Marcos,
Thank you. I have get it.
2007/11/23, Marcos Verissimo Alves <[EMAIL PROTECTED]>:
>
> The variable cell option is useful when you want to determine the lattice
> constants in an automatic fashion, but sometimes you don't want to relax
> the cell. For example, you could want to determine changes in the band gap
> or how the chemical bonds change under uniaxial stress. Then you would
> keep lattice vectors fixed and let the atomic positions relax. If you want
> the DOS for the equilibrium geometry, you must at least relax the atomic
> coordinates at the equilibrium value of the lattice constant, doesn't
> matter how you obtained the lattice constant. If you do a variable-cell
> calculation, in the end you obtain both - the relaxed geometry at the
> equilibrium lattice constant.
>
> For some structures, a variable cell calculation is not necessarily worth
> the effort. For example, take a simple metal or a zincblende structure.
> For those structures, which can be described by one atom in the unit cell
> (two in the case of the ZB), you can simply perform a series of six ir
> seven fixed cell calculations around the equilibrium lattice constant -
> they will do only one entire SCF cycle because of symmetry, and a
> Murnaghan fit will give you the bulk modulus, perhaps in less steps than a
> variable-cell would need to find the minimum energy structure. So, how you
> obtain the lattice constants is not important, what matters for the DOS is
> that the atomic positions be relaxed at the desired lattice constant.
>
> Cheers,
>
> Marcos
>
>
>
> Vous avez écrit / You have written / Lei ha scritto / Você escreveu...
> Ñ»ª
> >> Marcos,
> >
> >
> > In the letter you have said that both bands and DOS should be
> > calculated from the relaxed structure. Did you mean that the
> > MD.VariableCelloption be True? Or others?
> >
> > Thank you .
> >
> >
> >
> >
> >
> > --
> > Best Wishes!
> > F.H.Yang
> >
>
>
> --
> Dr. Marcos Verissimo Alves
> Post-Doctoral Fellow
> Unité de Physico-Chimie et de Physique des Matériaux (PCPM)
> Université Catholique de Louvain
> 1 Place Croix du Sud, B-1348
> Louvain-la-Neuve
> Belgique
>
> ------
>
> Gort, Klaatu barada nikto. Klaatu barada nikto. Klaatu barada nikto.
>
--
Best Wishes!
F.H.Yang
