Total magnetization as printed in the code is equivalent to nelup-neldw. You cannot get an odd total magnetization if you have an even number of electrons.
The total magnetization is only the same as the magnetization on cobalt if cobalt is the only atom in your simulation cell with a magnetic moment. If, as you have indicated, you have created a surface with an odd number of electrons, it must also have a net spin or, alternatively, charge transfer may occur between Co and the surface so that the surface has no net spin. Either way, mulliken population analysis, projections onto a localized basis, or some alternative approach is required to approximately identify the magnetic moment on each atom. If you provide the reference you're comparing to, it might be clearer what you are trying to compare against. Hope that helps, Heather Kulik Stanford University On Monday, August 8, 2011 at 4:16 PM, Izaak Williamson wrote: > Thank you Heather and Jia for your replies. For a single Co atom, I do get > the total magnetization of 3 but when Co is on a surface, its spin > polarization comes out to be 2, which is S=1. The all-electron calculations > were also done using pbe functional and on the same surface and with the same > coverage of Co. > > Heather, what do you mean by locally? I do calculate the spin polarization > from pdos. The whole system does have an even number of electrons but why > does it result in S=1? Any explanation would be appreciated. > > -- > Izaak Williamson > Research Assistant > Physics Department > Boise State University > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20110808/a3ebde1e/attachment.htm
