Electric field application shows no effect (a flat potential profile)

[nuha_sahir]

Hello everyone.

I am facing a very weird issue when applying an electric field (along the z-direction) in my system. For specific materials I am studying, such as GaN, SnSe, and Germanene, applying an electric field does not change the total energy, nor does it produce a ramp in the planar average potential from the LOCPOT. From my understanding, when an electric field is applied, the planar average potential should create a ramp with a slope corresponding to the applied electric field, but this is absent in my system. When I apply the same settings to other materials (such as hBN and GeSe), they show both a change in total energy and a ramp in the planar average potential, which makes me believe the error is not in how I am applying the electric field. This is the INCAR for when I apply the electric field:

Code: Select all

INCAR:
SYSTEM = SnSe

!!SYSTEM INITIALIZATION!!
ISTART = 0
ICHARG = 2

!!ELECTRONIC STEPS!!
ALGO = Fast    
ENCUT = 500
ISMEAR = 0 
SIGMA = 0.05
EDIFF = 1E-6
NSW = 1000

!!IONIC STEPS!!
PREC = Normal
IBRION = 2
POTIM = 0.08
EDIFFG = -0.01
NELM = 300
NELMIN = 5

!!CELL RELATED!!
ISIF = 2
ISYM = 0
LREAL = Auto
SYMPREC = 1E-4

!!CORRECTIONS!!
IVDW = 11
IDIPOL = 3
LDIPOL = .TRUE.
DIPOL = 0.5 0.5 0.5
!!LMONO = .TRUE.

!!MAGNETIZATION!!
ISPIN = 2 ! if graphene sheet 1, otherwise double check
!!NUPDOWN = 0
MAGMOM = 168*0.7 ! depends on ISPIN
!!NELECT = 448
EFIELD = -0.50

!!GENERAL/CALCULATION OPTIMIZATION!!
LWAVE = .FALSE.
LCHARG = .FALSE.
LVHAR = .TRUE.
NCORE = 2

I have tried changing the way the electric field and the dipole corrections are applied to see if it creates any changes. Some of the things that I did:
1. Commented out LDIPOL, DIPOL and MAGMOM. LMONO = .TRUE. and ISPIN = 1 (the same lack of effect)
2. Tried using EFEILD_PEAD (this showed the same lack of effect, but this was expected as it is usually used with bulk systems and not monolayers).

I have attached a minimal reproducible example for SnSe. I haven't changed the parameters that I used (in case this issue is because of any of those parameters), but the system is optimized, therefore, it will converge quickly. I have a few specific questions, though:
1. Is there any case where the planar averaged potential from the LOCPOT might produce a flat ramp even in the presence of an electric field?
2. Could certain input parameters clash together and result in no effect for some materials (based on their chemistry)?

Any help with this issue would be much appreciated, as I do not understand why I am observing this result with only certain materials. Physics-wise, their behavior should be similar to that of the other materials in which an electric field effect was observed, so any guidance would be much appreciated.

Thank you in advance!

[max_liebetreu]

Hello,

Thank you for reaching out to us on the VASP Forum!
Apologies for the delayed response - we are investigating the issue and will get back to you as soon as possible.

Best regards,
Max

[max_liebetreu]

Hello,

After looking at your example, we suspect what you are seeing is the expected physical reaction: Your system's dipole moments cancel out - your total dipole moment is 0.
Therefore, when you are applying an electric field in z direction, some dipoles are affected one way, others in the exact opposite way, and the total dipole moment is still 0.

This is likely responsible for the behavior you are seeing; we checked for your linked example of SnSe, where we didn't observe any unexpected results, physics-wise.
Please let us know if this also explains the differences for the other materials you studied!

Best regards,
Max