The ABINIT Users Mailing List ( CLOSED )

[Stephane BECU <stephane.becu@st.com>]

Hi Pierre,

I'm glad to hear from you! In fact the supercell "only" contains 40 atoms (16 
Al + 24 O)... The thing is I do not know if I have to imput all the 40 
positions in my input file...

Stéphane

-----Original Message-----
From: Pierre Boulenc [mailto:pierre.boulenc@isen.fr] 
Sent: Thursday, March 16, 2006 11:49 AM
To: forum@abinit.org
Subject: Re: [abinit-forum] Vacancies implementation in Al2O3


Dear Stephane,

How big is your supercell for gamma-Al2O3?
Starting with the cubic spinel unit cell (which has 16 * 83.33 % 
octahedral Al, 8 tetrahedral Al and 32 O atoms), you need at least 3 
unit cells to have 8 vacancies in your whole supercell. That would bring 
you up to 160 atoms in your supercell.
I think that's quite a lot, and if you plan to calculate dielectric 
properties, you will need well-enough converged results which may take a 
huge time for such a number of atoms.

Regards,

Pierre


Stephane BECU a écrit :

>Dear abinit users,
>
>I try to build a supercell for calculating dielectric properties of 
>gamma-Al2O3. I begin with an ideal spinel unit, typified by MgAl2O4. Then I 
>do some modification of this ideal unit, one of it is to remove two Al atoms 
>to create two vacancies. The thing is I do not know how to declare the 
>vacancies:
>--> do I have to count them like the other atoms in natom? what type 
>--> (typat) of atoms do I have to define? how define their positions?
>
>Thanks for your help,
>
>Stéphane Bécu
>
>  
>

-- 

Pierre Boulenc

----------------------
IEMN / ISEN / Physique
----------------------
ISEN
41 Boulevard Vauban
59046 LILLE CEDEX
Tel: 03 59 57 44 12
Bureau: C602