[Wannier] Projections

[Jonathan Yates]

On 21 May 2010, at 00:11, Mark Mazar wrote:

> Hello,
> 
> I was wondering, is it possible to find wannier functions with bond
> projections?
> 
> For example, one of the features of my system is a Ta-C bond. How can
> I specify the projections in the *.win file in order to capture the
> bonding behavior between these two atoms?

Mark,

I think it is important to realise that the initial projections you specify in the *.win file are simply guesses for the MLWF. Using a good guess will speed up the minimisation - but they should not determine the final form of the MLWF.

 For a given set of bloch states the MLWF should be well defined (as the set of WF which minimise the quadratic spread). That statement needs a few qualifications to be precise - but it does means that you can't determine the form of the MLWF yourself.

 If you obtain the MLWF from the set of valence states then they probably will tell you something about the bonding in your material. (eg see the pictures of MLFW in Si and GaAs on wannier.org - also look at the Psi_k article linked from the site).

 Hope that's of some help.

 Jonathan
 


-- 
Department of Materials, University of Oxford, Parks Road, Oxford, OX1 3PH, UK
tel: +44 (0)1865 612797                http://users.ox.ac.uk/~oums0549/