Dipole Correction in charged bulk KTaO3 systems

[Bmodak]

Dear Sir,
I have some confusion related to Dipole Correction in charged bulk KTaO3 systems.
I have consider two different situations for charged oxygen vacancy defects
First:
NELECT=
IDIPOL = 4
DIPOL= coordinate of charged vacancy
EPSILON= Dielectric constant of the system

Second:
NELECT=
IDIPOL = 4
DIPOL= coordinate of charged vacancy
LMONO= True
EPSILON= Dielectric constant of the system

I am getting different results in the two cases. I have confused which one I should use?

Thanks and Regards
Brindaban

[henrique_miranda]

Did you have a look at the specific documentation for the two variables in our wiki?
https://www.vasp.at/wiki/wiki/index.php/LMONO
https://www.vasp.at/wiki/index.php/LDIPOL

In particular in LMONO:

for LDIPOL=.TRUE., VASP automatically determines whether the system is charged and activates the monopole corrections automatically

Could you post here the lines after "Dipole corrections" in your OUTCAR files for the two calculations?

[Bmodak]

Thank you very much Sir for your reply. Please find the output for the two cases for KMgF3 with 222 supercell containing +1 charged F vacancy
INCAR: Case1
NELECT = 312
IDIPOL = 4
DIPOL = 0.75000 0.50000 0.750000
EPSILON = 6.97

OUTPUT
DIPCOR: dipole corrections for dipol
direction 1 min pos 25, direction 2 min pos 1, direction 3 min pos 25,
dipolmoment -0.000013 -0.000016 -0.000013 electrons x Angstroem
Tr[quadrupol] -1909.146698

energy correction for charged system 0.374237 eV
dipol+quadrupol energy correction -17.199756 eV
added-field ion interaction 0.000000 eV (added to PSCEN)

energy without entropy= -206.42577598

INCAR: Case2
NELECT = 312
LMONO = .TRUE.
IDIPOL = 4
DIPOL = 0.75000 0.50000 0.750000
EPSILON = 6.97

OUTCAR
DIPCOR: dipole corrections for dipol
direction 1 min pos 25, direction 2 min pos 1, direction 3 min pos 25,
dipolmoment -0.000013 -0.000016 -0.000013 electrons x Angstroem
Tr[quadrupol] -1909.146698

energy correction for charged system 0.374237 eV
dipol+quadrupol energy correction 0.000000 eV
added-field ion interaction 0.000000 eV (added to PSCEN)

energy without entropy= -189.22602027

I want to calculate formation energy for +1 charged F vacancy. I am in confusion which case I should consider?

Regards
Brindaban Modak

[henrique_miranda]

What is happening is that when LMONO=.TRUE. the Dipole and quadrupole energies are set to zero.

[Bmodak]

Dear Sir
Thank you very much for your reply.
Please advice me whether I should use LMONO= TRUE tag in this case?
Is their any procedure to choose LMONO= TRUE or IDIPOLE= 4 Tag depending on the systems?
In this case, one vacancy containing singly positive charge, so looks like monopole.

Regards
Brindaban

[henrique_miranda]

Please advice me whether I should use LMONO= TRUE tag in this case?

This depends on the type of calculation you want to do.
Please read the documentation (and papers) carefully to know what corrections are implemented in VASP and their meaning:
https://www.vasp.at/wiki/index.php/Mono ... orrections

More specific to your question (defects in solids) you might read the information here:
https://www.vasp.at/wiki/index.php/Dipo ... _in_solids

If something is not clear for you in the documentation please let us know so we can update it

Code: Select all

Is there any procedure to choose LMONO= TRUE or IDIPOLE= 4 Tag depending on the systems?

In the documentation above you should find what flag you should use depending on the system you want to study.
If you mean an automatic procedure to determine which flags to use, then the answer is not.
The user should understand how the corrections are computed, their meaning and whether they should be included or not.