Bandgap problems of bulk GaAs using G0W0

Queries about input and output files, running specific calculations, etc.


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
Ken-Ming Lin
Newbie
Newbie
Posts: 2
Joined: Tue Oct 29, 2013 3:30 am

Bandgap problems of bulk GaAs using G0W0

#1 Post by Ken-Ming Lin » Thu Nov 07, 2013 5:07 am

Dear All,

I am running the G0W0 for the bulk GaAs by VASP 5.3.3 and try to reproduce the results of M. Shishkin and G. Kresse, Phys. Rev. B 74, 035101 (2006). I used three steps as the VASP guide and the same parameters of the paper except NBANDS (e.g., I used NBANDS=168 and those of paper PRB 74, 035101 is 160.). However, I can not get the same results. My procedures are as follows,

1. The standard groundstate calculation with few unoccupied orbitals only.

INCAR
System = Bulk GaAs GS occupied orbitals (From VASP Guide p133)
ISMEAR = 0; SIGMA = 0.05
EDIFF = 1E-8
ENCUT = 310

POSCAR
GaAs_Bulk
5.650000000
0.00000000 0.50000000 0.50000000
0.50000000 0.00000000 0.50000000
0.50000000 0.50000000 0.00000000
1 1
Direct
-0.1250000 -0.1250000 -0.1250000
0.1250000 0.1250000 0.1250000

KPOINTS
6x6x6
0
G
6 6 6
0 0 0

POTCAR
PAW Ga_d_GW 06Jul2010
PAW As_GW 12Mar2012

2. The second calculation of a large number of unoccupied orbitals. After the first step calculation was finished, I copied all the files to a new empty directory and replaced the INCAR as follows,

INCAR
System = Bulk GaAs GS occupied orbitals (From VASP Guide p133)
ALGO = Exact
NELM = 1
NBANDS = 168 ! The NBANDS of paper PRB74, 035101 is 160.
ISMEAR = 0; SIGMA = 0.05
LOPTICS = .TRUE.
ENCUT = 310

3. After the second step calculation was finished, I copied all the files to another new empty directory and replaced the following INCAR to perform G0W0 calculation.

INCAR
System = Bulk GaAs G0W0 Phys. Rev. B 74, 035101 (2006)
ALGO = GW0; NOMEGA = 300
NBANDS = 168
ISMEAR = 0; SIGMA = 0.05
LOPTICS = .TRUE.
ENCUT = 310
CSHIFT = 0.1
ENCUTGW = 250

My results for Gamma point are as follows,
k-point 1 : 0.0000 0.0000 0.0000
band No./ KS-energies/ QP-energies/ Ga-3d_LDA[1]/ Ga-3d_GW[1]
5/ -14.947/ -16.552/ -14.94/ -13.61
6/ -12.779/ -12.621/ -12.79/ -12.76
9/ 0.00/ 0.00/ 0.00/ 0.00
10/ 0.337/ 1.229/ 0.31/ 1.42
13/ 3.669/ 4.171/ 3.65/ 4.40
[1] PRB 74, 035101 (2006).

The energies of the second and third columns are my results. The fourth and fifth columns are from the TABLE III in the paper PRB 74, 035101. All my energies are refered to the energy of band no. 9. My KS-energy are almost the same with those of the paper. Nevertheless, the QP-energy are different. Does any body have suggestions?



Thanks,

Ken-Ming Lin
Last edited by Ken-Ming Lin on Thu Nov 07, 2013 5:07 am, edited 1 time in total.

giacomo giorgi
Full Member
Full Member
Posts: 122
Joined: Tue Mar 10, 2009 2:04 am

Bandgap problems of bulk GaAs using G0W0

#2 Post by giacomo giorgi » Tue Nov 12, 2013 6:21 pm

Hi,
what about using
ICHARG=11
in step 2?
As far as I know indeed the second step should be NSCF.
Anyway, I don't know if this will fix your problem.

Cheers,
Giacomo
Last edited by giacomo giorgi on Tue Nov 12, 2013 6:21 pm, edited 1 time in total.

Post Reply