The ABINIT Users Mailing List ( CLOSED )

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

On Thu, 9 Oct 2003 12:26 am, homegchiche@yahoo.fr wrote:
> I have another question about the study of vacancy with abinit. I would
> like to know if the present code could give directly self consistent
> Vacancy-formation energy.
>

Not sure what you mean by directly.

The most direct way I know is to first do a calculation for the pure 
substance. This is a small calcn., since you just use the primitive unit 
cell, usually one atom per cell. This gives total energy per atom, say 
E_pure. Then do a supercell calculation with N>>10 sites, where one atom has 
been removed. This gives energy, say E_scell, for the N-1 atom supercell. 
Then the energy of formation of the vacancy is

E_scell - (N-1)*E_pure

Make sure the two calculations use compatible parameters. In particular, 
number of k-points should be proportional to volume of reciprocal unit cell 
(inversely prop. to volume of supercell). You can try atuto generating k-mesh 
keeping kptrlen the same for the two calcns. (though I found that this fails 
someimes, may be a bug?)

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