After the discovery of the cuprate superconductors1 the elctronic structure of 3d-transition metals (TM) compounds became a focal point in solid state physics. Particularly, long standing questions regarding the physics of correlation effects are, nowadays, one of the more important issues to be addressed in order to understand the electronic mechanisms that are responsable for the superconducting phenomena2. An appropriate description of the electronic structure of late (Mn-Cu) 3d TM compounds does not exist and different approaches are used to interpret particular behavior of these compouds. The one electron-band picture, which ignores the correlation effects by using an average potential, gives results in contradiction with the experiments. Discrepancy, reported first by Boer and Verwey3, was partially resolved by Mott4 and Hubbard5. They pointed out that if the exchange and Coulomb energies (U), involved in charge fluctuations between d-orbitals of TM ions in different sites, are larger than the one-electron dispersion with band width (w), the one-electron description of solids is expected to break down.