[Pw_forum] problems with Fluor pseudopotentials

[valemar at cab.cnea.gov.ar]

Dear all
I've problems with the adsorption of a AlF3 molecule on Cu(100). The
relaxation calculation never end.

I 've doubts about if the pseudopotential of Fluorine is correct. I proved
with a simple test system like  fluorine bulk and calculate the cell
parameter.
The structure of bulk fluorine used in the calculation is a cubic Pm-3n
(223) spacial group with 8 atoms in the 2a y 6d positions.

The lattice parameter experimental data is 6,67 A but mi result is around
4,23 A. I proved with higher ecutrho and both pseudopotential present in
http://www.quantum-espresso.org/pseudo.php and the result is the same;
around 4 A.

Has anyone work with these pseudopotential for Fluorine? These is some
problem in yours calculations?

My input file is:

 &control
 prefix='fluor',
 outdir='./output'
 pseudo_dir = '/home/valeria/QE_4.3.2/espresso-4.3.2/pseudo'
 calculation = 'scf'
 tprnfor = .TRUE.
 tstress = .TRUE.
 /
 &system
 ibrav= 1, celldm(1) = 12.00, nat= 8, ntyp= 1,
 ecutwfc = 30
 ecutrho = 400
 occupations = 'smearing'
 smearing = 'mv'
 degauss = 0.02
 /
 &electrons
 conv_thr = 1.D-5
 /
&ions
 /
ATOMIC_SPECIES  F 18.9984  F.pbe-n-van_ak.UPF

ATOMIC_POSITIONS crystal
  F        0.000000000   0.000000000   0.000000000
  F        0.500000000   0.500000000   0.500000000
  F        0.250000000   0.500000000   0.000000000
  F        0.750000000   0.500000000   0.000000000
  F        0.000000000   0.250000000   0.500000000
  F        0.000000000   0.750000000   0.500000000
  F        0.500000000   0.000000000   0.250000000
  F        0.500000000   0.000000000   0.750000000

K_POINTS automatic 6 6 6 1 1 1


Thank you in advance

Valeria Martin
DAEE-Bariloche Atomic Center-CNEA