[Pw_forum] U of vanadium

[Peng Chen]

Dear All,

I did LDA+U calculation on Ni3V2O8, and got the U of V (vanadium) -1.7 eV.
I am not sure if it is reasonable.  I checked the occupation number of
vanadium. All d orbitals are filled almost evenly. Whereas, I expected to
see only 3 occupied orbitals. I tried to set inital occupation number
by starting_ns_eigenvalue. But the results are the same after
iterations. Dr. Kulik mentioned that "if 3d and 4s are comparably filled
and their energies are close, then 3d and 4s shells should be included
directly in the U calculation". So I am wondering if this is the case for
vanadium and  can QE calculate U with both 3d and 4s shells? Thank you in
advance for any help you can provide!

  spin  1
    eigenvalues:
  0.441  0.448  0.480  0.505  0.513
    eigenvectors:
  0.813  0.183  0.000  0.004  0.000
  0.012  0.018  0.028  0.211  0.731
  0.001  0.001  0.576  0.248  0.174
  0.001  0.001  0.394  0.533  0.071
  0.174  0.797  0.001  0.004  0.024
    occupations:
  0.442  0.003  0.002 -0.003 -0.002
  0.003  0.509 -0.006 -0.001  0.011
  0.002 -0.006  0.492 -0.013 -0.002
 -0.003 -0.001 -0.013  0.496  0.001
 -0.002  0.011 -0.002  0.001  0.448



-- 
  Best Regards.
        Peng
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20120924/02a7df81/attachment.html>