October 6, 2008
University of Delaware's Robert
Research News editor Eric Smalley carried out an email conversation
Birkmire, a professor of materials science and engineering and
physics at the University of Delaware.
Birkmire is an authority on inorganic solar cells, including
crystalline silicon and amorphous and polycrystalline thin-film solar
cells. He's director of the Institute
of Energy Conversion, a US Department of Energy Center of Excellence
for Photovoltaic Research and Education.
Birkmire grows and analyzes thin-film semiconductors for photovoltaic
and opto-electronic devices. He tunes the growth process to improve
film properties and device performance. He's also finding ways to
use sensor technologies to improve thin-film semiconductor manufacturing.
Birkmire received the World Renewable Energy Network Pioneer
Award at the World Renewable Energy Congress in 2002.
Birkmire has written more than 170 technical publications
and has eight U.S. patents. He received a PhD in physics from the
University of Delaware in 1976.
ERN: What are the important or significant trends you
see in energy research?
RB: The efforts to develop environmentally benign sources
of energy, driven by the realization that energy is the root cause
of climate change and that the supply of oil is limited, is the focus
of energy research.
Unfortunately, instead of being the leader, the United States
has been slow to respond to the need for alternative energy sources.
This is especially critical, particularly since the United States
is responsible for about 25 percent of the world�s energy consumption.
ERN: What would you like to see happen? Is this different
from today's national and global energy research priorities?
RB: A better global recognition of the interrelationships
between energy, environment, population and security is needed to
accelerate implementation of alternate energy sources.
Further, nearly 35 percent of the world�s population does
not have access to electricity. Providing clean forms of electricity
to underdeveloped countries needs to be a priority of the industrialized
nations. Providing electricity will provide clean water supplies and
improve agriculture, thus reducing famine and increasing the overall
quality of life.
ERN: What's the general focus of your research, and
how does it relate to energy?
RB: The Institute of Energy Conversion (IEC) at the
University of Delaware has been involved in the development of thin
film photovoltaic technology for over 36 years and was designated
as a Department of Energy Center of Excellence for Photovoltaic Research
and Education in 1992.
The mission of IEC is to develop the fundamental science and
engineering base required to improve photovoltaic device performance,
develop processing technologies, and effectively transfer these laboratory
results to large-scale manufacturing.
Currently, our research activities include
1) Developing copper-indium-diselenide-based (CuInSe2) materials
and devices on flexible substrates and wide bandgap materials and
devices for improved module performance and tandem cell applications.
2) Expanding the fundamental understanding of the device operation
and processing treatments on doping the cadmium telluride (CdTe),
improving it�s electronic properties and formation of a low resistance
3) Developing all back contact crystalline silicon heterojunction
solar cells and providing support for industries in amorphous silicon.
4) In conjunction with industry, evaluating transparent encapsulating
materials for flexible photovoltaic modules.
ERN: What's your view of the major thrusts in photovoltaics
research: dye-sensitized, organic semiconductor, thin-film, multijunction
and variations involving nanotechnology?
RB: In my opinion organic and organic/inorganic hybrid
(dye-sensitized) photovoltaics are a long way off in the future (if
ever) [for] providing large amounts of electricity. However, for specialty
applications that do not require high efficiencies and long lifetime
this type of device could be acceptable.
Thin film technologies are emerging as real alternatives to
traditional crystalline silicon technologies. First
Solar has demonstrated how economies of scale can reduce the manufacturing
cost of thin film cadmium telluride photovoltaics and is selling 10-
to 11-percent efficient modules.
Amorphous-silicon is on the verge of a major manufacturing
expansion with the entrance of Applied Materials and Oerlikon as key
suppliers of manufacturing equipment. If the expansion of amorphous-silicon
facilities proceeds at the announced rate, there will not be sufficient
silane production capacity to accommodate the demand.
Copper-indium-selenide-based photovoltaics has demonstrated
the highest efficiency, [about] 20 percent, but has been slower to
develop manufacturing capacity. Copper-indium-selenide-based modules
have the potential to be the highest-performing thin film module in
There is renewed interest in multijunction-III-V-based devices
with the recent development of over 40 percent efficient devices tested
under concentration. These structures have the potential to be used
in power generating arrays.
Nanotechnology is the buzzword of science these days but in
photovoltaics there is no demonstration of a solar cell that takes
advantage of quantum effects as the size is reduced to below [about]
30 nanometers. Nanoparticle technology may find use as a precursor
material for thin film synthesis and there is potential to develop
new materials for the future (beyond 2020).
ERN: Thin-film solar cells seem to be particularly
important, especially for ease of manufacturing. Tell me about your
lab's research in thin silicon. How does thin silicon compare to other
thin-film solar cell technologies?
RB: In the 1990�s, we investigated issues related to
amorphous silicon single and multijunction device fabrication in collaboration
with the US industry.
Our recent efforts in thin film silicon were focused on metal
induced crystallization of amorphous silicon using a cheap low temperature
substrate. We are no longer pursuing this, in part because of funding,
and in part because we could not see a pathway to growing a thin crystalline
silicon film suitable for fabricating a solar cell with reasonable
performance. Thin film silicon is the least advanced of the thin film
technologies and is not an ideal material to use for thin film solar
cells since it has an indirect bandgap thus requiring elaborate light
ERN: Tell me about cadmium telluride and copper indium
diselenide solar cells. What's their potential and how do they compare
RB: First Solar is the largest manufacturer of thin
film cadmium telluride modules in the world. Their primary market
appears to be power arrays due to the perceived potential environmental
impact of the cadmium in the module.
To alleviate this issue, an insurance policy that guarantees
reclamation of the modules at the end of their life is part of the
module cost which is currently less than crystalline silicon modules.
However, if there are alternative module technologies at nearly the
same cost, this could present a major obstacle for cadmium telluride
in the future.
Additionally, based on laboratory research there is not a
clear path to improving the cadmium telluride module performance to
15 percent since the highest efficiency laboratory cells are [about]
However, copper-indium-diselenide-based modules have the demonstrated
potential of reaching a 15 percent module based on laboratory cells
results with [about] 20 percent efficiency but developing manufacturing
has proved challenging.
Global Solar, Showa
Solar each have manufacturing facilities over 20 megawatts [per
year] with plans to increase capacity in the next few years. Further,
copper-indium-diselenide-based modules are being made on flexible
substrates which can have important applications in building integrated
photovoltaics and reduce transportation cost compared to glass module
-- large manufacturing facilities with distributed module fabrication
facilities. The unanswered question is the eventual cost of the copper-indium-diselenide-based
In the future, crystalline silicon will continue to dominate
the high efficiency module market but Cadmium telluride and copper-indium-diselenide-based
modules will be cheaper.
ERN: What's the potential for improving the efficiency
of amorphous silicon solar cells?
RB: Several groups have reported small area amorphous
silicon/nano-crystalline silicon or micromorph cells and mini-modules
with stable efficiency greater than 10 percent so the potential is
there. The challenge is to increase the deposition rate of the nano-crystalline
ERN: Has development of traditional crystalline silicon
solar cells plateaued, or is there more research to be done?
RB: There is more research and development needed.
As wafers become thinner, the processing of the wafer to fabricate
a device becomes more challenging, particularly, where high temperature
processing is used.
The heterojunction structures currently based on amorphous
silicon, where all the process is done at temperatures below 200�
Celsius has the potential for higher efficiency and reduced manufacturing
cost by coupling thin film and crystalline silicon technologies and
is particularly attractive in an all back contact configuration.
ERN: What are the milestones to watch for in photovoltaics
RB: The milestones to look for in the near future would
1) Readily available > 20 percent crystalline silicon modules.
2) Bring on-line large capacity amorphous silicon facilities.
3) A large scale copper indium diselenide module manufacturing
facility at the 100 megawatt scale.
4) Any photovoltaics technology on a flexible substrate at
the100 megawatt scale.
ERN: How useful are solar concentrators for photovoltaics?
RB: Solar concentrators are required for multijunction
III-V solar cells because of the device size, manufacturing cost and
the enhanced performance. Low-level concentration where tracking is
not required could be a cost effective approach for high performance
crystalline silicon solar cells.
ERN: Why is there a significant gap between photovoltaic
efficiencies achieved in the laboratory and those achieved by industry?
RB: In the laboratory, there are processing conditions
you can use that are not acceptable for a commercial product and,
in the laboratory.
Yield is not an issue. Lab cells are aggressively optimized
with some process condition �on the edge� that is not acceptable for
high yield. For thin films, lab cells can use higher cost substrates
that are not practical for manufacturing. Plus the size of the champion
laboratory cell or module is in general smaller than a commercial
module. Additionally, there are inherent area-related losses going
from the laboratory cell/module to a commercial module.
ERN: How much of the US and world electricity needs
can be met by photovoltaics?
RB: This is always an interesting question. In the
2020 to 2030 time frame, I think photovoltaics could generate 20 percent
of the electricity. Research from the National Renewable Energy Laboratory
(NREL) indicates that photovoltaics will begin to impact base-load
generation at about a 10 percent level.
ERN: What are the important social questions related
RB: When over two billion people have no electricity
and live in poverty -- this a big societal issue. The access to electricity
improves agriculture and quality of life (water for irrigation and
clean water to drink) and provides a means of communication with the
ERN: What are your thoughts on the state of public
understanding of energy and energy research?
RB: In the US, I suspect that the average person is
confused. There is no leadership at the national level and the mis-information
that is spread by the existing energy companies and special interests
groups is exceptionally misleading. On the other side, media hype
based on press releases or other non-refereed reports can lead to
unrealistic expectations such as solar cells that can be painted on
your house being available soon.
ERN: What could be done to improve the pursuit of energy
research in terms of business trends, politics, and/or social trends?
RB: Leadership at the national level is needed to address
short- and long-term solutions to the energy future coupled with the
reality that energy and the changes in climate are directly connect
to burning fossil fuels. Thus, any energy plan/policy needs to be
coupled to climate change and reduction of carbon dioxide (Co2) emissions.
ERN: In a perfect world how would we get our electricity?
RB: A mix of renewable energy source that have minimal
impact on the environment and are safe and secure.
ERN: In terms of energy and anything affected by energy,
what will be different about our world in five years? In 10? In 20?
RB: In the US, the next five years will be the transition
period where energy efficiency is a primary focus of energy policy
along with the emergence of clean energy technology into the marketplace.
In ten years a measurable fraction of our energy will come
from renewable sources and competing technologies will start to define
the energy for the future.
By 2030, there will be a clear pathway to how energy will
be produced -- photovoltaics will be one of the major contributors
in the energy mix.
ERN: What do you imagine you will be working on in
five years? 10 years?
RB: I suspect that I will still be working in the area
of photovoltaics on ways to improve the performance and on emerging
technologies. One of the key areas for photovoltaics will be to address
the storage issues which I have an expanding interest in becoming
ERN: What got you interested in science and technology?
RB: I grew up in an exciting time in the development
of science and technology. I used to watch a TV show "Watch Mister
Wizard" which demonstrated the wonders of science. It was an era when
Albert Einstein was a hero, the launching of Sputnik in 1957 and the
inspiration of John Kennedy to go to the moon.
ERN: What's the most important piece of advice you
can give to a child who shows interest in science and technology?
RB: Convince the child that science is not difficult
as some make it out to be and it is fun and rewarding to understand
how things work whether from a fundamental science point of view or
an engineering perspective.
ERN: What's the most important piece of advice you
can give to a college student who shows interest in science and technology?
RB: Go for it! If you have the interest and have the
aptitude for math (the language of science and engineering), science
and engineering will be a rewarding and exiting career.
ERN: What books that have some connection to science
or technology have impressed you in some way, and why?
RB: There was a book by George Gamow, �Mr. Tompkins
in Wonderland�, which is a great way to learn about modern physics.
There is a series of books that followed on science.
Is there a particular image (or images) related to science or technology
that you find particularly compelling or instructive?
RB: At IEC we deposited 110 feet of copper indium germanium
diselenide (CuInGeSe2) on a 10-inch wide polyimide web coated with
molybdenum in our multi-source in-line evaporator. The average cell
efficiency of small area devices made on the web is [about] 10 percent.
I believe that this technology has the potential to reach 15 percent
at the module level and will be one of the competitive technologies
for the future.
Back to ERN
October 6/13, 2008