Ground-state phases of the frustrated spin-1/2 $J_{1}$--$J_{2}$--$J_{3}$ Heisenberg ferromagnet ($J_{1}<0$) on the honeycomb lattice with $J_{3}=J_{2}>0$

TL;DR

This study investigates the ground-state phases of a frustrated spin-1/2 Heisenberg model on a honeycomb lattice with ferromagnetic and antiferromagnetic interactions, revealing a direct first-order transition between ferromagnetic and striped phases.

Contribution

It provides the first detailed comparison of ground-state phase transitions in the ferromagnetic versus antiferromagnetic J1-J2-J3 honeycomb models using advanced computational methods.

Findings

Identifies a quantum phase transition at J2 ≈ 0.1095 between FM and striped order.

Finds no intermediate disordered phase in the FM model, unlike the AFM case.

Shows a significant difference in critical points between FM and AFM models.

Abstract

We study the ground-state (gs) properties of the frustrated spin-1/2 $J_{1}$--$J_{2}$--$J_{3}$ Heisenberg model on a honeycomb lattice with ferromagnetic (FM) nearest-neighbor ($J_{1}=-1$) exchange and frustrating antiferromagnetic (AFM) next-nearest-neighbor ($J_{2}>0$) and next-next-nearest-neighbor ($J_{3}>0$) exchanges, for the case $J_{3}=J_{2}$. We use the coupled cluster method in high orders of approximation, complemented by the exact diagonalization of a lattice with 32 sites, and calculate the gs energy, magnetic order parameter, and spin-spin correlation functions. We find a quantum phase transition between regions characterized by FM order and a form of AFM ("striped") collinear order at $J^{c}_{2} \approx 0.1095 \pm 0.0005$. We compare results for the FM case (with $J_{1}=-1$) to previous results for the corresponding AFM case (with $J_{1}=+1$). While the magnetic order parameters behave similarly for the FM and the AFM models for large values of the frustration parameter $J_{2}$, there are considerable differences between them for $J_{2}/|J_{1}| \lesssim 0.6$. For example, the quasiclassical collinear magnetic long-range order for the AFM model (with $J_{1}=+1$) breaks down at $J^{c_{2}}_{2} \approx 0.60$, whereas the "equivalent" point for the FM model (with $J_{1}=-1$) occurs at $J^{c}_{2} \approx 0.11$. Unlike in the AFM model (with $J_{1}=+1$), where a plaquette valence-bond crystal phase intrudes between the two corresponding quasiclassical AFM phases (with N\'eel and striped order) for $J^{c_{1}}_{2} < J_{2} < J^{c_{2}}_{2}$, with $J^{c_{1}}_{2} \approx 0.47$, we find no clear indications in the FM model for an intermediate magnetically disordered phase between the phases exhibiting FM and striped order. Instead, the evidence points strongly to a direct first-order transition between the two ordered phases of the FM model.