Could mean 35% efficiency boost for solar

[stuartiannaylor]

http://www.worldofrenewables.com/renewables_news/photovoltaics/pv_tech_innovation/scientists_generate_two_energetic_electronic_states_from_one_pho.html

[KLD]

Wow, cunning plan. 

What is the limiting factor in the efficiency of current pV cells? Naively I would have thought that splitting the energy of a single photon and distribute it onto two electrons wouldn't gain anything, but that depends a lot upon how much of the photon's energy is used in a standard pV cells (1 photon --> 1 electron). Anybody in the know?

Klaus

[stuartiannaylor]

Ooooof thats a question. Just a guess with my football knowledge of the quantum. Is it basically that atoms are mainly space and photons and stuff largely can pass straight through. So if you are stuck with a probabilty of an atom strike then wakeing the nieghbour who might of been missed increases efficiency.

That is probably totally wrong but you never know

Nope totally wrong

http://en.wikipedia.org/wiki/Shockley%E2%80%93Queisser_limit

Gooooaaaalllll

[KLD]

Oh, many more variables 
So, we have the distribution of photon energies as they arrive from the sun, and there is an almost fixed value of photon energy needed to lift an electron across the bandgap in a semiconductor. Photons of an energy below the bandgap will not do anything, and those above the bandgap will waste some part of their energy. The new material described in the link in the OP will take photons of at least twice the bandgap energy, and convert them into two photons. Could work! Where can you by the shares? 

Klaus

[Ivan]

I can't see any great advantage (because you can't adjust the sun to produce only high energy photons). The real trick would be to add together any excess energy from photons which have displaced an electron and sum these excesses to use for displacing further electrons. That would be breaking a lot of rules, hence the current relatively low limit of PV efficiency.

[KLD]

Ivan,

Could you elaborate a little on this? My whole "knowlegde" here is limited to what's stated on Wikipedia... Bit thin, in other words.
The minimum photon energy required to lift a photon from the valence band into the conduction band (for silicon) is about 1.1eV, that corresponds to light of a wavelength of 1100nm. Higher photon energies are wasted into heat (so, sadly, not single "leftover energy packets" per absorbed photon, which you'd like to collect together and make one new, high energy photon from). The wavelength distribution of sunlight provides many more photons at higher energy. If you could make use of the photons at below 550nm to make two electrons, might that not add quite a bit to the efficiency of the panel?

Klaus