[QE-users] Dense k Point Interpolation to Arbitrary k Points for a Band Structure

[Lorenzo Paulatto]

> But I don't know how QE transforms the k vector corresponding to K and G to a single number. I tried to read the source code but didn't understand it.
Hello,
it is just the length of the path. If you have two points, k and k', the length will be zero at the first point, and |k-k'| at the second, if you add a third point k'', then you also add a length |k'-k''| and so on. There are some special rules to detect equivalent points and changes of directions, but they do not matter here.

To make it even simpler, if you see that the crossing appear at i.e. 30% of the segment between k and k', you can just add a point k + 0.3(k'-k) = 0.7k + 0.3k'
hth
--
Lorenzo Paulatto - Paris
On Mar 23 2021, at 5:19 am, Truman Ng Yu <ngtruman at nus.edu.sg> wrote:
>
> Hi Lorenzo,
>
> Thanks for your reply. I just set lsym=.false. and no_overlap=.false. and ran bands.x again as you suggested. I'm waiting for the results now.
>
> Just in case it doesn't work, may I have more details on how to find out the intersection points by hand? In the output of the original band structure calculation, K is located at 0.000 and G at 0.750 along the horizontal axis. But I don't know how QE transforms the k vector corresponding to K and G to a single number. I tried to read the source code but didn't understand it.
>
> In the bands calculation, the specific k points are chosen unlike the scf calculation where it's automatic. May I know how the grid is formed based on the specific k points? Then, I'd be able to check that the intersection points lie on the grid.
>
> Thanks.
>
> Regards,
> Truman
> National University of Singapore
> From: users <users-bounces at lists.quantum-espresso.org> on behalf of Lorenzo Paulatto <paulatz at gmail.com>
> Sent: Tuesday, March 23, 2021 12:53 AM
> To: Quantum ESPRESSO users Forum <users at lists.quantum-espresso.org>
> Subject: Re: [QE-users] Dense k Point Interpolation to Arbitrary k Points for a Band Structure
>
>
>
> - External Email -
>
>
>
>
>
> Hello Truman,
> the bands code decides if bands cross or avoid using different criteria: per default symmetry, otherwise overlap of wfcs. You may getting a better result by changing criteria. Otherwise, you'll have to find the intersection point by hand, but if it incommensurate with the grid, it will always show as a discontinuity if you zoom in close enough
>
> regards
> --
> Lorenzo Paulatto - Paris
>
> On Mar 22 2021, at 4:27 pm, Truman Ng Yu <ngtruman at nus.edu.sg> wrote:
> >
> > Hi QE users,
> >
> > I'm using QE 6.7MaX. I plotted a band structure with the following k point coordinates.
> >
> > K_POINTS {crystal_b}
> > 2
> > 0.375 0.375 0.750 2000 !K
> > 0.000 0.000 0.000 2000 !G
> >
> >
> > I'd like to include specific k points in between K and G such that the expected band crossing at the red dot in the figure below is captured in the band structure. How do I choose the k points such that this is possible?
> >
> >
> >
> > Thanks.
> >
> > Regards,
> > Truman
> > National University of Singapore
> > _______________________________________________
> > Quantum ESPRESSO is supported by MaX (https://ddec1-0-en-ctp.trendmicro.com:443/wis/clicktime/v1/query?url=www.max%2dcentre.eu&umid=61254bc7-7283-4c83-82cf-caf8de7ff73d&auth=8d3ccd473d52f326e51c0f75cb32c9541898e5d5-f0f4a8ee4cc782ae5521d01e88692d06d7daac84)
> > users mailing list users at lists.quantum-espresso.org
> > https://ddec1-0-en-ctp.trendmicro.com:443/wis/clicktime/v1/query?url=https%3a%2f%2flists.quantum%2despresso.org%2fmailman%2flistinfo%2fusers&umid=61254bc7-7283-4c83-82cf-caf8de7ff73d&auth=8d3ccd473d52f326e51c0f75cb32c9541898e5d5-867140e5c3e6b62f783eb9e91cb5fa75247edb38
>
>
>
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20210323/992797bf/attachment.html>