[Ifeffit] FEFF: problem with calculations of La and Eu XANES spectra with UNFREEZEF card

[Павел Засимов]

Dear XAFS community,

I have got a problem with FEFF9 calculations of La and Eu XANES spectra with UNFREEZEF card (The calculation don't converge, but it runs without any problems if I do not use UNFREEZEF card). This problem does not depend on the chemical form of the element (i.e. it occurs in La2O3, EuPO4,  LaTp3, EuTp3, where Tp is trispyrazolylborate) and seems to be element-specific: the same feff.inp files where La/Eu are replaced with Ce or U are processed by FEFF program without the problem. 

To note, SCF settings were the same as recommended in the FEFF9.6 manual: SCF 3.7 0 30 0.05 10.

Logfile (log1_er.dat), as well as an example of an input file (feff_er.inp), are attached to this mail.

The fragment of logfile reporting about the error is shown below:

Electronic configuration
iph il N_el
0 0 1.683
Found bad counts.
Occupation number in getorb is 4.000
Will repeat this iteration.
0 1 7.372
0 2 10.521
Found bad counts.
Occupation number in getorb is 1.000
Will repeat this iteration.
0 3 0.727
...
3 3 0.000
Core-valence separation
Bad counts found repeatedly - pot will exit now.

Did anyone notice anything similar?

Thanks for any help,
Zasimov Pavel
-------------- next part --------------
A non-text attachment was scrubbed...
Name: feff_er.inp
Type: application/octet-stream
Size: 15771 bytes
Desc: not available
URL: <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20200512/b95a8b42/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: log1_er.dat
Type: application/octet-stream
Size: 24370 bytes
Desc: not available
URL: <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20200512/b95a8b42/attachment-0003.obj>