Solar panels are becoming an increasingly common sight on rooftops. These panels are made up of photovoltaic cells, which absorb the photons from sunlight and energize electrons in the cell’s material, creating electricity. The current maximum efficiency of commercial photovoltaic cells is, however, about 20 percent. This low efficiency results from the fact that only photons with a certain amount of energy—that is to say, only part of the solar spectrum—can sufficiently energize the electrons to form the current; other photons are essentially wasted.
Sunlight can also be converted to thermal energy, or heat, which can then be used to generate electricity as well. The advantage is that none of the spectrum is wasted; all of them can be converted to heat. Generating electricity from solar-thermal energy, however, usually requires a large-scale system, with an array of mirrors that reflect and concentrate sunlight onto tanks or pipes filled with water or other liquids. The heated liquid is used to generate steam, which turns a turbine, generating electricity. Although the efficiency of these systems is higher than photovoltaic cells—around 30 percent in some cases—they can’t be scaled down to rooftop applications.
To overcome the various drawbacks of photovoltaic and solar-thermal systems, a team of researchers at Massachusetts Institute of Technology has created a new device that combines elements of both, which they describe in a January 19 paper in Nature Nanotechnology. (Scientific American is part of Nature Publishing Group.) Their invention is known as a solar thermophotovoltaic device. Whereas other researchers have built them before, the new device is the most efficient one yet, says Evelyn Wang, an associate professor of mechanical engineering at M.I.T. and the paper’s senior author. Despite that advance, though, the device only achieves about 3 percent efficiency. “There is really much more potential in this technology,” Wang says. “This is just a starting point.”
Still, the achievement is noteworthy. Two years ago, Alejandro Datas, a researcher at Polytechnic University of Madrid’s Solar Energy Institute who was not involved in the study, built a solar thermophotovoltaic device that achieved about 1 percent efficiency. He says that tripling the efficiency in that time is significant.
To build their device, the M.I.T. scientists used carbon nanotubes, which are extremely effective absorbers of sunlight; they approach theoretical “blackbody absorbers” that take in 100 percent of light shone on them. “That’s the best kind of absorber you can have,” Wang says. The scientists shone concentrated sunlight onto the carbon absorber, heating it to about 1,000 degrees Celsius. The absorber is attached to a photonic crystal composed of a stack of silicon and silicon dioxide layers that begins to glow at such high temperatures. The glowing crystal emits photons, which travel to the underlying photovoltaic cell. Unlike regular sunlight, however, most of the photons emitted by the crystal have enough energy to excite the electrons and generate a current. By first converting the sunlight to heat and then back into light, the device fine-tunes the energy of photons absorbed by the photovoltaic cell, maximizing the electricity-generating potential.
Because the carbon nanotubes are such efficient absorbers of sunlight, they don’t waste any of the spectrum, converting nearly all of it into heat energy. And because the sunlight is also transformed into heat, that energy can be stored more easily than the direct electricity that photovoltaic cells produce, Wang says. “You can store the energy using thermal or chemical means,” she adds, such as by using a chemical such as molten salt that liquefies when heated and then gives off that heat when it later solidifies.
Andrej Lenert, a PhD student at M.I.T. and the lead author of the paper, notes that “anytime you go through this thermal conversion process, it lends itself to the possibility of storing that energy as heat.” This capability allows solar energy stored as heat to later be converted to electricity, say at night or when the sun isn’t shining. Storing electricity from conventional photovoltaic cells requires batteries, which are impractical at rooftop scales and expensive at larger scales.
Besides the increase in efficiency, Lenert thinks his group’s work will provide a framework for future advances in solar thermophotovoltaics. “The experimental procedures that we’ve established, and the methodologies … I think will benefit the community going forward,” he says.
The big step, of course, will be to surpass the 20 percent efficiency mark set by photovoltaic cells. Wang thinks they are well on their way. Part of the problem, she says, is scale. The device they built is about the size of a fingernail; because the area is small relative to the length of the edges, more heat is lost through inevitable dissipation. Increasing the size will exponentially increase the area compared with the length, reducing heat loss. “If we can scale up…, then we can get over 20 percent efficiencies,” Wang says.