trying to run elfield; failing miserably.

Phonons, DFPT, electron-phonon, electric-field response, mechanical response…

Moderators: mverstra, joaocarloscabreu

Locked
johnbrehm
Posts: 12
Joined: Mon Sep 25, 2017 9:30 pm

trying to run elfield; failing miserably.

Post by johnbrehm » Mon Sep 25, 2017 9:37 pm

I tried the tutorial tffield_6.in and it worked fine.
So, I branched out and changed to the real system I was working on: 40 atoms, monoclinic, with 156 val. bands.
Miserable failure: it did the polarization calc fine, but then hung for hours after that.

So, I decided to break up the ttfield_6.in.
1. I did the polarization cal and it ran fine. I copied over the WFK and all other output files to a new dir.
2. Then, in the new dir,I made a new in-file that just did this:

getwfk 1
berryopt 4
efield 0.0001 0.0001 0.0001

Should have run, no?

No.
Here is the error I get:
initberry: Reduced electric field (ebar)
red_efieldbar(1:3) = 0.001172310 0.002030586 0.002419097

initberry: for direction 1, nkstr = 8, nstr = 64
initberry: for direction 2, nkstr = 8, nstr = 64
initberry: for direction 3, nkstr = 8, nstr = 64

initberry: COMMENT -
As a rough estimate,
to be below the critical field, the bandgap of your system
should be larger than 0.53 eV.


================================================================================

Computing the polarization (Berry phase) for reciprocal vector:
0.12500 0.00000 0.00000 (in reduced coordinates)
0.01066 -0.00000 0.00145 (in cartesian coordinates - atomic units)
Number of strings: 64
Number of k points in string: 8

--- !BUG
src_file: berryphase_new.F90
src_line: 937
mpi_rank: 0
message: |
For k-point # 1,
the determinant of the overlap matrix is found to be 0.
...


leave_new: decision taken to exit ...


Now, I do have a gap -- of 1.2 eV. so that's not the issue.

Can someone please help.
Tnx.
JB

mverstra
Posts: 655
Joined: Wed Aug 19, 2009 12:01 pm

Re: trying to run elfield; failing miserably.

Post by mverstra » Thu Oct 12, 2017 7:41 am

Please read the netiquette and get back to us with many more details. Which version of abinit which system are you in reduced dimensions?

Basically it can not find a neighborhood of inequivalent kpoint neighbors to perform the berry phase strings. This used to happen if your cell is very anisotropic and the nearest k in one direction is further than the second or third neighbor in another direction.
Matthieu Verstraete
University of Liege, Belgium

Locked