LMBJ for 1D system

[subrata]

Hi,
I have a problem with LMBJ calculations for 1D systems. For example

C6H4
1.00
11.0000000 0.00000000 0.00000000
0.00000000 18.0000000 0.00000000
0.00000000 0.00000000 4.25370598
C H
6 4
Direct
1.66417228E-03 2.35353783E-03 3.46985529E-03
1.24619855E-03 -6.47786558E-02 0.171479195
1.23689917E-03 -6.47874400E-02 0.495433241
1.65815942E-03 2.35325214E-03 0.663318634
2.24087597E-03 6.94881454E-02 0.171472147
2.24125758E-03 6.95001483E-02 0.495434552
1.07410457E-03 -0.118248492 4.89231870E-02
1.05037668E-03 -0.118312985 0.617444932
2.87996116E-03 0.122985058 4.90232334E-02
2.88981409E-03 0.123051353 0.617333829

The band gaps seem very wrong. Any suggestions on how to do the correct calculations

Subrata

[max_liebetreu]

Hello subrata,

Thanks for reaching out to us on the VASP Forums!
We can gladly take a look - please provide a minimal reproducible example containing the relevant files for your calculation.
Could you please offer some more details about the issue you are seeing with the bandgaps?

Best regards,
Max

[subrata]

Here is the INCAR file for LMBJ calculation

Code: Select all

EDIFF = 1E-5
ISMEAR = 0
SIGMA = 0.05
NELM = 1000
ALGO = A
IMIX = 1
AMIX = 0.4
AMIN = 0.01
LASPH = .TRUE.
METAGGA = LMBJ
SMBJ = 2.0
RSMBJ = 7.0
CMBJA = 0.488
CMBJB = 0.5
CMBJE = 1.0

and Here is the POSCAR file

Code: Select all

C6H4
1.00
11.0000000 0.00000000 0.00000000
0.00000000 18.0000000 0.00000000
0.00000000 0.00000000 4.25370598
C H
6 4
Direct
1.66417228E-03 2.35353783E-03 3.46985529E-03
1.24619855E-03 -6.47786558E-02 0.171479195
1.23689917E-03 -6.47874400E-02 0.495433241
1.65815942E-03 2.35325214E-03 0.663318634
2.24087597E-03 6.94881454E-02 0.171472147
2.24125758E-03 6.95001483E-02 0.495434552
1.07410457E-03 -0.118248492 4.89231870E-02
1.05037668E-03 -0.118312985 0.617444932
2.87996116E-03 0.122985058 4.90232334E-02
2.88981409E-03 0.123051353 0.617333829

The band gap I am getting from LMBJ OUTCAR is

Code: Select all

val. band max:       46.4875371038   @ k =   0.0000   0.0000   0.0000
 cond. band min:      26.6582385172   @ k =   0.0000   0.0000   0.3000
 fundamental gap:    -19.8292985866
 Fermi energy: ********************

I am using regular PBE C and H pseudopotential

[max_liebetreu]

Thank you, that helps us understand your setup.
I will discuss this with my colleagues, and we will get back to you on Monday.

[manuel_engel1]

Hello Subrata,

I have taken a look at your calculation with the information you have shared. Unfortunately, I am unable to reproduce your problem. The energies on my end appear reasonable. Please, upload as a zip archive the following files of the failing run:

• INCAR

• POSCAR

• KPOINTS

• OUTCAR

The OUTCAR file in particular will be required for me to understand what goes wrong in your calculation.

[subrata]

I am attaching full files for a look.

[manuel_engel1]

Thank you very much. I was able to confirm that this is an issue with VASP 6.4.3 that does not exist in the latest release version VASP 6.6.0. I'm very sorry for the inconvenience. If possible, I would urge you to upgrade your VASP version to the latest release. We are currently investigating whether this is a bug that exists also in other versions. If we find out more details, I will post them here as well.