Page 1 of 1
elph spectral function
Posted: Sun Mar 06, 2022 6:26 pm
by sitangshu
Hi!
While testing with a BN monolayer geometry (k-sampling 4 4 1), I computed the elph fan+dw self energies. Although not fully converged, the quasi-particle corrections seems OK at the direct point (k-point: 3). The problem is with the spectral function at the same point.
The o-sf* file shows: Spectral function at this point integrates to: 0.7199301E-152 . This results in Im|G| to be of some very low value...
Am i doing any error in the input script? I am attaching the files here...
Thanks.
Sitangshu
Re: elph spectral function
Posted: Mon Mar 07, 2022 9:39 am
by sitangshu
Another thing: Is the run command ypp_ph -g d implemented in 5.0.4? I checked after generating the matdyn modes on a 6 6 1 grid, but the double-grid options did not come...
Regards
Re: elph spectral function
Posted: Tue Mar 08, 2022 10:44 am
by claudio
Dear Sitangshu Bhattacharya
we are releasing a new version of Yambo, within a couple of weeks, that will includes ypp -g d
and do not have any problem with spectral functions, I just tested it again to be sure.
it will be online very soon, before the Yambo school
with best regards
Claudio Attaccalite
Re: elph spectral function
Posted: Thu Mar 10, 2022 5:02 am
by sitangshu
Dear Professor Claudio,
Many thanks for your reply. Please also include the followings in the next release:
% INTERP_Grid
-1 |-1 |-1 | # Interpolation BZ Grid
%
#PrtDOS # Print Exciton Density of States
in the ypp -e i. It seems these two are also not there....
Regards,
Sitangshu
Re: elph spectral function
Posted: Thu Mar 10, 2022 8:48 am
by claudio
Thank you
I will fix it too
Claudio
Re: elph spectral function
Posted: Wed Jun 15, 2022 9:01 am
by sitangshu
Hello all,
I found that the ypp interpolator (ypp -e i) is showing the following error while computing the exciton dispersion. The BSE for all q points (BSEQptR) ran perfectly fine. This error is generated only when I change the value of INTERP_Grid to say 36|36|1 and add printdos, otherwise if I leave it like -1 |-1| -1| and comment printdos there generates no error.
Any suggestions?
<---> [INTERPOLATION] Number of shells: 422
<---> Interpolation@excitons (star) | | [000%] --(E) --(X)
<---> Interpolation@excitons (star) |########################################| [100%] --(E) --(X)
<---> Interpolation@excitons (H fill) | | [000%] --(E) --(X)
<---> Interpolation@excitons (H fill) |########################################| [100%] --(E) --(X)
<---> Interpolation@excitons (computing engre) | | [000%] --(E) --(X)
<---> Interpolation@excitons (computing engre) |########################################| [100%] --(E) --(X)
<---> Interpolate@excitons Fourier | | [000%] --(E) --(X)
<---> Interpolate@excitons Fourier |########################################| [100%] --(E) --(X)
<---> [INTERPOLATION@excitons] Errors: Max, Avg, Avg_rel = 1.84768867E-13 3.27612206E-14 1.47285198E-13
<---> Number of Q/K-points in the circuit : 238
<---> Exciton bands
<---> Interpolate@excitons Fourier | | [000%] --(E) --(X)
<---> Interpolate@excitons Fourier |########################################| [100%] --(E) --(X)
<---> Exciton Grid
<---> BZ -> IBZ reduction | | [000%] --(E) --(X)
<---> BZ -> IBZ reduction |########################################| [100%] --(E) --(X)
<---> [GRID_INTERP] Points (IBZ) : 127
<---> (BZ) : 1296
<---> [ERROR] Allocation of BSE_interp_E failed with code 5014
[ERROR] STOP signal received while in[07] Excitons Interpolation
[ERROR]Attempt to allocate an allocated object
Re: elph spectral function
Posted: Thu Jun 16, 2022 8:32 am
by claudio
Dear Sitangshu
I tried to reproduce your error with the last version of the GPL
https://github.com/yambo-code
but I did not get any error.
Please try again with the last version, and in any case use the same version
of Yambo for the BSE and the interpolation
best
Claudio
Re: elph spectral function
Posted: Mon Jul 18, 2022 3:13 am
by sitangshu
Hello Professor Attacallite,
Hope you are doing great!
I am bit confused with the version you are mentioning. Is this the one (
https://github.com/yambo-code/yambo/tree/5.1) ?
# Version 5.1.0 Revision 21419 Hash (prev commit) a5ad0d553
# Branch is
# MPI+SLEPC+HDF5_MPI_IO Build
#
http://www.yambo-code.org
I am still getting the same situation ...
Regards,
Sitangshu
Re: elph spectral function
Posted: Sun Jul 24, 2022 2:53 pm
by claudio
Dear Sitangshu
I tried the last github version
https://github.com/yambo-code/yambo
but also the 5.1 should be fine.
https://github.com/yambo-code/yambo/tree/5.1
let me know
best
Claudio