Partial Disorder and Metal–Insulator Transition in the Periodic Anderson Model on a Triangular Lattice

Abstract

The ground state of the periodic Anderson model on a triangular lattice is systematically investigated by mean-field approximation. We found that the model exhibits two different types of partially disordered states: one is at half filling and the other is at other commensurate fillings. In the latter case, the kinetic energy is lowered by forming an extensive network involving both magnetic and nonmagnetic sites, in sharp contrast to the former case in which nonmagnetic sites are rather isolated. This spatially extended nature of nonmagnetic sites yields a metallic partially disordered state by hole doping. We discuss the mechanism of the metal-insulator transition by the change in electronic structure.KEYWORDS: periodic Anderson model, geometrical frustration, triangular lattice, partial disorder, metal-insulator transitionGeometrical frustration is the conflict of competing interactions originating from the structure of the underlying lattice.1,2) The competition leads to degeneracy in low energy states, which is the source of fascinating phenomena, such as unusual orderings and exotic ground states. One of the peculiar orderings under geometrical frustration is a partial disorder (PD), which is characterized by the coexistence of magnetic order and paramagnetic or nonmagnetic sites. This offers an intriguing example of selforganization to relieve geometrical frustration.PD was first theoretically proposed in the localized Ising spin model on a triangular lattice, 3) and later experimentally observed in d-electron compounds. 4,5) In localized spin systems, PD is mainly driven by the entropic effect, and hence it is stable only at finite temperatures. On the other hand, a metallic PD was also found in some f -and d-electron compounds. [6][7][8] In stark contrast to the PD states in insulating spin systems, the metallic PD states persist down to the lowest temperature, which suggests that itinerant electrons play a key role in stabilizing the PD states. In particular, in the f -electron compounds where itinerant electrons are coupled with localized f moments, the Kondo singlet formation 9,10) is expected to be relevant to the vanishment of moments at nonmagnetic sites. In spite of extensive studies, the stabilization mechanism and the nature of these metallic PD states are not fully understood yet.In the present study, we theoretically examine the possibility of PD in itinerant electron systems by focusing on the role of coupling between itinerant and localized electrons. Similar attempts were made for an effective pseudospin model, 11,12) the Kondo necklace model, 13-15) the Kondo lattice model, 14,16) and the periodic Anderson model 17,18) on frustrated lattice structures. However, the obtained PD states have been all insulating thus far, although metallic PD states are experimentally observed. It is highly desired to establish a theory to describe metallic PD states not only for understanding the experimental results but also for paving the way to explore unusual magnetotransport properties associated with t...

Charge Order with a Noncoplanar Triple-Q Magnetic Order on a Cubic Lattice

“…Meanwhile, in the periodic Anderson model, the possibility of ferromagnetic CO was discussed in infinite and three dimensions [15,16]. CO was also studied on geometrical frustrated lattices; for example, on a triangular lattice, CO was discussed associated with partial magnetic disorder in the Kondo lattice model [17,18] and the periodic Anderson model [19,20].…”

Dynamical mean-field study of partial Kondo screening in the periodic Anderson model on the triangular lattice

“…Within the present study we focus on magnetic phases with collinear moments for simplicity even though non-collinear magnetic phases may be expected in the case of the triangular lattice due to geometrical frustration. In fact, previous Hartree-Fock (HF) calculations at and off half-filling 17,18 suggest that a "classical" non-collinear 120 • antiferromagnetic phase is realized in a certain range of the phase diagram. We expect that, by enforcing collinearity, the 120 • phase is replaced by a collinear "↑, ↑, ↓" antiferromagnetic phase which has also been found within HF theory.…”

“…We expect that, by enforcing collinearity, the 120 • phase is replaced by a collinear "↑, ↑, ↓" antiferromagnetic phase which has also been found within HF theory. 17,18 A. Phase diagram Fig.…”

Realistic many-body approaches to materials with strong nonlocal correlations

“…1a) alleviates the frustration and thus allows the remnant spins (on sites B and C) to order magnetically via the RKKY coupling. This is an interesting compromise between a local and a nonlocal correlation effect which has attracted considerable attention in the past [43,44] and which has been studied in the PAM on the level of the static mean-field (Hartree-Fock) approximation [45,46]. The dynamical meanfield study discussed here correctly includes all local fluctuations and also captures the Kondo effect.…”