My Community

Hello,

This is a technical question about how VASP deals with atoms that move from one side of the cell to the other (usually because of periodic boundary conditions and relaxation of initial and final anchor configurations) in a Nudged Elastic Band run.

I have found a couple of other posts where the answer to this question was no, but I noticed that, at least from the ones I could find, the posts were a few years old (2006, 2007). Is this still the case for vasp 4.6?

Thanks!

yes. due to the periodic boundary conditions, positions like -0.1 and 0.9 are completely equivalent (in E,...). So, the atoms are placed inside the unit cell (0 ->1 in direct coordinates) in CONTCAR. The only scenario where this has to be taken care of is when one does the interpolation of the atomic positions for the geometries of the IMAGES in the NEB. Therefore, please always check the initial- and the final state positions of the ions before you interpolate them.