Strange phonon spectrum after interpolation

[since2012]

Dear all,

I am calculating the EPC of my system. After wannierization, I fonund a small part of phonon is too stange when I plotted the band structure and phonon, as shown in the figure below,


Why it happened?

BTW, the version I used is v5.8， and the dense k-mesh and q-mesh are 48*48*48 and 24*24*24, respectively.

Thanks and regard!

Jianguo Si

[hpaudya1]

Hi Jianguo Si,

Could you double check your 'selecq.fmt' file and make sure all the q-points along the high symmetry path are included? It looks like those q-points are excluded in your calculation.

Happy EPWing,
Hari

[since2012]

Hi Hari,

Thanks for you kindly reply.

I have checked selecq.fmt and find that it inculde all q points (total 13824 points with a 24*24*24 dense q-mesh), as shown in the following figure. Besides, I check the band structure calculated with same high symmetry path, it is consistent well with the DFT one. So I'm still confused.

s1.png (25.27 KiB) Viewed 1539 times

Thanks again!
Jianguo

[stiwari]

Hi Jianguo,

I think the problem comes from the fsthick. Increase the fsthick size or use etf_mem=0 in your epw input file. If that does not help, please check the q-point path that it does not have any discontinuity.

Best regards,
Sabya.

[since2012]

Hi Jianguo,

I think the problem comes from the fsthick. Increase the fsthick size or use etf_mem=0 in your epw input file. If that does not help, please check the q-point path that it does not have any discontinuity.

Best regards,
Sabya.

Hi Sabya,

Thanks for your kindly relpy, I will try with your suggestion. Hope for better results.

Thanks again and respect!

Jianguo