Problem with Sm

Message

mattis · #1 Post by **mattis** » Tue Jun 19, 2007 1:58 pm

Hi
I have a problem. I try to calculate both CeO2 and SmO2, both bulk, using PBE. With the same INCAR, POSCAR, and KPOINTS files CeO2 converges nice and rapid, but for SmO2 everything goes wrong. I am kicked out in the first electronic loop and gets the errors: ZPOTRF failed, and error ZHEGV

the INCAR looks like:
System = Sm02 fluorite

ISTART = 0 ! start from scratch

#Ionic relaxation
NSW = 0 ! maximum number of ionic steps
#ISIF = 2 ! relax ions only
#IBRION = 2 ! C-G algorithm
#EDIFFG = -1E-2 ! forces less than 0.01eV/A

#electronic relaxation

NELM = 40 ! max nuber electronic steps
NELMIN = 8 ! min number electronic steps
IALGO = 38 ! Blocked Davidsson
ENCUT = 600 ! ev cut off

Anyone knows what I am doing wrong?

Best

M.

#2 Post by **admin** » Wed Jun 20, 2007 10:11 am

Please check if it works if you use the Sm_3 PP (the Sm f-electrons frozen in the core).

mattis · #3 Post by **mattis** » Wed Jun 20, 2007 11:33 am

PBE Sm_3 worked fine for SmO2!

But why is it that for example Ce and Pr works with +4 potentials in REO2 while Sm does not?
In case of a +4 potential all of them should adopt a +3 valence anyway right?

Thanks.

M.