Difference of Magnetism in new abinit

Running Abinit Ground state
1

Hey all, I am re-running a friend of mine’s file from around 2019 (he gave me permission to post it here). It is a spc band structure calculation. Back when he ran it, he got around 3 bohr magnetrons of magnetism. However, when I ran the exact same file. It gave me close to none. I’ve pasted the input file below.

ndtset 2

kptopt1 1

nshiftk1 1

shiftk1 0.5 0.5 0.5

ngkpt1 8 8 1

prtden1 1

toldfe1 1.0d-10

iscf2 -2

getden2 -1

kptopt2 -4

nband2 31

ndivsm2 100

kptbounds2 0 0.0 0.0 # Gamma point

0.5 0 0.0 # X point

0.5 0.5 0 # S point

0 0.5 0 #Y

0 0 0 #Gamma

tolwfr2 1.0d-12

enunit2 1

spinat 0 0 0

0 0 0

0 0 0

0 0 0

0 0 0

0 0 0

0 0 0

0 0 0

0 0 4

nsppol 2

occopt 4

getwfk -1

pp_dirpath “$ABI_PSPDIR/PAW/ATOMICDATA” # This is the path to the directory were

pseudos “N.GGA_PBE-JTH.xml,P.GGA_PBE-JTH.xml,Cr.GGA_PBE-JTH.xml”

acell 9.88145956449322E+00 9.12729744269756E+00 4.01631205296322E+0 # This is equivalent to 10.18 10.18 10.18

rprim 1.0 0.0 0.0 # In lessons 1 and 2, these primitive vectors

0.0 1.0 0.0 # (to be scaled by acell) were 1 0 0 0 1 0 0 0 1

0.0 0.0 1.0

ntypat 3 # There is only one type of atom

znucl 7 15 24 # The keyword “znucl” refers to the atomic number of the

pawovlp -1

natom 9 # There are two atoms

typat 41 42 3 # They both are of type 1, that is, Silicon.

xcart

2.45234748316333E+00 5.97584012196829E+00 2.19157965884472E+00

3.75766000190541E+00 1.40812918057154E+00 2.78620633981975E+00

6.63044117147932E+00 3.61321256514787E+00 -8.49965661913236E-01

8.62352347240942E+00 8.53656046424190E+00 1.04860152428114E+00

2.99406500406500E+00 4.02658465464344E+00 4.49904156136230E+00

5.83673994647639E+00 8.42139105894501E-01 2.57636912305955E-01

1.05572121344649E+00 8.70832800754168E+00 3.33440173591544E+00

8.42383152280269E+00 6.07128325136989E+00 -1.60754074816541E+00

4.32468308608378E+00 6.42086375122143E+00 -8.66888573273084E-01

ecut 28
pawecutdg 56 # Maximal kinetic energy cut-off, in Hartree

nstep 1000 # Maximal number of SCF cycles

diemac 12.0

2

Hi there,

Can you provide a log file to your trial?
Preemptively to narrow down a potential problem, can you try a run with a noncollinear setting with the following flags active (with a minimal k-point setting)?

 nspinor 2            
 nspden  4           
 nsppol  1

As a side note, is the structure you show in the input the correct one? Because I see you use a k-point sampling of 8x8x1 while the lattice is roughly 9x9x4 Bohr in dimensions. Was the last dimension supposed to be 4.01631205296322E+01 instead of 4.01631205296322E+0 what you actually show?

Let me know!
Bogdan