Current Control of Structural and Physical Properties in Spin-Orbit- Coupled Mott Insulators

Electrical current as a means to control structural and related physical properties has been recognized only recently. The application of small electrical currents in sensitive detector and control applications, and in information technologies, is often preferable to other external stimuli. However, until recently it has not been widely accepted that electrical current can readily couple to the lattice, orbital, and spin degrees of freedom. Mounting experimental evidence has indicated that a combination of strong spin-orbit interactions and a distorted crystal structure in magnetic Mott insulators may be sufficient for electrical current to control structural and related properties. Current control of quantum states in 4d- and 5d-transition metal oxides has therefore rapidly expanded as a key research topic. This chapter presents two model systems, Ca2RuO4 and Sr2IrO4, in which applied current effectively controls the lattice, and thus the physical properties.

Lattice-Driven Ruthenates

Ruthenates have extended 4d-electron orbitals and comparable Coulomb, crystalline electric field, and spin-orbit interactions, as well as significant p-d orbital hybridization and spin-lattice coupling. The physical properties of ruthenates are highly susceptible to even slight lattice distortions; as a result, external magnetic field, pressure, electrical current, and chemical doping can generate disproportionate responses in structural as well as other physical properties, which can lead to unusual ground states or phenomena. Examples of the unusual, strong coupling of the ruthenates to external stimuli include negative volume thermal expansion via orbital and magnetic order in doped Ca2RuO4, colossal magnetoresistivity via avoiding a spin-polarized state and quantum oscillations in Ca3Ru2O7, and pressure-induced transition from ferromagnetism to antiferromagnetism in Sr4Ru3O10.

Single-Crystal Synthesis

Growing single crystals of 4d- and 5d-transition metal oxides is often difficult, as they tend to form incongruently, as well as having high vapor pressure and high melting points. Two crystal growth techniques are commonly used for transition metal oxides—flux and floating-zone techniques; each has advantages and disadvantages. An established capability in both techniques makes it possible to grow single crystals of almost all stable materials. Some basic aspects of both techniques are discussed, and a few general remarks on crystal growth of 4d- and 5d-transition metal oxides are presented. Crystal structures of most 4d- and 5d-transition metal oxides are inherently distorted. An innovative “field-altering” technique is under development, in which an applied magnetic field aligns magnetic moments and, through strong spin-orbit interactions and magnetoelastic coupling, alters crystal structures at high temperatures. Preliminary results show that a field-altering technology is highly effective for resolving physical properties of spin-orbit-coupled oxides.

Introduction

The fundamental and technological importance of transition metal oxides, and the relationship of the present work to previous monographs dealing with transition metal oxides are reviewed. The relatively abundant 3d-transition metal oxides are contrasted with the rarer 4d- and 5d-transition metal oxides that exhibit a unique interplay between spin-orbit, exchange, crystalline electric field and Coulomb correlations. The combined effect of these fundamental interactions yields peculiar quantum states and empirical trends that markedly differ from those of their 3d counterparts. General trends in the electronic structure are related to generalized phase diagrams of the magnetic and insulating ground states. The intriguing absence of experimental evidence for predicted topological states and superconductivity in these materials are discussed.

Magnetic Frustration

Spins often prefer to anti-align with their neighbors in antiferromagnetic correlation. Materials with triangle lattices exhibit energetic degeneracy among the possible rearrangements of anti-aligned spins, which is denoted geometric frustration that is associated with strongly depressed transitions to magnetic order. Honeycomb iridates and ruthenates, pyrochlore systems, and double-perovskite iridates all feature triangular lattices as primary building blocks of their structures. Another frustration mechanism evolves from the Kitaev’s exact solution of a spin-liquid model on a honeycomb lattice with strong spin-orbit interactions. The protracted search for a Kitaev spin liquid has recently focused on the honeycomb itidates Na2IrO3 and Li2IrO3. A newer kind of quantum liquid has been identified in the magnetic insulator Ba4Ir3O10, where Ir3O12 trimers form an unfrustrated square lattice.

Spin-Orbit Interactions in Ruddlesden-Popper Phases Srn+1IrnO3n+1 (n = 1, 2, and ∞)

The Ruddlesden-Popper phases Srn+1IrnO3n+1 (n = 1, 2, and ∞) have been intensively studied, and exhibit many novel behaviors and ground states driven by a rare interplay between strong spin-orbit and Coulomb interactions. One key empirical trend is that most iridates are antiferromagnetic insulators, contrary to conventional wisdom. The spin-orbit-coupled Mott state does not always closely track the magnetic state in iridates. Often, chemical doping can effectively induce a metallic state. Defying expectations, Sr2IrO4, which is the prototypical spin-orbit-coupled Mott insulator, does not become superconducting upon electron doping, but remains insulating under applied pressures extending into the Mbar range, highlighting the extraordinary susceptibility to the lattice degrees of freedom, which is at the heart of the physics driving the iridates.