ABINIT Discussion Forums

Dear all,

I have a problem while calc. el-phon coupling in Ce. I follow the input from the tutorial, and since imaginary Al fcc has the same structure as gamma-Ce, I left most of the variables the same, like the ngkpt and qpt. I used also ngkpt 4x4x4 and qpts from 2x2x2 q point mesh (gamma, (1/2,0,0) and X). Gamma point perturbations pass ok, but for qpt (1/2,0,0) I get a warning

Phonon wavevector (reduced coordinates) : 0.50000 0.00000 0.00000
Phonon energies in Hartree :
0.000000E+00 0.000000E+00 0.000000E+00
Phonon frequencies in cm-1 :
- 0.000000E+00 0.000000E+00 0.000000E+00
chkph3 : WARNING -
The dynamical matrix was incomplete : phonon frequencies may be wrong ...

for all 3 directions. For X point second derivatives unconverge wildly.

Can anyone help me with this issue? 2x2x2 mesh is a submesh of the 4x4x4, so why does it call a warning?

Thank you all in advance!

Yours,
Igor Lukacevic

Hi Igor (again),

The warning is just a warning - you are doing 1 perturbation at a time, so the dynamical matrix will always be incomplete inside a given dtset (only 1 row filled, and not enough syms to complete all of it) - not a problem. Run anaddb at the end with the DDB files you get to see if anything is missing.

If it is, you may have fewer symops with Ce because of magnetism or spin orbit. Try a normal run with 2x2x2 kpts ground state and kptopt 1 (shiftk 0 0 0) to get the qpts you need to use. You should see an extra one or 2 with respect to Al.

matthieu