Bosonic Kane-Mele Model: Topological Boson Systems

• The bosonic Kane-Mele model is a family of lattice Hamiltonians where bosonic particles mimic the topological behavior of the electronic Kane-Mele model through engineered spin-dependent phases.
• It is realized in various platforms—including spin-orbit-coupled ferromagnets, Bose-Hubbard settings, and moiré exciton systems—each revealing unique chiral edge modes and thermal/spin responses.
• This framework unifies topological band theory with strongly interacting boson physics, offering insights into quantum spin Hall analogues, frustrated magnetism, and the emergence of chiral spin states.

Searching arXiv for the specified topic and closely related papers. Found the central papers and a closely related variant on the bosonic Kane-Mele-Hubbard setting. Retrieving an additional directly relevant arXiv result on the bosonic Kane-Mele-Hubbard model. The bosonic Kane-Mele model denotes a class of bosonic lattice Hamiltonians whose single-particle or quasiparticle band structure reproduces the defining topology of the electronic Kane-Mele model, typically on the honeycomb lattice, with opposite effective spin-orbit-coupling signs in two spin or pseudospin sectors. In the literature summarized here, this structure appears in three distinct settings: bosonic spinons in spin-orbit-coupled ferromagnets, the bosonic Kane-Mele-Hubbard model and its Mott-phase spin reduction, and topological moiré excitons in semiconductor heterostructures. Across these realizations, the central theme is that bosonic degrees of freedom can inherit Kane-Mele-type band topology, edge states, and transverse thermal or spin responses, while the many-body phase need not be intrinsically topologically ordered (Kim et al., 2016).

1. Defining structure and conceptual scope

The canonical Kane-Mele construction is the spinful version of the Haldane model. In its bosonic analogues, the role of fermionic electrons is taken by bosons such as spinons or excitons, while the role of intrinsic spin-orbit coupling is played by a spin-dependent next-nearest-neighbor phase or an equivalent pseudospin-dependent hybridization pattern. The resulting bands are topological in the sense that each spin or pseudospin sector realizes a Haldane-type model with opposite sign, yielding a bosonic analogue of a quantum spin Hall structure (Kim et al., 2016).

Within the works considered here, the phrase “bosonic Kane-Mele model” does not refer to a single microscopic Hamiltonian. Rather, it refers to a family of effective descriptions. In localized spin systems, the mean-field spinon Hamiltonian resembles the Kane-Mele model for bosonic spinons. In the bosonic Kane-Mele-Hubbard model, complex spin-dependent hopping phases and onsite interactions define the microscopic lattice model, whose strong-coupling limit maps to a frustrated spin model. In moiré semiconductors, two interlayer exciton states play the role of an internal pseudospin, and moiré-modulated hybridization generates a honeycomb-like effective model with topological flatbands (Plekhanov et al., 2017).

A concise comparison of the three realizations is given below.

Platform	Bosonic degree of freedom	Principal outcome
Honeycomb spin-orbit-coupled ferromagnet	Bosonic spinons / magnons	Haldane-type magnon bands and Kane-Mele-type spinon bands
Bosonic Kane-Mele-Hubbard model	Two-component bosons	Mott-phase frustrated XY physics and an emergent chiral spin state
TMD moiré heterostructure	Interlayer excitons	Topological flatbands realizing a bosonic Kane-Mele model

A plausible implication is that the term is best understood as an effective topological universality class for bosons rather than a unique microscopic model.

2. Localized-spin realization on the honeycomb lattice

A direct route to a bosonic Kane-Mele structure was proposed for spin-orbit-coupled ferromagnets on the honeycomb lattice with the spin Hamiltonian

$$H = -J\sum_{\langle i, j \rangle} \mathbf{S}_i \cdot \mathbf{S}_j - K\sum_{\langle i, j \rangle} S^z_i S^z_j + D\sum_{\langle\langle i, j \rangle\rangle} \nu_{ij} \hat{z} \cdot (\mathbf{S}_i \times \mathbf{S}_j).$$

Here $J$ is the nearest-neighbor Heisenberg exchange, $D$ is the next-nearest-neighbor Dzyaloshinskii-Moriya interaction, and $\nu_{ij}$ encodes the lattice-dependent sign. In the ordered phase, the Holstein-Primakoff transformation yields an effective magnon Hamiltonian

$$H_{\text{m}} = (3 J S + B) \sum_i d_i^\dagger d_i - J S \sum_{\langle i, j \rangle} (d_i^\dagger d_j + \text{h.c.}) - D S \sum_{\langle\langle i, j \rangle\rangle} (i \nu_{ij} d_i^\dagger d_j + \text{h.c.}),$$

which is equivalent to the Haldane model for electrons, except that the effective particles are bosonic magnons (Kim et al., 2016).

In that ordered regime, the DM term opens gaps at the Dirac points, and the magnon bands acquire nonzero Chern numbers $C^\pm = \pm 1$. This establishes bosonic topological magnon bands together with the associated chiral edge state. The significance of this mapping is that a standard topological band construction, originally formulated for electronic Chern bands, emerges from a localized-spin Hamiltonian without itinerant charge degrees of freedom.

The same work then extends beyond the ordered phase by switching from magnons to Schwinger bosons. This shift is essential because the magnon description breaks down at higher temperatures or in the quantum-disordered regime. The bosonic Kane-Mele model therefore enters not as a low-energy spin-wave approximation alone, but as a mean-field topological description of bosonic spinons applicable to both ordered and disordered phases (Kim et al., 2016).

3. Schwinger-boson formulation and bosonic spin Hall topology

In the Schwinger-boson representation,

$$S_i^+ = c_{i\uparrow}^\dagger c_{i\downarrow}, \qquad S_i^z = (c_{i\uparrow}^\dagger c_{i\uparrow} - c_{i\downarrow}^\dagger c_{i\downarrow})/2,$$

with the constraint $\sum_s c_{is}^\dagger c_{is} = 2S$. Under mean-field Hartree-Fock decoupling, the spin Hamiltonian reduces to a quadratic bosonic Hamiltonian for spinons. The resulting structure closely resembles the Kane-Mele model: $$H_{\text{s}}^{\text{MF}} = -\eta J \sum_{\langle i, j \rangle, s} (c_{is}^\dagger c_{js} + \text{h.c.}) + \frac{D}{2} \sum_{\langle\langle i, j \rangle\rangle, s} [i\nu_{ij} s \zeta_{-s} c_{is}^\dagger c_{js} + \text{h.c.}] + \cdots .$$ The mean fields $\eta$ and $J$0 encode short-range spin correlations, and the DM term plays the role analogous to intrinsic spin-orbit coupling (Kim et al., 2016).

The topological content is sector-resolved: each spin sector corresponds to a Haldane-type model, but with opposite DM sign for spin up and spin down. The spinon bands therefore have nonzero Chern numbers $J$1, and protected edge states follow from this band topology. In the language used in the source, this signifies a bosonic quantum spin Hall effect analogue.

The transport signature highlighted in this framework is the spin Nernst effect. Because the Berry curvatures satisfy

$J$2

a longitudinal temperature gradient generates a transverse spin current without charge current in the insulating system. The spin Nernst conductivity is written as

$J$3

where $J$4 and $J$5 is the Bose distribution. The reported temperature dependence peaks at intermediate temperatures where spinons are thermally excited but not yet entirely disordered (Kim et al., 2016).

A common misconception is that this immediately implies a bosonic topological insulator in the many-body sense. The cited result is narrower and more precise: the mean-field spinon bands are topologically nontrivial and produce a spin Nernst response. Whether the fully interacting phase is topologically ordered is a separate question.

4. Bosonic Kane-Mele-Hubbard model and Mott-regime reduction

A distinct but related framework is the bosonic Kane-Mele-Hubbard model on the honeycomb lattice,

$J$6

Here $J$7 and $J$8 are first- and second-neighbor hoppings, $J$9 and $D$0 are intra- and inter-species interactions, and $D$1 encodes spin-dependent hopping phases (Plekhanov et al., 2017).

In the strong-interaction Mott-insulating regime at one boson per site, perturbation theory maps this problem to an effective spin-$D$2 frustrated XY model,

$D$3

with $D$4 and $D$5. For $D$6, one has $D$7, yielding a pure frustrated XY model. The frustration parameter is $D$8 (Plekhanov et al., 2017).

Bosonic dynamical mean-field theory yields a phase diagram containing a uniform superfluid, a chiral superfluid, and a Mott insulator. Within the Mott region, the magnetic character depends on frustration: for small $D$9 there is ferromagnetic order in the $\nu_{ij}$0-plane; for large $\nu_{ij}$1 there is $\nu_{ij}$2 spiral order; and for intermediate $\nu_{ij}$3 an unconventional non-coplanar phase appears, termed the chiral spin state. The reported intermediate regime is $\nu_{ij}$4.

This Mott-phase construction is related to the bosonic Kane-Mele model in a specific sense: the microscopic bosonic hopping problem carries Kane-Mele-type complex phases, while the strong-coupling limit transfers that frustration into an effective spin model. This suggests that the bosonic Kane-Mele structure can act as a parent model for unconventional magnetic phases rather than only for topological band transport.

5. Chiral spin state: symmetry breaking without topological order

The chiral spin state in the Mott phase of the bosonic Kane-Mele-Hubbard model is characterized numerically by bosonic DMFT and exact diagonalization on small clusters. In the intermediate frustration regime, the ground state lacks coplanar $\nu_{ij}$5 magnetic order and responds strongly to perturbations that break parity $\nu_{ij}$6 or time-reversal $\nu_{ij}$7, indicating spontaneous breaking of both symmetries (Plekhanov et al., 2017).

Its defining order parameter is finite scalar spin chirality,

$\nu_{ij}$8

for $\nu_{ij}$9 forming a triangle. The state also shows antiferromagnetic correlations between sublattices, remains translationally invariant, and has a gapped excitation spectrum. The physical mechanism identified in the source is the interplay of strong interactions and complex hopping from spin-orbit coupling, which causes frustration and suppresses classical magnetic order.

A central issue is whether this state is a chiral spin liquid with intrinsic topological order. The calculations summarized in the source argue against that interpretation. Exact diagonalization gives a doubly, not fourfold, quasi-degenerate ground state on torus geometries, and the non-Abelian Berry-curvature calculation of the Chern number,

$$H_{\text{m}} = (3 J S + B) \sum_i d_i^\dagger d_i - J S \sum_{\langle i, j \rangle} (d_i^\dagger d_j + \text{h.c.}) - D S \sum_{\langle\langle i, j \rangle\rangle} (i \nu_{ij} d_i^\dagger d_j + \text{h.c.}),$$0

yields $$H_{\text{m}} = (3 J S + B) \sum_i d_i^\dagger d_i - J S \sum_{\langle i, j \rangle} (d_i^\dagger d_j + \text{h.c.}) - D S \sum_{\langle\langle i, j \rangle\rangle} (i \nu_{ij} d_i^\dagger d_j + \text{h.c.}),$$1. Moreover, the ground-state manifold does not connect under flux insertion as expected for a topologically ordered phase. The conclusion stated in the source is therefore sharp: the chiral spin state is gapped and chiral, but topologically trivial (Plekhanov et al., 2017).

This distinction addresses an important conceptual confusion. Broken $$H_{\text{m}} = (3 J S + B) \sum_i d_i^\dagger d_i - J S \sum_{\langle i, j \rangle} (d_i^\dagger d_j + \text{h.c.}) - D S \sum_{\langle\langle i, j \rangle\rangle} (i \nu_{ij} d_i^\dagger d_j + \text{h.c.}),$$2 and $$H_{\text{m}} = (3 J S + B) \sum_i d_i^\dagger d_i - J S \sum_{\langle i, j \rangle} (d_i^\dagger d_j + \text{h.c.}) - D S \sum_{\langle\langle i, j \rangle\rangle} (i \nu_{ij} d_i^\dagger d_j + \text{h.c.}),$$3, finite chirality, and a gap do not by themselves establish intrinsic topological order. In this case, the many-body phase is chiral but not a chiral spin liquid.

6. Moiré excitons and topological flatband realization

A more recent realization places the bosonic Kane-Mele model in semiconductor moiré heterostructures. The proposed platform is a trilayer heterostructure formed by a twisted $$H_{\text{m}} = (3 J S + B) \sum_i d_i^\dagger d_i - J S \sum_{\langle i, j \rangle} (d_i^\dagger d_j + \text{h.c.}) - D S \sum_{\langle\langle i, j \rangle\rangle} (i \nu_{ij} d_i^\dagger d_j + \text{h.c.}),$$4 homobilayer stacked on a monolayer $$H_{\text{m}} = (3 J S + B) \sum_i d_i^\dagger d_i - J S \sum_{\langle i, j \rangle} (d_i^\dagger d_j + \text{h.c.}) - D S \sum_{\langle\langle i, j \rangle\rangle} (i \nu_{ij} d_i^\dagger d_j + \text{h.c.}),$$5, with the moiré pattern supplied by the twist in the top two layers. The lowest-energy excitons are interlayer excitons, with the electron in MoSe$$H_{\text{m}} = (3 J S + B) \sum_i d_i^\dagger d_i - J S \sum_{\langle i, j \rangle} (d_i^\dagger d_j + \text{h.c.}) - D S \sum_{\langle\langle i, j \rangle\rangle} (i \nu_{ij} d_i^\dagger d_j + \text{h.c.}),$$6 and the hole in either of the two WSe$$H_{\text{m}} = (3 J S + B) \sum_i d_i^\dagger d_i - J S \sum_{\langle i, j \rangle} (d_i^\dagger d_j + \text{h.c.}) - D S \sum_{\langle\langle i, j \rangle\rangle} (i \nu_{ij} d_i^\dagger d_j + \text{h.c.}),$$7 layers, producing two states $$H_{\text{m}} = (3 J S + B) \sum_i d_i^\dagger d_i - J S \sum_{\langle i, j \rangle} (d_i^\dagger d_j + \text{h.c.}) - D S \sum_{\langle\langle i, j \rangle\rangle} (i \nu_{ij} d_i^\dagger d_j + \text{h.c.}),$$8 and $$H_{\text{m}} = (3 J S + B) \sum_i d_i^\dagger d_i - J S \sum_{\langle i, j \rangle} (d_i^\dagger d_j + \text{h.c.}) - D S \sum_{\langle\langle i, j \rangle\rangle} (i \nu_{ij} d_i^\dagger d_j + \text{h.c.}),$$9. Because interlayer excitons have greatly reduced optical dipole moment relative to intralayer excitons, their lifetimes can reach up to nanoseconds (Xie et al., 2024).

The single-exciton effective Hamiltonian for valley $C^\pm = \pm 1$0 in the center-of-mass momentum basis is

$C^\pm = \pm 1$1

Here $C^\pm = \pm 1$2 is the single-exciton dispersion, $C^\pm = \pm 1$3 is the moiré potential, and $C^\pm = \pm 1$4 is the moiré-modulated interlayer hybridization. The nontrivial topology originates from the spatially modulated hybridization of $C^\pm = \pm 1$5 and $C^\pm = \pm 1$6, which induces a layer-pseudospin winding analogous to spin-orbit coupling in the Kane-Mele model (Xie et al., 2024).

Twist angle and perpendicular electric field are the principal tuning parameters. Small twist angles enlarge the moiré period, strengthen the moiré potentials, and narrow the bands. A magic angle near $C^\pm = \pm 1$7 minimizes the bandwidth of the lowest moiré exciton band. The electric field controls the detuning between $C^\pm = \pm 1$8 and $C^\pm = \pm 1$9,

$$S_i^+ = c_{i\uparrow}^\dagger c_{i\downarrow}, \qquad S_i^z = (c_{i\uparrow}^\dagger c_{i\uparrow} - c_{i\downarrow}^\dagger c_{i\downarrow})/2,$$0

with $$S_i^+ = c_{i\uparrow}^\dagger c_{i\downarrow}, \qquad S_i^z = (c_{i\uparrow}^\dagger c_{i\uparrow} - c_{i\downarrow}^\dagger c_{i\downarrow})/2,$$1. Near resonance, strong hybridization produces band inversion and enables topological phases (Xie et al., 2024).

The resulting lowest exciton bands are isolated and topological. Their topology is characterized by

$$S_i^+ = c_{i\uparrow}^\dagger c_{i\downarrow}, \qquad S_i^z = (c_{i\uparrow}^\dagger c_{i\uparrow} - c_{i\downarrow}^\dagger c_{i\downarrow})/2,$$2

and the calculations reported in the source give $$S_i^+ = c_{i\uparrow}^\dagger c_{i\downarrow}, \qquad S_i^z = (c_{i\uparrow}^\dagger c_{i\uparrow} - c_{i\downarrow}^\dagger c_{i\downarrow})/2,$$3 and $$S_i^+ = c_{i\uparrow}^\dagger c_{i\downarrow}, \qquad S_i^z = (c_{i\uparrow}^\dagger c_{i\uparrow} - c_{i\downarrow}^\dagger c_{i\downarrow})/2,$$4 for the two lowest bands in an appropriate field-and-angle regime. A Wannier construction maps these bands to an effective honeycomb-lattice bosonic Kane-Mele tight-binding model,

$$S_i^+ = c_{i\uparrow}^\dagger c_{i\downarrow}, \qquad S_i^z = (c_{i\uparrow}^\dagger c_{i\uparrow} - c_{i\downarrow}^\dagger c_{i\downarrow})/2,$$5

confirming the topological flatband interpretation. The source further states that interaction estimates of $$S_i^+ = c_{i\uparrow}^\dagger c_{i\downarrow}, \qquad S_i^z = (c_{i\uparrow}^\dagger c_{i\uparrow} - c_{i\downarrow}^\dagger c_{i\downarrow})/2,$$6-$$S_i^+ = c_{i\uparrow}^\dagger c_{i\downarrow}, \qquad S_i^z = (c_{i\uparrow}^\dagger c_{i\uparrow} - c_{i\downarrow}^\dagger c_{i\downarrow})/2,$$7 meV dominate over the moiré bandwidth in experimental parameter regimes, placing the system in a Bose-Hubbard regime with topological flatbands (Xie et al., 2024).

7. Experimental signatures, limitations, and broader implications

The localized-spin proposal identifies artificial honeycomb arrays and chromium trihalides such as CrBr$$S_i^+ = c_{i\uparrow}^\dagger c_{i\downarrow}, \qquad S_i^z = (c_{i\uparrow}^\dagger c_{i\uparrow} - c_{i\downarrow}^\dagger c_{i\downarrow})/2,$$8 as feasible routes, with the DM interaction potentially induced by proximity to heavy elements such as Pt. The proposed signatures include thermal Hall or spin Nernst effects and the detection of chiral bosonic edge modes through nonlocal thermal transport or magnonic edge states (Kim et al., 2016).

In ultracold-atom realizations of the bosonic Kane-Mele-Hubbard model, the experimentally relevant observables for the chiral spin state include scalar spin chirality, current correlations, local densities, and the response to $$S_i^+ = c_{i\uparrow}^\dagger c_{i\downarrow}, \qquad S_i^z = (c_{i\uparrow}^\dagger c_{i\uparrow} - c_{i\downarrow}^\dagger c_{i\downarrow})/2,$$9- or $\sum_s c_{is}^\dagger c_{is} = 2S$0-breaking perturbations. The main limitation established in the source is that the chiral spin state, despite finite chirality and a gap, is not topologically ordered and has Chern number zero (Plekhanov et al., 2017).

In the moiré exciton setting, the predicted signatures include topological edge states detectable via local optical probes or spatially resolved photoluminescence, as well as incompressible behavior and spectral jumps at fractional filling in the search for bosonic fractional quantum anomalous Hall states. The source states that the bosonic Kane-Mele Bose-Hubbard model realized in this solid-state platform is predicted by prior theory and numerics to host fractional quantum anomalous Hall states at fillings such as $\sum_s c_{is}^\dagger c_{is} = 2S$1, $\sum_s c_{is}^\dagger c_{is} = 2S$2, and non-Abelian states at $\sum_s c_{is}^\dagger c_{is} = 2S$3 (Xie et al., 2024).

Taken together, these results place the bosonic Kane-Mele model at the intersection of topological band theory, frustrated magnetism, and bosonic many-body physics. One branch emphasizes topological quasiparticle bands and transverse responses, another shows how a Kane-Mele-type parent model generates frustrated Mott magnetism, and a third provides topological flatbands in a solid-state excitonic platform. A plausible synthesis is that the bosonic Kane-Mele model is not a single phase but a unifying framework for bosonic systems in which Kane-Mele-type topology can manifest either as band topology, as a precursor to correlated phases, or as a tunable flatband platform for strongly interacting bosonic topological matter.