ELECTRICAL & COMPUTER ENGINEERING

A Bright Future for Solar Energy

Georgia Tech is playing an important role in photovoltaics' status as a leading contender in the search for clean, renewable energy sources.

By Amanda Crowell

MIKE ROPP, A DOCTORAL STUDENT in Georgia Tech's School of Electrical and Computer Engineering (ECE), has just climbed nearly 150 feet of ladders to the barrel-vaulted roof of the Georgia Tech Aquatic Center. Wind whips menacingly over the sides and a stunning view of the Atlanta skyline lies to the south.

photo by Stanley Leary Reseachers have reduced the time required to produce solar cells without losing efficiency. (200-dpi JPEG version - 362k)

"Welcome to my laboratory," he quips, flashing a ready grin and spreading his arms expansively.

And what a laboratory it is.

Spread over nearly three-quarters of an acre is what is believed to be the world's largest solar-powered energy system connected to a power grid and located on a single rooftop. The 342-kilowatt photovoltaic system — which converts sunlight into electricity — serves as both a research model and a supplementary power source for the Aquatic Center.

It is also one of many projects conducted under the Georgia Institute of Technology's University Center of Excellence for Photovoltaics Research and Education (UCEP), which is designed to help make photovoltaics (PV) a leading contender in the search for clean, renewable energy sources for the future.

Established in 1992 by the U.S. Department of Energy and supported by the DOE's Sandia National Laboratories, UCEP is one of only two national centers of excellence in PV research. (The second is at the University of Delaware.)

Researchers are charged with advancing PV research, producing cheaper and more efficient solar cells, and training the next generation of PV scientists — all with an eye toward giving the United States a competitive edge in photovoltaics.

"I think the main reason the DOE decided to make us a university center of excellence was there was no other university at the time, other than the University of Delaware, that could do research all the way from photovoltaic materials to materials characterization, modeling, process development, fabrication, testing and analysis of cells," says Dr. Ajeet Rohatgi, who directs UCEP and is a Regents' Professor and Georgia Power Distinguished Professor in ECE. "Large grid-connected PV systems on campus make us even more unusual. There are very few places that have everything going on in one place."

Dr. Joseph R. Romm, acting assistant secretary for the DOE's Office of Energy Efficiency and Renewable Energy, also notes that Georgia Tech "has an unusually strong interdisciplinary emphasis and a commitment to sustainable development."

"There's also a good healthy emphasis on education," he says. "All of that adds up to the perfect setting for a center of excellence."

Although proponents of photovoltaics say it's an ideal technology to supplement or replace traditional energy sources, PV power currently is less efficient (defined as the amount of energy a system produces divided by the energy that goes into it) and about four times more expensive. But 20 years ago, PV power was 50 times as expensive as traditional energy sources.

UCEP researchers have made major contributions to bringing down this cost by designing and testing new PV systems and developing cheaper, more efficient solar cell technologies.

Olympic Legacy

In the area of new PV systems, the Georgia Tech Aquatic Center is a standout example. Built to host swimming and diving events for the 1996 Summer Olympic and Paralympic Games,

photo by Stanley Leary The Georgia Tech Aquatic Center's roof holds a 342-kilowatt photovoltaic (PV) system, which will provide significant, long-term data on how to build and maintain large-scale PV structures. (200-dpi JPEG version - 362k)

it is a lasting legacy for the campus and should provide significant, long-term data on how to build and maintain large-scale PV structures.

"The goal is to get a better understanding of how these systems work — their performance, their reliability and our modeling capability to predict their performance," says Rohatgi, who designed the $5.2 million system with Dr. Miroslav M. Begovic, also an ECE professor, and Richard Long, project support manager in Georgia Tech's Office of Facilities.

Funding came from Georgia Tech, Georgia Power Company and the DOE.

"We realize that photovoltaics is a technically viable source for supplying future energy needs, and we wanted to help in the demonstration of that," explains Chuck Huling, who coordinates research for Georgia Power. "The Olympics provided a wonderful opportunity to demonstrate this renewable technology to a local, national and international audience."

During its first year, the system operated close to the efficiency level expected, although actual energy production was lower than predicted. Reasons included unexpected down time, periodic shutdowns for experiments and higher- than-usual temperatures during some months, which decreased the system's efficiency.

From July 1996 to June 1997, the system produced 333.3 megawatt hours of electricity, which is 81.5 percent of the 409 megawatt hours predicted and enough energy for about 28 average Georgia homes.

The rooftop system features a solar array made up of 2,856 photovoltaic modules, each with 72 multicrystalline silicon solar cells connected in series. A power conditioning system, or inverter, converts the array's direct current (DC) power to utility-compatible alternating current (AC) power, and a data acquisition system stores performance and meteorological information every 10 minutes.

Why Photovoltaics? The bottom line for renewable energy is not that it's a matter of if. It's a matter of when. When proponents of photovoltaics — the direct conversion of sunlight into electricity — argue their case, they note that two billion people in the world don't have access to electricity and that most conventional energy sources cause pollution, deplete natural resources or contribute to global warming. Photovoltaics (PV), or solar power, offers a clean, renewable alternative. The U.S. Department of Energy is supporting extensive research in this area, including establishment of Georgia Tech's University Center of Excellence for Photovoltaics Research and Education (UCEP). PV power operates on a simple principle: a cell is created from a semiconductor material like silicon. When sunlight hits the cell, photovoltage on an electric current is created, which flows through an external circuit and produces energy. Several cells can be wired together and encased in clear glass or plastic to form a panel or module. These can be connected into arrays — to collect and produce more power — then placed atop a building and either connected to an existing electrical system or linked to batteries. The process is silent and self-contained, with no moving parts, no emissions and sunlight as energy source. Compared to burning coal, for example, DOE officials estimate that every gigawatt hour of PV-generated electricity prevents the emission of up to 1,000 tons of carbon dioxide. Solar power also is versatile enough to supply nearly any energy need, from lighting and small appliances for a single home to water-pumping systems for farms or industrial activities for whole villages. Although PV power currently is less efficient and more expensive than conventional energy sources like coal, oil, natural gas and nuclear power, its advantages already make it the preferred choice in many everyday applications. Examples include calculators, U.S. Coast Guard navigational beacons, highway emergency telephones, traffic warning signs, satellites and remote cabins and farms. It's also economically competitive in some parts of the United States now — including Hawaii, where electricity is very expensive; Massachusetts and New York, where energy costs are high and local governments often support solar power; and California and Arizona, which have large remote areas and much sunlight. To help make photovoltaics more competitive, government, private industry and utility company partners have built or proposed dozens of projects, from large-scale power plants to programs that encourage home owners to install rooftop PV systems. Worldwide demand for solar power grew 290 percent from 1987 to 1995. But for such advances to continue, sustained commitment is needed. Federal funding for renewable energy sources, high during the oil crisis of the 1970s, fell sharply in the early 1980s and only began rebounding in the past decade. "We're at the point where we're ready to reap large returns on the investments that we've made over the years," says Dr. Joseph R. Romm, acting assistant secretary for the DOE's Office of Energy Efficiency and Renewable Energy. "You cannot profit optimally if you focus on lab work, then throw the results over the fence, assuming that the marketplace will pick them up. There needs to be a partnership with the private sector, and that's what we're doing at the DOE. "The bottom line for renewable energy, really, is not that it's a matter of 'if,'" he adds. "It's a matter of when and who profits."

Researchers also built a 4.5-kilowatt, AC array at the entrance to the Callaway Student Athletic Complex. It differs from the Aquatic Center system in that each module converts the solar-generated DC power to AC power itself, which reduces costs and simplifies installation.

"While UCEP has long been in the forefront of research in developing world-record efficient hardware, the PV systems will help us in becoming an authority in design and help assess the cost/benefit of the yet-to-be-built systems of the future," Begovic says.

New Processes and Materials

But for photovoltaics to truly compete with conventional energy sources, production costs must be reduced, so Georgia Tech researchers are exploring several innovative techniques.

One is rapid thermal processing (RTP), which researchers recently used to fabricate for the first time a silicon solar cell with the same 19 percent efficiency rating as cells produced by conventional furnace processing, but in half the time — 81/2 hours rather than 17.

Conventional solar cell production generally involves three trips into a high-temperature furnace, and each step lasts one to three hours. The cells also must be cleaned between each step. With RTP fabrication, the front and back of the cell are formed simultaneously by a rapid thermal diffusion process that takes three minutes, and an oxide is grown on the front of the cell by a five- minute rapid thermal oxidation (RTO) process.

Industrial manufacturers often delete the oxidation process, called passivation, to save money and increase output. Georgia Tech's RTO process offers a time-saving way to include this performance-enhancing step.

Once a solar cell is created, metal contacts are added to extract the electrical power from the cell. This step is the most time-consuming; in RTP fabrication, for example, it accounts for 80 percent of the production process. The common techniques of evaporation and photolithography give good resolution and conductivity, but Rohatgi says many commercial manufacturers have switched to a quicker method called screen printing, which produces less efficient cells.

In 1996, Georgia Tech researchers successfully integrated screen printing with RTP, slashing cell production time to 11/2 hours. Since then, they've raised cell efficiency from 14.7 percent to 16.3 percent and outlined modifications for future increases.

"If we can make the solar cells very fast compared to what's being done out in industry today, without sacrificing the cell performance, that will obviously reduce the use of chemicals, gases and manpower, and it will increase the production capacity and throughput," Rohatgi explains. "This should result in significant reduction in the cost of solar cell modules."

Researchers also are experimenting with a technique they call "Simultaneously Diffused, Textured, In-Situ Oxide AR-coated Solar Cell Process" or STAR. In this process, a single high- temperature furnace step can provide front and back surface diffusions simultaneously, in addition to front and back in-situ oxide surface passivation. The cell is textured and has an anti- reflection (AR) coating, to trap more light in the cell.

So far, researchers have created cells with 20.1 percent efficiency. And although the STAR process is not as fast as RTP cell fabrication, Rohatgi says STAR is compatible with high- throughput machinery commonly used by the solar industry today, while RTP currently isn't.

Another way to reduce the cost of photovoltaics is to make solar cells from less expensive materials. UCEP researchers are working with several promising silicon materials, including float zone, Czchralski, cast multicrystalline, EFG sheet and dendritic web silicon, and currently hold the record for high-efficiency multicrystalline silicon cells — 18.6 percent.

Crystalline silicon is used in about 80 percent of the solar cell modules produced today, Rohatgi says. The other 20 percent are made from amorphous silicon and thin film materials like cadmium telluride.

Industry's Importance: Today and Tomorrow

Rohatgi attributes part of UCEP's success to close working relationships with more than two dozen U.S. solar manufacturers, including industry leaders like Solarex Corp., Siemens Solar Industries and ASE America Inc.

photo by Gary Meek Research done at Georgia Tech could help lower the cost of producing photovoltaic arrays. (200-dpi JPEG version - 283k)

"To make our processing more manufacturable, we try to do applied research that can be easily transferred to industry," Rohatgi says. "That is part of the mandate from the DOE. Our job is not to just do blue-sky type research, [but to] focus on research that can lead to commercially viable solar cells."

So far, UCEP is having no trouble meeting that mandate. Researchers hold patents for seven production techniques and have applied for several others. They've published over 100 papers in peer-reviewed journals and both refereed and non-refereed conference proceedings. UCEP also includes an Educational Support Program (ESP) Laboratory, where solar cells are fabricated and/or tested for other universities, and lab space in both ECE and the Microelectronics Research Center.

Besides reducing solar costs and improving technologies, Rohatgi says future successes also will depend on transferring these new techniques from the laboratory to the production line.

"The next step would be to scale up some of the novel technologies we're developing to a larger scale — making larger-area cells, then transferring this know-how to industry," Rohatgi says. "Only then will industry get excited about it and be able to use it."

Further information is available from Dr. Ajeet Rohatgi, School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0250. (Telephone: 404/894-7692) (E-mail: ajeet.rohatgi@ee.gatech.edu) You can find the Georgia Tech Aquatic Center on the Web at http://www.ece.gatech.edu/users/2648/index.html.

Contents | Research Horizons | GT Research News | GTRI | Georgia Tech

Send questions and comments regarding these pages to Webmaster@gtri.gatech.edu

Last updated: Dec. 3, 1997