[Error] Search for G=G1-G2

[Sheleon]

Hi,everyone,
Recently, when i try to calculate the GW correction for my system, i got "[Error] Search for G=G1-G2" error for COHSEX calculation, and " [Error] STOP signal received while in : [04] Bare local and non-local Exchange-Correlation.[Error][NetCDF] Not a directory". Pleas help me out.
This is my abinit input file:

Code: Select all

# Crystalline KBBF
# Generation of the KSS file

# Number of datasets: 2-step calculation
ndtset 2

# Step 1: determine the ground state
iscf1    3                    # Algorithm: Anderson mixing of the potential
nband1   55                    # Number of bands
tolwfr1  1.0d-16              # Tolerance on wavefunctions to stop SCF cycles
prtden1  1                    # Store electronic density in a file

# Step 2: obtain the Kohn-Sham band structure
getden2    1                  # Read electronic density from step 1
iscf2     -2                  # Non self-consistent calculation
tolwfr2    1.0d-16            # Tolerance on wavefunctions
nband2    205                  # Number of bands for the system
nbandkss2 200                 # Number of bands to output to the KSS file
kssform2   3                  # Format of the KSS file
symmorphi2 0                  # Disallow non-symmorphic operations (required)
istwfk2    *1               # Wavefunction storage mode for each k-point#Definition des atomes
#Definition of lattice parameters
#********************************
            acell      1.0642642742E+01  1.0642642742E+01  1.0642642742E+01 Bohr
            rprim      4.4648943606E-01  0.0000000000E+00  8.9478890443E-01
                      -2.2324471803E-01  3.8667119415E-01  8.9478890443E-01
                      -2.2324471803E-01 -3.8667119415E-01  8.9478890443E-01
 #Definition atomic positions and types
#*************************************
            natom           9
           ntypat           5
            typat      1  2  2  3  3  3  4  5  5
             xred      0.0000000000000   0.000000000000    0.00000000000000
                       6.9823682468E-01  6.9823682468E-01  6.9823682468E-01
                       3.0176317532E-01  3.0176317532E-01  3.0176317532E-01
                       1.8740623445E-01  8.1259376555E-01  5.0000000000E-01
                       8.1259376555E-01  5.0000000000E-01  1.8740623445E-01
                       5.0000000000E-01  1.8740623445E-01  8.1259376555E-01
                       5.0000000000E-01  5.0000000000E-01  5.0000000000E-01
                       7.9528661468E-01  7.9528661468E-01  7.9528661468E-01
                       2.0471338532E-01  2.0471338532E-01  2.0471338532E-01
            znucl       19.00000    9.00000    8.00000    5.00000    4.00000
#Parameters of the SCF cycles
#****************************
   iscf        7
   nstep       200

#Pland wave basis and k-point grid
#*********************************
   ecut        45
   ecutsm      0.5
   ngkpt       6 6 6 
   nshiftk     1
   shiftk      0.0 0.0 0.0
                      
# add to conserve old < 6.7.2 behavior for calculating forces at each SCF step
 optforces 1

and this is my Cohsex input file:

Code: Select all

em1s                         # [R Xs] Static Inverse Dielectric Matrix
gw0                          # [R GW] GoWo Quasiparticle energy levels
cohsex                       # [R Xp] COlumb Hole Screened EXchange
HF_and_locXC                 # [R XX] Hartree-Fock Self-energy and Vxc
EXXRLvcs=  35 Ry    # [XX] Exchange RL components
Chimod= "Hartree"            # [X] IP/Hartree/ALDA/LRC/BSfxc
% BndsRnXs
   1 | 30 |                 # [Xs] Polarization function bands
%
NGsBlkXs=  30 Ry    # [Xs] Response block size
% LongDrXs
 1.000000 | 0.000000 | 0.000000 |        # [Xs] [cc] Electric Field
%
% GbndRnge
   1 | 30 |                 # [GW] G[W] bands range
%
%QPkrange                    # [GW] QP generalized Kpoint/Band indices
  1| 32|  1|30|
%
%QPerange                    # [GW] QP generalized Kpoint/Energy indices
  1| 32| 0.0|-1.0|
%

last is my GW0 input file:

Code: Select all

em1d                         # [R Xd] Dynamical Inverse Dielectric Matrix
gw0                          # [R GW] GoWo Quasiparticle energy levels
ppa                          # [R Xp] Plasmon Pole Approximation
HF_and_locXC                 # [R XX] Hartree-Fock Self-energy and Vxc
EXXRLvcs=  35 Ry    # [XX] Exchange RL components
Chimod= "Hartree"            # [X] IP/Hartree/ALDA/LRC/BSfxc
% BndsRnXp
   1 | 200 |                 # [Xp] Polarization function bands
%
NGsBlkXp=  35 Ry    # [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 | 200 |                 # [GW] G[W] bands range
%
GDamping=  0.10000     eV    # [GW] G[W] damping
dScStep=  0.10000      eV    # [GW] Energy step to evalute Z factors
DysSolver= "n"               # [GW] Dyson Equation solver (`n`,`s`,`g`)
%QPkrange                    # [GW] QP generalized Kpoint/Band indices
  1| 32|  15|25|
%
%QPerange                    # [GW] QP generalized Kpoint/Energy indices
  1| 32| 0.0|-1.0|
%

Thanks in advace.


Sheleon

[Daniele Varsano]

Dear Sheleon,
please post also your report files.
Anyway, looking at your inputs, do you really need 30/35 Ry in your Response Block size, it looks me an uncommon enormous value.
Best,
Daniele

[Davide Sangalli]

Dear Sheleon,
your input file results in a quite cumbersome calculation.
Can you reproduce the error in a faster calculation, i.e. lower convergence parameters both for pw and yambo, or it just shows up when you increase the parameters ?

Best regards,
Davide

[Sheleon]

Dear Davide,
When i reduced the value of the parameter to "EXXRLvcs=20Ry, NGsBlkXs=10Ry", everything goes fine, So, is it a limitation of these parameters or i just used them in a physical meaningless way?

Thanks,
Sheleon

[Daniele Varsano]

Dear Sheleon,
most probably you used a too big value, and the difference G=G1-G2, was not present in the g-vector database. In order to use such a big value may be you should calculate your ground state with a bigger cutoff. Anyway, you can check your convergence using a still lower value, I do not know what system you are dealing with, but that parameter should be over-converged.
Best,
Daniele

[Sheleon]

Thank both of you very much~

Sheleon