Solar efficiencies of photovoltaics in relation to the band gap energies?

[cohen990]

I'm reading a paper on photovoltaic hydrogen production and the author claims that for reasonable solar efficiencies the band gap must be less than 2.0eV. This I understand.

My question is how will it effect the photon absorption if the band gap is much smaller.

Say the band gap energy is 0.5eV and a photon of energy 6eV strikes the semi--conductor. Will the photon be absorbed or is its energy too high?

For efficient absorption of solar light, there is clearly an upper limit for band edge energy but is there a lower limit?

Basically I want to know if I can make the following statement - "the lower the band edge energy, the greater the absorption."

 

[mfb]

Usually, an energy above the band gap is fine, as long as it is not really too large - 6eV might be fine, but it can depend on the band structure of the semiconductor.
Another issue is that you would waste 5.5 eV in this setup. If the band gap is too small, you can catch more photons, but the energy gained per photon is lower.