Light from the sun is made up of a number of different wavelengths ranging from 1 nanometer to 1 millimeter. Solar cells aren’t reactive to the whole spectrum, but only a small portion of it, so the rest of the spectrum outside of the reactive range is essentially wasted, absorbed by the glass and structure an converted to heat instead of electricity.
Although photovoltaic solar panels have been around for decades, they haven’t really improved on their efficiency all that much. Only about 20% of solar energy they capture is converted to electricity, while the rest is converted to heat.
The overall result is that solar panels take up more space than they really have to. Researchers have been working on increasing efficiency, and Harry Atwater, Professor of Applied Physics and Materials Science at California Institute of Technology [CalTech], is revisiting an old approach.
Solar cells can be made to react to different wavelengths of light, so the question is, “What if one could split the light into these different wavelengths and feed it to specific solar cells reactive to those wavelengths?” Various techniques have been attempted, but none with very much success or simplicity. Prisms can do the job, but end up being bulky and inefficient because of flaws in the optics. Another approach is to stack different reactive layers, and efficiencies as high was 43% have been reported with this method. The stacked method is expensive and power output is limited by the performance of the least efficient layer.
Professor Atwater is approaching the same theory, using nanostructured materials to split sunlight into eight or ten colors to convert more of the the solar spectrum into electricity, possibly more than 50%. The nanomaterials used can split and guide light using features smaller than the wavelengths needed, and because of the nature of the material, can be made flat.
Currently, production of nanostructured materials needed for such solar panels is only just starting to come online. A solar panel using the new technology might not look much different than other solar panels. Twice the efficiency is nothing to sniff at, but the pricing on such a device might be prohibitive until mass production can begin.