The ABINIT Users Mailing List ( CLOSED )

[verstraete@pcpm.ucl.ac.be]

Hello Prasenjit Sen

I've looked at your outputs; a few comments:

1) the convergence of Etot with the number of kpoints is not monotonous. 
Your comparison of 2x2x1 and 4x4x1 implies that you're not converged yet 
(by more than 10 mHa).

2) Your cell is not primitive, and you could halve it without any loss of
precision (doubling the number of kpts along that direction). I presume
you want to do alternated dimers later, but you can always double the
energy. When you impose chkprim=0, abinit doesn't find the space group
(although it may not be output because you're using version 4.2).

3) I can't see the difference in symmetry between what you call the symm 
and the asymm configurations. Given that the latter, for the same 2x2x1 
grid, is lower in energy, I would say that it's the correct one, but you 
should converge both wrt nkpt. Am I right in thinking that all you changed 
was the starting point? Either you have 2 local minima, but the 
configurations look very close, or you're not converged yet wrt something. 
It looks like the forces and displacements are small enough... Your sub 
layers are also a bit different in the 3 cases.

Bye

Matthieu


-- 
===================================================================
Matthieu Verstraete               mailto:verstraete@pcpm.ucl.ac.be 
PCPM, Boltzmann, pl. Croix du Sud, 1            tel: 010/ 47 33 59
B-1348 Louvain-la-Neuve  Belgium                fax: 010/ 47 34 52