[ase-users] Using VASP/NEB module - atoms moving acorss the PBC

[Eric Hermes]

I see the issue you're talking about and I've implemented a hackish fix locally. I hesitate to push the fix to SVN until I understand what's going on in this routine a little bit more. Particularly, I'm not sure where else in the ASE code base the "fit0" function is called, and currently the fit0 function is not passed enough information to perform the minimum image convention check. I've attached the diff of my hackish fix, though.

In addition, I suspect that this routine does not work correctly for SS-NEB (where the lattice vectors can change over the band), and I don't know enough about SS-NEB to fix it. It would be useful if someone who knows more about SS-NEB could chime in and explain e.g. how the distance between images is defined when the unit cell vectors can fluctuate.

Eric

From: Ambarish Kulkarni [mailto:arkulkarni245 at gmail.com]
Sent: Friday, August 28, 2015 4:47 PM
To: Eric Hermes <ehermes at chem.wisc.edu>
Cc: ase-users at listserv.fysik.dtu.dk
Subject: Re: [ase-users] Using VASP/NEB module - atoms moving acorss the PBC

Just a follow up.

By using find_mic in the tangent calculation, the issues with atoms moving about the PBC are indeed fixed.

However, I think the function fit0(E, F, R) in neb.py also needs to be modified to check for minimum image conventional distances. For atoms that are on different sides of the PBC, the current version calculates very large values of the reaction coordinates.

Thank you,

Regards,
Ambarish


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/ase-users/attachments/20150831/2029d108/attachment.html>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: neb_temp_fix.txt
URL: <http://listserv.fysik.dtu.dk/pipermail/ase-users/attachments/20150831/2029d108/attachment.txt>