Note on the band theory of magnetism

Abstract

The conclusions given by Matt is in a recent study of an exactly solvable, one-dimensional model are criticized. A correct Hartree-Fock treatment of the model leads to an interaction independent spin susceptibility, in agreement with the exact result. Therefore here, at least, Hartree-Fock theory is vindicated rather than vitiated; and the paradox concerning correlations, posed by Mattis, is spurious. Finally, the model is shown to be not pertinent to the band theory of magnetism.