EPW Forum

Dear all,

I am reading the "electron-phonon interactions from first principles", but with several conceptual puzzles:
(1) Why are the formalisms based on BvK supercell, instead of unit cell as in DFPT calculations?

(2) In the paper, "Carriers in semiconductors are typically confined within a narrow energy range near the band extrema; consequently, it is expected
that the dominant electron-phonon scattering mechanisms will involve long-wavelength phonons (q?0)." Does this statement hold only for semiconductors with band extrema at Gamma point?

Can anyone help explain it to me?

Dear yuan,

I can try:
1) The number of BvK supercell is directly linked with the number of q-points needed to sample your primitive cell within the DFPT formalism.
Whenever implemented, the DFPT formalism is far less computationally demanding.

2) It does hold. Band extrema are in k-space while here you are talking about q-space.
Regardless where the band extrema is in k-space, you will have a divergence as q tend to 0 in polar materials in q-space.

Hope it helps,
Best,
Samuel

Dear Samuel,
Got it. Thanks a lot for your help.
Best