The ABINIT Users Mailing List ( CLOSED )

[Xavier Gonze <gonze@pcpm.ucl.ac.be>]

Dear Zhufeng,

In your input file, you required only a few perturbations to be calculated, 
because
of your choice of the input variable :
rfatpol3   1 2
In order to compute the full dynamical matrix, it should be :
rfatpol3   1 6
because you have 6 atoms per unit cell.
The code does not complain about you not willing to compute all perturbations 
(you can
use mrgddb to complete the DDBs, using another run), but it gives you 
correctly the warning that
the dynamical matrix is incomplete, and that the phonon frequencies delivered
at that place might not be what you want.
Actually, you computed only the reponse to Zr displacements, so that is it 
sure that
the frequencies are wrong. This warning has nothing to do with
cutoff energy, k-points and convergence tolerance ...

Xavier

zfhou@mail.edu.cn wrote:
> Dear abinit users,
>
> I have tried to calculate the dielectric tensor of HfO2 with tetragonal 
> structure containing 12
> atoms or 6 atoms, using Hartwigsen-Goedecker-Hutter LDA psp for Hf and O. In 
> my calculation, the
> cutoff energy is 35 Ha and the k-point mesh is 7 x 7 x 5. However the warning 
> "The dynamical matrix
> was incomplete : phonon frequencies may be wrong ..." occured in output file. 
> Except the
> increasment of cutoff energy, k-points and convergence tolerance, is there 
> any other rule-of thumb
> to get rid of this warning?
>
> I have tested to increase the k-point mesh and convergence tolerance, this 
> warning can't be
> removed.
>
> In addtion, I have used the same pseduopotentials and cutoff energy to 
> calculate the cubic HfO2.
> Although I got a larger dielectric tensor than the values in 
> PRB69,184301(2004)(G.-M.Rignanese, et
> al), there was no such waning in the calculation.
>
> Your suggestions and helps will be greatly appreciated! zhufeng
>
> ----------input file for tetragonal HfO2-------- ndtset  3
>
> #Ground state calculation kptopt1   1 tolvrs1   1.0d-18 iscf1   5
>
> #Response Function calculation : d/dk rfelfd2   2 rfdir2   1 1 1 nqpt2   1 
> qpt2   0.0 0.0 0.0
>
> getwfk2   -1 kptopt2   2 iscf2  -3 tolwfr2   1.0d-18
>
> #Response Function calculation : electric field perturbation and phonons 
> rfphon3   1 rfatpol3   1 2
>  rfelfd3   3 rfdir3   1 1 1
>
> nqpt3   1 qpt3   0.0 0.0 0.0
>
> getwfk3   -2 getddk3   -1
>
> kptopt3   2 tolvrs3   1.0d-10 iscf3   5
>
> #Definition of the unit cell acell    6.7261613019E+00  6.7261613019E+00  
> 9.7644094669E+00 angdeg
> 90.0  90.0  90.0
>
> #Definition of the atom types spgroup  137 ntypat 2 znucl 72  8 natom 6 typat 
> 2*1 4*2 xred
> 0.0000000000E+00  0.0000000000E+00  0.0000000000E+00 5.0000000000E-01  
> 5.0000000000E-01
> 5.0000000000E-01 0.0000000000E+00  5.0000000000E-01  1.9835156639E-01 
> 0.0000000000E+00
> 5.0000000000E-01  6.9835156639E-01 5.0000000000E-01  0.0000000000E+00 -1.9835156639E-01 
> 5.0000000000E-01  0.0000000000E+00  3.0164843361E-01
>
> #Definition of the k-point grid ngkpt 7 7 5
>
> #Definition of the SCF procedure ecut  35.0 nstep 60 diemac 3.0 nband 60 ixc  2 
> ----------------------END-------------------------------------