only gamma point of qmesh is being sampled, regardless of QPOINTS mesh

[preston_vargas]

In an attempt to calculate the Phonon DOS for primitive BCC Nb, I noticed that only the gamma point of the q-mesh is being calculated. The provided q-mesh is being read in and appears in the OUTCAR, but the phonon frequencies remain identical across all qpoints, resulting in a single spike in the phon-DOS at 0 THz.

What might be causing this behavior?

[ajinkya_hire1]

I am facing the same issue for my test Nb calculation (qmesh- 3x3x3). Here's the relevant part of the OUTCAR file-

Phonon DOS integrates to: 3.00000000000000
the correct value is: 3.00000000000000

--------------------------------------------------------------------------------

Phonons

q-point No. q_x q_y q_z q-path length
1 0.000000 0.000000 0.000000 0.000000
frequencies
branch index f[THz] f[2PiTHz] f[cm-1] f[meV]
1 0.000000 0.000000 0.000000 0.000000
2 0.000000 0.000000 0.000000 0.000000
3 0.000000 0.000000 0.000000 0.000000

q-point No. q_x q_y q_z q-path length
2 0.333333 0.000000 0.000000 0.142006
frequencies
branch index f[THz] f[2PiTHz] f[cm-1] f[meV]
1 0.000000 0.000000 0.000000 0.000000
2 0.000000 0.000000 0.000000 0.000000
3 0.000000 0.000000 0.000000 0.000000

q-point No. q_x q_y q_z q-path length
3 0.333333 0.333333 0.000000 0.284012
frequencies
branch index f[THz] f[2PiTHz] f[cm-1] f[meV]
1 0.000000 0.000000 0.000000 0.000000
2 0.000000 0.000000 0.000000 0.000000
3 0.000000 0.000000 0.000000 0.000000

q-point No. q_x q_y q_z q-path length
4 -0.333333 0.333333 0.333333 0.284012
frequencies
branch index f[THz] f[2PiTHz] f[cm-1] f[meV]
1 0.000000 0.000000 0.000000 0.000000
2 0.000000 0.000000 0.000000 0.000000
3 0.000000 0.000000 0.000000 0.000000

[fabien_tran1]

Hi,

There are two ways to compute the phonon dispersion:

Phonons from finite differences: As indicated, a supercell needs to be considered in order to get the phonon frequencies at q-points other than only GAMMA.

Phonons from DFPT: In principle, this method allows to calculate the phonon dispersion on a q-point mesh (not only at GAMMA), but the current implementation is not complete, which is the reason for: "Mind: Only zone-center (Γ-point) frequencies are calculated.". Thus, a supercell is also needed with DFPT.

In your case, the problem is that you are not using a supercell, but just the primitive one-atom cell, so that the phonon frequency is calculated only at the GAMMA point.