Abstract
Previous neutron experiments on polycrystalline samples of have led to a series of conflicting proposals, including a spin-glass state, for the magnetic structure below the ordering temperature of ∼150 K. Our experiments on a stoichiometric single crystal show that the principal interaction is a commensurate antiferromagnetic ordering of the Fe sublattice, with a moment at 4.2 K of 1.08(2) per Fe atom in the basal plane of the tetragonal structure. However, symmetry arguments suggest that the Fe sublattice has a weak ferromagnetic component also in the basal plane. Experiments in a magnetic field with polarized neutrons establish that the ferromagnetic U moment is 0.47(2) per U atom. In a magnetic field applied in the [010] direction (basal plane) the Fe sublattice antiferromagnetism is aligned perpendicular to the field in the basal plane, i.e., in the direction [100]. Combining the neutron and magnetization results shows that the weak Fe ferromagnetic component in zero field is ∼0.3 so that the canting angle of the Fe moments is 16°. Relatively small fields are sufficient to cause a further canting of the Fe moments towards the field direction; for example at 4.6 T this canting is 25°. Polarized-neutron experiments in the paramagnetic state show that the Fe susceptibility is almost isotropic; however, the response of the U 5f electrons is much smaller along the c axis, so that it is the hybridization between the Fe 3d and U 5f electrons that gives rise to the measured anisotropy in this material. The weak ferromagnetism of the Fe sublattice, which may be a consequence of the interaction between the U 5f and Fe 3d electrons assures that the two sublattices develop their moments in the same manner. This is different from the situation in the (R=rare earth) compounds, in which the rare earth develops its full moment only at a lower temperature.
DOI:https://doi.org/10.1103/PhysRevB.55.14370
©1997 American Physical Society