[Pw_forum] grid shift

[dekura at cmp.sanken.osaka-u.ac.jp]

Dear Fan

I think it is reasonable that use of different grid gives different total
energy because the calculated KS energies are different, and hence
averaged total energy is different in each selected grid if you don't have
sufficient grid for k-points leading to converged energy.

So that I think you should check the convergency of total energy with
increasing mesh of k-point.

In my guess, If you have dense grid, you can have same energy not
depending on grid offsets.

Best Regards
Haruhiko Dekura






> Dear?everyone:
>
> I have some questions about grid offsets.
>
> In?our input file, the k-points are automatically generated with uniform
> grid,?and the grid offsets?are either set as 0 (no offset) or 1 (grid
> displaced by 1/2 step). But the problem is that?we got different results
> with different offsets.
>
> For example, we tested the adsorption of?one oxygen atom on three
> low-index Pt crystal planes, namely, Pt(100), Pt(110) and Pt(111).?With
> grid shift (0 0 0), the adsorption energy of O on Pt(110) is slightly
> larger than on Pt(100), but much larger than on Pt(111). However, with
> grid shift (1 1 1), the order of adsorption energy is
> Pt(110)>Pt(111)>Pt(100).
>
> Which result is more reasonable?
>
> Best Regards & Merry Christmas,
>
> Fan Yang
> *******************************************
> Fan Yang
> PH.D Candidate?in Electrochemistry
> College of Chemistry and Molecular Science
> Wuhan University, 430072, Hubei Province, China
> E-mail:shrek_826 at yahoo.com.cn
> *******************************************
>
>
>       ___________________________________________________________
>   好玩?�等�?，?箱?�全新上?！
> http://card.mail.cn.yahoo.com/_______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>