Ising-Kondo lattice with transverse field: A possiblef-moment Hamiltonian forURu2Si2

Abstract

We study the phase diagram of the Ising-Kondo lattice with transverse magnetic field as a possible model for the weak-moment heavy-fermion compound URu 2 Si 2 , in terms of two low-lying f singlets in which the uranium moment is coupled by on-site exchange to the conduction electron spins. In the mean-field approximation for an extended range of parameters, we show that the conduction electron magnetization responds logarithmically to f -moment formation, that the ordered moment in the antiferromagnetic state is anomalously small, and that the Néel temperature is of the order observed. The model gives a qualitatively correct temperature-dependence, but not magnitude, of the specific heat. The majority of the specific heat jump at the Néel temperature arises from the formation of a spin gap in the conduction electron spectrum. We also discuss the single-impurity version of the model and speculate on ways to increase the specific heat coefficient. In the limits of small bandwidth and of small Ising-Kondo coupling, we find that the model corresponds to anisotropic Heisenberg and Hubbard models respectively. PACS numbers: 71.10. Fd, 71.27.+a, 71.70.Ch, 75.30.Mb Γ t1 1 higher singlets doublets and t2 J J J -+ z Γ FIG. 5. Spectrum of the crystal-field Hamiltonian, showing states related to the low-lying singlets by application of the operator J.

Electronic structure theory of the hidden-order materialURu2Si2

“…27 Other theories on the basis of CEF considerations for a U 4+ 5f 2 ionic state have also been developed. 16,49 Electric quadrupolar ordering should be detectable with RXS, 145 yet experimental studies gave a null result. 151,152 Quadrupolar order requires an E 1 -E 1 scattering channel to be detected, whereas higher order multipoles require at least an E 2 optical transition.…”

“…Single-ion theories of the HO, such as, e.g., quadrupolar or octupolar ordering, are based on the assumption of localized 5f's. 14,16,19,21,23,26,28,[49][50][51][52] This important issue of the degree of 5f localization has been controversially discussed recently. Several theories adopt the picture of localized 5f states from the outset, however, an examination of the grounds for this is needed.…”

Ferromagnetic instability at the edge of intermediate valence

“…They can be classified into two groups depending on whether the order parameter (OP) is (A) a magnetic dipole m [12,13,14,15,16,17] or (B) some other degree of freedom ψ [18,19,20,21,22,23,24,25,26,27,28,29]. In the former group, the AF order is regarded as intrinsic and the models involve a mechanism for decreasing the effective g value to account for the difference between µ o and ∆S.…”

Colloquium: Hidden order, superconductivity, and magnetism: The unsolved case ofURu2Si2

“…Walker and Buyers (1995) however noted that dipole order cannot be induced as a secondary OP by quadrupolar order and that the proposal would not be compatible with the polarized neutron scattering (see also Santini (1998); Santini and Amoretti (1995)). As mentioned before, there is to date no experimental evidence of such localized CEF levels and splittings.A different CEF interpretation was proposed by Barzykin and Gor'kov (1995) who suggested that U ion could be in an intermediate valence configuration, dominantly U 4+ , but with a small admixture of half-integer spin configurations (5f 1 or 5f 3 ) being responsible for the small moment.The first suggestion of an itinerant-localized duality model was made by Sikkema et al (1996), who developed an Ising-Kondo lattice model, in which they adopting the U 4+ two singlet CEF level ground state (variant A) with additional on-site exchange coupling to the conduction electron spins, leading to Kondo screening. In this model the HO is explained as small moment antiferromagnetism, but the appearance of such a state is now regarded as being parasitic to the HO.…”