Hi,
I am performing some simulations for Er2O3 and found that the best agreement with experimental data are found when I use UNFREEZEF card.
With this card I find some coditions I cannot acchieve convergence, I atach the inp and output files.
At the end of the output fila I find:
Electronic configuration
   iph    il      N_el
     0     0    1.601
     0     1    1.494
     0     2   10.380
Found bad counts.
Occupation number in getorb is     1.000
Will repeat this iteration
     0     3    0.656
     1     0    0.371
     1     1    6.451
     1     2    1.822
     1     3    9.026
     2     0    1.985
     2     1    5.571
     2     2    0.000
     2     3    0.000
              Core-valence separation
WARNING: fatal error in subroutine corval. Try
to reduce ca1 in SCF card. If does not help,
SEND bug report to AUTHORS
CORVAL-1

I tried to change the ca en SCF card and no good results.
Thanks in advance, euG