Bader analysis for muffin tin radius

[halil54]

Dear all,

I have some problems to calculate my muffin tin radius in LMTO calculations. I am thinking that, at least, Bader analysis can give me some idea regarding the muffin tin radius approximation. First of all, do you think is that right ?

In this respect, I have calculated my charge density in VASP with core density included and perform:

chgsum.pl AECCAR0 AECCAR2

Then I used the bader code:

bader CHGCAR -ref CHGCAR_sum

That gives me the AVF.dat, ACF.dat and BCF.dat files. My structure is Laves YFe2.

Currently, I am not really able to get how I can interpret my results, but I attached them in here. Can you give some interpretations based on that results. In addition to this I have 0 distance in BCF.dat file. Can you give me an idea why it is like that ?

Regards,
Halil

[graeme]

The data in the ACF.dat looks fine. You can see an electronic charge of 9.69 on the first species and 8.65 on the second species. If you subtract the number of valence electrons for each atom type, you will have the net charge on each species.

I don't really understand the question about the muffin tin model. For that, you would typically integrate charge in a spherical volume around each atom. This is an alternative approach to the Bader calculation.