[Wien] How to set parameters in virtual crystal approximation?

[Peter Blaha]

No, you can't use VCA for something like  Pb(1-x)SnxTe (x=0.3 )

VCA in WINE2k is only possible between neighboring elements, like Na-Mg or
Ba-La. 
In principle you can simulate  Pb(1–x)GaxTe, but with Tl instead of Ga.
However, VCA is only "good" if these elements do not have
"active" electronic states at EF (like Ba/La, which are basically Ba2+/La3+
ions, which just donate a certain number of electrons to the rest of the
system).

Use some supercell modells (for Pb(1-x)SnxTe (x=0.3 )).

If x is small, but changes the number of valence electrons, you can use
the "rigid-band" model, i.e. calculate the pure compound, and only for the 
final DOS you change NE in case.in2 accordingly.


>      I am a new user of WIEN2K.I want to use the
>  virtual crystal approximation (VCA)to caclute the next systems, I don't
>  know how to set the value Z and NE in case.struct and case.in2.
>  Could someone tell me about it.
>  No.1   Pb(1-x)SnxTe (x=0.3 )
>  No.2   Pb(1–x)SixTe (x = 0.00005),
>  No.3   Pb(1–x)GaxTe (x = 0.001)
>  
>  No.4   For PbTe I select  ENERGY to separate core and valence states -8.0 Ry,
>  so the NE is 36 in pbte.in2. If the carrier concentration is 4xE-4/unit cell,
>  for p-type pbte is it right to set the NE=35.9996 and set Z
>  value of Pb in case.struct to 81.9996? Is this the only method?
>  for n-type is it right to set NE=36.0004, Z=82.0004? Is this the only method?
>  
>  Z value :
>  Pb:82.0       Sn:50.0      Te:52.0      Si:14.0      Ga:31.0
>  
>  
>  Joey Davis
>  
>  
> 		
> ---------------------------------
> How low will we go? Check out Yahoo! Messenger’s low  PC-to-Phone call rates.


                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------