When I do calculations of systems with Eu ([Xe].4f7.6s2) using the standard verion of potential in potpaw_PBE (no spin polarization, no spin-orbit coupling), the OSZICAR looks like this:
...
DAV: 56 -0.246222022248E+03 0.36040E+00 -0.15198E+01 9549 0.147E+00 0.414E+00
DAV: 57 -0.246584488452E+03 -0.36247E+00 -0.23453E+01 9416 0.190E+00 0.521E+00
DAV: 58 -0.246222272317E+03 0.36222E+00 -0.15184E+01 9541 0.147E+00 0.415E+00
DAV: 59 -0.246584466878E+03 -0.36219E+00 -0.23454E+01 9417 0.190E+00 0.521E+00
DAV: 60 -0.246222580465E+03 0.36189E+00 -0.15175E+01 9550 0.147E+00
1 F= -.24622258E+03 E0= -.24622258E+03 d E =0.000000E+00
The calculation converges with dE~0.36 which is much larger than EDIFF (=10^-4)!
Could someone tell me what could be the problem and how should I treat the f-electrons in this case?
Thanks.
Eu - convergence problem
Moderators: Global Moderator, Moderator
-
user
- Newbie
- Posts: 30
- Joined: Wed Mar 23, 2005 10:18 pm
- License Nr.: 74
Eu - convergence problem
Last edited by user on Mon Apr 17, 2006 3:04 pm, edited 1 time in total.
-
admin
- Administrator
- Posts: 2921
- Joined: Tue Aug 03, 2004 8:18 am
- License Nr.: 458
Eu - convergence problem
this depends on the system you calculate. Please check whether the f-electrons are localized and whether it may be more reasonable to use the Eu_2 PP (f-electrons in the core)
Last edited by admin on Wed Apr 26, 2006 10:05 am, edited 1 time in total.