Electronic States of Single-Component Molecular Conductors [M(tmdt)2]

TL;DR

This paper theoretically investigates the electronic states of single-component molecular conductors [M(tmdt)2] (M= Ni, Au, Cu), revealing how orbital interactions and Coulomb effects influence their magnetic and electronic properties.

Contribution

It develops a multiorbital model based on first-principles calculations to explain the electronic and magnetic behaviors of [M(tmdt)2] compounds, including new insights into their orbital interactions.

Findings

Orbital energy differences control electronic states near the Fermi level.

Coulomb interactions lead to various magnetic structures in the Hubbard model.

Explanation of the dual state in Cu(tmdt)2 and scenarios for antiferromagnetic transition in Au(tmdt)2.

Abstract

The electronic states of isostructural single-component molecular conductors [M(tmdt)2] (M= Ni, Au, and Cu) are theoretically studied. By considering fragments of molecular orbitals as basis functions, we construct a multiorbital model common for the three materials. The tight-binding parameters are estimated from results of first-principles band calculations, leading to a systematic view of their electronic structures. We find that the interplay between a p\pi-type orbital (L) on each of the two ligands and a pd\sigma-type orbital (M\sigma) centered on the metal site plays a crucial role: their energy difference controls the electronic states near the Fermi energy. For the magnetic materials (M= Au and Cu), we take into account Coulomb interactions on different orbitals, i.e., we consider the multiorbital Hubbard model. Its ground-state properties are calculated within mean-field approximation where various types of magnetic structures with different orbital natures are found. An explanation for the experimental results in [Cu(tmdt)2] is provided: The quasi-degeneracy of the two types of orbitals leads to a dual state where localized M$\sigma$ spins appear, and L sites show a nonmagnetic state owing to dimerization. On the other hand, [Au(tmdt)2] locates in the subtle region in terms of the degree of orbital mixing. We propose possible scenarios for its puzzling antiferromagnetic phase transition, involving the M\sigma orbital in contrast to previous discussions mostly concentrating on the L sector.