[Wien] core charge leaks out of MT-sphere

[Xavier.Rocquefelte@cnrs-imn.fr]

Dear Colleague,

You could also change the muffin tin radius related to silicon atoms  
(RMT) in the case.struct file.

Concerning the optimization of the number of k-points, you simply  
should use such a strategy:
1. do an SCF calculation with few k points
2. when converged, increase the number of k points gradually
3. plot the total energy (:ENE) as a function of k-points. It will  
allow you to see when your results are converged, then choose the  
smallest k-mesh leading to good compromise between accuracy (you need)  
and computer efficiency.

Concerning the last question, yes there is other parameters. Did you  
read carefully the userguide? It is very nicely written and will be  
the best way to inform you. Finally, you could also look at the very  
good book written by David Singh "Planewaves, Pseudopotentials, and  
the LAPW Method".

With my best regards

Xavier


Kathirvel vel <kathir.phy at gmail.com> a écrit :

> Dear Sir/Madam,
>           Now i started to do calculation using wien2k for Silicon system.
> Here i am facing some problems and i am having doubts also.
>
> While initialize calculation in "X lstart:- shows its default value (-6.0
> Ry) for energy to separate core and valence states". but this value made to
> leaks out the some core charges of MT - sphere for Silicon element. So i
> increased the energy in order to bring the charges inside the MT - sphere
> and checked, it  optimized with  "-7.1 Ry"value. My question is " is this
> the way to bring the core charges inside the MT - sphere" ? or is there any
> other way to bring the core charges inside the sphere?
>
> And also Please clear my following doubts.....
> 1. How to optimze the k-points with energy?
> 2. Is there any other parameters which we have to optimize? if yes, tel me
> how?
>
>
> --
> With Regards
>
> V. Kathirvel
> Research Scholar
> Crystal Growth Centre
> Anna University
> Chennai
> Tamil Nadu - 600 025
> India.
>