I'm trying for the first time the SO coupling capabilities of Abinit and stumble upon some issue.
1) I studied a perovskite material with norm-conserving GGA (fhi) psps. I would like to treat the Pb with SO coupling. Is it mandatory to use HGH psp? It would make the calculations a lot heavier.
I tried with the GGA psps to see:
Code: Select all
nspinor 2 # Spin-polarized (collinear) calculation.
nsppol 1 # Calculation with spin up and sping down that cannot be disantengled.
nspden 4 # Vector magnztization.
so_psp 1 1 1 1 1 # Pb 4*other psp 1==> automatic treatment
paral_kgb 0 # To be sure ...
autoparal 0 # To be on the safe side.
The scf cycle seems to work:
Code: Select all
ETOT 1 -128.77044980075 -1.288E+02 1.106E-02 1.880E+03
ETOT 2 -128.95234512014 -1.819E-01 4.890E-06 2.658E+02
ETOT 3 -128.96664200898 -1.430E-02 7.374E-04 5.352E+01
ETOT 4 -128.97321356721 -6.572E-03 3.226E-04 1.292E+01
ETOT 5 -128.97433018342 -1.117E-03 5.321E-05 7.140E-01
ETOT 6 -128.97441044789 -8.026E-05 4.097E-06 6.116E-02
ETOT 7 -128.97441768406 -7.236E-06 2.032E-07 2.482E-03
ETOT 8 -128.97441776467 -8.061E-08 2.187E-09 2.746E-04
ETOT 9 -128.97441777455 -9.875E-09 3.611E-10 5.096E-05
ETOT 10 -128.97441777734 -2.788E-09 1.140E-10 1.632E-05
ETOT 11 -128.97441777863 -1.295E-09 1.714E-11 4.851E-07
ETOT 12 -128.97441777864 -1.089E-11 4.624E-13 3.662E-08
ETOT 13 -128.97441777864 -1.961E-12 1.106E-13 6.907E-09
ETOT 14 -128.97441777865 -1.506E-12 2.444E-14 8.435E-10
ETOT 15 -128.97441777864 1.677E-12 1.120E-15 1.837E-10
ETOT 16 -128.97441777864 -5.969E-13 1.996E-16 5.538E-11
ETOT 17 -128.97441777865 -9.095E-13 6.924E-17 1.514E-11
ETOT 18 -128.97441777864 2.274E-13 1.840E-17 4.168E-12
ETOT 19 -128.97441777864 7.674E-13 3.535E-18 1.126E-12
ETOT 20 -128.97441777864 -2.842E-14 8.201E-19 3.397E-13
ETOT 21 -128.97441777865 -1.165E-12 4.452E-19 1.016E-13
ETOT 22 -128.97441777865 1.990E-13 9.470E-20 2.249E-14
ETOT 23 -128.97441777864 3.979E-13 2.586E-20 6.971E-15
ETOT 24 -128.97441777865 -3.979E-13 9.846E-21 1.581E-15
ETOT 25 -128.97441777865 -9.663E-13 2.157E-21 3.655E-16
ETOT 26 -128.97441777865 -1.137E-13 5.759E-22 1.001E-16
ETOT 27 -128.97441777865 2.558E-13 1.874E-22 2.022E-17And there is twice as much eigenenergies (as it should).
The issue is that I get a crash (abinit version 7.6.1) while writing the _DEN file:
Code: Select all
Cartesian components of stress tensor (hartree/bohr^3)
sigma(1 1)= -2.13553508E-07 sigma(3 2)= 7.26873239E-08
sigma(2 2)= -4.17058858E-07 sigma(3 1)= 1.38768846E-08
sigma(3 3)= -4.89733734E-07 sigma(2 1)= -5.54571747E-08
1
2
3
4
5
6
7
8
9
10
11
12
ioarr: writing density data
ioarr: file name is bandstructureo_DS1_DEN
forrtl: Remote I/O error
forrtl: severe (30): open failure, unit 9, file /mnt/fhgfs/sponce/perovskite/bandstructure_SO/bandstructureo_DS1_DENIs there a reason for that? Is it due to the fact that I'm not using a HGH psp?
Thank you!
Samuel Poncé.