Dear All,<br><br>I have a question regarding spin density in<br>collinear magnetic calculation. I searched in<br>the archive for an answer to my question, but<br>could not find. The question is: Is it possible<br>to obtain the difference between spin-up and spin-dn
<br>charge nonintegral (say 0.56, 3.48) instead of integers?<br>In what circumstances, I may expect that situation?<br>Is it due to occupancy of ghost bands i.e. bands whose<br>energies Ei > Ef (Fermi energy).<br><br>Thanks a lot for your answers,
<br clear="all"><br>Indrajit <br>