Insulating charge transfer ferromagnetism

We propose a mechanism for insulating ferromagnetism in the honeycomb Hubbard model of semiconductor moiré superlattices. The ferromagnetism emerges at critical charge transfer regime, stabilizing the quantum anomalous Hall state without Hund’s coupling. We further note the ferromagnetic exchange applies to general charge transfer systems when breaking particle-hole symmetry.

💡 Research Summary

The authors address a puzzling observation of insulating ferromagnetism (FM) at filling ν = 1 in twisted MoTe₂ moiré superlattices, where conventional mechanisms such as Hund’s coupling or metallic Nagaoka ferromagnetism are absent. They propose a new mechanism that operates in a charge‑transfer insulating regime of a honeycomb Hubbard model with a tunable sub‑lattice potential difference Δ (controlled by a vertical electric field). In the strong‑coupling limit (U ≫ |t|), electrons occupy the lower‑energy A sub‑lattice, leaving a massive spin degeneracy. When kinetic hopping is turned on, this degeneracy is lifted and an effective spin Hamiltonian emerges.

Through third‑order perturbation theory the authors derive an effective Heisenberg model H_eff = −J ∑⟨i,j⟩_A S_i·S_j with an exchange constant J that contains three distinct contributions:

1. Conventional super‑exchange (∝ t_AA²/U);
2. A mixed kinetic‑super‑exchange term (∝ t_AB² t_AA/(Δ U));
3. A ring‑exchange term (∝ t_AB² t_AA/Δ²).

Crucially, the mixed and ring‑exchange terms involve an odd number of hopping amplitudes. When the nearest‑neighbor hopping within the A sub‑lattice, t_AA, is negative, the signs of the two paths (direct t_AA versus indirect t_AB‑t_AB) become opposite, leading to constructive interference that makes J positive. Positive J corresponds to ferromagnetic coupling, whereas for Δ ≫ U the first term dominates and J becomes negative, yielding the familiar 120° antiferromagnetic (AFM) order of a triangular lattice. Thus the phase diagram is governed by two dimensionless parameters, Δ/U and t_AB²/(t_AA Δ), and is universal—independent of lattice details—once U_A = U_B = U.

To validate the analytical picture, the authors perform exact diagonalization (ED) on a 3 × 3 periodic cluster and density‑matrix renormalization group (DMRG) calculations on a 3 × 12 cylinder. They compute spin gaps (Δ_s) and spin‑spin correlation functions χ(i,j). The numerical results reproduce the analytical phase boundary: for intermediate Δ/U (≈ 1) the system exhibits a vanishing spin gap and a pronounced Γ‑point peak in χ(k), indicative of ferromagnetism; for both small and large Δ/U the spin gap opens and the K‑point dominates, signalling AFM order. The agreement improves for larger U, confirming the perturbative regime.

The study is extended to a Kane–Mele–Hubbard model with complex second‑nearest‑neighbor hopping t_AA = |t_AA| e^{iϕ}. Although SU(2) spin symmetry is broken, the Γ‑point dominance of χ(k) persists across ϕ, showing that the ferromagnetic region is robust against spin‑orbit‑type complex hoppings. Energy differences between S_z = 3/2 and S_z = 1/2 sectors reveal a small in‑plane ferromagnetic anisotropy, while the AFM side displays weak out‑of‑plane canting reminiscent of Dzyaloshinskii–Moriya interactions.

In the discussion, the authors emphasize that breaking particle‑hole symmetry (via Δ) allows odd‑order hopping products to appear in the effective exchange, a situation forbidden in the particle‑hole symmetric half‑filled bipartite Hubbard model where Lieb’s theorem guarantees only antiferromagnetism. Hence, any charge‑transfer insulator where a sub‑lattice potential imbalance can be engineered should be susceptible to this ferromagnetic exchange. They suggest experimental platforms such as R‑stacked moiré bilayers, twisted transition‑metal dichalcogenides, and other 2D heterostructures where gating can tune Δ, providing a practical route to engineer insulating ferromagnets and, consequently, quantum anomalous Hall states without relying on Hund’s coupling.

Overall, the paper delivers a comprehensive theoretical framework—supported by rigorous analytical derivations and state‑of‑the‑art numerics—for a novel ferromagnetic exchange mechanism in insulating charge‑transfer systems, opening new avenues for designing spin‑polarized quantum phases in van‑der‑Waals materials.