[Wien] ask spin-polarized nonmagnetic state again

[Peter Blaha]

I don't quite understand your problems.
You want to perform a non-magnetic calculation. Non-magnetic means, that 
spin-up = spin-dn; and this is what we ensure by this trick when copying 
vectorup to dn.
Even when Vup = Vdn you can of course apply LDA+U (it may redistribute 
some states (euqlly for both spins).

>      Thank you for your reply. I have read the script of runsp_c_lapw. But I
> meet some problems. In your reply, "the up and dn densities are the same",
> however, I think it leads to V(up)=V(dn) and we can't do the LDA+U. Is there
> any statistical method in the script and why can you copy the case.vectorup
> to vectordn simply? I just wonder the theoretical foundation
> 
> Thank you very much.
> 
>  
> 
>  
> 
>  
> 
>  
> 
>  
> 
> Yours
> 
>  
> 
>  
> 
> Kongshi
> 
>  
> 
>  
> 
> 


                                      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/
--------------------------------------------------------------------------