Hello, I've been generating vacuum potentials for a few work function calculations using VASP and p4vasp (20-30 angstrom slabs). I expected to see a rapid increase in average potential from the surface followed by a plateau that spans the bulk of the vacuum followed by a rapid decline in average potential as the periodic boundary of the slab cell is approached (returning to the potential within the bulk). That is present, but also in the middle of the gap the maximum/minimum potential depart wildly from the average and there is a small dip (around 0.1 eV) in the average before returning to the pseudo-plateau. What is the cause for this behavior in the vacuum potential, and from where should I take readings for the vacuum potential: the adjacent pseudo-plateau, or the reading at the bottom of the "meniscus"? Is this an artifact caused by an insufficient vacuum size? Thanks in advance
-Sean
Hi Sean,
Hard to visualize what exactly the anomaly you're observing. If you have a 'step' in the middle of your vacuum potential - this is due to the dipole correction in VASP (https://www.vasp.at/wiki/index.php/LDIPOL). For discussion, see https://www.materialssquare.com/blog/20-dipole-correction-an-offsetting-effect-of-the-artificial-electric-field
I've had the same problem. I guess it's because your unit cell or ions have not been relaxed.
Those oscillations are usually due to a poor choice of the vacuum thickness, causing the wave function of one side of the slab to overlap with that of the other side. Increasing the vacuum thickness should fix it.
- Badges
- Science topic
- Similar topics
- Psychology
- Mental Processes
- Learning