Searching for Better Solar Solutions
|The sun has always been humanity’s ultimate power plant. It fueled the crops that allowed the earliest civilizations to form. It drove the winds that propelled people across oceans in search of new lands. It nourished the growth of plants and algae that, squeezed together under rock and dirt for millions of years, turned into the fossil fuels that are the lifeblood of our societies today.
For more than four billion years, the sun has blasted Earth’s surface with more than 120,000 trillion watts of energy every year—enough to take care of current global power consumption many thousand times over. And yet, capturing even a tiny fraction and turning it directly into electricity or fuel has been an enormous challenge. Today, more than half a century after the invention of the silicon photovoltaic cell, solar energy contributes less than one percent to the United States energy portfolio.
However, the photovoltaic market is gaining ground. “If you look at the solar industry, it’s been growing between 40 and 50 percent per year during the last decade, and as fast as 70 percent per year for the last several years,” says Reuben Collins, physics professor and associate director of the Colorado School of Mines Renewable Energy Materials Research Science and Engineering Center. Collins is one of many scientists around the world who are ramping up their efforts to chip away at photovoltaic cells’ two main shortcomings: high cost and low efficiency.
With prices currently between 25 and 50 cents per kilowatt hour, solar electricity is roughly five times more expensive than coal. Solar’s sassy price tag is the result of two things: conventional photovoltaic cells convert only a fraction of the sunlight that hits them into electricity, and the materials they are made from are very expensive to produce. For example, more than 80 percent of the solar cells that currently keep light bulbs, laptops and TVs running are made from slices of bulk silicon that consists of either single or multiple crystals—structures in which all atoms are arranged in perfect order. Making these materials is pricey, says Craig Taylor, physics professor, director of the Colorado School of Mines Renewable Energy Materials Research Science and Engineering Center and associate director of the Colorado Energy Research Institute. “The manufacturing process involves many steps and it also wastes quite a bit of silicon.”
The most efficient and widely used photovoltaic cells are made from slices of monocrystalline silicon about 300 micrometers thick. Although a depth of only about 10 micrometers of silicon is needed to create an electric current, the practicalities of slicing and engineering the material require additional depth. However, a major thrust is currently under way to make so-called thin film solar cells, which use about 90 percent less silicon. “That’s one of the primary ways to take the cost out,” says Richard Ahrenkiel, research professor of metallurgical and materials engineering.
Thin film cells work just like their thicker cousins: high-energy photons, the energy carriers in light, hit the solar cell, kick electrons away from the atoms they belong to and push them over an energy barrier, called a band gap. Each dislodged electron leaves a hole behind into which another electron can fall, which in turn opens up an empty spot for yet another electron. Because cells are designed such that electrons move in one direction, this game of musical chairs ultimately results in an electrical current, with displaced electrons collected and conducted out of the cell through the negative terminal and back into the cell on the positive side. Put an electrical device into the circuit, and the current is put to work.
Imperfections or blemishes in the crystalline structure, such as a misplaced atom or a distortion in the lattice, act like obstacles, hindering the flow of electrons; instead of making it across the band gap, they lose energy and fall back into their original position without producing current, Taylor says. In fact, keeping these charge carriers—electrons and the holes they leave behind—apart from one another long enough to siphon off an electric current is essential for all types of solar cells. This so-called carrier lifetime has to be at least a few tens of microseconds long for the cell to be useful. “All of the photovoltaic technologies critically depend on this number,” Ahrenkiel points out. The carrier lifetime in a thin film cell will depend on its structure, which is influenced by how the film was grown.
Elaborate growing processes usually produce better material, but they also drive up cost. “Anybody can make an expensive material with a long carrier lifetime,” Ahrenkiel explains. The trick is to make cheap materials with long enough carrier lifetimes to yield adequate currents. Within the Colorado Collaboratory Center for Revolutionary Solar Photoconversion, Ahrenkiel and his students have started a project to characterize thin films grown with different techniques to find out which produce good films and which don’t. Taylor, who is involved in a similar project, explains that one of the main drivers of cost is the speed of the deposition process: slower means more expensive but usually also better quality, he says. That’s why “there is a big push among a lot of the thin film companies to increase the growth rate and keep the quality of their product the same.” If companies could grow their films twice as fast, they would automatically double their production. “That would be a huge benefit to them even if the efficiency stayed exactly the same.”
Imperfections are not only made during the manufacturing process, they creep into any crystal over time. Constantly blasted by high-energy sunlight, atoms are pushed around and structural defects result. “That’s a critical problem for really all solar cells,” Taylor says. That’s why one question needs to be answered for all new materials: can they maintain their efficiency after operation for long periods of time? Taylor and his team are trying to provide answers by mimicking this process in the lab. They shine high-intensity light onto the materials constantly for several days and then use various techniques to examine whether any damage was done. Those materials that hold up best are the top candidates for photovoltaics.
But no matter the quality of the material, ultimately it is the laws of physics that dictate what is possible—and they put a firm ceiling on solar cell efficiencies. Photovoltaic cells only convert between 8 and 25 percent of solar radiation into electricity because they can use only a narrow window of the solar energy spectrum. Photons carrying just the right amount of energy needed to excite electrons over the band gap are converted into electrical current quite efficiently. But sunlight is made of a large variety of photons: those of blue light, for example, carry more energy that those of red light. Photons without sufficient energy may impact an electron, but if it’s not kicked over the band gap it produces no current. And photons carrying enough energy to bounce two or three electrons across the gap, can only transfer their energy to one—the residue is converted to heat. This means that “you are basically throwing away a fair fraction of the energy that you absorb,” Taylor points out.
Altogether, this puts a theoretical limit of about 30 percent on efficiency. Stacking various materials—each with different band gaps—on top of each other can push that limit up by an additional 10 percent or so. This technique is often employed in photovoltaic cells used in space, but it’s “way too expensive for most terrestrial applications,” Taylor says. For a long time scientists thought that this would be as good as it could possibly get, but it has recently become clear that the weird laws of quantum mechanics ruling the nanoscale world may open a door to get around the fundamental limits.
An electron in a solar cell that is kicked into a higher energy or excited state can shake off its excess energy in a number of different ways: it can jump across an energy barrier and flow as an electrical current, but it may also give off heat or emit light. When using nanoscale materials, such as tiny dots or wires, scientists at Mines are working on ways to manipulate the ways in which excited electrons give off their energy, nudging more of them into producing current, rather than heat or light. In the world of quantum mechanics it’s even possible for one photon to kick more than one electron loose. So, by fine-tuning nanostructures, scientists may be able to “optimize the amount of energy that goes into photovoltaic conversion and minimize the energy that gets lost,” says Collins. That way, much higher efficiencies than the limit of 30 percent are theoretically possible.
However, Collins cautions that nanostructure-based photovoltaic cells are very much a “down-the-road kind of thing.” Even though “there is a lot of interest, and if it works, it’s got a great deal of promise.” Scientists have yet to figure out how to make the process work, Collins adds, although they are certainly trying. Collins, along with Taylor and Pauls Stradins at the National Renewable Energy Laboratory, have started to look at nanowires, tiny strands of silicon and other materials that have a small enough diameter to follow the bizarre rules of the nanoworld, but at the same time provide a path for the flow of a current.
If scientists manage to boost solar cells’ efficiencies and bring down the cost of solar power dramatically, there will be another major limit to its usefulness—storage. No efficient options currently exist for storing solar power on a large scale. Even though many scientists agree that solar energy will likely become an important complement in our mix of future energy sources, “unless we have some real breakthroughs in distribution and storage solar power is not going to be the be-all, end-all,” Ahrenkiel says.
The storage problem may be worked out in the future but for photovoltaics to become a major player in our energy portfolio “there is still a lot of work that has to be done,” Collins says. And that requires large research investments. Recent global growth in the solar industry was primarily due to government action, such as legislation requiring utilities to invest in renewable energy technology, or tax deductions to consumers who install photovoltaic cells on their homes. An alternative market mechanism under consideration is a carbon tax, which would raise the price of coal-generated electricity. But even with such a greenhouse gas penalty, the cost of conventional photovoltaics will likely remain high compared to coal; that’s why looking into new and innovative approaches, such as nanostructure-based solar cells, is critical, Taylor says. Empirical laws that forecast future progress based on history — such as Moore’s law which predicts how much faster computer chips get each year — currently cast a gloomy outlook for photovoltaic cells, he says. Right now, photovoltaics make up about 0.01 percent of the total power mix and looking at solar cells’ analogue of Moore’s law shows that “in 2050, we might be at one percent or so.” But a breakthrough in efficiency could be a game changer because it would likely let the price of solar power plummet. And “if you get the cost to where you are competitive with coal-fired power plants, then you have a chance."
Concentrating Solar Power
Ask someone what they think the world’s largest solar power station looks like, and most people will tell you that it’s probably a huge array of shiny photovoltaic cells. But the facility located in California’s Mojave Desert has nothing to do with photovoltaics. Instead, it’s a solar thermal system, consisting of nine individual plants with a collective capacity of 354 megawatts, which have been churning out electricity since the mid-80s. The Solar Energy Generating Systems power plants provide about half a million Californians with power, and unlike photovoltaic cells, they are doing it without expensive semiconductors. Instead, the system uses mirrors to focus sunlight onto long tubes containing a synthetic oil which is used to boil water. The steam is then used to spin a turbine to produce electricity.
A few such concentrating solar power (CSP) systems are scattered across the world, but so far the technology makes up a minute part of global energy production. The main obstacle to large-scale implementation is cost. Even though CSP power is less expensive than photovoltaic electricity, it still doesn’t reach the low price of coal. “As long as there is no penalty for carbon emissions, you just can’t touch coal,” says Thomas McKinnon, professor of chemical engineering at Mines. Another issue is access to power lines, which generally don’t exist in places where CSP technologies work best: deserts. Unlike photovoltaic cells, CSP systems must have direct sunlight, because the light needs to be focused. Even a thin cloud layer can interfere with the process. “For example, a state like Florida has a lot of sunlight, but they also have a lot of high diffuse clouds; that reduces the efficiency of CSP.” That pretty much limits the places in the United States where CSP could be successful to the Southwest, but that’s actually all that would be needed. “There is more than enough to power the entire country with just what’s in the Desert Southwest,” McKinnon says.
Because CSP is generating hot fluids, it also allows for some fairly easy storage options. For example, it is possible to heat molten salt when the sun is shining and store it in insulated tanks. When electricity is needed, the hot salt is used to heat water to drive the steam turbine generators. In addition to that, CSP technology is already developed. “That means you could deploy hundreds of megawatts of CSP fairly fast,” McKinnon says. And it looks like that might be beginning to happen. For example, Xcel Energy plans to build a 200-megawatt CSP plant in Colorado by 2015, and the Israeli company Solel Solar Systems intends to have its new Mojave Solar Park—a more than 553-megawatt facility—up and running in the Mojave Desert by 2011.