Electric Field step by step???

[slavik]

Hi,
I was looking through the forum about electric field calculation and I saw that it is possible to apply electric field in VASP by using EFIELD, IDIPOL, LDIPOL. I couldn’t find any details about the calculation itself.
I would appreciate if somebody dive the path how to implement such calculation, I mean what is first what is second , and how to calculate the band structure with electric field (do I do relaxation with electric field or I apply electric field for self-consistent calculation).
And another question which is big issues how to improve the convergence for electric field calculation.
Thank you

Yaroslav

[admin]

there is not much more to say about it. the sequence of the tags in INCAR is not of relevance, but please mind that electric fields should only be applied to slabs, molecules and small clusters in order to achieve convergence. the nuber of k-points along the direction in which translational symmetry is broken should be 1.
if the calculation does not converge though, please proceed as you would do for any non-converging calculation: try a different electronic relaxation algorithm, decrease mixing, test for convergence of the k-point set (of course in the directions perpendicular to the field) and the basis set size

[lcyin]

I think only apply the E-field during SCF calculation is enough. Anyway, you can have a try with or without E-field when you do relaxation, then do SCF with E-field, then check the difference.