Antiferromagnetic Kitaev Interaction in $f$-Electron Based Honeycomb Magnets

TL;DR

This paper predicts that certain $f$-electron honeycomb magnets can host antiferromagnetic Kitaev interactions, offering new possibilities for realizing Kitaev spin liquids beyond known $d$-electron materials.

Contribution

It introduces a theoretical framework showing that $A_2$PrO$_3$ compounds with $f$-electrons exhibit dominant antiferromagnetic Kitaev interactions, contrasting with previously known ferromagnetic cases.

Findings

$A_2$PrO$_3$ compounds have honeycomb layers with Pr$^{4+}$ ions.

Effective interactions are predominantly antiferromagnetic Kitaev type for light alkali metals.

This provides a new platform for exploring Kitaev spin liquids.

Abstract

We theoretically propose a family of $f$-electron based magnets that realizes Kitaev-type bond-dependent anisotropic interactions. Based on {\it ab initio} calculations, we show that $A_2$PrO$_3$ ($A$: alkali metals) crystalize in a triclinic structure with honeycomb layers of edge-sharing PrO$_6$ octahedra. Each Pr$^{4+}$ cation has a $4f$ electron in the $\Gamma_7$ doublet, which comprises a spin-orbital entangled Kramers pair with the effective moment $J_{\rm eff}=1/2$. By using the Wannier orbitals from the {\it ab initio} calculations, we find that the effective interactions between the $J_{\rm eff}=1/2$ moments are predominantly of {\it antiferromagnetic} Kitaev type for light alkali metals $A$=Li and Na, in stark contrast to the ferromagnetic ones in $4d$- and $5d$-electron based materials. Our finding would provide a playground for the Kitaev spin liquids that is hard to be accessed by the candidates ever discovered.