[Pw_forum] spin dope

[Paolo Giannozzi]

On Jun 27, 2007, at 6:54 , Srijan Kumar Saha wrote:

>       It's written in the input documentation that nelup and neldw  
> are REAL.
>       The sum must yield nelec that is also real. Then why
>       tot_magnetization (nelup - neldw) is INTEGER ( default = -1  
> [unspecified] )?

because whoever implemented this feature didn't need a fractional
total magnetization

>       In the code, total magnetization is divided by  omega/ (nr1  
> nr2 nr3).
>       What does it mean???? I mean what is omega????

omega is the volume of the unit cell, (nr1 nr2 nr3) is the total number
of grid points in real space, omega/(nr1 nr2 nr3) = Delta V in the
integration

Paolo
---
Paolo Giannozzi, Democritos and University of Udine, Italy