The electronic structure of some transition metal alloys

Thermal neutron scattering experiments have provided detailed information on the distributions of magnetic moment in a number of disordered ferromagnet binary alloys. The general features of these distributions together with saturation magnetization data are discussed and compared with various simple theories. Attention is focused on dilute alloy systems. After an introduction the paper is divided into four sections, the first of which deals with alloys which tend to follow the Slater-Pauling curve. Here a simple Thomas-Fermi treatment due to Friedel suggests that magnetic moment changes, largely confined to the minor constituent (solute) sites, should occur with a sign dependent on the nature of the density of states at the Fermi level in the pure major constituent (solvent). Comparison with experiment shows qualitative agreement except in the case of Fe-based alloys containing transition element solutes from the right of Fe in the periodic table. This discrepancy is examined and an explanation put forward. The next section outlines a discussion of the electronic structure of alloys of transition elements with non-transition metal solutes. The view is taken that the electronic configuration of a solute atom is roughly similar to the configuration found in the pure non-transition metal: it follows that no partially filled d orbitals are expected at solute sites. Use of a simple Thomas-Fermi model based on this assumption indicates that some of the electric screening associated with a non-transition metal solute takes place in the surrounding transition metal slovent. Additional electrons introduced in this way into the solvent occupy mainly d states and cause a reduction in magnetic properties. This reduction together with the total loss of d-state effects from the solute sites themselves can account qualitatively for the changes observed in Ni, Pd and Fe-based alloys with non-transition elements. The fourth section deals with the transition metal alloys which show marked departures from Slater-Pauling behaviour, e.g. NiCr. An explanation for these alloys has been provided by Friedel's bound impurity state model and the mechanism suggested by Comly, Holden and Low to account for the similarity in shape of the magnetic disturbances observed in different systems. The final section discusses ferromagnetic alloys of PdFe and PdCo. The giant moments associated with the Fe and Co solutes result from a widespread polarization of the Pd solvent contiguous to the solute atoms. This polarization can be interpreted with the use of a non-local exchange-enhanced susceptibility function for the Pd host. With increasing solute content this function becomes modified to an extent dependent on the shift of d holes from one spin direction to the other, i.e. on the mean polarization of the Pd.