Electronic and magnetic properties of a hexanuclear ferric wheel

TL;DR

This study investigates the electronic and magnetic properties of a hexanuclear ferric wheel using advanced computational methods, revealing the importance of functional choice in accurately predicting magnetic exchange interactions.

Contribution

It compares different computational approaches, showing that hybrid functionals like B3LYP best match experimental magnetic exchange data, unlike Hartree-Fock or LDA/PBE methods.

Findings

B3LYP provides the best agreement with experimental exchange coupling.

Hartree-Fock incorrectly predicts the sign of exchange coupling.

LDA and PBE overestimate exchange interactions due to orbital delocalization.

Abstract

The electronic and magnetic properties of the hexanuclear ferric wheel [LiFe6(OCH3)12-(dbm)6]PF6 have been studied with all-electron Hartree-Fock and full-potential density functional calculations. The best agreement for the magnetic exchange coupling is at the level of the B3LYP hybrid functional. Surprisingly, the Hartree-Fock approximation gives the wrong sign for the exchange coupling. The local density approximation and the gradient corrected functional PBE strongly overestimate the exchange coupling due to the too large delocalization of the d-orbitals. These findings are supported by results from the Mulliken population analysis for the magnetic moments and the charge on the individual atoms.