VASP Forum

The potential profile from LVHAR = .True. has a slope in vacuum even with dipole correction for a slab. I'm studying different kinds of heterostructures, and only this one has this problem. I've never encountered this problem before and couldn't figure out where I did wrong here. All settings are consistent with my previous calculations.

All input files and some output files are attached.

Hello,

Thanks for posting on the VASP forum. From what I can see, there doesn't seem to be an obvious error in the input files. However, there are two things I would suggest to see if they make the difference:

• Increase ENCUT. The default in this case is very low and your 300 eV is just barely above the it. This will make the calculation more expensive but it's good to check if it is converged.

• Increase the vacuum size. Compared to the vacuum, your slab is pretty wide. It's possible that there is still some spurious interaction happening in the vacuum layer.

Let me know what you find.

Dear manuel,
Thanks very much for you suggestions. Both increasing ENCUT from 300 to 400 eV and vacuum region from 20 to 25 Å works.

Since usually 20Å vacuum region is enough, and here ENCUT is relatively small, which is just slightly larger than ENMAX, so I think increasing ENCUT is better here.