I know it has been discussed earlier but due to ambiguities, still I want to make it clear. I'm trying to apply external electric field to a 2D system (e.g., graphene). After getting relaxed structure, next, in the SCF calculation, I just need to add the following three tags
#Electric Field:
LDIPOL = .TRUE.
IDIPOL = 3
EFIELD = 0.025
and similarly, in the band structure calculation (reading WAVECAR and CHGCAR from SCF).
Is it the right way to apply electric field in VASP? If no, please share the correct method. If yes, the please share your experience with spin-orbit calculations and its convergence.
Thank you in advance.
Hi Shahid,
The dissertation that is at the first link I provided includes more detailed instructions on implementing the method using VASP. Specifically they have provided their VASP input files in Appendix A. I believe that may be what you are looking for. I have included the link again below for convenience.
https://curate.nd.edu/show/bc386h4634k
for 2D case (sheet) you only want to have x-y components. SO coupling will give a rise to magnetic anisotropy so you need to setup non-collinear calculation. you also want to use LSDA+U calculation. please check the manual how toset this up. and don't forget to recompile the software with proper flags for NCL.
Thanks! How about method of applying Efield, as you didn't comment on that, is that correct? Also, I didn't get why we need to use LSDA+U for SOC calculation of graphene. Would you kindly elaborate a bit?
for correct application of e-field please consult vasp manual. I only commented that for 2D case you will most probably want to apply the field only in lateral directions. Regarding LSDA+U, your system is strongly correlated (magn anisotropy, "surface" states). here you can read more about this:
http://cms.mpi.univie.ac.at/vasp/vasp/On_site_Coulomb_interaction_L_S_DA_U.html
http://www.fplo.de/workshop/ws2008/koepernik2_fplo2008.pdf
http://users.phys.psu.edu/~crespi/docs/CuO2LDA+U.pdf
http://arxiv.org/abs/0808.1706
Thanks for the explanation.
I wish someone who applied Efield in his work could comment on this! It would be great help for others. I used manual to write the above mentioned input tags.
Thanks,
I have included some links to works that have used VASP to simulate an external electric field. I hope they are helpful.
https://curate.nd.edu/show/bc386h4634k
http://dx.doi.org/10.1103/PhysRevB.64.125403
It appears that the DOI link to the paper was converted to a link to the article on ResearchGate. I have attached a link to the journal page that will hopefully not get converted.
http://journals.aps.org/prb/abstract/10.1103/PhysRevB.64.125403
That also seems to have been converted. Apparently ResearchGate does not want you to leave their website. I will try once more with the links as text. Hopefully one of these will get through. Also, the reference is: Feibelman, P. J. Surface-diffusion mechanism versus electric field: Pt/Pt(001). Physical Review B 64, 125403 (2001).
http://dx.doi.org/10.1103/PhysRevB.64.125403
http://journals.aps.org/prb/abstract/10.1103/PhysRevB.64.125403
Hello Michael, Thanks for your input. I have gone through the PRB paper. However, a more recent example is the following paper:
https://arxiv.org/ftp/arxiv/papers/1403/1403.6189.pdf
The point is that in many papers, authors have used electric field to tune the band gap or self-diffusion barrier (like in the PRB paper you mentioned). However, it is unclear about how to properly set tags and procedure to apply it in VASP. I wish some of the authors could answer this.
Hi Shahid,
The dissertation that is at the first link I provided includes more detailed instructions on implementing the method using VASP. Specifically they have provided their VASP input files in Appendix A. I believe that may be what you are looking for. I have included the link again below for convenience.
https://curate.nd.edu/show/bc386h4634k
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Materials
- Carbon Materials
- Graphite
- Graphene