As I understand it, a solar panel will respond to a particular frequency (or multiples of that frequency) by kicking off electrons and thereby getting a potential. Efficient solar panels have more than one “receptor” tuned to a different frequency of light. Is this accurate? Can a solar panel be tuned to respond to various IR wavelengths?
Thanks,
Rob
Most solar panels use visible light, not IR, because it has more “energy” in the wavelenghts.
The material you use when building the PV determines what wavelengths it reacts to. So usually the companies will mix different elements to get maximum output.
Basically, the photon needs to have enough energy to knock an electron from the valence band to the conduction band of the semiconductor. (I.e. the photon energy must be greater than the bandgap.) Which is why solar panels don’t respond to far-IR. (This is also why yo can’t build a thermal infrared camera using a silicon-based detector; you need to use a semiconductor with a smaller bandgap to detect long wavelengths, like InGaAs and HgCdTe.)
If the photon has more energy than necessary for this, the extra energy is wasted - i.e. one blue photon will deposit as much energy as one red photon, even though the blue photon has more energy. So if you were designing a PV cell powered by a monochromatic light of known wavelength, you could “tune” the cell by choosing a semiconductor whose bandgap is close to that wavelength.
There are other sources of inefficiencies too. It’s all included in the Shockley-Queisser Limit.
The response of crystalline solar cells does peak well in the IR range, at around 1,000 nm (visible is 400-700 nm), while amorphous solar cells have a response peaking in the visible range, so crystal structure (or lack of) affects the spectral response. Note that even crystalline solar cells make good use of sunlight because the spectrum extends well into the IR range and they have a much wider response range.
In general, the response depends on the band-gap of the semiconductor used; photons with insufficient energy will be unable to knock electrons loose so response decreases at longer wavelengths; at shorter wavelengths it is because the energy in each photon increases relative to the number (i.e. for 1 W of light energy, twice as many photons are available at 1000 nm than at 500 nm). Also, solar cells are encased in glass, which absorbs shorter wavelengths (explaining why you can’t get a sunburn through glass, unless it is quartz, but nobody uses that).
You’re going to have to use small words here. As I understand it, the photon’s energy must match the energy required to free the electron from the atom. Are you saying that for a silicon atom, that energy is greater than what can be achieved by IR light?
I had heard that if the photon has double the energy, it would knock off two electrons. Is that not accurate?
What I am curious about is why we don’t have PV panels that generate current in response to ambient heat photons that must be bouncing around all over the place. Since I doubt I am the first one to think of that, why is it not possible? Because typical photons are not energetic enough?
What about quantum dots? I understand that by playing with the particle size you can tune these like antennae.
Thanks,
Rob
So, why can’t you make a solar panel that will work with heat photons rather than light? Not enough power to make it worthwhile?
Thanks,
Rob
Essentially, that is the reason; while it is basically infrared radiation, the energy and power are both too low to be practical (note the part about band-gap and photons with energy below that not contributing; power also scales with temperature by the fourth power, so a 5,000 K source emits 16 times more energy than at 2,500 K, and over 77,000 times more than at 300 K, near “room temperature”). In order to be useful (for silicon), the temperature needs to be in the 2-3,000 kelvin range, as shown on this graph (as previously posted, silicon doesn’t accept photons with wavelengths longer than about 1 micron).
Now, it is possible to convert heat to electricity, just not with photovoltaic cells; what you are looking for is a thermoelectric generator; however, they are rather inefficient and generally only suitable for low power loads, especially if the temperature is low, refer to the 4th power law above (some satellites however used thermoelectric generators with nuclear fission as a heat source); heat engines and Stirling engines are a much better choice (but need a generator to make electricity).
