🔬 mesoscale physics

Stabilizing and detecting electron fractionalization in the Hofstadter-Hubbard model

This paper demonstrates that increasing orbital flux in the half-filled triangular-lattice Hofstadter-Hubbard model stabilizes a $\mathrm{U}(1)_2$ chiral spin liquid phase where electrons fractionalize into gapped spinons and chargons, producing distinct spectral signatures that link band topology to intrinsic topological order.

Original authors: Jiahao Yang, Zeyu Han, Gang v. Chen

Published 2026-10-22

📖 7 min read🧠 Deep dive

CC BY 4.0

Original authors: Jiahao Yang, Zeyu Han, Gang v. Chen

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the world of modern physics, scientists have long been fascinated by materials that conduct electricity in strange, protected ways. Imagine a highway where traffic flows perfectly in one direction without ever crashing or slowing down, no matter how many potholes appear in the road. This is the essence of a topological insulator, a state of matter where the rules of the road are written into the very geometry of the electrons' paths. For decades, physicists understood how these highways worked when electrons moved freely, ignoring each other. But the real frontier lies in what happens when these electrons are forced to interact, pushing and shoving against one another with great intensity. In these crowded, chaotic conditions, the neat, predictable highways can dissolve into something far more exotic: a state where the electron itself seems to break apart, shedding its identity to become a collection of independent, fractional pieces.

This breakdown of the electron is not just a theoretical curiosity; it is a potential key to building a new kind of quantum computer that is immune to errors. However, finding and proving that this fractionalization actually happens in a real material has been a monumental challenge. The particles that emerge from this breakup, specifically the ones carrying spin but no electric charge, are invisible to the standard tools used to measure electricity. They do not flow like current, so they leave no trace on a voltmeter. To catch them, researchers must look for subtle fingerprints in the way energy is absorbed and released, searching for a pattern that reveals the electron has split into distinct, interacting parts.

A team of researchers has now taken a significant step toward solving this puzzle by simulating a specific, highly controlled environment where these effects can be observed. They focused on a theoretical model of electrons arranged in a triangular grid, a shape that naturally encourages complex interactions. By introducing a magnetic field that threads through the loops of this grid, they created a scenario where the electrons are forced to choose between staying together as a solid block or breaking apart into a fluid-like state. Their work suggests that by carefully tuning the strength of this magnetic field, it is possible to stabilize a mysterious state of matter known as a chiral spin liquid. In this state, the electron does indeed fractionalize, splitting into a neutral piece that carries spin and a charged piece that carries the electric charge, both of which become trapped in energy gaps that prevent them from moving freely.

The researchers used a sophisticated computational method to map out exactly how this system behaves as they changed the strength of the electron interactions and the magnetic flux. They found that when the magnetic field is strong enough, it acts as a stabilizing force, widening the window where this exotic fractionalized state can exist. In this regime, the electron is no longer a single, coherent particle. Instead, it exists as a combination of two separate entities: a spin-carrying particle and a charge-carrying particle. These two pieces are linked, but they move independently, and their combined behavior creates a unique signature. When the researchers calculated what an experiment would see if they tried to knock an electron out of this material, they found a very specific pattern. Instead of seeing a single, sharp peak of energy that would indicate a normal, intact electron, they saw a broad, complex structure with multiple peaks.

This complex structure is the result of the two fractional pieces moving together. The researchers discovered that the magnetic field controls how these pieces arrange their energy levels. In some cases, the charged piece has a few distinct energy levels, and when it combines with the spin piece, it creates a series of sharp, resolved peaks in the energy spectrum. In other cases, the charged piece has a smoother distribution of energy, resulting in a single, broad hump. This difference is crucial because it means that by measuring the energy spectrum of the material, scientists can tell not only that the electron has broken apart, but also exactly how the magnetic field is influencing that breakup. The study shows that the magnetic flux is not just a background condition but a precise control knob that determines the detailed pattern of the fractionalization.

The team also compared this fractionalized state to two other possibilities that could exist in the same material: a standard topological insulator where electrons remain whole, and a state where electrons lock into a rigid, repeating pattern of bonds. They found that the fractionalized state has a unique signature that distinguishes it from both. Unlike the standard insulator, which shows a sharp, coherent signal, the fractionalized state shows the broad, incoherent spread of energy caused by the electron splitting. Unlike the rigid bond state, which breaks the symmetry of the grid and creates a patchwork of different regions, the fractionalized state remains uniform across the entire material. This uniformity, combined with the specific energy pattern, provides a clear way to identify the state in an experiment.

Perhaps most importantly, the researchers identified a way to confirm the existence of this state using transport measurements. While the neutral pieces of the broken electron do not carry electric current, they do carry heat. The study suggests that if this state exists, it will conduct heat in a quantized, perfectly precise way along the edges of the material, a signature that is distinct from any other known state. This thermal signature, combined with the energy spectrum patterns, offers a complete toolkit for experimentalists. They can now look for these specific fingerprints in real materials, such as the twisted layers of atoms known as moiré superlattices, which are currently being studied in laboratories around the world.

The findings also shed light on how the material transitions from a normal state to this exotic one. Depending on the strength of the magnetic field, this transition can happen in two different ways. In some cases, the change is abrupt, like water suddenly freezing into ice, where the properties of the material jump instantly from one value to another. In other cases, the change is smooth and continuous, like a magnet slowly losing its strength as it heats up. The researchers found that the magnetic field determines which of these two paths the material takes. This distinction is vital for understanding the fundamental nature of the transition and for predicting how the material will behave under different conditions.

By linking the behavior of the electrons to the geometry of the magnetic field, this work bridges the gap between the simple, free-moving world of single particles and the complex, entangled world of strongly interacting matter. It provides a concrete roadmap for how to create and detect a state where the electron loses its individual identity. The study does not claim to have discovered this state in a physical sample yet, but it provides the precise theoretical predictions needed to find it. It tells experimentalists exactly what to look for: a uniform material that shows a broad, multi-peaked energy spectrum and a quantized flow of heat along its edges. If these signatures are found, it will confirm that the electron can indeed be split into fractional parts, opening the door to a new era of quantum materials where topology and strong interactions work together to create entirely new forms of matter.

Technical Summary: Stabilizing and detecting electron fractionalization in the Hofstadter–Hubbard model

Problem Statement
The paper addresses the challenge of understanding how noninteracting band topologies, specifically Chern insulators (CIs), evolve under strong electron-electron correlations. A central question is whether strong on-site Coulomb repulsion can drive a system into an intrinsic topological order characterized by long-range quantum entanglement and fractionalized excitations, such as a chiral spin liquid (CSL). While theoretical proposals suggest that the triangular-lattice Hubbard model with zero flux might host a CSL, this scenario remains unsettled. The authors investigate whether the introduction of large orbital magnetic fluxes, naturally realizable in moiré superlattices, can stabilize and broaden the CSL phase, and how to experimentally distinguish this fractionalized state from competing trivial insulators (like valence-bond solids) and the parent Chern insulator.

Methodology
The authors study the half-filled triangular-lattice Hofstadter–Hubbard model using self-consistent slave-rotor mean-field theory.

Model: The Hamiltonian includes nearest-neighbor hopping with Peierls phases induced by an external orbital magnetic flux ( $\Phi_{\triangle}^{ext}$ ) and an on-site Hubbard repulsion $U$ . The system is analyzed at half-filling ( $n=1$ ).
Slave-Rotor Decomposition: Physical electrons are fractionalized into neutral spin-1/2 fermionic spinons ( $f_{i\sigma}$ ) and spinless charged rotors ( $X_i$ ). This introduces an emergent local $U(1)$ gauge redundancy.
Mean-Field Treatment: The physical electron hopping is decoupled into mean-field bond fields: $\chi_{ij}$ for spinon hopping and $\Delta^*_{ij}$ for rotor kinetic energy. The total Hamiltonian separates into decoupled spinon and rotor sectors.
Phase Diagram Construction: The ground-state energies of nonmagnetic mean-field states—Chern insulator (CI), $U(1)_2$ chiral spin liquid (CSL), and plaquette valence-bond solid (VBS)—are compared across various external fluxes ( $\Phi_{\triangle}^{ext} = \pi/2, \pi/4, \pi/6$ , etc.) and interaction strengths ( $U/t$ ).
Spectral Calculations: The single-electron spectral function ( $A_e(k, \omega)$ ) and density of states (DOS) are calculated. In the CSL phase, where rotors are gapped and uncondensed, the electron spectrum is derived from the convolution of the spinon and rotor propagators.

Key Results

Stabilization of the CSL: Increasing the orbital flux systematically stabilizes the $U(1)_2$ CSL phase. As $\Phi_{\triangle}^{ext}$ increases, the upper boundary of the CSL phase (separating it from the VBS) shifts to significantly larger values of $U/t$ , while the lower boundary (separating it from the CI) remains largely unchanged. This expands the intermediate-coupling window where the CSL exists.
Nature of the Mott Transition: The transition between the CI and CSL depends on the flux:
- For $\Phi_{\triangle}^{ext} = \pi/4$ and $\pi/6$ , the transition is first-order. It involves a discontinuous jump in the rotor condensate weight ( $Z$ ), rotor gap, and flux partitioning between spinons and rotors. The spinon band Chern number changes from $C_f=4$ (or 6) in the CI to $C_f=2$ in the CSL via an energetic level crossing of distinct self-consistent solutions.
- For $\Phi_{\triangle}^{ext} = \pi/2$ , the transition is continuous. Both phases share the same spinon band Chern number ( $C_f=2$ ) and flux partitioning. The transition is driven by the continuous vanishing of the rotor condensate, analogous to the superfluid-Mott transition in the Bose-Hubbard model.
Electron Fractionalization Signatures: In the CSL phase, the electron is fractionalized into gapped spinons and gapped chargons (rotor excitations).
- The electron spectral function lacks coherent quasiparticle peaks, which are present in the CI phase.
- Instead, the spectrum exhibits weakly dispersing ridges and multi-peak structures in the density of states. These features arise from the convolution of narrow spinon bands with gapped rotor excitations.
- The specific pattern of these peaks depends on the flux partitioning. For example, at $\Phi_{\triangle}^{ext} = \pi/6$ , the rotor spectrum reorganizes into narrow groups, leading to resolved electron-DOS peaks, whereas at $\pi/2$ , a broad maximum is observed.
Distinguishing Competing States:
- CSL vs. CI: The CSL is distinguished by the absence of coherent quasiparticle peaks (due to $Z=0$ ) and the presence of incoherent continuum features.
- CSL vs. Plaquette VBS: While both exhibit incoherent spectra, the CSL remains spatially uniform with a non-zero spinon Chern number ( $C_f=2$ ) and $U(1)_2$ topological order. In contrast, the plaquette VBS breaks translational symmetry (spatial bond order) and is topologically trivial ( $C_f=0$ ).

Significance and Claims
The paper claims to identify orbital flux as a crucial control parameter for stabilizing and tuning the nature of electron fractionalization in correlated Chern systems.

Theoretical Contribution: It demonstrates that large orbital fluxes can stabilize the $U(1)_2$ CSL in the triangular-lattice Hofstadter–Hubbard model, a scenario that is difficult to realize in the zero-flux limit. It provides a unified low-energy description (Chern-Simons-Higgs framework) for the continuous CI-CSL transition at $\pi/2$ flux.
Experimental Relevance: The results provide specific spectral fingerprints (weakly dispersing ridges and multi-peak DOS structures) for angle-resolved photoemission spectroscopy (ARPES) and scanning tunneling microscopy (STM) to detect fractionalization.
Transport Signatures: The paper suggests that the nature of the transition (continuous vs. first-order) imprints distinct transport signatures. Specifically, the continuous transition at $\pi/2$ predicts a universal jump in longitudinal resistivity and finite Hall resistivity at criticality, whereas first-order transitions may show phase coexistence and hysteresis.
Platform: The work highlights moiré superlattices (e.g., twisted WSe $_2$ or TMD heterobilayers) as realistic platforms where large orbital fluxes and tunable interactions ( $U/t$ via electric displacement fields) can be simultaneously accessed to explore these fractionalized topological phases.

The authors emphasize that while single-particle probes can reveal the incoherent nature of the CSL, distinguishing it from the VBS requires combining spectral data with real-space imaging (to detect translational symmetry breaking) or thermal Hall measurements (to detect chiral edge modes).

More like this