My Community

Hi

I am a new user to VASP so I have some basic questions.

I want to investigate the adsorption of a gas molecule on a semiconductor surface.

Firstly, should I use ISYM = 2 or ISYM = 0?

Secondly, for the surface, should I use ISMEAR = -5 or can I use MP with a proper sigma?

And what about the whole system, which ISMEAR should I use?

Thank you

1) the choice of ISYM depends on whether you want to keep the symmetry of your input POSCAR throughout the whole relaxation process, (the forces are symmetrized if ISYM = 2) or if you want to give the system additional freedom to decrease ist symmetry.
Please note that the k-point (and hence the requested CPU time and memory) set may be significantly larger if symmetry is switched off (depending on how much symmetry can be found for the original input geometry)
2) you can of course use the Methfessel Paxton BZ integration for semiconductors as well. Please be careful with your choice of SIGMA if you do so, it should be smaller that 1/2 of the gap size in any case. I would recommend to use the same type of BZ integration for all requested calculations (clean and covered surface), and possibly do ONE accurate run for the DOS with ISMEAR = -5 and an increased k-point set (based on the scf-CHGCAR) after the system has converged