Table of Contents
Fetching ...

Robust semiclassical magnetization plateau in the kagome lattice

Gabriel Capelo, Eric C. Andrade

TL;DR

The paper addresses the robustness of the $1/3$ magnetization plateau in the kagome $J_1$-$J_2$ Heisenberg model under a magnetic field. It combines classical analysis, thermal Monte Carlo, and both linear and nonlinear spin-wave theories to show that thermal and quantum fluctuations select the collinear uud state via order-by-disorder, producing a plateau whose width is only weakly sensitive to $J_2$. A notable finding is the magnetization jump at saturation that occurs only at $J_2=0$ due to a flat magnon band, a feature that disappears with finite $J_2$ or at finite temperature. Overall, the semiclassical framework reliably captures the plateau physics and provides a useful lens for interpreting experiments and numerical studies of kagome magnets.

Abstract

Inspired by recent observations of the $1/3$ magnetization plateau in kagome-based magnets, we investigate the $J_1-J_2$ Heisenberg model on the kagome lattice under the influence of an external magnetic field. Although the classical ground state at zero field depends on the sign of $J_2$, we find a robust $1/3$ semiclassical magnetization plateau in both cases. The mechanism that stabilizes this plateau is analogous to that observed in the triangular lattice, where quantum fluctuations select a collinear state from the degenerate classical manifold. We calculate the plateau width, which shows a weak dependence on $J_2$, using nonlinear spin-wave theory. Additionally, we find that a straightforward approach based on linear spin-wave yields quantitatively accurate results even for $S=1/2$. Furthermore, we identify a magnetization jump at the saturation field when $J_2=0$; this jump is related to the presence of a flat band and disappears for $J_2 \neq 0$. Our study demonstrates that a semiclassical approach effectively captures the $1/3$ plateau in the kagome lattice and serves as a valuable tool for interpreting experimental and numerical results alike.

Robust semiclassical magnetization plateau in the kagome lattice

TL;DR

The paper addresses the robustness of the magnetization plateau in the kagome - Heisenberg model under a magnetic field. It combines classical analysis, thermal Monte Carlo, and both linear and nonlinear spin-wave theories to show that thermal and quantum fluctuations select the collinear uud state via order-by-disorder, producing a plateau whose width is only weakly sensitive to . A notable finding is the magnetization jump at saturation that occurs only at due to a flat magnon band, a feature that disappears with finite or at finite temperature. Overall, the semiclassical framework reliably captures the plateau physics and provides a useful lens for interpreting experiments and numerical studies of kagome magnets.

Abstract

Inspired by recent observations of the magnetization plateau in kagome-based magnets, we investigate the Heisenberg model on the kagome lattice under the influence of an external magnetic field. Although the classical ground state at zero field depends on the sign of , we find a robust semiclassical magnetization plateau in both cases. The mechanism that stabilizes this plateau is analogous to that observed in the triangular lattice, where quantum fluctuations select a collinear state from the degenerate classical manifold. We calculate the plateau width, which shows a weak dependence on , using nonlinear spin-wave theory. Additionally, we find that a straightforward approach based on linear spin-wave yields quantitatively accurate results even for . Furthermore, we identify a magnetization jump at the saturation field when ; this jump is related to the presence of a flat band and disappears for . Our study demonstrates that a semiclassical approach effectively captures the plateau in the kagome lattice and serves as a valuable tool for interpreting experimental and numerical results alike.

Paper Structure

This paper contains 14 sections, 33 equations, 8 figures.

Figures (8)

  • Figure 1: (a)Kagome lattice showing its three sublattices, $A$, $B$, and $C$. Here, $J_1$ is the nearest-neighbor exchange coupling and $J_2$ the next-nearest neighbor; (b) Network connecting next-nearest neighbours forming a kagome supperlattice; Semiclassical ground states of the $J_1-J_2$ model: (c) $\vec{Q}=0$ for $J_2>0$, and (d) $\sqrt{3} \times \sqrt{3}$ for $J_2 < 0$. The shaded green area indicates the magnetic unit cell; (e) Possible semiclassical states of the $J_1-J_2$ Heisenberg model in a kagome lattice for a finite field. The three colors represent the three different spins on an elementary triangle. From left to right: Y state, up-up-down (uud) state, V state, and polarized state.
  • Figure 2: Magnon dispersions $\omega_{\textbf{k}}$, obtained with the linear spin-wave theory, for the $J_1-J_2$ Heisenberg model in the kagome lattice inside the polarized phase at $h=h_c$. The dispersions are plotted along the indicated path in the Brillouin zone for different values of $J_2$: (a) $J_2<0$, (b) $J_2=0$, and (c) $J_2>0$.
  • Figure 3: Classical MC results for the magnetization $m/S$ along the field direction, and its derivative $\chi=dm/dh$, as a function of the field $h$ for $k_{B}T=0.07J_1$ and $L=24$. We have $h_c/S=6J_1$ for $J_2\le0$ and $h_c/S=6\left(J_1+J_2\right)$ for $J_2>0$. For $J_2 > 0$ we renormalized the $\chi$ curve by $\left(J_1 + J_2\right)/J_1$ to account for the change in $h_c$ due to $J_2$.
  • Figure 4: (a) Ground-state energy $E_{\text{gs}}$ as function of the magnetic field $h$ obtained with LSWT, Eq. \ref{['eq:Enegs']}, for $S=1/2$ and $J_2/J_1=-0.2$; The uud energy was calculated with Eq. \ref{['eq:linextrap']}. (b) Same as (a), for $S=1/2$ and $J_2/J_1=0.2$; (c) Magnetization along the field direction $m$ as a function of $h$, obtained with Eq. \ref{['eq:mh-dEgsdh']}, for $S=1/2$ and different values of $J2$: $J_2/J_1=-0.2$, $J_2/J_1=0$, and $J_2/J_1=0.2$; (d) Same as (c) for $S=1$. The $1/3$ plateau was inserted phenomenologically as discussed in Sec. \ref{['sec:MagnetizationPlateau']}.
  • Figure 5: Critical fields $h_1$ and $h_2$ for the $\vec{Q}=0$ state as function of $J_2$ for: a) spin $S=1/2$; and $b)$ spin $S=1$. The open symbols indicate the result of the angle renormalization given by Eqs. \ref{['eq:chubukov1_def']}, and \ref{['eq:chubukov2_def']}. The solid symbols with dashed lines represent the result of the gap closing calculated with Eq. \ref{['eq:defhcgap']}. The yellow star on the $J_2$ axis indicates the breakdown of LSWT stability for $h=0$, see Sec. \ref{['sec:unstable']}; Critical fields of the plateau phase as a function of $1/S$ for: c) $J_2\geq 0$; and d) $J_2\leq 0$. The dashed red (black) lines correspond to results with the phenomenological method discussed in Sec. \ref{['sec:MagnetizationPlateau']} for $J_2/J_1=\pm0.1 (J_2/J_1=\pm 0.2)$. The solid symbols indicate the result of the angle renormalization given by Eqs. \ref{['eq:chubukov1_def']}, and \ref{['eq:chubukov2_def']}. The gray lines following the solid symbols are a guide for the eye.
  • ...and 3 more figures