Spin-polarized calculations
Posted: Fri Jul 23, 2010 2:51 am
hi,
I am running calculations for a metallic system with an odd number of electrons. Both spin-polarized and unpolarized calculations give the same total energies, with the magnetic moment of the system equal to zero. Since the system I am considering is a metal, it does not matter where this "extra" electron goes and the occupancies will be partial, I suppose. Now, if it were a semiconductor or an insulator, would you expect the spin-polarized and unpolarized calculations to give different answers and a magnetic moment greater than zero? Is it always the case, and if not (i.e. both spin polarized and unpolarized calculations give the same total energies), where would this extra electron go? What does it mean? Also, what does it mean to have an occupation number of, let's say, 1.923 (in OUTCAR) for a given kpoint and a band number? I thought it should be always 1 or 2.
I appreciate your reply.
I am running calculations for a metallic system with an odd number of electrons. Both spin-polarized and unpolarized calculations give the same total energies, with the magnetic moment of the system equal to zero. Since the system I am considering is a metal, it does not matter where this "extra" electron goes and the occupancies will be partial, I suppose. Now, if it were a semiconductor or an insulator, would you expect the spin-polarized and unpolarized calculations to give different answers and a magnetic moment greater than zero? Is it always the case, and if not (i.e. both spin polarized and unpolarized calculations give the same total energies), where would this extra electron go? What does it mean? Also, what does it mean to have an occupation number of, let's say, 1.923 (in OUTCAR) for a given kpoint and a band number? I thought it should be always 1 or 2.
I appreciate your reply.