Fixed atoms move

Message

hortophilus · #1 Post by **hortophilus** » Tue Sep 30, 2008 4:48 am

I have a system composed of 20 Fe atoms in a 5-layer 2x2 stack (bcc) with the bottom three layers frozen using "F F F" after the atom's coordinate listing in the POSCAR file. However, when I look at the result, the "fixed" atoms have moved by almost 0.1 Angstrom! Further, it is not a constant translation of all atoms. Am I missing something or is there a bug in VASP that moves these atoms? I am very concerned because the system is tangibly changed by such movement.

forsdan · #2 Post by **forsdan** » Tue Sep 30, 2008 5:25 am

1. Have you checked if you have added the Selective dynamics tag correctly to the POSCAR-file?

2. Check if you do relax the volume/shape of the unit cell. Since the "F F F" locks the direct coordinates the positions might change if the basis vectors change during the run.

If none of the above helps, please provide your INCAR+POSCAR files so that we can give you a more detailed response.

Cheers
/Dan

<span class='smallblacktext'>[ Edited Tue Sep 30 2008, 07:28AM ]</span>

hortophilus · #3 Post by **hortophilus** » Wed Oct 01, 2008 12:45 am

I did not have the selective dynamics tag, so I will try it with the tag.

#4 Post by **admin** » Wed Oct 01, 2008 9:25 am

this explains the behaviour. If Selective dynamics is not set, the Ts and Fs are ignored.