The ABINIT Users Mailing List ( CLOSED )

[Steven Homolya <Steven.Homolya@spme.monash.edu.au>]

On Mon, 6 Oct 2003 homegchiche@yahoo.fr wrote:

> 
> Dear Abinit users, Actualy i studie the migration of vacancies in
> nickel.Now I try to simulate the presence of one vacancy in nickel
> substracte, I don't know the input variable wich refers to this. I
> don't know if necessary to give the cordonate of this vacancy. any one
> have an exemple about this? Thanks for your time Megchiche
> 

Vacancy is like any other type of substitutional defect (well, almost).  
You need a big enough "super"-cell of the "host" (hh) substance in which
you replace one lattice site with defect or vacancy (i.e., remove hh
atom from that site).

You'll need to remember that you're always computing properties of an 
infinite, periodic structure.

How big a supercell you need, you'll have to figure out by converging
whatever property you're interested in with respect to supercell size ...
if that is in fact possible. If not, you might have to face limitations of
hardware/software and extrapolate or give up.

For vacancy in fcc Ni, you might want to explore the following supercell
configurations:

8 site fcc supercell  ---  7 Ni : 1 vacancy
16 site bc supercell  --- 15 Ni : 1 vacancy
27 site fcc supercell --- 26 Ni : 1 vacancy
32 site sc supercell  --- 31 Ni : 1 vacancy
64 site fcc supercell --- 63 Ni : 1 vacancy

Always make sure the property you're interested in is well converged wrt.  
numerical parameters (kptrlen, ecut, etc), and the supercell is geometry
optimised (see vars. optcell, ionmov and of course the tutorials).

hope this helps,

Steve

-- 
Steven Homolya
School of Physics and Materials Engineering
Monash University, VIC 3800
Australia
Tel: +61 3 9905 3694
Fax: +61 3 9905 3637