The ABINIT Users Mailing List ( CLOSED )

[Takeshi NISHIMATSU <t_nissie@yahoo.co.jp>]

Tom,

You should contine to discuss it in the
forum, so that someone can correct us.

If you can find/define/approximate transition path(s) between
two phases easily, you may have to trace the path(s) under
some pressures. See a thread around:
https://listes-2.sipr.ucl.ac.be/abinit.org/arc/forum/2005-05/msg00010.html

> Dear Professor Takeshi,
> 
> Thank you for the reply. The thing is that if i do the relaxation with
> restricted cell shape and use strtarget to set the stress tensor, i find it 
> is
> hard to find the exact intermediate structure. For example, i want to catch 
> the
> phase transtion between hcp and omega structures, these two are all 
> hexagonal,
> but the numbers of atom in the primitive cell are different. How should i
> handle this situation, should i construct two supercells with the same 
> number
> of atoms in it and let abinit do the rest relaxation job? Could you 
> enlighten
> me more specifically? Thank you very much.
> 
> sincerely,
> tom
> 6/6/2009

Good luck,
-- 
 Takeshi
 love && peace && free_software
 http://loto.sourceforge.net/feram/   Fast MD program for perovskite-type 
ferroelectrics