Fe dimer

Total energy, geometry optimization, DFT+U, spin....

Moderator: bguster

Locked
mverstra
Posts: 655
Joined: Wed Aug 19, 2009 12:01 pm

Fe dimer

Post by mverstra » Sun Feb 21, 2010 1:38 pm

Hello Keitel

1) for a molecule, you need to avoid metallic preconditionning of the screening, so
diemac 1.0

2) for your dimer you are not setting spinat, which is probably the source of your slow convergence.

3) you should do the ecut convergence first, then the relaxation, and eventually check at the end that the ecut is still ok (forces stay small with a larger ecut). Doing a relaxation for each ecut is a bit wasteful.

4) the dilatmx is useless as your unit cell is fixed, as should be chkprim (your unit cell should be primitive anyway).

Matthieu

Hi people,

I am newer in Abinit issues so forgive my silly questions.
I am trying to optimize the ecut for an Fe-Fe molecule, but
it lasts too much i guess and the convergence threshold
will never reach. The SCF steps are 200, and the energy
seem to not converge.

At the end I show my .in.
Any theoretical (that is books) and technical (ticks) recommendations
are always welcome.

Thanks in advance

# iron dimer
#Define the different datasets

prtgeo 15 # more than 1
ndtset 10
udtset 5 2

chkprim 0
dilatmx 1.2

#Definition of the unit cell
acell 16 16 16 # The keyword "acell" refers to the
# lengths of the primitive vectors (in Bohr)
ecut:? 15 ecut+? 5

rprim 1 0 0 0 1 0 0 0 1 # This line, defining orthogonal primitive vectors,
# is commented, because it is precisely the default
value of rprim

#Definition of the atom types
ntypat 1 # There is only one type of atom
znucl 26 # The keyword "znucl" refers to the atomic number of the
# possible type(s) of atom. The pseudopotential(s)
# mentioned in the "files" file must correspond
# to the type(s) of atom. Here, the only type is Hydrogen.

#FIRST DATASET 2 ATOMS
natom?1 2
ionmov?1 3
ntime?1 30
tolmxf?1 5.0d-4
xcart?1 -1.9 0.0 0.0
1.9 0.0 0.0
ecutsm?1 0.5
toldff?1 5.0d-5
nband?1 10


#SECOND DATASET 1 ATOM
natom?2 1
nsppol?2 2 #spin polarized calculation
nband?2 10
toldfe?2 1.0d-6
xcart?2 0.0 0.0 0.0
spinat?2 0.0 0.0 1.0
ecutsm?2 0.5

typat 1 1


#Definition of the k-point grid
kptopt 0 # Enter the k points manually
nkpt 1 # Only one k point is needed for isolated system,
# taken by default to be 0.0 0.0 0.0

#Definition of the SCF procedure
nstep 200 # Maximal number of SCF cycles

diemac 1.0d+6 # Although this is not mandatory, it is worth to
# precondition the SCF cycle. The model dielectric
# function used as the standard preconditioner
# is described in the "dielng" input variable section.
# Here, we follow the prescriptions for molecules
# in a big box

ixc 8
Matthieu Verstraete
University of Liege, Belgium

Locked