When light is absorbed by a semiconductor, it will only lead to generation of EHPs if the light has energy greater than or equal to the bandgap. Now, all the sources I've been able to find online say that any excess energy in this process (ie, E>Eg) is simply reabsorbed into the lattice as the electrons and holes return to the conduction/valence levels. My question is: is this an immediate/definite process? Is there some sort of "diffusion length" of higher-energy electrons? If we're using an indirect-bandgap material like silicon, does this change the issue at all? Basically: are we forever stuck generating electrons of only energy Eg in semiconductors, or is there another way? Thanks for any thoughts!