[Pw_forum] choosing ecutwfc through vc-rx

Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu
Sun May 31 19:10:07 CEST 2009


On Sun, 2009-05-31 at 06:22 +0330, Mehrnaz Anvari wrote:
> Dear all
> I have some fundamental questions.
> 1.If  I  want to achieve the equilibirium value of lattice paramrters
> (such as a, c,...) is it possible to replace bulk modulues
> calculations with vc-rx?

in principle, yes. but there are some caveats. for vc-relax you have
to converge the stress tensor and that will usually take a higher
wavefunction cutoff and if you use ultrasoft psps particularly a
higher density cutoff. so for all simple systems where the symmetry
defines all/most atom positions, just running a series of single point
total energy calculations should be much less computational effort
and is trivially parallizable.

> 2.when I use vc-rx for my material most of the time after 300 stpes
> the stress becomes approximately zero(-0.6-0.6) but energy doesn't
> converge. I don't pay attention just use the cell parameters & atomic
> positions to do scf calculations. I tried to achieve convergency in
> scf calculation with changing ecutwf & k-points. Now I think it
> couldn't be true & I have to optimize my ecutwfc during vc-rx
> calculation (Is it true?) what about k-point ,where could I optimize
> it? 
> 

wavefunction cutoff and k-point grid can be converged fairly
independently. however, it is usually more efficient to first
converge the wavefunction (and density) cutoff with just one
or a few k-points and then crank up the k-points and make a
confirmation test on wavefunction. if you use smearing, then 
you also have to consider the smearing parameter vs. number
of k-points relation.

cheers,
   axel.

> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.



More information about the Pw_forum mailing list