[QE-users] Wrong Fermi level with smearing compared to tetrahedra

[Nicola Marzari]

Hi,

Fermi energy is not really defined in an insulator/semiconductor; at 0 
temperature any value in the gap a is good fermi energy, but better to 
talk about the chemical potential. At temperature is raised, it would be 
approximately start in the middle of the gap, and move away from that 
depending on how the conduction and valence DOS are. Maybe read the 
Ashcroft Mermin chapter on homogeneous semiconductors (27, I think).

A typical good choice for smearing (0.02 Ry, 0.15 A^-1 density of 
k-points) would give you the fermi energy in the middle of the gap, 
approximately, if you use gaussian smearing in an insulator. mv and mp 
should not be used for an insulator/semiconductor.

Also, mv has positive occupations, so something seems wrong with your 
plot - are you sure you are not using mp?

			nicola

On 27/06/2020 14:56, 후신 부지드 wrote:
> Dear QE users,
> 
> 
> I tried to search a solution for this problem and looked to the previous 
> posts but didn't work with me
> 
> 
> 
> I made scf calculation for bulk SnS2 with dft-d3 using occupations= 
> 'tetrahedra' I get the the fermi level Ef on top of the valence band as 
> reported in the literature,
> 
> but when I try to use smearing the Ef will shifts about 1 eV to locate 
> in the middle of the gap
> 
> I tried gaussian smearing and fd using a huge k points, but eventhough 
> the energy converges to the value in tertahedra case, the Ef always 
> shifts by ~ 1eV.
> 
> The only exception is mv smearing with 0.01 degauss it gives the right 
> Ef, (but it has a tail with finite dos which is not seen in the gaussian 
> case with wrong Ef, pls see the attached image) and when I try mv with 
> 0.005 degauss it also show wrong Ef even with huge k points.
> 
> 
> Does anyone know why I'm getting this shift in Ef even though the energy 
> is converged and the kpoint is large enough
> 
> 
> Here is the input file for tetrahedra and smearing:
> 
> 
> 
> Tetrahedra:
> 
> 
> &control
>      calculation = 'scf'
>      restart_mode='from_scratch'
>      prefix = 'tet'
>      outdir = '/home/outdir'
>      pseudo_dir = '/home/pseudo'
>      verbosity = 'high'
> /
> 
> &system
> 
>      ibrav = 0,
>      nat   = 6,
>      ntyp  = 2,
>      ecutwfc=80
>      ecutrho=560
>      vdw_corr = 'dft-d3'
>      occupations='tetrahedra'
>      nbnd=52
>      lda_plus_u=.true.
>      Hubbard_U(1)=9.0,
>      Hubbard_U(2)=9.0,
> 
> /
> 
> &electrons
>      conv_thr=1e-6
> /
> 
> ATOMIC_SPECIES
>   Sn    118.71000  Sn.pbe-dn-kjpaw_psl.1.0.0.UPF
>   S      32.06600  S.pbe-n-kjpaw_psl.1.0.0.UPF
> 
> ATOMIC_POSITIONS (angstrom)
> 
> Sn      -0.000001900   2.129228760   7.368473930
> Sn       1.843967305   1.064612836   1.474204489
> S        1.843967305   1.064612836   8.846101191
> S       -0.000001900   2.129228760   2.951831855
> S       -0.000000000  -0.000000000  -0.005322640
> S       -0.000000000   0.000000000   5.888946695
> 
> CELL_PARAMETERS (angstrom)
> 
>   3.687930810   0.000000000  -0.000000000
> -1.843965405   3.193841769  -0.000000000
> -0.000000000  -0.000000000  11.788538670
> K_POINTS (automatic)
>   20 20 6 0 0 0
> 
> 
> 
> Smearing
> 
> 
> 
> &control
>      calculation = 'scf'
>      restart_mode='from_scratch'
>      prefix = 'smea'
>      outdir = '/home/outdir'
>      pseudo_dir = '/home/pseudo'
>      verbosity = 'high'
> /
> 
> &system
> 
>      ibrav = 0,
>      nat   = 6,
>      ntyp  = 2,
>      ecutwfc=80
>      ecutrho=560
>      vdw_corr = 'dft-d3'
>      occupations='smearing',smearing='gaussian',degauss=0.01
>      lda_plus_u=.true.
>      Hubbard_U(1)=9.0,
>      Hubbard_U(2)=9.0,
> /
> 
> &electrons
>      conv_thr=1e-6
> /
> 
> ATOMIC_SPECIES
>   Sn    118.71000  Sn.pbe-dn-kjpaw_psl.1.0.0.UPF
>   S      32.06600  S.pbe-n-kjpaw_psl.1.0.0.UPF
> 
> ATOMIC_POSITIONS (angstrom)
> 
> Sn      -0.000001900   2.129228760   7.368473930
> Sn       1.843967305   1.064612836   1.474204489
> S        1.843967305   1.064612836   8.846101191
> S       -0.000001900   2.129228760   2.951831855
> S       -0.000000000  -0.000000000  -0.005322640
> S       -0.000000000   0.000000000   5.888946695
> 
> CELL_PARAMETERS (angstrom)
> 
>   3.687930810   0.000000000  -0.000000000
> -1.843965405   3.193841769  -0.000000000
> -0.000000000  -0.000000000  11.788538670
> K_POINTS (automatic)
>   32 32 10 0 0 0
> 
> 
> Houcine BOUZID
> Sungkyunkwan University,
> 
> 
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
> 


-- 
----------------------------------------------------------------------
Prof Nicola Marzari, Chair of Theory and Simulation of Materials, EPFL
Director, National Centre for Competence in Research NCCR MARVEL, EPFL
http://theossrv1.epfl.ch/Main/Contact http://nccr-marvel.ch/en/project