ABINIT Discussion Forums

Respected Sir
I want to perform an atomic position relaxation.so, for this i have first done ionmov 2 and optcell 0,second ionmov 2 and optcell 2 (using the atomic positions from step 1).
Sir,Please kindly guide me what are those atomic positions from the output file of step 1 which i must use in the next step2?
The output file is attached for your kind reference.
Many thanks in advance
waiting for your response
With Best Regards

Well, I'm not sure I fully understand your question, but here goes:

1) with ionmov 2 and optcell 0, the atom positions are relaxed within a fixed cell geometry, consistent with the input space group. You can find the new, relaxed atom coordinates at the the bottom of the output file, as "xred" (presuming you are using reduced coordinates).

2) Then, for ionmov 2, optcell 2, copy the xred output coordinates from file 1.out to the input file and start your full cell+coordinate relaxation.

But looking at your attached output file, I have a couple of comments/questions:

First, it looks like all your atoms are on special positions (high symmetry), so the force on each one vanishes by symmetry--that's why your optcell 0 calculation only did a single iteration, it is impossible to lower the energy by moving atoms because none of the atoms have any degree of freedom to move, without breaking symmetry. So, in this case step 1 from above was not necessary.

Second, why are you running with a non-primitive cell? You seem to be forcing it to use the conventional cubic cell but the primitive cell would be much smaller and faster to compute. But maybe you are planning down the road to introduce defects or something like that.

Hello Sir,
Thank you very much for your response now i can understand why there are not many changes in my xred after relaxation too.Many thanks.
The answers to your questions are:
1)i choose these set of atomic positions because i required to generate a rocksalt fcc cubic structure.but got the following error that time:
chkprimit : ERROR -
According to the symmetry finder, the unit cell is
NOT primitive. The multiplicity is 2 .
The use of non-primitive unit cells is allowed
only when the input variable chkprim is 0.
Action : either change your unit cell (rprim or angdeg),
or set chkprim to 0.
So, i tried using chkprim 0 and got this non-primitive unit cell after running this input variable.
2) yes sir actually i will require to introduce substitution of one atom of my unit cell by some other atom at last.
so,i will be thankful to you if you Kindly refer me some article/weblink in abinit itself where i can find detail how abinit generate the other atomic positions of a crystal structure other than which user provide through input file because i still have confusions regarding mentioning atomic positions in the input file.
Regards
Mukul