Regarding "Holes" in bands, and Photons

Question

So from learning Band theory, and PN Junction and such, I've learned that photons are created when "holes" are filled in a band, and this is what can create light (Isn't this how LEDs work?)

Anyways, my question is - How come when Electrons move between conduction bands light isn't produced? Or is it and it's just so small we can't tell?

Because the conduction band is technically still an orbital, which means it can have "holes", right?

Secondary Question: Is there a "Point" in the orbitals in which the "Grip" a nucleus has on an electron becomes basically insignificant. Like I know it gets less and less as the Bands go outward, but is the conduction band basically a point where it's just "decided" that the electron at that band will probably not be attracted to the nucleus more than the thermal energy produced in a normal environment is much more?

Answer 1

When an electron is promoted from the valence band to the conduction band it leaves a hole in the valence band. When the electrons falls back down from the conduction band to the valence band (to fill the hole) energy is released, but not necessarily as light. A photon can be emitted only when the material has a direct band gap i.e. when the electron in the lowest energy state in the conduction band and the hole in the highest energy state in the valence band have the same momentum. If this isn't the case (and for most semiconductors it isn't) the electron must interact with the crystal lattice, and energy is lost to the lattice rather than as a photon. It actually takes careful design to make diodes emit light!

Because the conduction band is technically still an orbital, which means it can have "holes", right?

No, by definition you can only have a hole in an energy band that is full e.g. the valence band. A hole is what is left behind when you remove an electron from a full band.

Re your last question, it depends on how localised the electron wavefunctions are. You're quite correct that the inner electrons on the silicon atoms are localised and can't jump from atom to atom. When the average distance of the electron from the nucleus is comparable to the interatomic spacing the electrons feel an "average" force from all the nuclei in the crystal and they may form delocalised wavefunctions. Exactly when this happens will depend on the fine detail of the atoms and the crystal structure.