Electric field in the standard code

Message

jaimeoviedo · #1 Post by **jaimeoviedo** » Tue Apr 01, 2008 10:08 am

Hello,

I am trying to apply an external electric field in VASP.
I read in previous mails that I should include the lines

EFIELD = XXX ( V/A)
LDIPOL = .TRUE.
IDIPOL = 3 (z direction)

When I performed the calculation at several field intensities
I got exactly the same numbers. I did get a dipole correction,
etc but no external field. I could not find any mention to it in
the OUTCAR file. When I had a looked in the source files
I could not find any word like "efield". My question is, Is it
implemented in the standard version (from 4.6) this kind
of calculations or are they included in separate (additional)
source files?

Thank you very much for your help

Jaime

#2 Post by **admin** » Tue Apr 08, 2008 12:55 pm

applying an external is standard in vasp.4.6 (all versions from up to at least 4.6.12), however, fields must not be applied to bulk systems
please look for EFIELD (written in capital letters) in OUTCAR and the code
the results (bond lengths, total energies, dipole moments) differ if
you apply external fields of different strength and/or orientation. (please test eg for a small molecule like CO)

eariel · #3 Post by **eariel** » Wed Nov 26, 2008 10:29 am

Hi,
I need the charge to be calculated self consistently with the external field. Can VASP do this calculation?
Thanks
Eduardo

#4 Post by **admin** » Wed Nov 26, 2008 3:10 pm

frg213 · #5 Post by **frg213** » Sat Aug 21, 2010 1:43 pm

Hello,
I also tried to reach the ground state of a nanoribbon with electric field. However I found it extremely difficult to converge (even after 450 iterations, it did not converge). My system consists of 36 atoms. My INCAR is roughly as below:
SYSTEM =
Startparameter:
PREC = ACCURATE
#ISTART = 0
ICHARG = 2
IALGO = 38
ISYM = 2
ISPIN = 2
MAGMOM = 1 0 0 0 0 0 0 -1 1 0 0 0 0 0 0 -1 0 0 0 -1 0 0 0 -1 0 0 0 0 0 0 0 0 0 0 0 0

Electronic Relaxation:
EDIFF = 1E-04
LREAL = AUTO
NELM = 1000
NELMIN= 6

Ionic relaxation:
#EDIFFG = -1E-02
NSW = 0
#IBRION = 2
#ISIF = 2

DOS related values:
ISMEAR = 0
SIGMA = 0.10
LORBIT = 11

Electric Field parameters:
EFIELD = 0.03
LDIPOL = .TRUE.
IDIPOL = 1

Write flags:
LWAVE = .TRUE.
LCHARG = .TRUE.

Is it because EDIFF is too large? However when I check the OSZICAR, there's simply no trend for the energy to converge. It fluctuates very violently like even several hundreds of eVs.
Thank you!