Yambo Community Forum

Posted: **Thu Jun 22, 2017 2:38 pm**

dear all!

I have read this thread: viewtopic.php?f=14&t=1071&p=4870&hilit=invalid#p4870

But I still cannot perform band-structure interpolation for calculations with SOC included. Maybe someone can let me know if there is another "trick" to force YPP to obey my will...?

so basically I have run the whole thing without SOC and I get amazing band-structures thanks to the h-BN tutorial and yambo (Tool: yambo 4.1.2 rev.14036)

when SOC is switched on (using rel pp generated using ONCV), the GW calculation runs smooth but the interpolation does not work (after removal of TR), the error is

Code: Select all

 <---> [01] CPU structure, Files & I/O Directories
 <---> [02] Y(ambo) P(ost)/(re) P(rocessor)
 <---> [03] Core DB
 <---> :: Electrons             : 44.00000
 <---> :: Temperature       [ev]: 0.000000
 <---> :: Lattice factors [a.u.]: 11.27810  11.27810  11.27810
 <---> :: K points              : 120
 <---> :: Bands                 :  400
 <---> :: Symmetries            :  8
 <---> :: RL vectors            : 138745
 <---> [04] K-point grid
 <---> :: Q-points   (IBZ): 120
 <---> :: X K-points (IBZ): 120
 <---> [05] CORE Variables Setup
 <---> [05.01] Unit cells
 <---> [05.02] Symmetries
 <---> [WARNING]Spatial Inversion not found among the given symmetry list
 <---> [05.03] RL shells
 <---> [05.04] K-grid lattice
 <---> [05.05] Energies [ev] & Occupations
 <---> [06] Interpolation tool
 <---> Number of K-points in the circuit :  6
 <---> [INTERPOLATION] Number of shells:  1538
 <---> [INTERPOLATION] Errors: Max, Avg, Avg_rel =   4.02331352E-06  4.12009342E-07  1.51901331E-03
 <---> Spinor-factor |                                        | [000%] --(E) --(X)
 <20s> Spinor-factor |########################################| [100%] 19s(E) 19s(X)
 <20s> Magn-factor |                                        | [000%] --(E) --(X)
 <40s> Magn-factor |########################################| [100%] 20s(E) 20s(X)
 <40s> [SPN] Kpts covered exactly  [o/o]: 38.46154
 <40s> [Interpolate] Nighbours         : 1
 <40s> [Interpolate] Kpts covered [o/o]: 61.53846
 <40s> [SPN] Kpts covered exactly  [o/o]: 38.46154
 <40s> [Interpolate] Nighbours         : 1
 <40s> [Interpolate] Kpts covered [o/o]: 61.53846
 <40s> [MAG] Kpts covered exactly  [o/o]: 38.46154
 <40s> [Interpolate] Nighbours         : 1
 <40s> [Interpolate] Kpts covered [o/o]: 61.53846
 <40s> [MAG] Kpts covered exactly  [o/o]: 38.46154
 <40s> [Interpolate] Nighbours         : 1
 <40s> [Interpolate] Kpts covered [o/o]: 61.53846
 <40s> [MAG] Kpts covered exactly  [o/o]: 38.46154
 <40s> [Interpolate] Nighbours         : 1
 <40s> [Interpolate] Kpts covered [o/o]: 61.53846[pnel-dft:01605] *** Process received signal ***
[pnel-dft:01605] Signal: Segmentation fault (11)
[pnel-dft:01605] Signal code: Invalid permissions (2)
[pnel-dft:01605] Failing at address: 0x628b0d
[pnel-dft:01605] [ 0] /lib/x86_64-linux-gnu/libpthread.so.0(+0x11390)[0x7f60bcaf1390]
[pnel-dft:01605] [ 1] /lib/x86_64-linux-gnu/libc.so.6(+0x14dfed)[0x7f60bc863fed]
[pnel-dft:01605] [ 2] /opt/yambo-4.1.3/bin/ypp[0x501532]
[pnel-dft:01605] [ 3] /opt/yambo-4.1.3/bin/ypp[0x441b40]
[pnel-dft:01605] [ 4] /opt/yambo-4.1.3/bin/ypp[0x4394b6]
[pnel-dft:01605] [ 5] /opt/yambo-4.1.3/bin/ypp[0x40748b]
[pnel-dft:01605] [ 6] /opt/yambo-4.1.3/bin/ypp[0x404a62]
[pnel-dft:01605] [ 7] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf0)[0x7f60bc736830]
[pnel-dft:01605] [ 8] /opt/yambo-4.1.3/bin/ypp[0x404e89]
[pnel-dft:01605] *** End of error message ***
Segmentation fault

I have 44 electrons and 44 bands, I am interested in bands around the band-gap (the E0 gap is only 0.5 eV, with GW it opens to 1.xx, which is in good agreement with other papers).

Code: Select all

electrons                    # [R] Electrons (and holes)
bnds                         # [R] Bands
cooIn= "rlu"                 # Points coordinates (in) cc/rlu/iku/alat
BANDS_steps=10                # Number of divisions
% INTERPGrid
-1 |-1 |-1 |                             # Interpolation BZ Grid
%
ShellFac= 20.00000           # The bigger it is a higher number of shells is used
%QPkrange                    #  generalized Kpoint/Band indices
  1|120|  40|50|
%
%BKpts                       # K points of the bands circuit
0.00000 |0.00000 |0.00000 |
0.00000 |0.50000 |0.00000 |
0.50000 |0.50000 |0.00000 |
0.00000 |0.00000 |0.00000 |
0.50000 |0.50000 |0.50000 |
0.00000 |0.50000 |0.00000 |
%

Unit cell is cubic (as you can see above) CsPbBr3. I use a 5 atoms basis. The error does not depend on the k-circuit, even two points segfault. I made sure not to run ypp in parallel.

I compiled using internal netcdf and for reference I also tried two more versions of yambo/ypp that came with QE 5.4 and QE 6.0; exactly the same problem!

Any hint is very welcome! Thank you for the time and patience here on the board!

Yours,
Chris

Yambo Community Forum

ypp -s b: band structure in the case of SOC

ypp -s b: band structure in the case of SOC