[Wien] About the occupation of atoms!!

[Hong Jiang]

Hi,
I guess your question is a little puzzling. What do you mean by
"occupation of atoms"? Do you mean the occupation numbers in case.inst?
If I understand your question correctly, you are trying to calculate an
alloy with fractional number of atoms. But a normal DFT calculation can
only treat integral number of atoms. To model alloys with fractional
number of atoms you have to model a supercell with consistent ratios. In
your case, you need to model a supercell of at least Mo24Si15C3
(probably more is needed). You can also use the so-called virtual
crystal approximation to model alloys in a cheaper and approximate way.

-Hong

bluemoon wrote:
> Dear All,
> I have changed the occupation of atoms to get a typical compound of
> Mo4.8Si3C0.6 (Mn5Si3 type; C at 0,0,0 and 0,0,1/2; the occupation of C
> is 0.6; Parthe et al. Acta Cryst. 1965 19:1031).
> After the optimization of the crystal structure, the rusult shows the
> total energy of Mo10Si6C2 (i.e. the all atoms in the studied structure
> / 18 atoms in a unit cell) but not Mo9.6Si6C1.2. I have calculated the
> total energy of Mo10Si6C2 before that job.
> Can the calculated method only deal with the full occupation (1.0)?
> I have heard that somebody have done such job (change the occupation
> of atoms) for a particular purpose but I have not known which the
> calculated method he chose.
> I want to know if anybody has done such job by Wien2K. And I will be
> grateful if anybody gives me some advice.
> Best wishes!
> Guangming Cheng
> Shenyang National Laboratory for Materials Science,
> Chinese Academy of Sciences,
> 72 Wenhua Road, Shenyang, China, 110016
> ------------------------------------------------------------------------
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>