[Pw_forum] Does dens grid contain the whole carse grid in calculation of EPI?

whwaniphy whwaniphy at gmail.com
Wed Feb 6 08:37:12 CET 2013


Hi Quantum espresso users!

I want to learn to calculate electron-phonon constant. The example of Al in exapmel07 shows that it is convenient to use a sparse k-point (nk1*nk2*nk3) grid to calculate the phonons (nq1*nq2*nq3), and use a dense k-point grid (nk1fit*nkfit2*nkfit3) to calculate the sum at the Fermi energy. The latter MUST contain all k and k+q grid points used in the subsequent electron-phonon calculation. A puzzle rises. Dose the vector “k” mentioned above belong to sparse grid or dense grid? The former means i/nk1+j/nq1=k/nkfit and latter means i/nkfit1+j/nq1=k/nkfit1 where i,j,k are integer. Which is right? I found the PWSCF won’t produce Error but produce different results in two cases. .
Suggestions welcome!

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20130206/f8c4aa3f/attachment.html>


More information about the users mailing list