ANADDB calculation of silicon

Moderators: mverstra, aromero

ANADDB calculation of silicon

Postby Aiswarya Samal » Tue Mar 26, 2019 12:08 pm

Dear all,
I am using abinit 8.8.4 and I am following the tutorial on electron-phonon coupling.I got the ddb files and gkk files and merged them following the tutorial.But when I am doing the ANADDB calculation I am facing the following errors.
src_file: read_gkk.F90
src_line: 666
mpi_rank: 0
message: |
gkk element 4 1 1 1 was not found by symmetry operations on the irreducible perturbations and qpoints given
Here I am uploading my input files.
Attachments
teph_1.in
(3.94 KiB) Downloaded 51 times
Aiswarya Samal
 
Posts: 7
Joined: Mon Mar 18, 2019 2:56 pm

Re: ANADDB calculation of silicon

Postby mverstra » Tue Mar 26, 2019 6:50 pm

First off, you have 2 atoms, and left rfatpol 1 1. It should be rfatpol 1 2

You should also check that you have all of the q-points. Should be ok, but I suspect that you just copied from the FCC Al case: in the general case you need to redetermine them for every symmetry group / system. Also see the input variables qptopt 1 ngqpt 2 2 2 shiftq 0 0 0, and iqpt = 1, 2...

NB: by default, with an semiconductor, you will get 0 for all of the EPC quantities, as the Fermi level will be in the gap. Rigid band doping is possible, but not always very well defined. If you want temperature shifts of the gap, etc... you should try the other tutorials on the AHC theory and eig2rd
Matthieu Verstraete
University of Liege, Belgium
mverstra
 
Posts: 637
Joined: Wed Aug 19, 2009 12:01 pm

Re: ANADDB calculation of silicon

Postby Aiswarya Samal » Wed Mar 27, 2019 10:22 am

I have some doubts.
How to choose this q-points for bulk or surface structure of several materials??
Aiswarya Samal
 
Posts: 7
Joined: Mon Mar 18, 2019 2:56 pm

Re: ANADDB calculation of silicon

Postby mverstra » Wed Mar 27, 2019 4:09 pm

The q points should be a homogeneous grid (here 2x2x2) and usually centered at Gamma (shiftq 0 0 0). You can reduce them by symmetry and then complete the dynamical matrices again inside anaddb.

To get the irreducible list of q, run a normal ground state (in another directory with low ecut and nband, just to get the list of wavevectors) with kptopt 1, nshiftk 0 0 0 and ngkpt 2 2 2. The resulting kpt list in the output/log can be used as q-points in the DFPT calculation (the symmetries are the same for q and k reduction).

An alternative is to use the qptopt shiftq ngqpt iqpt variables I mentioned in the previous post. Abinit then generates the q points automatically, and you just have to set iqpt 1 for Gamma, iqpt 2 for the next non zero q, etc... If you go too high in iqpt the code stops (or use previous method to see how many irred q you have).

For a surface you will have a 2D grid e.g. 2x2x1
Matthieu Verstraete
University of Liege, Belgium
mverstra
 
Posts: 637
Joined: Wed Aug 19, 2009 12:01 pm


Return to Response calculations

Who is online

Users browsing this forum: No registered users and 3 guests

cron