[Pw_forum] generating k-point weights

[shobhana at jncasr.ac.in]

>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Mon, 03 Aug 2009 13:47:54 +0200
> From: "Lorenzo Paulatto" <paulatto at sissa.it>
> Subject: Re: [Pw_forum] generating k point weights
> To: "PWSCF Forum" <pw_forum at pwscf.org>
> Message-ID: <op.ux24d4ila8x26q at paulax>
> Content-Type: text/plain; charset=utf-8; format=flowed; delsp=yes
>
> In data 31 luglio 2009 alle ore 16:25:39, Manoj Srivastava
> <manoj at phys.ufl.edu> ha scritto:
>
>> Dear All,
>>  I was wondering if someone can tell me, how to generate k point
weights
>> in the BZ. Reference of a paper would be great!
>
> Dear Manoj,
> you can find the algorithm tha tgenerates points and weights in
> pwtools/kpoints.f
> The obvious reference is the original article from Monkhorst and Pack:
Phys. Rev. B 13, 5188 - 5192 (1976), although I don't know if they give
an
> explicit formulation for the weights there (weights depend on how many
symmetry-equivalent copies of a point are present in the set, not on the
generation algorithm).
>
> cheers
>
>
Hello, I would like to point out that pwscf generates k-points slightly
differently from the way it was done by Monkhorst and Pack in their
original paper. For them, the k-point grid either included or did not
include Gamma (zone center) depending on whether the number of divisions
was odd or even. However, in PWscf, 'unshifted' grids always include
Gamma, and 'shifted' ones do not, regardless of whether the number of
divisions is odd or even.
To get weights: you have to see how many k points in the full BZ are
equivalent to the given point in the IBZ. I have a slide illustrating this
in my recent talk at the UCSB Q-E summer school, see my first lecture,
slides available at http://www.quantum-espresso.org/wiki/index.php/QESB09

Shobhana Narasimhan
Theoretical Sciences Unit
Jawaharlal Nehru Centre for Advanced Scientific Research
Bangalore