Solar Thermophotovoltaic Cells Can Generate Electricity at Night

Photovoltaics could have a sunny outlook as an energy source if they can overcome a couple of deficiencies. First and foremost, they don’t generate power at night. And second, inexpensive single-junction PV cells only absorb a narrow band of wavelengths from sunlight, as shown here:


Researchers at MIT may have solved both of these problems with one device: the solar thermophotovoltaic (STPV) cell. The idea is to concentrate sunlight onto a solar cell, grab most of the solar spectrum, and convert it to the wavelength that photovoltaics prefer. The multilayer STPV cell incorporates an absorber made of carbon nanotubes that absorb most of sun’s spectrum and convert it to heat, and an emitter made of of a silicon photonic crystal that glows when heated. The emitter’s output is tuned to the wavelength at which PV cells operate most efficiently:


(Those drawings didn’t come from the researchers; they’re my graphical interpretations, not drawn to any particular scale.)


STPV makes more solar energy available for conversion by tuning the energy to match the PV cell’s prefered wavelength. In other words, the PV cell is receiving more solar energy at the right wavelength, allowing it to produce more electricity with the same overall solar input.  


What happens at night? Heat can be stored more easily than electricity, so panels made of solar thermophotovoltaic (STPV) cells can generate electricity at night, assuming they hold on to the heat that they absorbed during the day. The researchers are a little sketchy about how the heat would be stored. I’m wondering if the emitter material can have a relatively large thermal mass, so it heats up slowly and retains its heat after the sunlight disappears. That way, instead of getting a burst of energy when the sun is shining and nothing after sunset, you’d get a more steady energy production day and night.


STPV isn’t exactly a new idea; the MIT researchers simply improved its performance. Past attempts at using STPV cells have produced experimental efficiencies of around 1%. MIT’s team brings that up to 3.2%. They believe that with slight improvements, STPV will soon achieve around 20% efficiency, making it competitive with today’s PV panels. Theoretically STPV could reach 80% efficiency, according to Dr. Evelyn Wang, associate professor of mechanical engineering at MIT, one of the lead researchers on this project. But even at 20%, the ability to store energy as heat and convert it to electricity later gives STPV an edge over a 20% efficient PV panel.


Video: MIT


You still can’t capture lightning in a bottle, but with solar thermophotovoltaics, maybe you’ll be able to catch a ray of sunshine to use on a cloudy day.