R. Wenham, “Improved Apparatus for Light Induced Plating of Solar Cells”, AU2010/001421, filed on 25 October 2010.
Metal Adhesion”, International Patent Application Australian Provisional Application 2010903869, filed
PCT/AU2010/000215, filed on 24 February 2010. on 27 August 2010. S.R. Wenham, B. Tjahjono, A. Lennon and Y. Yao,
“Method and Apparatus for Light Induced Plating
P. Hamer and S.R. Wenham, “Method of Fabricating of Solar Cells”, International Patent Application PCT/
a P-N Junction”, Australian Provisional Application AU2010/001461, filed on 3 November 2010.
2010905292, filed on 1 December 2010.

124
8.4 Papers in refereed Scientific and
Technical Journals
P. Aliberti, S.K. Shrestha, R. Teuscher, B. Zhang, M.A. D. Di, I. Perez-Wurfl, A. Gentle, D-H. Kim, X. Hao, S. He, J. Jannsens, J. Wong and A. Sproul, “The
Green and G.J. Conibeer, ”Study of Silicon Quantum L. Shi, G. Conibeer and M.A. Green, “Impacts of Influence of Base Doping Density on the Performance
Dots in a SiO2 Matrix for Energy Selective Contacts Post-metallisation Processes on the Electrical and of Evaporated Poly-Si Thin-Film Solar Cells by
Applications”, Solar Energy Materials & Solar Cells, Vol. Photovoltaic Properties of Si Quantum Dot Solar Solid-Phase Epitaxy”, Thin Solid Films, Vol. 519, pp.
94, pp. 1936-1941, 2010. Cells”, Nanoscale Research Letters, Vol. 5 (11), pp. 475-478, 2010.
1762-1767, 2010.
P. Aliberti, Y. Feng, Y. Takeda, S.K. Shrestha, M.A. S. He and A. Sproul, “Optical Properties of Evaporated
Green and G. Conibeer, “Investigation of Theoretical D. Di, I. Perez-Wurfl, G. Conibeer and M. A. Green, Poly-Si Thin-Films on Glass”, Thin Solid Films, Vol. 519,
Efficiency Limit of Hot Carriers Solar Cells with a Bulk “Formation and Photoluminescence of Si Quantum pp. 351-356, 2010.
Indium Nitride Absorber”, Journal of Applied Physics, Dots in SiO2/Si3N4 Hybrid Matrix for All-Si Tandem
Vol. 108, 094507 (10 pp.), 2010. Solar Cells”, Solar Energy Materials and Solar Cells, Vol. S. He, J. Wong, D. Inns, B. Hoex, A. Aberle and A.
94 , pp. 2238-2243, December 2010. Sproul, “Influences of Oxygen Contamination on
Y. Augarten, T. Trupke, M. Lenio, J. Bauer, J. Weber, Evaporated Poly-Si Thin-Film Solar Cells by Solid-
M. Juhl and O. Breitenstein, “Calculation of C. Flynn, D. König, I. Perez-Wurfl, M.A. Green and G. Phase Epitaxy”, Thin Solid Films, Vol. 518, pp. 4351-
Quantitative Shunt Values Using Photoluminescence Conibeer, “Correlation between Fixed Charge and 4355, 2010.
Imaging”, Progress in Photovoltaics (re-submitted Capacitance Peaks in Silicon Nanocrystal Metal-
January 2011). Insulator-Semiconductor Devices”, Semiconductor G. Jin, P. I. Widenborg, P. Campbell and S. Varlamov,
Science and Technology, Vol. 25, 045011 (8pp), 2010. “Lambertian Matched Absorption Enhancement
G. Barber, T. Mallouk, S. Zakeer, M. Graetzel, M.A. in PECVD Poly-Si Thin Film on Aluminium
Green, A. Lee, P.G. Hoertz, N. Abrams, J. Mikulca and P. C. Flynn, D. König, M.A. Green and G. Conibeer, Induced Textured Glass Superstrates for Solar Cell
Liskat, “Tandem Hybrid Concentrator Using Dye Cells “Modelling of Metal-Insulator-Semiconductor Applications”, Progress in Photovoltaics: Research and
and Silicon”, Journal of Physical Chemistry Letters, Vol. Devices Featuring a Silicon Quantum Well”, Physica E, Application, Vol. 18, pp. 582-589, 2010.
2, pp. 518-585. 2011. Vol. 42, pp. 2211-2217, 2010.
D. König, K. Casalenuovo, Y. Takeda, G. Conibeer,
G. Barber, S-H. Lee, P. Liska, Paul, M. Graetzel, S.M. M.A. Green and A. Ho, “43% Composite Split- J.F. Guillemoles, R. Patterson, L.M. Huang and M.A.
Zakeeruddin, M.A. Green, A. Ho-Baillie, J. Mikulca, Spectrum Concentrator Solar Cell Efficiency”, Progress Green, “Hot Carrier Solar Cells: Principles, Design and
T. Mallouk, P. Hoertz and N. Abrams, “Utilization of in Photovoltaics: Research and Applications, Vol. 18, Materials”, Physica E: Low-Dimensional Systems and
Direct and Diffuse Sunlight in a Dye-Sensitized Solar pp. 42-47, 2010. Nanostructures, Vol. 42, pp. 2862-2866, 2010.
Cell – Silicon Photovoltaic Hybrid Concentrator
System”, Journal of Physical Chemistry Letters M.A. Green, K. Emery, Y. Hishikawa and W. Warta, A. Lahreche, Y. Beggah and R. Corkish, “The Effect
(submitted 9 February 2011). “Solar Cell Efficiency Tables (Version 35), Progress in of Electron Range on Electron Beam Induced
Photovoltaics, Vol. 18, pp.144-150, 2010. Current Collection and a Simple Method to Extract
I. Brazil and M.A. Green, “Measuring Strain Changes an Electron Range for Any Generation Function”,
during Production of Thin Film Crystalline Silicon M.A. Green, K. Emery, Y. Hishikawa and W. Warta, Ultramicroscopy (submitted July 2010).
on Glass Photovoltaic Modules”, Journal of Materials “Solar Cell Efficiency Tables (Version 36), Progress in
Science: Materials in Electronics, Vol. 21, No. 11, pp. Photovoltaics, 2010, Vol. 18 (5), pp. 346-352, 2010. M. Lenio, A. J. Lennon, A. Ho-Baillie and S. R. Wenham.
1207-1211, 2010. “Effect of Electroless Nickel on the Series Resistance
M.A. Green, “Enhanced Evanescent Mode Light of High-Efficiency Inkjet Printed Passivated Emitter
I. Brazil and M.A. Green, “Investigating Polysilicon Trapping in Organic Solar Cells and Other Low Index Rear Contacted Solar Cells”. Solar Energy Materials
Thin Film Structural Changes during Rapid Thermal Optoelectronic Devices”, Progress in Photovoltaics, and Solar Cells, Vol. 94 (12), pp. 2102-2107,
Annealing of a Thin Film Crystalline Silicon on Glass Vol. 19, pp. 473-477, 2011. December 2010.
Solar Cell”, Journal of Materials Science: Materials in M.A. Green, “Learning Experience for Thin-Film Solar
Electronics, Vol. 21 (10), pp. 994-999, 2010. D. Lin, L. Ma, G. Conibeer and I. Perez-Wurfl,
Modules: First Solar, Inc. Case Study”, Progress in “Study on electrical properties of Si quantum dots
R. Clady, M.J.Y. Tayebjee, P. Aliberti, D. König, N.J. Photovoltaics, Vol. 19, pp. 498-500, 2011. based materials”, Physica Status Solidi (accepted
Ekins-Daukes, G.J. Conibeer, T.W. Schmidt and M.A. M.A. Green, “Is Sour Crude or Sour Gas a Potential September 2010).
Green, “Femtosecond Investigation of High Energy Source of Se and Te?”, Progress in Photovoltaics
Carrier Relaxation Processes in GaAs and InP for L. Ma, D. Lin, G. Conibeer and I. Perez-Wurfl,
(accepted 1 February 2011). “Introducing Dopants by Diffusion to Improve the
Hot Carrier Solar Cell Applications”, Progress in
Photovoltaics (accepted October 2010). M.A. Green, “Analytical Treatment of Trivich-Flinn Conductivity of Silicon Quantum Dot Materials in 3rd
and Shockley-Queisser Photovoltaic Efficiency Limits Generation Photovoltaic Device”, Physica Status Solidi
R. Clady, T. Schmidt, M. Tayebjee, P. Aliberti, G. Using Polylogarithms” Progress in Photovoltaics: (accepted September 2010).
Conibeer, D. Konig, M.A. Green and N.J. Ekins-Daukes, Research and Applications (accepted 11 March 2011).
“Interplay Between the Hot Phonon Effect and B. Mitchell, T. Trupke, J. Weber and J. Nyhus, “Bulk
Intervalley Scattering on the Cooling Rate of Hot P.J. Gress, P.I. Widenborg, S. Varlamov and A.G. Aberle, Minority Carrier Lifetimes and Doping of Silicon
Carrier in GaAs and InP”, Progress in Photovoltaics: “Wire Bonding As a Cell Interconnection Technique Bricks from Photoluminescence Intensity Ratios”,
Research and Applications (accepted 10 March 2011). for Polycrystalline Silicon Thin-film Solar Cells on Journal of Applied Physics (accepted March 2011).
Glass”, Progress in Photovoltaics: Research and E.J. Mitchell, S. Lindekugel, M. Künle, K. Schillinger,
G. Conibeer, M.A. Green, D. Konig, I. Perez-Wurfl, S. Applications, Vol. 18, pp. 221-228, 2010.
Huang, X. Hao, D. Di, L. Shi, S. Shrestha, B. Puthen- S. Janz and S. Reber, “c-Si Wafer-equivalent Epitaxial
Veettil, Y. So, B. Zhang and Z. Wan, “Silicon Quantum Z. Hameiri, L. Mai, A.B. Sproul and S.R. Wenham, Thin-film Solar Cells on Isolating Substrates”, Solar
Dot Based Solar Cells: Addressing the Issues of “18.7% Efficient Laser Doped Solar Cell on CZ p-type Energy Materials and Solar Cells, Vol. 95, Issue 4, pp.
Doping, Voltage and Current Transport”, Progress in Silicon”, Applied Physics Letters, Vol. 97, pp. 222111- 1163-1167, 2011 (doi:10.1016/i.solmat.2010.12.046).
Photovoltaics: Res. Appl. 18 (2010) In press, Published 222113, 2010. E.J. Mitchell, N. Brinkmann and S. Reber, “Progress
online in Wiley Online Library (wileyonlinelibrary. with Epitaxy Wrap-through Crystalline Silicon Thin-
Z. Hameiri, L. Mai and S.R. Wenham, “Advantages of
com). DOI: 10.1002/pip.1045 (accepted film Solar Cells”, Progress in Photovoltaics: Research
Photoplating for Laser Doped Solar Cells”, Progress
September 2010). and Applications (accepted December 2010).
in Photovoltaics: Research and Applications (DOI:
G. Conibeer and S. Huang, “International Journal of 10.1002/pip.1072), 2010. Z. Ouyang, O. Kunz, A. B. Sproul and S. Varlamov,
Nanoparticles”, ed Chang Hui Ye, Nanomaterials for “Influence of the Absorber Doping for p-type
Z. Hameiri, T. Puzzer, L Mai, A.B. Sproul and S.R.
Third Generation Photovoltaics (in press), 2010. Evaporated Polycrystalline Silicon Thin-Film Solar
Wenham, “Laser Induced Defects in Laser Doped
G. Conibeer, R. Patterson, L. Huang, J-F. Guillemoles, Solar Cells”, Progress in Photovoltaics: Research and Cells On Glass, Journal of Applied Physics, Vol. 109,
D. König, S. Shrestha and M.A. Green, “Modelling Applications (DOI: 10.1002/pip.1043), 2010. p.060104, 2011.
of Hot Carrier Solar Cell Absorbers”, Solar Energy Z. Ouyang, S. Pillai, F. Beck, O. Kunz, S. Varlamov, K.
Z. Hameiri, T. Puzzer, L Mai and S.R. Wenham, “SR.
Materials and Solar Cells, Vol. 94, pp. 1516-1521, 2010 Catchpole, P. Campbell and M.A. Green, “Effective
Influence of Laser Power on the Properties of Laser
(doi:10.1016/j.solmat.2010.01.018). Light Trapping in Polycrystalline Silicon Thin-Film
Doped Solar Cells”, Solar Energy Materials and Solar
Richard Corkish, Anna Bruce and Deo Prasad, Cells, 95, pp. 1085-1094, 2011. Solar Cells by Means of Rear Localised Surface
“Integrated Solar Photovoltaics for Buildings”, Review Plasmons”, Applied Physics Letters, Vol. 96, p.
of Architecture and Building Science (ISSN: 1225- 261109, 2010.
1666), Vol. 54 (1), pp.86-92, 2010.

125
 Z. Ouyang, X. Zhao, S. Varlamov, Y. Tao, J. Wong C.Y. Tsao; J. W. Weber; P. Campbell; G. Conibeer; D. K. S. Wang, B. S. Tjahjono, J. Wong, A. Uddin and S.
and S. Pillai, “Nanoparticles Enhanced Light- Song and M.A. Green, “In Situ Low Temperature R. Wenham, “Sheet Resistance Charaacterization
Trapping in Thin-Film Silicon Solar Cells”, Progress in Growth of Poly-Crystalline Germanium Thin Film of Laser-dope on Crystalline Silicon Wafers for
Photovoltaics (submitted November 2010). on Glass by RF Magnetron Sputtering”, Solar Energy Photovoltaic Applications”, Solar Energy materials
Materials and Solar Cells, Vol. 94, pp. 1501-1505, 2010. and Solar Cells, Vol. 95, pp. 974-980, 2011.
R. Patterson, M. Kirkengen, B. Puthen-Veettil, D.
Konig, M.A. Green and G. Conibeer, “Phonon Lifetimes C.Y. Tsao, J. Wong, J-L. Huang; P. Campbell, D. Song K. S. Wang, B. S. Tjahjono, A. Uddin and S. R. Wenham,
in Model Quantum Dot Superlattice Systems with and M.A. Green, “Structural Dependence of Electrical “A Method to Characterize the Sheet Resistance of
Applications to the Hot Carrier Solar Cell”, Solar Properties of Ge Films Prepared by RF Magnetron a Laser Doped Line on Crystalline Silicon Wafers for
Energy Materials & Solar Cells, Vol. 94, pp. 1931- Sputtering”, Applied Physics A, Vol. 102, pp. 689- Photovoltaic Applications”, Applied Physics Letters,
1935, 2010. 694, 2010. Vol. 98, pp. 094105 (1-3), 2011.

S. Pillai and M.A. Green, “Plasmonics for Photovoltaic C.Y. Tsao, P. Campbell, D. Song and M.A. Green, J. Wong, J. L. Huang, S. Varlamov, M. A. Green,
Applications”, Solar Energy Materials and Solar Cells, “Influence of Hydrogen on Structural and Optical R. Evans, M. Keevers and R. Egan, “Structural
Vol. 94, pp. 1481-1486, 2010. Properties of Low Temperature Polycrystalline Ge Inhomogeneities in Polycrystalline Silicon On
Films Deposited by RF Magnetron Sputtering”, Glass Solar Cells and Their Effects on Device
S. Pillai, I. Perez-Wurfl, G.J. Conibeer and M.A. Green, Journal of Crystal Growth, Vol. 312 (9), pp. 2647- Characteristics”, Progress in Photovoltaics: Research
“Surface Plasmons for Improving the Performance of 2655, 2010. and Applications, DOI: 10. 1002/pip.1089, 2011.
Quantum Dot Structures for Third Generation Solar
Cell Applications”, Physica Status Solidi, Vol. 8, pp. C.Y. Tsao, J.L. Huang, X.J. Hao, P. Campbell and J. Wong, J.L. Huang, B. Eggleston, M.A. Green, O. Kunz,
181-184, 2011. M.A. Green, “Heavily Boron-Doped Hydrogenated R. Evans, M. Keevers and R.J. Egan, “Lifetime Limiting
Polycrystalline Ge Thin Films Prepared by Recombination Pathway in Thin-Film Polycrystalline
B. Puthen-Veettil, D. König, G. Conibeer and M.A. Cosputtering”, Electrochemical and Solid-State Silicon on Glass Solar Cells” Journal of Applied
Green, “Impact of Disorder in Double Barrier QD Letters, Vol. 13 (10), pp. H354-H356, 2010. Physics, Vol. 107, p.123705, 2010.
Structures on Energy Selectivity Investigated by Two
Dimensional Effective Mass Approximation”, Energy T. Trupke, J, Nyhus and J.Haunschild, “Luminescence Y. Yao, A. Sugianto, A. Lennon, B. Tjahjono and S.
Procedia, Vol. 2, p. 213, 2010. Imaging for Inline Characterisation in Silicon Wenham, “Use of ICP-MS and OES Measurements
Photovoltaics”, Phys.Stat.Sol. (RRL) (published online to Characterise Light-induced Plating for Silicon
Santosh K. Shrestha, Pasquale Aliberti and Gavin J. 2 February 2011). Solar Cells”, Solar Energy Materials and Solar Cells
Conibeer, “Energy Selective Contacts for Hot Carrier (submitted January 2011).
Solar Cells”, Solar Energy Materials and Solar Cells, Vol. A. Uddin, C. B. Lee and T. G. Anderson, “Initial Rise of
94, pp. 1546-1550, September 2010. Transient Electroluminescence in Doped Alq3 Films”, G. Zatryb, A. Podhorodecki, X.J. Hao, J. Misiewicz,
Physica Status Solidi A, Vol. 207, No. 10, pp. 2334- Y.S. Shen and M.A. Green, “Quantitative Evaluation
R.A Sinton and T. Trupke, “Limitations on Dynamic 2338, 2010 of Boron-Induced Disorder in Multilayers Containing
Excess-Carrier Lifetime Calibration Methods”, Progress Silicon Nanocrystals in an Oxide Matrix Designed for
in Photovoltaics (accepted March 2011). A. Uddin and C. B. Lee, “Exciton Behaviour in Doped Photovoltaic Applications”, Optics Express, Vol. 18, No.
Tris (8-hydroxyquinoline) Aluminum (Alq3) Films), 21, pp. 22004-22009, 2010.
Y-H. So, M.A. Green, G. Conibeer, S. Huang and A. Physica Status Solidi C, Vol. 8 (1), pp. 80-83, 2011.
Gentle, “Size Dependent Optical Properties of Si B. Zhang, S. Shrestha, M.A. Green and G. Conibeer,
Quantum Dots in Si-rich Nitride/Si3N4 Superlattice A. Uddin, C. B. Lee and J. Wong, “Emission Properties “Surface States Induced High p-type Conductivity
Synthesized by Magnetron Sputtering”, Journal of of Dopants Rubrene and Coumarin 6 in Alq3 Films”, in Nanostructured Thin Film Composed of Ge
Applied Physics (accepted February 2011). Journal of Luminescence, Vol. 131, pp. 1037- nanocrystals in SiO2 Matrix”, Applied Physics Letters,
1041, 2011. Vol. 97, p. 132109, 2010.
Yong-Heng So, Shujuan Huang, Gavin Conibeer and
Martin A Green, “Formation and Photoluminescence Z. Wan, S. Huang, M.A. Green and G. Conibeer, B. Zhang, S.K. Shrestha, M.A. Green and G.J. Conibeer,
of Si Nanocrystals in Controlled Multilayer Structure “Residual Stress Study of Silicon Quantum Dot in “Size Controlled Synthesis of Ge Nanocrystals in
Comprising of Si-rich Nitride and Ultrathin Silicon Silicon Carbide Matrix by Raman Measurement”, SiO2 at Temperatures Below 400 degree C Using
Nitride Barrier Layers”, Thin Solid Films (accepted Physica Status Solidi C, Vol. 8 (1), pp. 185-188, 2011. Magnetron Sputtering”, Applied Physics Letters, Vol.
February 2011). 96, p. 26901, 2010.
Z. Wan, S. Huang, M.A. Green and G. Conibeer, “Rapid
A. Sugianto, L. Mai, M.B. Edwards, S.R. Wenham et Thermal Annealing and Crystallization Mechanisms B. Zhang, S. Shrestha, P. Aliberti, M.A. Green and G.
al., “Investigation of Al-Doped Emitter on N-Type Study of Silicon Nanocrystal in Silicon Carbide Conibeer, “Characterisation of Size-Controlled and
Rear Junction Solar Cells”, IEEE Transactions on Matrix”, Nanoscale Research Letters, 6:129, 2011. Red Luminescent Ge Nanocrystals in Multilayered
Electron Devices, Volume: 57, Issue: 2, pp. 525-529, Superlattice Structure”, Thin Solid Films, Vol. 518, pp.
February 2010. S. Wang, A. Lennon, B. Tjahjono and S. Wenham,
“Overcoming Overplating Problems for PECVD SiNx 5483-5487, 2010.
Y. Takeda, T. Motohiro, D. König , P. Aliberti, Y. Feng, S. Passivated, Laser-doped p-type Multicrystalline B. Zhang, Y. Yao, R. Patterson, S. Shrestha, M. A. Green
Shrestha and G. Conibeer, “Practical Factors Lowering Silicon Solar Cells, Solar Energy Materials and Solar and G. Conibeer, “Electrical Properties of Conductive
Conversion Efficiency of Hot Carrier Solar Cells, Cells (submitted March 2011). Ge Nanocrystal Thin Films Fabricated by Low
Applied Physics Express”, Applied Physics Express 3, Temperature in Situ Growth”, Nanotechnology, Vol.
p. 104301, 2010. 22, p. 125204, 2011.

8.5 UNREFEREED Papers in Scientific and

Technical Journals
Richard Corkish, “Our Time to Shine: Australia’s T. Trupke and W. McMillan, “Photoluminescence
Burgeoning Solar Energy Industry”, Ecogeneration, Imaging Speeds Solar Cell inspection”, Laser Focus
Issue 57, pp. 24, 26, March/April 2010. World, 12/2010.
http://ecogeneration.com.au/news/our_time_
to_shine_australias_burgeoning_solar_energy_ T. Trupke, “Photolumineszenz Bildgebung für die
industry/012453/ (invited). Silizium-Photovoltaik”, Photonik 1/2011, p.38, 2011.

126
8.6 Conference Papers AND
OTHER PRESENTATIONS
P. Aliberti, Y. Feng, R. Clady, M.J.Y. Tayebjee, T.W. Richard Corkish, “Antenna collection of solar energy”, Martin A. Green, “Price and Supply Constraint on
Schmidt, S.K. Shrestha, M.A. Green, G.J. Conibeer, “On The First International Symposium on Solar Energy Te and In Photovoltaics”, 35th IEEE Photovoltaic
Efficiency of Hot Carriers Solar Cells with a Indium Technology, Yeungnam University, Korea, 16 Specialists Conference, Honolulu, Hawaii, 20-25
Nitride Absorber Layer”, 5th World Photovoltaic Solar November 2010 (invited). June, 2010.
Energy Conference, Oral Presentation, Proceedings of
European Photovoltaic Conference, Valencia, Spain, Richard Corkish, “Renewable Energy in Australia”, Martin A Green, “The International Marketing Trend
6-10 September 2010. Australian Conservation Foundation, Central Coast of Silicon Solar Cells”, World PV Forum, Chengdu, 28-
Branch, Gosford, 18 September 2010 (invited). 30 September 2010 (invited).
P. Aliberti, B. Zhang, M.A. Green. G. Conibeer, S.
Shrestha, A. Hsieh, R. Li and B. Puthen-Veettil, Richard Corkish, “Photovoltaics Technology and Martin A Green, “Possible Evolutionary Paths
“Investigation of Single Layers of Silicon Quantum Education at UNSW”, Cutting Edge Conference, MLC, for Photovoltaics”, Symposium C, E-MRS 2010
Dots in SiO2 Matrix for Energy Selective Contacts Burwood, 15 September 2010 (invited). Fall Meeting, Warsaw, Poland, 13-17 September
in Hot Carriers Solar Cells”, Australian Solar Energy 2010 (invited).
Richard Corkish, “Supporting Activities for Renewable
Society Conference, Solar2010, Canberra, 1-2 Energy Growth”, University of Delaware, 23 April Martin A. Green, “The Global Evolution of PV
December 2010. 2010 (invited). technology”, 2nd Chinese Renewable Energy
X. Bai, A. Sugianto, B. J. Hallam and S. R. Wenham, Conference & Solar Power Exhibition, Wuxi, 17-19
Richard Corkish, “Supporting Activities for Renewable September 2010 (invited).
“Improved Open Circuit Voltage for Commercial Energy Expansion in the Asia-Pacific Region”, Indo-
Grade Wafers by High Quality Hydrogen Passivation”, Australia Solar Energy Workshop, Amity Institute of Martin A. Green, “Principles and Perspectives of
Australian Solar Energy Society (AuSeS) Conference, Renewable & Alternative Energy, Amity University- Photovoltaics”, Walter Schottky Lecture 2010, RWTH
Solar2010, Canberra, 1-3 December 2010. Uttar Pradesh, India, 9-10 February 2010 (invited). Aachen, Germany, 21 October 2010 (invited).
G. Conibeer, M.A. Green, D. König, I. Perez-Wurfl, S. T. Curtis, R. Corkish and R.J. Fuller, “Solar Thermal Martin A. Green, “Technologies Shaping the
Huang, X. Hao, D. Di, L. Shi, S. Shrestha, B. Puthen- Drying of Canarium Indicum Nuts”, Solar 2010, 48th Photovoltaic Industry in the Coming 10 Years”,
Veetil, Y.H. So, B. Zhang and Z. Wan, “Silicon Quantum Annual Conference of the Australian Solar Energy Lecture, Solland Solar, Aachen, Germany, 22 October
Dot Based Solar Cells: Addressing the Issues of Society (AuSES), Canberra, 1-3 December 2010. 2010 (invited).
Doping, Voltage and Current Transport”, 5th World
Photovoltaic Solar Energy Conference, Valencia, R. Dumbrell, P. Hamer, A. Lennon and B. Tjahjono, Martin A. Green, “High Efficiency Silicon and Third
Spain, 6-10 September 2010. “Emitter etch-back application for selective emitter Generation Solar Cells”, Colorado Nanofabrication
silicon solar cells”, Proceedings of SOLA 2010 Laboratory Symposium, Colorado, USA, 24-26
G. Conibeer, “Third Generation Photovoltaics”, (AuSES), Australian Solar Energy Society Conference, October 2010 (plenary).
Australian Solar Energy Society Conference, Canberra, 1-3 December 2010.
Solar2010, Canberra, December 2010. Martin A. Green, “Solar PV – The Real Answer to
H. Cui, M.A. Green, P. Campbell, M. Wolf and G. Jin, Future Energy?”, AIE National Conference, Adelaide,
Richard Corkish, Stephen Bremner, Anna Bruce, Evatt “Enhanced Absorption of SPC Poly-Si Thin Films 14-16 November 2010 (invited).
Hawkes, Merlinde Kay, Alison Lennon, Alistair Sproul, on Aluminium Induced Textured (AIT) Glass with
Ted Spooner, Santosh Shrestha, Geoff Stapleton, Submicron Texture Feature Size”, 25th European Martin A. Green, “Photovoltaic Physics: Light
Ashraf Uddin, Muriel Watt and Darcy Wentworth, Photovoltaic Solar Energy Conference and Exhibition Emission, Ultimate Efficiencies, Light Trapping”,
“Renewable Energy Education at UNSW”, Solar (25th EU PVSEC) / 5th World Conference on Australian Institute of Physics (AIP) 19th National
2010, 48th Annual Conference of the Australian Photovoltaic Energy Conversion (WCPEC-5), Session: Conference, Melbourne, 5-9 December 2010 (invited).
Solar Energy Society (AuSES), Canberra, 1-3 3BV.3.59, Abstract No. [2008], Valencia, Spain, 6-10 Martin A. Green, “Photovoltaics Solar Conversion
December 2010. September, 2010 (poster presentation). for Sustainable Energy Supply”, First International
Richard Corkish, “Renewable energy engineering Dawei Di, Ivan Perez-Wurfl, G. Conibeer and Conference on Sustainable Urbanization (ICSU 2010),
education at UNSW”, World Green Energy Forum, M.A. Green, “Fabrication and Characterisation of Hong Kong, 15-17 December 2010 (invited).
Gyeongju, Korea 17-18 November 2010 (invited, Silicon Quantum Dots in SiO2/Si3N4 Hybrid Matrix”, P. Gress, P. Widenborg, G. Jin and S. Varlamov,
presentation slides published in proceedings). Proceedings of SPIE – Optics and Photonics – Next “Incorporation of Interconnection Technique into
Richard Corkish, “Photovoltaics technology Generation (Nano) Photonic & Cell Technologies for the Cell Design of Thin-Film Polycrystalline Silicon
development at UNSW”, World Green Energy Forum, Solar Energy Conversion, San Diego, August 2010. Solar Cells On Glass”, 25th EUPVSEC, pp. 3580-3583,
Gyeongju, Korea 17-18 November 2010 (invited, Hua Fan, Richard Corkish, Mary-Louise McLaws Valencia, Spain, 2010.
presentation slides published in proceedings). and Jian Gang Zhang, “Solar Electrolysis System B. Hallam, N. Western, S. Wenham, “Large Increase in
Richard Corkish, “Advanced silicon photovoltaics for Remote Emergency Production of Disinfectant Bulk Lifetime of Commercial Grade CZ Silicon Wafers
technology and education at UNSW, Australia”, Solutions”, Solar 2010, 48th Annual Conference of the Through Hydrogenation”, Australian Solar Energy
International Photovoltaic Solar Energy Conference Australian Solar Energy Society (AuSES), Canberra, Society (AuSeS) Conference, Solar2010, Canberra, 1-3
and Exhibition (IPVSEE 2010) 27-29th September 1-3 December 2010. December 2010.
2010, Beijing (invited). M.A. Green, G. Conibeer, D. König, S. Shrestha, S. B. Hallam, A. Sugianto, B. Tjahjono, N. Kuepper,
Richard Corkish, “Australia-based Research for Huang, P. Aliberti, L. Treiber, R. Patterson, B. Puthen- N. Western, G. Xu, S. Wenham “Deep Junction
Photovoltaics Industry Expansion in the Asia- Veettil, A. Hsieh, Y. Feng, A. Luque, A. Marti, P.G. Formation for Silicon Solar Cells Through the Use
Pacific Region”, Ecogen 2010, 5-8 September 2010, Linares, E. Cánovas, E. Antolin, D. Fuertes Marrón, C. of Laser Doping”, Australian Solar Energy Society
Sydney (invited). Tablero, E.Hernandez, J-F. Guillemoles, L. Huang, A. Le (AuSeS) Conference, Solar2010, Canberra, 1-3
Bris, T. Schmidt, R. Clady and M. Tayebjee, “Hot Carrier December 2010.
Richard Corkish, “Australia-based Education for Solar Cells: Challenges and Recent Progress”, 35th
Renewable Energy Industry Expansion in the Asia- IEEE Photovoltaic Specialists Conference, Honolulu, P. Hamer, S. Wenham, Xue Bai, B. Hallam, “Application
Pacific Region”, Ecogen 2010, 5-8 September 2010, Hawaii, 20-25 June, 2010. of Boron Depletion Techniques to Low Resistivity
Sydney (invited). Silicon for Photovoltaics”, Australian Solar Energy
M. A. Green, G. Conibeer, D. König, S. Shrestha, S. Society (AuSeS) Conference, Solar2010, Canberra, 1-3
Richard Corkish, “Advanced PV Cell Research Huang, P. Aliberti, L. Treiber, R. Patterson, B. Puthen December 2010.
Directions & Education at UNSW”, SOLARCON India Veettil, A. Hsieh, A. Luque, A. Marti, P.G. Linares, E.
2010, 28-30 July 2010, Hyderabad, India (invited Cánovas, E. Antolín, D. Fuertes Marrón, C. Tablero, E. A. Hsieh, S. Shrestha, M. A. Green and G. Conibeer,
keynote, slides distributed to participants). Hernández, J-F. Guillemoles , L. Huang, T. Schmidt, “Investigation of Si/SiO2 Quantum Well Structures for
R. Clady and M. Tayebjee, “Recent Progress with Energy Selective Contacts in Hot Carrier Solar Cells”,
Richard Corkish, “Photovoltaics R&D at UNSW”, Hot Carrier Solar Cells”, 25th European Photovoltaic 5th World Photovoltaic Solar Energy Conference,
in KIER’s PV R&D Outcome Exchange 2010, pp. Solar Energy Conference and Exhibition (25th EU Valencia, Spain, 6-10 September 2010.
129-151. Daejeon, Korea 28 April 2010 (invited, PVSEC) / 5th World Conference on Photovoltaic
opening session). A. Hsieh, S. Shrestha, M.A. Green and G. Conibeer,
Energy Conversion (WCPEC-5), Session: 1AP.1.1,
“Investigation of Si/SiO2 Quantum Well Structures
Richard Corkish, “Photovoltaics Education and Abstract No. [1191], Valencia, Spain, 6-10 September,
for Selective Energy Contacts in Hot Carrier Solar
Technology at UNSW”, 2nd PV Summit Asia, 13-14 April 2010 (plenary).
Cells”, Australian Solar Energy Society Conference,
2010, Beijing. Day 2, Paper #1. http://www.merisis- Solar2010, Canberra, December 2010.
asia.com/pv2010/Speakers.html (invited).

127
 J. Huang, J. Wong, M. Keevers and M. Green, R. Patterson, B. Puthen Veettil, M. Kirkengen, D. A. Uddin and C. C. Teo, “High Efficiency Self-assembly
“Cell performance limitation investigation of Koenig, M.A. Green and G.J. Conibeer, Poster CdSe/ZnS Quantum dots Light-Emitting Devices in
polycrystalline silicon thin-film solar cells on glass by Presentation 1DV.3.77, 5th World Photovoltaic Organic Matrix”, IEEE International Nano-Electronics
transmission electron microscopy”, 25th EUPVSEC, pp. Solar Energy Conference, Valencia, Spain, 6-10 Conference, 3–8 January 2010, City University of
3565-3567, Valencia, Spain, 2010. September 2010. Hong Kong.

G. Jin, P. Gress, S. Varlamov and P. Widenborg, “Recent B. Puthen Veettil, R. Patterson, D. König, G. Conibeer A. Uddin and C.B. Lee, “Exciton Behaviors in Doped
Progress in High Rate PECVD SPC Poly-Si Thin Film and M. A. Green, “Modelling the Effects of Doping in Tris (8-hydroxyquinoline) Aluminum (Alq3) Films”, 9th
Solar Cells at UNSW”, 25th EUPVSEC, pp. 3577-3579, Quantum Dot Structures for Tandem Cells”, Poster International Conference on Excitonic and Photonic
Valencia, Spain, 2010. Presentation 1DV.3.64, 5th World PVSEC, Valencia, Processes in Condensed and Nano Materials,
Spain, 6-10 September 2010. (EXCON’10), 11–16 July 2010, Brisbane, Australia.
D. König, J. Rudd, G. Conibeer and M. Green,
“Introduction of Majority Carriers into Si and Ge Nano B. Puthen-Veettill, D. König, M.A. Green and G. S. Varlamov, Y. Tao, J. Wong, O. Kunz and R. Egan,
Crystals Embedded in SiO2 or Si3N4, Symposium J, Conibeer, “Analysis of QDs in Energy Selective “Crystallisation Kinetics of E-beam Evaporated Si
E-MRS Spring Meeting, Strasbourg, 7 - 11 June 2010, Contacts Using a 2-Dimensional Model”, Proc. E-MRS Films and Its Effects on Poly-Si Film and Solar Cell
Oral Presentation (invited). Spring meeting, Symposium B, June 2009, Energy Properties”, 25th EUPVSEC, pp. 2823-2827, Valencia,
Procedia, 2(1), 2010. http://www.elsevier.com/ Spain, 2010 (oral).
D. König, “Introduction of Majority Carriers into Si wps/find/journaldescription.cws_home/718157/
and Ge Nano Crystals by Modulation Doping of description#description Z. Wan, S. Huang, M.A. Green and G. Conibeer,
Embedding Dielectric”, Seminar at the Chemnitz “Raman Study of Stress Effect in Silicon Rich Carbide
University of Technology, Germany, 4 June F. Qi, G. Conibeer, S. Shrestha and S. Huang, “The (SiCx) Film by Furnace and Rapid Thermal Annealing
2010 (invited). Study of Potential Application of Group IV Alloys in for Photovoltaic Application”, 35th IEEE Photovoltaic
Photovoltaic Field”, Australian Solar Energy Society Specialists Conference, Honolulu, Hawaii, 20-25
D. König, “New Results on Si Nano Crystals: Interfaces, Conference, Solar2010, Australian Solar Energy June, 2010.
Doping, Energy Selective Contacts and Hot Carriers”, Society Conference, Canberra, 1-3 December 2010.
Seminar at the Institute of Microtechnology J. Wong, J.L. Huang, M. Keevers and M.A. Green, “Voc-
(IMTEK), University of Freiburg, Germany, 14 June Y. H. So, I. Perez-Wurfl, L. Shi, S. Huang, G. J. Conibeer Limiting Recombination Mechanisms in Thin Film
2010 (invited). and M. A. Green, “Silicon Nitride As Alternative Matrix Silicon on Glass Solar Cells”, 35th IEEE Photovoltaic
for All-Si Tandem Solar Cell”, Proceedings of 25th Specialists Conference, Honolulu, Hawaii, 20-25
Eunjoo Lee, Hyunwoo Lee, Ilhwan Kim, Junyoung European Photovoltaic Solar Energy Conference and June, 2010.
Choi, Dongjun Oh, Jimyung Shim, Kyungyeun Cho, Exhibition, Valencia, Spain, 6-10 September 2010.
Jisun Kim, Hae-Seok Lee, Soohong Lee, B. Hallam and D. Wang, B. Tjahjono, A. Uddin and S. Wenham,
S.R. Wenham, “The Potential Efficiency of Laser Doped A. Sugianto, J. Bovatsek, S. Wenham, B. Tjahjono, G. “Theoretical Analysis of Laser-Doped Semiconductor
Solar Cells Using Photoluminescence Imaging”, 35th Xu, Y. Yao, B. Hallam, X. Bai, N. Kuepper, C.M. Chong Fingers Silicon Solar Cell Design Using a Shading
IEEE Photovoltaic Specialists Conference, Honolulu, and R. Patel, “18.5% Laser-Doped Solar Cell on Cz and Resistive Power Loss Model” 25th European
Hawaii, 20-25 June, 2010. P-Type Silicon”, 35th IEEE Photovoltaic Specialists Photovoltaic Solar Energy Conference, 6-10
Conference, Honolulu, Hawaii, 20-25 June, 2010. September 2010, Valencia, Spain.
P.H. Lu, Y. Chen and A. Lennon, “Innovative Rear Point-
Contact Scheme for Silicon Solar Cells, Proceedings of A. Sugianto, A. Lennon, B. Tjahjono and S. Wenham, E. Wang, S. Mokkapati, F. J. Beck, Z. Ouyang,
SOLA 2010 (AuSES), Australian Solar Energy Society “Impact of Laser-Doped Line Conductivity on Ni S. Varlamov and K. Catchpole “Periodic silver
Conference, Canberra, 1-3 December 2010. Light-Induced Plating, Proceedings of the 25th nanoparticles fabricated with soft-stamp
European Photovoltaic Solar Energy Conference, nanoimprint on silicon solar cells”, 25th EUPVSEC,
W. McMillan, T. Trupke, J.W. Weber, M. Wagner, U. Valencia, Spain, 6-10 September 2010. pp.77-81, Valencia, Spain, 2010 (oral).
Mareck, Y.C. Chou and J. Wong, “In-line Monitoring of
Electrical Wafer Quality Using Photoluminescence”, Y. Tao, S. Varlamov, J. Wong, O. Kunz and R. Egan, Y. Yao, B. Zhang, S. Shrestha, G. Conibeer and M.A.
25th European Photovoltaic Solar Energy Conference, “Effects of SPC Temperature on Properties of Green, “Photovoltaic Effect in Ge Nanocrystals/c-
Valencia, Spain, September 2010. Evaporated Poly-Si Thin Films and Solar Cells”, 35th silicon Heterojunctions Devices”, 35th IEEE
IEEE PVSC, Hawaii, USA, 2010 (oral). Photovoltaic Specialists Conference, Honolulu,
W. McMillan, T. Trupke, J.W. Weber, S. Ha, B.S. You, W.Y. Hawaii, 20-25 June, 2010.
Park, K.Y. Lee and J. Nyhus, “Slip-line Detection on L.S. To, “Social and Institutional Aspects of
Mono Ingots and Wafers at Production Throughput Photovoltaic Projects in Western China”, Climate: Y. Yao, A. Lennon, B. Tjahjono and S. Wenham, “Use
Using Photoluminescence”, 25th European Science + Humanities Symposium, The Faculty Club, of Inductively-Coupled-Plasma Measurements to
Photovoltaic Solar Energy Conference, Valencia, Harvard University, 4 March 2010. Monitor Light-Induced Plating of Silicon Solar Cells”,
Spain, September 2010. Proceedings of the 25th European Photovoltaic
L.S. To, “Solar Photovoltaic Projects in Remote Areas Solar Energy Conference, Valencia, Spain, 6-10
G.A. Parnis, A.B. Sproul, “Fast Thermal Modelling of China”, Asia Pacific Week, Australian National September 2010.
Using Micro-Cap”, The Australian Solar Energy University, 11 February 2010.
Society (AuSES) Solar 2010 Conference, Australian H. Zhang, S. Huang and G. Conibeer, “Tandem
Solar Energy Society Conference, Canberra, 1-3 T. Trupke, B. Mitchell, J.W. Weber and J. Nyhus. Photoelectrochemical Cell for Direct Water Splitting”,
December 2010. “Bulk Minority Carrier Lifetime from Luminescence Australian Solar Energy Society Conference,
Intensity Ratios Measured on Silicon Bricks”, 25th Solar2010, Canberra, December 2010.
Rajesh S. Patel, J. Bovatsek, A. Sugianto, B. Hallam, European Photovoltaic Solar Energy Conference,
B. Tjahjono, “Laser Doped Selective Emitters: Valencia, Spain, September 2010. B. Zhang, S. Shrestha, S.J. Huang, P. Aliberti, M.A.
Comparing Mode-locked UV 355nm and CW Green and G. Conibeer, “Structure Studies of
532nm Laser Process”, 25th EU PVSEC, Valencia, 6-10 T Trupke, “Luminescence Imaging: A Powerful Multilayered Ge Nanocrystals Embedded in SiO2
September 2010. Characterization Tool for Photovoltaic Applications”, Matrix Fabricating Using Magnetron Sputtering”, Proc.
SPIE, Optics & Photonics, San Diego, August 1-5, E-MRS Spring meeting, Symposium B, June 2009,
R. Patterson, Z. Wan, B. Puthen-Veettil, D. König 2010 (invited). Energy Procedia, 2(1), 2010. http://www.elsevier.com/
and G. Conibeer, “Phonon Decay in Naostructure: wps/find/journaldescription.cws_home/718157/
Computational Study”, Proc. of 35th IEEE PVSC, T Trupke, “Luminescence Imaging for Inline
Applications in Photovoltaics”, 10th International description#description
Honolulu, Hawaii, 20 - 25 June 2010, proc. pp. 1843
(art.-no. 5615953). Workshop on Beam Injection Assessment of
Microstructures in Semiconductors, Halle, Germany,
July 2010. (invited).

8.7 OTHER REPORTS

R. Armstrong, M.A. Green, L. Vant-Hull, E. Palomino Muriel Watt and Richard Corkish, “Assessment of the
and T. Plant, “2010 Solar Program Peer Review Current Status and Future Development Prospects
Report: An Independent Evaluation of Program of Photovoltaics”, Report to Rio Tinto by New South
Activities for FY2009 and FY2010”, Solar Energy Global and IT Power, March 2010.
Technologies Program, US Department of Energy,
July 2010 (782pp.).

128
This report is available from:

ARC Photovoltaics Centre of Excellence ARC

PHOTOVOLTAICS
University of New South Wales
Sydney NSW 2052, Australia
CENTRE OF
EXCELLENCE
Phone: 2010/11
ANNUAL REPORT
+61 2 9385 4018

Fax:
+61 2 9662 4240

Email:
[email protected]

Web:
http://www.pv.unsw.edu.au

Director:
Stuart Wenham

Executive Research Director:

Martin Green

Report Coordinators:
Martin Green, Stuart Wenham, Joyce Ho and
Robert Largent

Graphic Design:
Peter Haywood ([email protected])

Roman Balla ([email protected])

Photos, figures and graphs:

Courtesy of Centre Staff and Others
ARC
PHOTOVOLTAICS
ARC PHOTOVOLTAICS
CENTRE OF
CENTRE OF EXCELLENCE
EXCELLENCE
2010/11
The University of New South Wales
ANNUAL REPORT
UNSW Sydney NSW 2052 Australia
Tel +61 2 9385 4018
Fax +61 2 9662 4240
Web www.pv.unsw.edu.au