EPW Forum

Dear Dr. Samuel,

I am newbie of EPW. I am interested in the example of B-doped diamond, which their band structure and phonon is shown in the unfolding BZ. I tried to look at the package and I can not find thany input and output for B-doped diamond.

Could you give me some hints how to redraw the phonon dispersion in unfolding BZ? Is it possible for me to get the input file for B-doped diamond?

Thank you for all your help in advance.

With best wishes,
Loc Duong

IBS Center for Integrated Nanostructure Physics
Room 86283, N-Center, Suwon Cammpus, Sungkyunkwan University, 2006 Seobu-ro, Jangan-gu, Suwon-si, Gyeonggi-do, Korea
Phone: +82-31-299-6993
Email: mambom1902@yahoo.com, mambomm1902@gmail.com

Dear Loc Duong,

You can find the B-doped diamond example in



However, if you are only interested in the phonon bandstructure, you do not really need to use EPW.

You can use the QE/PH package http://www.quantum-espresso.org/wp-content/uploads/Doc/INPUT_PH.html and use Fourier interpolation to have a nice bandstructure.

Best,

Samuel

Dear Samuel,

Thank you so much for your prompt reply. I did check twice but the example contain only for diamond. Could you upload it in the forum? It will be helpful for many people.

Actually, I am interesting in the electron-phonon coupling in two dimensional materials. What I am looking is the spectral function dependent on k-vector as Fig. 6 in the diamond example.

There is another thing I am looking for is the spectral function dependent on q-vector. Maybe there is information I can get there.

I just start with EPW. I think I need to step by step learning the code as well as the theory behind it.

With best wishes,
Loc

Dear Loc,

The example I mention is B-doped diamond using the virtual crystal approximation. You can see that the C psp is different from a normal one as it contains less electrons.

Are you looking to do an explicit B-doped calculation, i.e. supercell with douzens of atoms and then replace one by B?

You can do this but to achieve the experimental doping, it will be quite time consuming.

Best,

Samuel

Dear Samuel,

Thank you so much for your reply. This is the first time I hear about the virtual crystal approach. I am reading about it.

Technically, could you instruct me how to generate the virtual potential (e.g. the potential file of C in the case of B-doped)?

THank you again for all your help.

With best wishes,
Loc

Dear

The C psp for that specific doping of B can be found in #

I did not generate the psp myself.

I'm guessing it can be done with any psp generator (like APE, http://www.tddft.org/programs/APE/) by placing fractional electron while generating it.

Best,

Samuel

Dear Samuel,

Thank you so much for your kind reply. I will further check it.

With many thanks,
Loc

Dear EPW users,

For this example I'm trying to reproduce Figure 8 (c) from Giustino et al, Phys. Rev. B 76, 165108 (2007). However, when I plot :

ibnd jbnd imode
4 4 6

For all q-points in the fine grid, the curve is quite different. I'm using exacly the same parameters as in the original input files just added prtgkk = .true.
Does anybody could help me with some tips about how to reproduce this figure?

Thanks in advance,

Wellery.

Dear Wellery,

The prt_gkk routine does averages on degenerates bands and modes. In addition Fig 8 was done with a very old version of the code.

Can you try to replicate the similar plot we did during the el-ph school http://epw.org.uk/School2018/School2018 ... .Verdi.pdf

You will also need to do the Tue lesson at http://epw.org.uk/Documentation/School2018

Best wishes,
Samuel

Dear Dr. Samuel,

Thank you for your swift answer, I will follow the steps indicated by you, as soon as I have some results I will post again.

Best wishes,

Wellery.