A recent breakthrough has the potential for providing solar-generated electricity at large scales for prices competitive with electricity generated by fossil fuels.
By Daniel P. Duffy
If you are reading this on the Internet instead of a classic “dead tree” magazine format, you can thank the Palo Alto Research Center (PARC). A subsidiary of the Xerox Corp., PARC is first and foremost famous for developing the graphical user interface (GUI) employed by nearly all personal computer operating systems, be they Mac or IBM. It also gave the world the mouse control that operates most computers. And no, it’s not really true that Xerox completely ignored the potential for these breakthroughs and missed its big chance to dominate the future personal computer market. Computers were only a tiny fraction of the research budget at PARC. In those computer areas that were considered central to PARC’s focus and not sidelines (laser printing, optical disks, liquid crystal displays, pervasive computing), Xerox seized the opportunities as they were presented. So the whole tale is something of an urban myth—but that’s another story.
What concerns us today, in an age of ever-rising oil prices, is finding the key to unlocking the promise of cheap solar energy. PARC’s recent breakthrough in concentrated solar cell technology has the potential to provide solar-generated electricity at large scales for prices competitive with electricity generated by fossil fuels. The first solar cells achieved efficiencies of only 10% to 15% (that is, only that percentage of solar energy striking the solar cell actually became usable energy) and produced this energy at costs that made them suitable for only the most specialized applications, such as space missions. PARC, in concert with Sol Focus Inc., has developed a system of mirrors and lenses built into individual solar cells that concentrates the sunlight as much as 500 times, achieving efficiencies as high as 50%. This reduces cell size and production costs, bringing the overall cost per watt of electricity generated by this system to prices competitive with standard sources of electricity. “We value PARC’s commitment to developing clean technology and the collaborative way it innovates. We teamed with PARC to help improve our first product concept and to accelerate a second-generation product that promises even greater performance,” says Gary Conley, chief executive officer of Sol Focus. “The first-generation panels will break price barriers in the market, but the second-generation panels with PARC technology will change the market for solar dramatically.” Last year, Sol Focus’s solar cell design won the grand prize at the Department of Energy’s 18th National Renewable Energy Laboratory Industry Growth Forum, which also recognized Gary Conley as Clean Energy Entrepreneur of the Year.
The Basics of Photovoltaics
Photovoltaic (PV) systems convert light into energy—or, if you prefer, photons into electrons. The term itself combines the words denoting light (“photo”) and electricity (“voltaic”). The basic building block of any PV energy system is the solar “cell.” Cells are packaged together into modules that electrically connect a large number of individual cells into a single frame. Usually it’s the modules that are apparent to the user, like the individual module that powers a handheld calculator. Modules are further grouped together into arrays. These large panels or sheets are what typically get installed on a house rooftop or are used to construct a single solar collector. Multiple arrays are grouped together into systems, such as solar energy farms covering dozens or hundreds of acres.
Getting back to the solar cells, these are constructed of materials called semiconductors. A typical semiconductor material is silicon, a very common material found in everything from beach sand to computer chips. Simply put, a photon from sunlight strikes an atom of the semiconductor and is absorbed by the atom, one of many atoms arranged into a rigid and regular crystalline structure. The energy of the photon strike is absorbed by the atom and knocks an electron loose from its orbit (or shell) around the atom’s nucleus. The free-flowing electrons constitute the electrical energy generated by the solar cell. Solar cells also create their own electrical fields that direct the freed electrons into an electrical current, which is then used to power appliances and electrical devices. Simple, right? Solar cells should be cheap sources of energy.
Unfortunately, silicon in its pure crystalline form is a poor conductor of electricity. This is one instance where impurities are actually a good thing. Other elements mixed in with the silicon crystal framework cause irregularities that allow electrons to flow more freely by allowing for more free electrons. Silicon crystals with the necessary impurities are called “doped” semi-conductors. Since the excess of free electrons causes an overall negative charge, the process of creating these free electrons with non-silicon impurities is specifically called “n-doping.” Including impurities that reduce free electrons and result in a net positive charge is called “p-doping.” Put the two types in contact and you have an effective pathway for electrons to flow. This pathway is created by an electric field that occurs at the boundary, called a diode, which allows electrons to move only one way—toward the n-doped silicon. Connect a wire from one type of silicon to another, and a direct current is established when light strikes the silicon and frees up electrons. Add the following layers from top to bottom, and you have a basic PV cell: cover glass, anti-reflective coating, top contact grid to connect the circuit, n-doped silicon, p-doped silicon, and the back contact grid to complete the circuit connection.
Photovoltaic Limitations
Until recently, PV meant flat-panel cells and modules. While this allows for some saving in production costs due to inexpensive roll-to-roll fabrication, the material costs are much higher, since almost the entire cell needs to be lined with doped silicon. The doping often involves the introduction of relatively expensive materials, such as gallium arsenide or indium selenide.
Low efficiencies hamper the use of solar cells. Cutting-edge laboratory research has produced solar cells with efficiencies as high as 30% by using a multiple junction configuration. Simpler amorphous solar cells have efficiencies lower than 10%. Most commercially available solar cells have efficiencies of about 15%. And efficiencies often come at a steep price. A high-efficiency photovoltaic cell utilizing expensive materials can cost 100 times more than an easier-to-produce amorphous cell that is one-fourth as efficient. Therefore, four times as many amorphous cells will produce the same amount of electricity for a quarter of the total cost. Sometimes, quantity has a quality all its own.
How does the cost of electricity generated by solar cells compare with other commercially available sources of electricity? The answer is: That depends. It first depends on the location of the PV system. Those located in the sunny Southwest are going to produce more electricity than arrays costing the same amount to manufacture and install that are located in cloudy cold areas like the Pacific Northwest or New England. Any PV array costing the same that produces less electricity will have higher generating costs.
The other factor affecting costs is the overall system efficiency. The direct current produced by PV cells needs to be converted into alternating current to operate household appliances. Furthermore, the sun doesn’t shine all the time. Solar electricity needs to be stored in batteries for use at night and on cloudy days. Depending on the location, commercial-scale solar electricity can be produced at overall efficiencies between 5% and 20%.
The overall cost ranges between 60 cents per kilowatt-hour to 30 cents per kilowatt-hour. Commercially available electricity from traditional coal, natural gas, and nuclear electrical generating plants can run as low as 4 cents per kilowatt-hour to 5 cents per kilowatt-hour, equivalent to one-twelfth to one-sixth the cost of PV. The overall cost of electricity generated by solar cells is based on the amortization of their capital costs (their operating costs are minimal), so the operating lifetime of a solar cell is important in determining its cost-effectiveness. Most solar cells are warranted for 25 years but typically operate for 30 to 40 years.
PARC’s Solar Cell Innovations
PV energy technology therefore finds itself on the horns of a dilemma. On the one hand, it has cheap but inefficient simple cells. On the other hand, it has more efficient but considerably more expensive advanced cells. What’s needed is a design that utilizes the materials of the efficient cells while reducing their costs. This can’t be done by utilizing cheaper materials or by reducing the costs of the added impurities. Gallium arsenide and indium selenide are exotic materials, and they don’t come cheap. The situation requires a complete rethinking and redesign of the solar cell. PARC has done exactly that.
And it’s all done with mirrors.
The technique is called concentrator photovoltaics (CPV), and, as the name suggests, it concentrates incoming sunlight on a small area by means of mirrors and lenses. Concentrator cells have been around for a while, but PARC’s design represents a significant improvement in increased efficiency and lowered production costs. Generally speaking, CPV concentrates sunlight by using exterior mirrors, dish-shaped mirrors, and Fresnel lenses. This allows for the use of smaller PV receivers that require those expensive materials for high efficiency. By the focusing of diffused sunlight onto a small point, CPV systems require only 0.1% of the receiver material required by a flat PV panel of the same surface area. This is a 1,000-times reduction in the amount—and most importantly the cost—of the doped semiconductor material needed to construct a PV cell. The amount of semiconductor material utilized by these concentrator cells operates at twice the efficiency of flat PV. Altogether, the cost of PV electricity generation is cut more than in half from $7 per watt installed to only $2.50 to $3 per watt.
The PARC designs combined a concave primary mirror like a shallow bowl (mirrored surface on the inside of the bowl) that was mated to a convex secondary mirror, also shaped like a shallow bowl (but with the mirrored surface on the outside). The secondary mirror is placed inside and centered on the primary mirror. Both are affixed to a front window that lets in the sunlight through the annulus between the two circular mirrors. Sunlight bounces off of the primary mirror and is reflected back onto the secondary mirror. It then bounces off of the secondary mirror in a tight beam of concentrated sunlight directed at an optical rod. Light passes down the rod, and its energy is focused on a tiny dot of doped silicon semiconductor. This design allows for a wide optical acceptance angle of plus- or minus-1 degree, making solar tracking by the unit less critical. Overall, the design reducing weight (and material costs) is simple to install (reducing installation costs) and requires a simple manufacturing process (reducing fabrication costs). Reduced cost and high reliability from its simple and rugged design are inherent in this CPV.
In fact, its simple design is why it is so inexpensive to manufacture. Compared to other PV cells (and certainly compared to solar thermal units) its total parts bill of materials is relatively low. It is also inherently scalable, and production of modules, arrays, and even whole systems can be fabricated from the megawatt to the multiple gigawatts range. Production is simple and streamlined and can be manufactured by automated processes or assembled manually, an important consideration for nonindustrialized markets.
Sunlight conversion efficiencies increase and power production cost decreases even more when this new cell design is combined with other leading-edge technologies. These include efficient solar trackers and associated driving motors and support frames, the exotic impurities used to dope silicon semiconductors, and multijunction semiconductors. Multijunction semiconductors convert 40% of sunlight into electrical energy through a series of stacked layers of three to five different compound semiconductors. Each type of semiconductor is attuned to capturing a different part of the solar spectrum, allowing for use of a higher percentage of the sunlight. Now these can be combined with concentrator designs that allow for the economical use of these often expensive technologies. Not only do concentrators allow for the use of smaller semiconductors, they also increase the efficiency of the semi-conductor material that is being used. Multijunction cells that use concentrators can achieve efficiencies of 50%.
So who gets first crack at this new technology? Sol Focus will ship its first fully operational systems to customers in the city of Shanghai and in the states of Hawaii and California. Pilot-testing of these systems began toward the end of 2006. A version of the technology targeted at large power plants is expected to be commercially available within 2007. The mirrors that are the heart of the system are manufactured in the same way that automobile windshields are made. Given how widespread this particular automotive manufacturing process is, these can be made almost anywhere. PARC and Sol Focus are also anticipating the benefits of mass production and economics of scale. The current cost of the technology is $2.50 per watt. At mass production rates of 1 gigawatt of capacity per year, the price falls to 50 cents per watt. The second-generation CPV design could reduce costs to 32 cents per watt. These modules and arrays will be suitable for use on commercial and residential rooftops and should be available by 2008.
The Market Potential for Highly Efficient Solar Cells
The merging of high-efficiency semiconductor materials, sun tracking mechanisms, concentrator cell designs that allow for radical reduction in the amount of expensive semiconductor materials needed to produce electricity, and multijunction cells that capture the full solar spectrum results in the perfect technological storm. “Concentrating solar electric power is on the cusp of delivering on its promise of low-cost, reliable, solar-generated electricity at a cost that is competitive with mainstream electric generation systems,” says Vahan Garboushian, president of Amonix Inc. of Torrance, CA. “With the advent of multijunction solar cells, PV concentrator power generation at $3 per watt is imminent in the coming few years,” he adds.
“We have seen steady progress in photovoltaic concentrator technology. We are working with advanced multijunction PV cells that are approaching 38% efficiency, and even higher is possible over time. Our goal is to install PV concentrator systems at $3 per watt, which can happen soon at production rates of 10 megawatts per year. Once that happens, higher volumes are readily achieved. ... The first-generation panels will break price barriers in the market, but the second-generation panels with PARC technology will change the market for solar dramatically,” he says. “The current installed cost of the flat-plate photovoltaic systems is about $7 per watt, but our approach should produce electricity for about half that amount—or less.”
PARC’s clean technology initiative emphasizes market-focused research—and not just in the area of solar energy. PARC also researches similar technologies in the fields of clean water, energy efficiency, and power grid reliability. “We believe there are big market opportunities in clean technology, and we have a half-dozen additional projects under way that could be equally transformative,” says Scott Elrod, head of PARC’s clean technology initiative. Though oil prices (adjusted for inflation) have not risen as high as consumers and the media maintain, one thing is certain: Fossil fuel prices won’t be falling in real terms any time in the near future—if ever. Demand for energy by the emerging economies of India and China (it’s no coincidence than one of the first PARC systems will be shipped to Shanghai) ensures that, even if the demand does not result in a “peak oil” scenario.
Despite all its inherent shortcomings (the sun only shines for half a day, inclement weather can significantly reduce power output, the need to convert the direct current generated by the solar cells into alternating current with the resultant losses in efficiencies, and relatively high up-front capital costs) the potential for photovoltaic energy remains huge. Even without the major breakthroughs described above, all of America’s energy needs could be theoretically supplied by sunlight falling on an area 200 miles square. This equivalent 40,000 square miles (approximately the size of either Ohio or Kentucky) will, of course, not be contiguous. Instead, the organic, market-driven growth generated by the individual decisions of businesses, homeowners, developers, and communities will result in the development of solar farms and rooftop arrays that provide sufficient area for PV to make a significant contribution to our energy needs. Will PV one day provide for all our energy needs? Probably not. Will it be a major player in the energy markets of the near future? Most certainly it will.
Currently, solar energy of all kinds provides a mere 0.06% of America’s energy needs. But two trends have become apparent. First, the solar energy market is experiencing a 30% annual growth rate. What was an $11.2 billion market in 2005 is projected to grow to a $51 billion market by 2015. Secondly, within the solar energy market, photovoltaics are displacing traditional solar thermal systems as the application of choice.
Since CPV cells are still in the research-and-development stage, nearly all of the existing photovoltaic market is for traditional flat-panel cells. Last year, approximately 1,200 MW of flat PV cells were sold worldwide (equivalent to one-third of a day’s output from a typical nuclear power plant).
And North America is not the only market to benefit from these advances. In America we think of the Southwestern states of Arizona and New Mexico when we think about suitable locations for solar energy applications. But much of the world from West Africa to Central Asia is at least as suitable (in fact, if the growth rate of the German solar energy market is any indication, practical applications of solar power may not be as climate dependent as we once thought).
Furthermore, many of these regions are lacking in extensive (and capital intensive) electrical power infrastructure. Commercial developments in Africa could leapfrog the need for such infrastructure, bringing power directly to people and businesses. The history of megaprojects in the less advanced regions of the world has not been encouraging.
A scale-appropriate, slow, organic development of solar energy would be easier to manage, having fewer unforeseen social and environmental consequences and providing immediate benefits to the neediest areas. This approach is surely worth considering instead of more large dams or nuclear power plants.
Daniel P. Duffy, P.E., is employed by the URS Corp. in Akron, OH.
DE - March/April 2007
|
