[Wien] LSMO GGA+U

[Lukasz Plucinski]

Dear Wien experts,

I am trying to do my first "+U" calculation on the example of LSMO. I 
did regular GGA calculation few weeks ago with all parameters default, 
only 8 Ry cutoff and 0.05 mixing. The comparison of other published 
results is very good.

Today I tried to do GGA+U. After correcting simple mistakes (e.g. giving 
U in eV instead of Ry) I guess I am more or less ok now, however, I am 
still very far from the result of Fig. 2 from Chikamatsu et al. PRB73 
195195 (2006), actually my band structure still looks almost like 
regular GGA, similar to Kotani J. Phys.: Condens. Matter 21 (2009) 266002.

I use w2web. I am doing regular initialization but with 8 Ry cutoff, and 
0.05 mixing, then I also initialize for spin-polarized. When starting 
SCF cycle I am aksed for inorb and indm files, then I run SCF with 
"spin-polarized" and "Orbital pot LDU+U" checked. Then I do 
Tasks/Bandstructure and follow there, -orb switches are already there.

At the moment I am trying to switch "+U" of 2eV (0.15 Ry) only for d 
orbitals of Mn -- maybe this is my mistake ?

Here are me input files (case.struct file attached):

case.inorb
  1 1  0                     nmod, natorb, ipr
PRATT  1.0                    BROYD/PRATT, mixing
  1 1 2                          iatom nlorb, lorb
  0                              nsic 0..AFM, 1..SIC, 2..HFM
   0.15 0.00        U J (Ry)   Note: we recommend to use U_eff = U-J and J=0

case.indm
-9.                      Emin cutoff energy
 1                       number of atoms for which density matrix is 
calculated
 1  1  2      index of 1st atom, number of L's, L1
 0 0           r-index, (l,s)index  
 

I would appreciate some hint, so tomorrow I could continue with some 
fresh ideas :)

Regards,
Lukasz
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: LSMO_LDAU.struct
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20100621/a6eaf78f/attachment.bat>