[QE-users] NO2 and CO2.- radical anion relax calculation

[dv009200 at fh-muenster.de]

Dear all,

Although it seems like a simple system I'm unable to relax the NO2
molecule and the CO2.- radical anion to the desired and experimentally
matching bent geometries. In fact I always end up with linear geometries
and bond angles of 180° (which would make sense for the NO2+ ion and the
uncharged CO2 molecule) after the relax calculations. I suspect it has
something to do with the radical character of my system but I'm not sure
what the key factors would be here. Has anyone some ideas/hints how to
solve this issue or did I oversee something important when trying to relax
molecules with a radical character?
Below are my input files. I already start with a bent geometry, I also
tried higher cutoff energy for the wavefunction and electron density and a
dense Monkhorst-Pack k-point grid but nothing helped.

I'm sorry if in fact I overlooked something trivial.

Best Regards

Dominik

NO2:
 &control
    calculation='relax'
    restart_mode='from_scratch',
    pseudo_dir = '/home/dominik/codes/QE/pseudo/',
    outdir='/home/dominik/codes/QE/tempdir/',
    prefix='NO2'
 /
 &system
    a     = 20
    ibrav = 1
    nat   = 3
    ntyp  = 2
    ecutwfc = 60.0
    ecutrho = 720.0
    nspin=2
    tot_magnetization=1
    tot_charge=0
 /
 &electrons
    diagonalization='david'
    mixing_beta = 0.7
 /
 &ions
 /
ATOMIC_SPECIES
N    14.01  N.pbe-n-rrkjus_psl.0.1.UPF
O    15.99  O.pbe-n-rrkjus_psl.0.1.UPF
ATOMIC_POSITIONS {angstrom}
N        0.9765100663      1.1315577009      1.0181266829
O        2.2090488872      0.6751240038      1.0305014554
O        0.7401285527      2.2218712848      1.4396405521
K_POINTS  gamma

CO2.-:
 &control
    calculation='relax'
    restart_mode='from_scratch',
    pseudo_dir = '/home/dominik/codes/QE/pseudo/',
    outdir='/home/dominik/codes/QE/tempdir/',
    prefix='CO2.-'
 /
 &system
    a     = 20.0
    ibrav = 1
    nat   = 3
    ntyp  = 2
    ecutwfc = 60.0
    ecutrho = 720.0
    nspin=2
    tot_magnetization=1
    tot_charge=-1
 /
 &electrons
    diagonalization='david'
    mixing_beta = 0.7
 /
 &ions
 /
ATOMIC_SPECIES
C    12.01  C.pbe-n-rrkjus_psl.0.1.UPF
O    15.99  O.pbe-n-rrkjus_psl.0.1.UPF
ATOMIC_POSITIONS {angstrom}
C        0.9765100663      1.1315577009      1.0181266829
O        2.2090488872      0.6751240038      1.0305014554
O        0.7401285527      2.2218712848      1.4396405521
K_POINTS  gamma


Dominik Voigt M.Sc.
PhD Student
University of applied sciences Münster
Email: dv009200 at fh-muenster.de