Insulator-bad metal transition in RNiO

3 · 2025

Read Insulator-bad metal transition in RNiO by 3 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-07377.txt.

Introduction

A single author reconsiders a famous phase change in a family of materials called nickelates, with formula RNiO3. These compounds switch from insulator to poor metal as conditions change, and textbooks explain the switch with the Hubbard model and density functional theory. The author argues the insulating state instead comes from charge disproportionation, where nickel sites split into electron-rich and hole-rich centers with different spins.

Building on that picture, the paper develops a leaner model that tracks only the charge degrees of freedom, using a pseudospin formalism and an effective field method. The model yields two ordered phases: a high-temperature one with a fixed charge pattern, and a low-temperature quantum one where charge and spin blur between centers. Its phase diagram reproduces the main features seen in real nickelates.

Catalog record

CREATOR3
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE39 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-07377

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Insulator-bad metal transition in RNiO 3 Abstract I. Introduction. Nickelates RNiO 3 (R = rare earth or yttrium Y) demonstrate extremely unusual electrical and magnetic properties, firstly it is a weakly first-order metal-insulator transition (MIT) observed in orthorhombic RNiO 3 (R = Lu,…, Pr) upon cooling below T_{\mbox{{\footnotesize MIT}}} in the range from 130 K for Pr to \sim 550-600 K for heavy rare earths [ 1 , 2 , 3 ] . For more than three decades, the origin of MIT in nickelates has challenged the condensed matter research community. Nickelates RNiO 3 belong to a broad class of Jahn-Teller (JT) magnets [ 4 , 5 , 6 , 7 ] , compounds based on Jahn-Teller 3d - or 4d -ions with t_{2g}^{n_{1}}e_{g}^{n_{2}} -type configurations in highly symmetric octahedral, cubic or tetrahedral environments and with an orbital E_{g} -doublet in the ground state. All the JT magnets are strongly correlated systems in the sense that they cannot be adequately described within the framework of density functional theory (DFT), or more precisely DFA (density functional approximation), as it is an ill-defined questionable "starting" approximation for describing local and non-local correlations, which practically excludes their adequate consideration (see, e.g., [ 8 ] ). Even the use of the most advanced ab initio DFT-based methods as LDA+DMFT and GGA+DMFT does not provide an adequate self-consistent description of the lattice, orbital and spin ordering for such a prototypical JT insulator as KCuF 3 [ 9 , 10 ] . Much more problematic is the application of these methods to the description of JT magnets unstable to charge transfer, such as nickelates RNiO 3 , with their characteristic metal-insulator transitions. The traditional approach to describing MIT with a sharp change in the magnitude and temperature dependence of resistance assumes the implementation of a spontaneous "order-to-order" phase transition from the high-temperature coherent Fermi liquid phase that admits a more or less adequate DFT description to a low-temperature insulating phase with a charge ordering. A typical theoretical model of MIT is the Hubbard U - t model [ 11 ] , which takes into account the single-particle kinetic energy determined by the effective transfer integrals t and local correlations determined by the effective parameter U . As a rule, the transport integrals are found within the DFT-based tight-binding approximation and U is treated as a fitting parameter for d -electrons. However, for the vast majority of JT magnets, the high-temperature metal-like (bad metal) phase is fundamentally different from the coherent Fermi liquid and is more reminiscent of "… a nondegenerate gas of small polarons interacting through very weak Coulomb forces, but which would condense in some way, at a low enough temperature" (see review paper by Mott [ 12 ] ). Obviously, such a phase cannot be adequately described within the DFT-based approach. Despite the numerous experimental findings of the non-Fermi-liquid behaviour of the high-temperature bad metallic phase in nickelates, the MIT in RNiO 3 has been considered as a canonical example of a "band-width-driven" transition [ 11 ] usually described within the effective two-band Hubbard model (see, e.g., Refs. [ 13 , 14 ] ). However, the MIT in nickelates has been considered by different researchers with varying conclusions regarding the driving force behind the transition, from different pure electronic DFT-based mechanisms [ 13 , 14 , 15 , 16 ] to an unexpected result of only the coupling between lattice modes [ 17 ] . Variety of the most recent DFT-based approaches converged to a common picture with a key role played by the electron-lattice interaction, bond and charge disproportionation [ 2 , 18 , 19 ] . However, the question of the reliability of these results remains open [ 19 ] . Many unsolved and unsoluble problems give rise to serious doubts in quantitative and even qualitative predictions made within the DFT-based approaches. Despite the optimistic statements of many theorists (see, e.g., [ 16 ] ) and even experimentalists [ 2 ] , standard DFT-based methods (LDA, LDA+U, LDA+DMFT,…) do not and cannot describe all the complex of basic properties of nickelates, although it cannot be denied that they can give, at first glance, a quite plausible semi-quantitative description of some individual characteristics of the electronic structure of these strongly correlated systems. The adequate description of the MIT and other properties of nickelates requires going beyond Hubbard model and DFT-based methods. Today, few researchers doubt that the low-temperature insulating phase of nickelates is the result of charge disproportionation (CD), but the question of the structure of the CD phase, key interactions and the effective Hamiltonian remains open. We believe that the low-energy physics of the charge-disproportionated nickelates is determined by a system of charge triplets, including a bare JT-center [NiO 6 ] 8- , spin-triplet (S=1) electron [NiO 6 ] 10- and spinless (S=0) hole [NiO 6 ] 8- centers [ 5 , 6 ] . We argue that local ( U ), nonlocal ( V ) correlations, and two-particle transfer integral ( t_{b} ) govern the charge degree of freedom in nickelates. Making use of the charge triplet model with the pseudospin formalism and effective field approximation we develop a novel minimal U - V - t_{b} -model for nickelates that provides an adequate description of the insulator-to-bad-metal phase transition as melting of the CD phase and the famous T_{\mbox{{\footnotesize MIT}}} -R phase diagram found for RNiO 3 [ 2 , 3 ] . We show the existence of two types of CD-phases, high-temperature classical CO-phase with the G-type charge ordering of electron and hole centers, and low-temperature quantum CDq-phase with charge and spin density transfer between electron and hole centers, "uncertain valence" [NiO 6 ] 9±δ , or Ni 3±δ (0 \leq\delta\leq 1) and spin value 0 \leq\langle S\rangle\leq 1. The paper is organized as follows. First, in Sec. II we describe the disproportionation phenomenon in nickelates. Then, in Secs. III–V we introduce the charge triplet model, pseudospin formalism and effective Hamiltonian with a focus on the two-particle, or bosonic transfer. Finally, in Secs. VI and VII we present a short information on the effective field model and results of calculations of T - V and T - t_{b} phase diagrams within the U - V -model (atomic limit) and extended U - V - t_{b} -model. In Sec. VIII, we summarize our main results. In Supplemental Materials we present a more detailed textbook description of the effective field theory in relation to our model and a brief note on magnetic structure. II. Anti-JT disproportionation in nickelates. All JT configurations of d -ions include one e_{g} -electron or one e_{g} -hole over stable, fully or half-filled, shells. In this sense, they are similar to the configurations of the numerous family of ions with one ns -electron over filled shells, such as the 6s -electron in Hg + , Tl 2+ , Pb 3+ , Bi 4+ . These ionic configurations are unstable with respect to the disproportionation reaction, or even nonexistent (missing oxidation states [ 20 ] ).Thus, in BaBiO 3 , instead of the nominal valence 4+, bismuth prefers the stable valence states Bi 3+ and Bi 5+ with fully filled shells. However, unlike ions with ns -electrons, for JT

Read the complete text