Carrier relaxation within a quantum well

Hi

Carriers injected in the conduction band of a bulk semiconductor, either by optical or by electrical injection, are relaxing to the lowest free states near the band edge by transmitting their excess energy and momentum to phonons before they recombine with holes in the valence band. This is fine for me. However, I don't know if this is the same situation for a quantum well with its discrete subband structure. Regarding a single electron diffusing and scattering from the barriere states in the quantum well, how would the transfer path look like until the electron will occoupy a state in the lowest energy subband? Is there also a k-selection rule to consider for intersubband transitions?

Maybe I'm searching for the wrong key points, but so far a have not found a satisfying explaination for this process. I would be grateful if someone could explain the process for me.

Best regards

If continuous spetrum phonon for conduction band, I assume discrete spectrum phonon for quantum well.

anuttarasammyak said:

If continuous spetrum phonon for conduction band, I assume discrete spectrum phonon for quantum well.

Thank you for your reply. Unfortunately I don't get the point and how it would answer my question. Can you work it out in more detail?

Photons generetated by transfer of levels in conduction band have no minimum energy because of continuous energy spetre of electrons.
Photons generated by tansfer of levels in quantum well have discrete energy of
[tex]E_{ini}-E_{final}[/tex]

anuttarasammyak said:

Photons generetated by transfer of levels in conduction band have no minimum energy because of continuous energy spetre of electrons.
Photons generated by tansfer of levels in quantum well have discrete energy of
[tex]E_{ini}-E_{final}[/tex]

Maybe we're talking past each other, but I get that point already.

My question is about how the relaxation process between subbands occurs until the electron gets from the barriers to the deepest energetic state in the QW? Might you want to suggest, that the relaxation between different subbands within the same band (either conduction or valence band) within a QW is based on photon emission?