Electric field calculations

[karthik_hj]

I gave the electric field to an already relaxed structure and it shows 'electronic self consistency is not achieved'. I even tried NELM=300 but still the ground state energy has not been converged. If I put ISIF 2 tag along with electric field tag, calculations ends abruptly showing error. Is there anything else we can do. Would it be right if I take the CONTCAR after giving electric field and relax it again using ISIF tags?

Or is there another procedure to do the same

I put idipol ldipol and electric field tags in the INCAR . Should I put dipol tag and Efield_pead tag also

I want to find the effect of electric field on the electronic structure of a 2d metallic system

[henrique_miranda]

What is the error that you obtain when you say that the calculation ends abruptly?

Could you share the INCAR, POSCAR, KPOINTS, POTCAR and OUTCAR files?

[karthik_hj]

The error is gone now when i put dipol tag also along with isif-2. Is it the correct procedure? What does an efield_pead tag do ? Could you provide a workflow for the electric field calculations for a structure?

[henrique_miranda]

The error is gone now when i put dipol tag also along with isif-2. Is it the correct procedure?

I don't know why setting ISIF would have any influence on the self-consistent computation with an electric field.

What does an efield_pead tag do ?

It applies an electric field.
wiki/index.php/EFIELD_PEAD

Could you provide a workflow for the electric field calculations for a structure

I don't understand what you mean

[karthik_hj]

I gave electric field using
LDIPOL = .TRUE.
EFIELD = 1.0
IDIPOL = 3
DIPOL = 0.5 0.5 0.5
Is it necessary to give EFIELD_PEAD tag? What does it do?

What i mean by workflow is the procedure to do electric field calculations. Please see whether which workflow is correct

Workflow 1
Give Electric field tags in INCAR and relax the structure with NSW=0
Give Electric field tags in INCAR and find dos and similarly bandstructure NSW=0


Workflow 2
Give Electric field tags in INCAR and relax the structure with NSW=0
Do not give Electric field tags in INCAR to find dos and bandstructure

[henrique_miranda]

The EFIELD tag refers to the electric field applied using the method described in https://doi.org/10.1103/PhysRevB.46.16067
This works for slabs or molecules as mentioned on this page:
wiki/index.php/Monopole_Dipole_and_Quad ... orrections

The EFIELD_PEAD tag refers to the electric field applied using the method described in https://doi.org/10.1103/PhysRevB.63.155107
This also works for bulk systems and its described on this page:
wiki/index.php/Berry_phases_and_finite_ ... ric_fields

You should not mix these two INCAR tags.

Workflow 1 is the correct one.

[karthik_hj]

Can these methods be only used for insulators/semiconductors? Because when I used EFIELD_PEAD for metals ,VASP exited.

[henrique_miranda]

EFIELD_PEAD only works for systems with a gap:
https://www.vasp.at/wiki/index.php/EFIELD_PEAD

The maximum magnitude of the electric field you can apply is limited by the size of the gap of the system.