[Pw_forum] ZB ZnO vs WZ ZnO at HSE level

[hannu.komsa at epfl.ch]

Hi

It seems to me that:
1. You are using cubic cell of Zinc-blende structure so you should use  
ibrav=1.
2. You are using LDA PP with hse, which is dangerous. I think PBE PP  
would be more suitable, but more importantly
3. The last time I checked, NLCC doesn't work correctly with hybrids.  
It might be a good idea for somebody to add a warning for this.


Regards,
   Hannu-Pekka Komsa
   University of Helsinki


> 2) ZB-ZnO
> &system
>     ibrav=  2, celldm(1) =8.447075169
>     nat=  8, ntyp= 2, force_symmorphic=.true., nspin=2, tot_magnetization=0,
>     ecutwfc =80.,  nbnd = 50,
>     input_dft='hse', nqx1=2, nqx2=2, nqx3=2,
>     x_gamma_extrapolation = .TRUE.
> /
> &electrons
>     diagonalization='david'
>     mixing_mode = 'plain'
>     mixing_beta = 0.7
>     conv_thr =  1.0d-8
> /
> ATOMIC_SPECIES
>  Zn  65.38     Zn_pz-2_nlcc.cpi.UPF
>   O  15.9994   O.pz-mt.UPF
> ATOMIC_POSITIONS (crystal)
>   Zn    0.00000   0.00000   0.00000
>   Zn    0.00000   0.50000   0.50000
>   Zn    0.50000   0.00000   0.50000
>   Zn    0.50000   0.50000   0.00000
>   O     0.25000   0.25000   0.25000
>   O     0.25000   0.75000   0.75000
>   O     0.75000   0.75000   0.25000
>   O     0.75000   0.25000   0.75000
> K_POINTS automatic
> 8 8 8 0 0 0
>
> For this one I get (!!)
>
> highest occupied, lowest unoccupied level (ev):    36.4937   31.9513  (!!!)
> total energy              =    -200.32746186 Ry