Re: Is it possible to change the energy level of a photon?

Hi Thomas,

It is possible to do both of the things you describe: "combining" two photons to make a higher-energy one, and "shearing off" a high-energy photon to make a lower energy one. In practical terms, I do not think these are of any use for photovoltaics (solar cells). See what you think:

Atomic physicists sometimes deal with "two-photon" processes. This is just what it sounds like: an atom absorbs one low-energy photon, then another, and the net effect is the same as absorbing one higher-energy photon. In my lab, we use a powerful near-UV nitrogen laser to ionize argon atoms - each atom must absorb two photons, almost simultaneously, in order to ionize. However, the two-photon process happens only very rarely; when we shine a few milliwatts of laser energy, focused to a small point, onto our gas sample, we get only hundreds or thousands of ionizations. (The number of photons pair-absorbed is proportional to the square of the intensity.) Even if you tightly focus sunlight onto your silicon, you'll get similarly few instances of two red photons "combining" to make higher-energy excitations.

For photons with an energy higher than the solar cell's band gap, they do get absorbed, but the "excess" energy becomes heat rather than electricity. For example, you might have a blue photon with 3 eV of energy; if the material's bandgap is 2eV, then the photon creates 2 eV of electrical energy and 1 eV of heat. One way around this is to use multi-junction solar cells. If you put a 3-eV-bandgap layer in front of a 2-eV-bandgap layer, then the blue photon will deposit its full 3 eV into the first layer; a 2 eV green photon would pass through the first layer and deposit its full energy in the second. Indeed, this is exactly how some of the most efficient solar cells are made.

The other thing you can do with high-energy photons is absorb/reemit them. That is how glow-in-the-dark paint does; it absorbs high-energy photons (exciting some molecular energy levels, I don't know the details) and slowly reemits them at lower energies. Many lasers also work this way, using a UV or white light lamp to excite atoms, which then emit redder light. It may be possible to find some system which (for example) can absorb a 3 eV photon, and reemit two 1.5 eV photons. I'm sorry I can't propose one, since this is not my specialty, but you should read about laser transitions (try here) to see what I'm talking about. However, I think that all such systems will be fairly inefficient. (If nothing else, the re-emitted photons will come out in random directions; a layer of downshifting paint on top of your solar cell, even if the absorbtion/reemission were perfect, would emit 1/2 of its photons back into the sky).

Hope this helps! It's a very interesting topic, and it has been fun watching the efficiency climb up (and the price down) over the past ten or fifteen years. Unfortunately, it always looks like the cutting edge of efficiency is with the most delicate, complicated, and expensive cells - multilayer germanium or whatnot. I'm more excited about new technology on the cheap end: can we make a 5% efficient solar cell as cheap as a pane of glass? Or a 1% efficient cell as cheap as tinfoil? This may be possible someday with organic semiconductors.

-Ben