Yambo Community Forum

Dear developers,

I am trying to run GW+BSE optics calculation but I am getitng NaN results for E-E0 in GW step and after BSE calculation I am getting NaN result for dielectric fn, Here is my input and output. What could be the problem?

INPUT:

ppa # [R Xp] Plasmon Pole Approximation
gw0 # [R GW] GoWo Quasiparticle energy levels
HF_and_locXC # [R XX] Hartree-Fock Self-energy and Vxc
em1d # [R Xd] Dynamical Inverse Dielectric Matrix
EXXRLvcs= 2327 RL # [XX] Exchange RL components
VXCRLvcs= 2327 RL # [XC] XCpotential RL components
Chimod= "HARTREE" # [X] IP/Hartree/ALDA/LRC/PF/BSfxc
% BndsRnXp
1 | 48 | # [Xp] Polarization function bands
%
NGsBlkXp= 1 RL # [Xp] Response block size
% LongDrXp
1.000000 | 0.000000 | 0.000000 | # [Xp] [cc] Electric Field
%
PPAPntXp= 27.21138 eV # [Xp] PPA imaginary energy
% GbndRnge
1 | 45 | # [GW] G[W] bands range
%
GDamping= 0.10000 eV # [GW] G[W] damping
dScStep= 0.10000 eV # [GW] Energy step to evaluate Z factors
DysSolver= "n" # [GW] Dyson Equation solver ("n","s","g")
%QPkrange # # [GW] QP generalized Kpoint/Band indices
1|10|35|45|
%

-----------------------------------------------------
OUTPUT:

# GW [Newton Solver]
#
# == PPA-GN GW ==
#
# PPA imaginary pt [ev]: 27.21138
# GW SC iterations :0
# dS/dw steps : 2
# dS/dw step [ev]: 0.10000
# X G`s [used]: 1
# X G`s [disk]: 1
# X bands : 1 48
# X poles [o/o]: 100.0000
# X e/h E range [ev]:-1.000000 -1.000000
# X xc-Kernel : none
# X BZ energy Double Grid: no
# X BZ Double Grid points:0
# Sc/G bands : 1 45
# Sc/G damping [ev]: 0.10000
# Sc bands terminator : no
# Sx RL components : 2327
#
# QP @ K 1 - 10 : b 35 - 45
#
# K-point Band Eo E-Eo Sc|Eo
#
1.00000 35.00000 -2.02691 NaN NaN
1.00000 36.00000 -1.82585 NaN NaN
1.00000 37.00000 -1.33740 NaN NaN
1.00000 38.00000 -0.52741 NaN NaN
1.00000 39.00000 -0.15123 NaN NaN
1.00000 40.00000 -0.14942 NaN NaN
1.00000 41.00000 0.52830 NaN NaN
1.00000 42.00000 0.72093 NaN NaN
1.00000 43.00000 0.72755 NaN NaN
1.00000 44.00000 0.73204 NaN NaN
1.00000 45.00000 1.02920 NaN NaN
2.00000 35.00000 -1.86185 NaN NaN
2.00000 36.00000 -1.76993 NaN NaN

Dear Burak,
in order to spot the problem, can you post the report file of your GW calculation?
Please also note that:
NGsBlkXp
is a variable that needs to be converged and setting equal to 1 is not meaningful.
Best,
Daniele

Report file is attached.

Dear Burak,
the problem is in the correlation part of the self-energy, but it is not easy to understand what went wrong here.
Can you delete the ndb.pp and ndb.QP database and repeat the calculation by assigning the parallelization strategy explicitly in the input file?
X_all_q_CPU= "15.2.1.1" # [PARALLEL] CPUs for each role
X_all_q_ROLEs= "v.c.k.q" # [PARALLEL] CPUs roles (q,g,k,c,v)
SE_CPU= "30.1.1" # [PARALLEL] CPUs for each role
SE_ROLEs= "b.qp.q" # [PARALLEL] CPUs roles (q,qp,b)
Please report if the problem persists, in this case, we will need to reproduce your error in order to understand what is going wrong with your system.
Not related to your problem, please note that 48 bands are for sure not enough to converge the calculation.

Best,
Daniele

The problem persists. I am sending my QE input file. Pseudopotentails are from THEOS website (Mg.pz-n-nc.UPF, Mo.pz-n-nc.UPF, S.pz-n-nc.UPF)

Input:

&control
calculation = 'scf',
restart_mode='from_scratch',
prefix='MgMoS2',
pseudo_dir = './',
outdir='./tmp',
wf_collect=.TRUE.
/
&SYSTEM
ibrav = 0, A = 6.96300, nat = 16, ntyp = 3,
ecutwfc =100,
occupations='smearing', smearing='mv', degauss=0.02,
force_symmorphic=.true.
/
&electrons
mixing_beta = 0.7,
conv_thr = 1.0d-8,
/
&ions
/
&cell
/
ATOMIC_SPECIES
Mo 95.96000 Mo.pz-n-nc.UPF
S 32.06750 S.pz-n-nc.UPF
Mg 6.96750 Mg.pz-n-nc.UPF
ATOMIC_POSITIONS {angstrom}
Mo 1.142891345 4.311583032 2.787324966
Mo 2.324836502 1.626518592 3.150508287
Mo 4.085890120 4.498477814 3.084493884
Mo -0.618162274 1.439623810 2.853339369
S -1.002556973 3.687896197 4.344421389
S 4.470284818 2.250205428 1.593411864
S 2.505337064 3.527934066 4.608407784
S 0.962390783 2.410167558 1.329425470
S 0.742843274 0.649575048 4.672120673
S 2.724884572 5.288526576 1.265712579
S 4.305741419 0.432914016 4.235870582
S -0.838013573 5.505187608 1.701962671
Mg -2.481606891 4.498872805 0.149822993
Mg 5.949334738 1.439228819 5.788010260
Mg 0.707394535 4.515280348 -0.178369418
Mg 2.760333312 1.422821276 6.116202672
K_POINTS {automatic}
4 4 4 0 0 0
CELL_PARAMETERS {alat}
1.015500852 -0.054408086 -0.024962103
-0.508250859 0.880804401 0.005240757
-0.009227898 0.026411616 0.872490735

Dear Burak,
if you can upload a zip file containing the scf/nscf inputs and pseudopotential this will help a lot.
Thanks,
Daniele

Dear Daniele,

I tried sending tar or zip file but I am getting an error saying file is too large. How can I sen you the files?

Regards,

Dear Burak,
we need just the input files (scf and nscf) and the pseudos, it should not be that large.
Anyway, if it is too large, you can send them in different posts.
Daniele

QE inputs attached.

Mg pseudopotential attached.