The research demonstrates the potential for solar, or photovoltaic, cells that reduce wasteful heat and maximize the amount of the sun’s energy that is converted to electricity – a key step toward making solar energy more cost-competitive with conventional power sources.
The NREL research team, led by Arthur Nozik, included Randy Ellingson, Matt Beard, Justin Johnson, Pingrong Yu, and Olga Micic, and worked in collaboration with theorists Alexander Efros and Andrew Shabaev of the Naval Research Laboratory (NRL) in Washington DC.
The findings are further confirmation of pioneering work by Nozik, who in 2000 predicted that quantum dots could increase the efficiency of solar cells, through a process now termed “multiple exciton generation,” or “MEG”. Last year, Richard Schaller and Victor Klimov of Los Alamos National Laboratory in New Mexico were the first to demonstrate the electron multiplication phenomenon predicted by Nozik, using quantum dots
made from lead selenide.
“We have shown that solar cells based on quantum dots theoretically could convert more than 65 percent of the sun’s energy into electricity,
approximately doubling the efficiency of solar cells,” Nozik said. The best cells today convert about 33 percent of the sun’s energy into
electricity.
The NREL and NRL researchers’ paper also describes a new theoretical foundation for the multiple exciton generation process that is based on certain unique aspects of quantum theory.
The recent work demonstrates MEG in quantum dots of a second semiconductor material, lead sulfide.
The NREL/NRL work not only shows higher overall efficiency for multiple exciton generation, it also establishes that the process occurs with lower photon energies, meaning it could make use of an even greater portion of the sun’s light spectrum.
Beyond potential use for photovoltaic cells, similar quantum dot technology may someday be used in photoelectrochemical cells, which could become a clean and renewable way to produce hydrogen directly from water and sunlight.
SOURCE: NREL Press Release
I’d heard that nanostructures might have applications in photovoltaics, but hadn’t seen a good description of this before, thanks!
The main point here is dealing with the wide solar spectrum in an efficient manner. Semiconductor photovoltaics are characterized by an energy gap, and the energy captured from an absorbed photon in releasing an electron-hole pair is generally just that gap value. If the gap is too high, only high-energy photons can excite across it, and you miss out on most of the spectrum. But if the energy gap is too low, you do generate an electron-hole pair for a wider portion of the spectrum, but the energy that can be recovered from each photon is only that small gap energy, so efficiency is reduced there too.
The traditional way around this is to have a multi-junction material with perhaps 3 layers: one has a high gap and captures higher-energy photons, one has a medium gap, and one has a small gap.
The solution here though is quite different – you stick with the small gap material, but instead of generating just 1 electron-hole pair, you generate 2, or 3, or more, thus capturing substantially more of the energy of a higher-frequency photon.
Brilliant, I say!
Thanks for the comment. I guess it boils down to getting the best quantum efficiency from your materials. I always remember talking to Cambridge University’s Jeremy Sanders about this a long time ago when I was writing about his superb supramolecular work for New Scientist. One of the possible applications was as a novel photon-trapping system analogous to those used in photosynthesis. He pointed out that no matter how good the material, nothing would ever beat the direct heating of water in a metal container on a hot sunlight roof. I guess that’s fine for those in California or on the Mediterranean, but for those in more temperate climes we’ve got to exploit all the photons we can!
Actually, as soon as you’re over about 30% electrical conversion rate, you’re beating direct heating, because direct heating does not produce useful work – it produces heat. If all you want is heat, that’s fine, but different types of energy aren’t quite fungible because of that entropy thing – electric energy is about 3 times as useful as heat, roughly speaking (steam turbines are typically about 30-35% efficient in converting heat to electricity).
So this 60% potential here is really very, very impressive.
Yes, I was less than specific. Sanders was indeed talking about the simple process of heat production rather than producing energy useful for doing work. I think what he was alluding to was those people who stick photovoltaics on their roofs to produce electricity to heat their water, that surely introduces energy conversion inefficiencies
If you have electricity you can do better than direct heating using a heat pump. See here:
http://energyoutlet.com/res/heatpump/efficiency.html
I’m just showing myself up as a non-physicist aren’t I? Thanks for the additional explanation.