Hybridization-mediated anisotropic coupling in plutonium compounds

The magnetic behavior of a class of cerium and light actinide compounds containing moderately delocalized $f$ electrons has been explained on the basis of an anisotropic two-ion interaction that arises from the hybridization of band electrons and the $f$ electrons. This theory, first developed by Siemann and Cooper for cerium compounds using the treatment of Coqblin and Schrieffer for the hybridization, was later generalized by Thayamballi and Cooper to $f^n$ systems in the $L-S$ and $j-j$ coupling limits. We here extend the theory to the case of intermediate intraionic coupling and further include the possibility of long-period antiferromagnetic structures. In particular, we have considered the ${\mathrm{Pu}}^{3+}$($f^5$) ion in PuSb. The theory reproduces the experimentally observed magnetic behavior of PuSb quite closely, predicting a phase transition from a low-temperature ferromagnetic phase to a long-period antiferromagnetic phase at about 75 K, for a fitting to a Néel temperature of 85 K, with ordered moments close to the experimental values. However, while the modulation in the long-period antiferromagnetic phase has been experimentally observed to be longitudinal, the theory predicts a transverse modulation with moments aligned along the cube edge. We also present the $T=0\mathrm{}$ magnetic excitation spectrum in the ferromagnetic phase calculated on the basis of this theory using the random-phase approximation.