[Pw_forum] k points

[Giovanni Cantele]

On 12 May 2014, at 15:20, Caroline <katesedate at gmail.com> wrote:

> Dear all, 
> 
> I am trying to understand the significance of k points in the input file. Can you please help me with the following questions:
> 
> 1) How k points are selected? Dos it have to with the dimension of the crystal? Say for 2 dimension crystal only two lines in K point section is defined?
> 
> 2) Or k points have to do with the numerical solution of integrals? I am confused how to define these values, and how they are important. 
> 
> 3) Is there a relation between number of k points and number of atoms in the lattice crystal?
> 
> 
> I have read text books on this, and learned that k points need to selected in such a way that total energy is converged.
> 
> Please help me with this.
> Thanks a lot,
> C.
> 

no relation between the number of k-points and number of atoms. K-point grids are used to perform sums over the Brillouin zone and must be chosen
such that the quantity you want to calculate is converged. For a 3D crystal you need a 3D grid, whereas if you lower the dimensionality of the crystal to 2D, 1D or 0D
you will have 2D, 1D or 0D grids respectively (e.g. 4x4x1, 4x1x1, 1x1x1).

You should better investigate the role of sums/integrals over the Brillouin zone using solid state textbooks, as well as try to follow the tutorials you can find on the Quantum-ESPRESSO
web site, for example http://www.quantum-espresso.org/tutorials/shanghai-2013/ (but you can find many others at http://www.quantum-espresso.org/tutorials/).

What is really important is to monitor how any property you are interested in (e.g. formation energy, phonon frequencies, bulk moduli) changes upon increasing the k-point grid. Also consider
that different properties might require different grids to be converged. For example, charge density usually converges faster (with respect to k-point sampling) than phonon
frequencies or optical properties.

Giovanni


PS please kindly provide an affiliation in a future message, as usually requested to the users of this forum.

-- 

Giovanni Cantele, PhD
CNR-SPIN
c/o Dipartimento di Fisica
Universita' di Napoli "Federico II"
Complesso Universitario M. S. Angelo - Ed. 6
Via Cintia, I-80126, Napoli, Italy
e-mail: giovanni.cantele at spin.cnr.it
Phone: +39 081 676910
Skype contact: giocan74

ResearcherID: http://www.researcherid.com/rid/A-1951-2009
Web page: http://people.na.infn.it/~cantele