According to many energy experts, the world will need 30 terawatts (TW) of energy resources in place by the year 2050 to maintain economic growth and accommodate the increasing energy requirements of rapidly expanding economies such as China and India. This compares to today's 13 TW of continuous worldwide energy consumption, of which the U.S. uses one-fourth. The greatest part of this new energy will have to come from non-carbon sources to avoid irreversible and catastrophic climate change.
To accommodate this kind of explosive growth, a source of energy is required that is almost limitless, and yet does not add to greenhouse emissions. For this reason, many scientists are looking toward the sun for a solution.
More energy strikes the earth from sunlight in one hour than all the energy consumed in the planet in a year, according to Cal Tech materials chemist Nate Lewis, who co-chaired the Department of Energy's Solar Energy Workshop. He estimates that a minimum of 600 TW of useable solar energy is available worldwide. With a solar energy conversion rate of 10 percent, well within current capabilities, the sun could provide us with 60 TW of energy, about double the amount we will require by 2050.
Solar Energy Tipping Point
Another solar cell authority, Lawrence Kazmerski, director of the National Center for Photovoltaics at the National Renewable Energy Laboratory (NREL) at Golden,Colorado, came to Penn State to deliver the 2006 Taylor Lecture. On the 50th anniversary of the development of the first practical solar cell, at Bell Labs in 1956, Kazmerski expressed the belief that the use of solar energy worldwide is at a “tipping point” for explosive development. In Germany, for instance, solar panels have been installed on 700,000 roofs and a “feed-in” tariff has been established by the German government that guarantees a high rate of return to consumers for generating electricity and returning it to the energy grid. He warned, however, that the U.S. may be losing out as other countries forge ahead with government support. “The United States is losing photovoltaic market share,” Kazmerski told the audience. “We need to maintain consistent federal research funding or scientists will leave this field.”
He also expressed concern with the slow process of commercializing solar technology in this country. Holding a 1950s-era solar-powered transistor radio, Kazmerski said, “This radio was in production within a year of the development of the first solar cell. We don't do that anymore. We have to cut the time to market, which is currently eight years. We need to get going.”
Second Generation Solar Technology
A second generation of solar technology with either much greater conversion efficiency or lower materials cost is ready to come to the market, he said. This technology, which includes thin films, organics, concentrators, and thin silicon wafers, will have a disruptive effect on the energy field. This is a critical time for research and development in materials science and chemistry. “If we don't invest in 3rd and 4th generation solar cell research, in 25 years we won't own any of them.”
Greenhouse Gas Problems
Although energy security is a national concern, says Tom Mallouk, Dupont Professor of Materials Chemistry and Physics in Penn State's College of Science, the environmental problem from the accumulation of greenhouse gases may be more serious. With the growth in energy use, which is expected to double by 2050, we could be facing environmental disaster in 10 to 50 years, he believes. “For students thinking about what to do with your life, this is the most important problem.”
In his solar cell research, Mallouk and his Penn State colleagues are using an approach that involves photonic crystals. These optical nanostructures slow the motion of photons through the solar cell material so that the cell can absorb more of the red wavelengths of light. So far, this method has shown a 25 percent increase in the current 10 percent efficiency of traditional dye-sensitized solar cells. By contrast, silicon solar cells are already 25 percent efficient, but cost too much to mass produce. With Penn State colleague Joan Redwing, Mallouk is hoping to lower the materials cost by using silicon nanowires.
The Devil Is In The Details
Penn State professor of electrical engineering Craig Grimes is working to create scalable and affordable materials to capture some portion of Nate Lewis' estimated 600 TW of usable solar energy. His group at the Materials Research Institute is involved in the 3rd and 4th generation of solar cell technologies, using nanoscale architectures to improve the performance of less expensive materials.
“The point about our work is that it is a process that is tremendously scalable to industrial applications. The way we make our nanotube arrays is through anodization. You take a big vat of chemicals, put your materials in, and strike a DC potential across it. This is the same way they make Calphalon™ cooking utensils,” Grimes explains.
When it comes to solar cell technology, like many things, the devil is in the details, he says. “For example, you can have a great material that absorbs lots of sunlight and efficiently generates electron-hole pairs, but then the photogenerated holes destroy the material. Whatever materials you bring to the table in trying to solve the global energy-carbon balance, they need to be cheap, have fast charge transfer properties, and be stable against corrosion. The ability to get these properties comes back to the specific material and its architecture.”
Ordered Titanium Oxide Nanotubes
The architecture that Grimes' team first developed at Penn State in 2001 is made up of arrays of highly ordered titanium oxide nanotubes. Those original nanotubes were on the order of 400 nm long. Over the intervening years they figured out how to go from 400 nm to 360 microns. “Why that is significant,” Grimes says, “is that the longer the nanotube, up to a point, the better it is at light absorption. Part of this equation is that you need to absorb the light, and the light generates an electron hole pair. You need an architecture where the electron is happy going one way and the hole is happy going the other way. We've gotten to the point now where we can control the length of the nanotubes, the wall thickness, and the intrinsic geometry. We have the ability to precisely control the architecture, and we now have a wall thickness of 4 or 5 nanometers.”
Grimes uses the arrays in water photolysis, converting sunlight into chemical energy. By 2006, his team had achieved a 16.25% water photolysis photoconversion efficiency rate. “It's an amazing architecture for photoconversion,” Grimes says. “This is remarkable efficiency for such a readily synthesized, single bandgap material. The negative is that it responds to ultraviolet light. Why is that an issue? The energy of UV light is only 5% of the energy that trickles down through the atmosphere.”
To come up with an answer to this problem, Grimes and his team are varying the chemicals in the anodization bath to dope the nanotube arrays as they are being fabricated. Using boron, they have successfully doped the nanotube arrays and dramatically improved the properties. While boron hasn't shifted the bandgap into the visible spectrum, it has shifted the absorption characteristics and improved the properties. “If we can find the right formula, we might shift that 16.25% into the visible spectrum.”
Another Promising Approach
Recently, Grimes and Ph.D. candidate Haripriya E. Prakasam have been exploring another material with exciting possibilities: a form of iron oxide called hematite. With a bandgap of 2.2eV, hematite will absorb about 45% of the wavelengths of light. With the right kind of nanoscale architecture, Grimes believes that hematite might be made to deliver a similar 16% efficiency as titanium oxide, but in the much larger energy range of visible light.
Iron Oxide Solar Cells
Iron oxide makes up a big portion of the Earth's crust, and that makes it about as cheap a material for solar cells as you can get, according to Grimes. However, iron oxide is well known to have very poor charge carrier mobility, meaning the electrons and electron holes don't move very well. In the past, this has made iron oxide a poor candidate for photoconversion. Grimes and Prakasam think the answer to poor mobility may be a highly ordered nano architecture.
Making and Ideal Solar Cell and Increasing Efficiencies
“Right now we have a nanoporous surface in which the space between the channels becomes the wall thickness,”; Prakasam explains. “This can be 20 nm or 30 nm across. This is still not ideal. Having a shorter distance for the holes to travel is a must. If we can develop hematite tubes of 5-10 nm wall thickness, something we have accomplished in our TiO2 nanotube arrays, there is much more possibility of the hole reaching the electrolyte and the electrons reaching the contact without high recombination loss. With shorter nanotubes on a conducting substrate, we would have an ideal situation. We would have more surface for electron harvesting, and since it is oriented vertically, the electrons would have no trouble getting to the contact and the holes would be almost 100% assured of getting to the solution.”
This research is still in the beginning stages, with efficiencies of less than 1%. However, there are many stages of optimization ahead, Prakasam says. For Grimes, finding a way to produce cheap and abundant carbon-free energy should be a national priority. “If we're ever going to solve this 40-terawatt question, that is to say powering the world without cooking ourselves, it's going to have to be something cheap, efficient, and stable. As of today, we can only hope that somebody discovers something like this. That's why I think this work is pretty significant.”