Relaxtion of Specific Ions using VASP

Message

jlbettis · #1 Post by **jlbettis** » Wed Jul 06, 2011 5:43 am

Consider a simple perovskite, ABO3. Is VASP capable of keeping the B and O-site ions fixed, while allowing the A-site ions to relax into an optimum geometry?

Thanks for any responses in advance.

I think the answer is no.

alex · #2 Post by **alex** » Wed Jul 06, 2011 11:38 am

By symmetry definition this makes no sense.

jlbettis · #3 Post by **jlbettis** » Wed Jul 06, 2011 7:56 pm

POSCAR...
[ Edited Wed Jul 06 2011, 08:03PM ]

jlbettis · #4 Post by **jlbettis** » Wed Jul 06, 2011 8:02 pm

I was told that within the POSCAR file I can define which ions are fixed and with are not during my relaxation. I refer you to the VASP manual where the POSCAR file is described, it states "The seventh line switches to 'Selective dynamics' (only the first character is relevant and must be 'S' or 's'). This mode allows to provide extra flags for each atom signaling whether the respective coordinate(s) of this atom will be allowed to change during the ionic relaxation. "

The passage below is also taken from the VASP manual:

Cubic BN
3.57
0.0 0.5 0.5
0.5 0.0 0.5
0.5 0.5 0.0
1 1
Selective dynamics
Cartesian
0.00 0.00 0.00 T T F
0.25 0.25 0.25 F F F

Notice the red "F"'s theses are the tags necessary for fixing just one axis or all three.

And I do hope that the symmetry of the system will be lowered. Do you still think that this approach is nonsensical?

And thanks for the dialogue!!

[ Edited Wed Jul 06 2011, 08:05PM ]

jlbettis · #5 Post by **jlbettis** » Thu Jul 07, 2011 12:25 am

Ultimately my goal is to incrementally tilt the octahedra within the ab-plane, then relax the A-site cation.

alex · #6 Post by **alex** » Thu Jul 07, 2011 10:53 am

With the POSCAR above this is nonsense. You would have to move to a (larger) supercell if you want to study this effects you described. Then you rotate a bit and you succeed with your kind of relaxation ... (But it'll become symmetric again ...)

Cheers,

Alex

jlbettis · #7 Post by **jlbettis** » Thu Jul 07, 2011 5:21 pm

Ah,
The POSCAR above is nonsense to me as well! It isn't my system, I wish to study a perovskite, ABO3, and I do not want everyone to know the specifics =). I was just hoping to show that it is possible to keep certain ions fixed while allowing others to relax via the POSCAR "setup". As I stated above, I took the POSCAR above straight from the VASP manual.

But back to the symmetry argument....
(But it'll become symmetric again ...)
What do you mean by that, could you please elaborate.

[ Edited Thu Jul 07 2011, 05:26PM ]

alex · #8 Post by **alex** » Thu Jul 07, 2011 6:05 pm

For ABO3 the spinel structure should be more stable than a distorted spinel. So it will relax back.
You may introduce impurities (e.g. other atoms) to somehow fix a distortion.
You may also try on your own and yuo will see and learn.

Cheers,

alex

jlbettis · #9 Post by **jlbettis** » Thu Jul 07, 2011 7:05 pm

I thought that the AB2O4 is the spinel structure. Ideally i'd like to lower the symmetry of perovskite ABO3 from Pm3m to Pbnm. And your right I will definitely play with the system to see how it behaves. Thanks!

alex · #10 Post by **alex** » Fri Jul 08, 2011 11:04 am

yep, my mistake. sorry.