🔬 mesoscale physics

Non-Hermitian topology of quantum spin-Hall systems to detect edge-state polarization

This paper demonstrates that non-Hermitian topology and the non-Hermitian skin effect in quantum spin-Hall systems arise specifically from directional transport imbalances caused by spin-selective coupling or Zeeman fields, providing a sensitive transport-based method to detect intrinsic edge-state polarization.

Original authors: Raghav Chaturvedi, Ion Cosma Fulga, Jeroen van den Brink, Ewelina M. Hankiewicz

Published 2026-02-13

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Raghav Chaturvedi, Ion Cosma Fulga, Jeroen van den Brink, Ewelina M. Hankiewicz

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

Imagine a busy highway system built inside a tiny, invisible city made of atoms. This is a Quantum Spin-Hall (QSH) device. In this city, traffic doesn't just flow in one direction; it has a special rule: cars with "red" paint (spin-up) must drive clockwise, while cars with "blue" paint (spin-down) must drive counter-clockwise. They are like a pair of dancers who are perfectly synchronized but moving in opposite directions.

Usually, if you put a sensor at any point on this highway, the traffic flow looks perfectly balanced. If you send a red car in, a blue car comes out the other side with the same ease. The system is fair, symmetrical, and predictable.

The Big Discovery: When the Highway Gets "Unfair"

This paper explores what happens when we break that perfect balance. The researchers found that you can make the traffic flow become "non-reciprocal"—meaning it's easy to go one way but hard to go the other—without actually breaking the laws of physics or changing the road itself. You just have to change who is allowed to enter the highway.

Here is the breakdown of their findings using simple analogies:

1. The "Spin-Selective" Gatekeepers

Imagine the highway has entry ramps (the "leads" or contacts).

Normal Scenario: The ramps accept both red and blue cars equally. The traffic flows smoothly in both directions. The system is "Hermitian" (a fancy physics word for "balanced and fair").
The Experiment: The researchers changed the ramps so they only let red cars in easily, while blue cars struggle to get on.
The Result: Even though the road itself is still perfectly symmetrical, the traffic pattern becomes lopsided. It's easy to drive from Ramp A to Ramp B, but hard to go from B to A.
The "Skin Effect": This is the coolest part. When the traffic is unbalanced, the "eigenvectors" (which you can think of as the traffic density maps) stop spreading out evenly. Instead, they all pile up and stick to one specific side of the highway, like a crowd of people huddling against a wall. In physics, this is called the Non-Hermitian Skin Effect. The paper shows that this "crowding" is a very sensitive way to detect if your ramps are biased toward red or blue cars.

2. The Magnetic "Weather" (Zeeman Fields)

The researchers also tested what happens if they change the "weather" of the city using magnetic fields.

Sideways Wind (In-Plane Field): Imagine a wind blowing across the highway. It mixes up the red and blue cars, making them swap lanes. This slows down the traffic and makes it messy, but it remains fair. Red cars can still go as easily as blue cars. The "crowding" (Skin Effect) doesn't happen.
Up/Down Wind (Out-of-Plane Field): Now imagine a wind blowing straight up or down. This wind pushes the red cars off the road entirely, leaving only the blue cars to drive.
- The Result: Now you have a one-way street! The traffic becomes wildly unbalanced. The "crowding" (Skin Effect) appears instantly.
- The Lesson: Simply breaking the symmetry (having a magnetic field) isn't enough to create this weird "skin" effect. The field has to push the cars in a way that unbalances the two directions.

3. The "Noise" Test (Disorder)

Finally, they asked: "What if the road is bumpy or broken?"

If the road is bumpy in a way that mixes red and blue cars randomly, it acts like the "Sideways Wind." It destroys the special "crowding" effect and makes the traffic fair again, even if the total amount of traffic drops.
However, if you already had a biased ramp (only red cars allowed), adding a little bit of "bumpiness" (disorder) can actually fix the fairness, making the system behave normally again.

Why Does This Matter?

Think of this like a diagnostic tool for a doctor.

In the past, if you wanted to know if a material had "spin-polarized" edges (where red and blue cars are separated), you needed complex, expensive equipment.
This paper says: "No, just look at the traffic flow!" If you see the traffic piling up on one side of the highway (the Skin Effect), you know immediately that your system is biased.
It turns out that measuring this "piling up" is actually more sensitive than just measuring how much current flows. It's like noticing a single person standing in a corner tells you more about a room's atmosphere than counting how many people are in the room.

Summary

The paper reveals that in these quantum cities, directional imbalance is the key to unlocking a strange new state of matter. By tweaking how cars enter the highway or how the magnetic wind blows, you can force the traffic to stick to one side. This "Non-Hermitian Topology" isn't just a math trick; it's a practical, highly sensitive way to measure the invisible spin of electrons, helping us build better future electronics.

1. Problem Statement

While non-Hermitian topology has been extensively studied in systems with broken time-reversal symmetry (TRS), such as Quantum Hall and Chern insulators, its emergence in time-reversal-symmetric topological systems remains underexplored. Specifically, it is unclear whether Quantum Spin-Hall (QSH) insulators—characterized by helical edge states with opposite spins propagating in opposite directions—can exhibit non-Hermitian topological features in their multi-terminal transport properties without explicitly breaking TRS in the scattering region.

The central questions addressed are:

Can spin-selective coupling at contacts alone induce a non-Hermitian conductance matrix in a QSH system?
How do external perturbations (Zeeman fields) and disorder modify the non-Hermitian topology?
Can non-Hermitian signatures serve as a diagnostic tool for detecting the intrinsic spin polarization of edge states?

2. Methodology

The authors employ a combination of analytical modeling and numerical transport simulations:

Model: The system is described by the Bernevig–Hughes–Zhang (BHZ) model, a four-band tight-binding model for HgTe/(Hg,Cd)Te quantum wells. The Hamiltonian consists of two time-reversed sectors coupled via spin-orbit interaction.
Device Geometry: A square-shaped scattering region ( $L \times L$ ) is connected to $N$ semi-infinite leads arranged clockwise along the perimeter.
Transport Formalism: Simulations use the Landauer–Büttiker formalism at zero Fermi energy ( $E=0$ ) via the kwant package. The multi-terminal conductance matrix $G_{ij}$ is calculated, where $G_{ij} \propto (\delta_{ij}N_i - T_{ij})$ .
Perturbations:
- Spin-Polarized Leads: Achieved by shifting the onsite energy of the spin-up block in the leads ( $\mu_\uparrow \neq \mu_\downarrow$ ), creating a spin-selective injection/collection mechanism.
- Zeeman Fields: Both in-plane ( $B_x$ ) and out-of-plane ( $B_z$ ) magnetic fields are applied to break TRS and modify the edge state dispersion.
- Disorder: Two types are introduced:
  1. Spin-independent disorder: Preserves helical structure but introduces sample fluctuations.
  2. Spin-mixing disorder: Locally mixes helical partners, breaking the protection of edge states.
Non-Hermitian Diagnostics:
- Conductance Asymmetry: $\langle \Delta G \rangle = |G_{12} - G_{21}|$ .
- Polar-Decomposition Invariant ( $w_{PD}$ ): A topological invariant derived from the polar decomposition of the conductance matrix, distinguishing trivial ( $w=0$ ) from non-trivial ( $w=\pm 1$ ) phases.
- Non-Hermitian Skin Effect (NHSE): Analyzed via the Summed Probability Density (SPD) of the right eigenvectors of $G$ . Exponential accumulation of eigenvectors at the boundaries indicates the skin effect.

3. Key Contributions and Results

A. Emergence of Non-Hermitian Topology via Spin-Selective Coupling

Mechanism: The authors demonstrate that breaking TRS in the scattering region is not required for non-Hermitian topology. Instead, spin-selective coupling at the contacts is sufficient.
Result: When leads couple preferentially to one spin sector (e.g., $\mu_\uparrow \gg \mu_\downarrow$ ), the conductance matrix becomes non-Hermitian ( $G_{ij} \neq G_{ji}$ ), even though the underlying Hamiltonian remains Hermitian and TRS is preserved in the bulk.
Topology: This asymmetry drives the system into a non-trivial topological phase characterized by:
- Directional transport imbalance.
- Exponential localization of the conductance matrix's right eigenvectors at the boundaries (NHSE).
- A non-zero polar-decomposition invariant ( $w_{PD} \approx \pm 1$ ).

B. Zeeman Fields as a Probe of Edge-State Polarization

The study reveals a critical distinction between in-plane and out-of-plane magnetic fields:

In-Plane Field ( $B_x$ ): Breaks TRS and mixes helical partners, suppressing quantized conductance. However, it preserves reciprocity ( $G_{12} = G_{21}$ ) and the system remains topologically trivial (no skin effect).
Out-of-Plane Field ( $B_z$ ): Acts along the spin quantization axis of the edge states. It lifts the Kramers degeneracy, pushing one helical branch away from the Fermi level while leaving the other intact. This creates a directional imbalance between counter-propagating modes, inducing a non-Hermitian skin effect and non-reciprocal transport.
Significance: The orientation-dependent response proves that the emergence of non-reciprocity is directly linked to the intrinsic spin polarization of the helical edge states.

C. Sensitivity of Diagnostics

The paper compares standard conductance asymmetry ( $\Delta G$ ) with eigenvector-based metrics (SPD and $w_{PD}$ ).
Finding: The Non-Hermitian Skin Effect (SPD) and the polar-decomposition invariant are significantly more sensitive to small changes in lead polarization or Zeeman fields than the conductance elements themselves. They provide a sharper, more disorder-resilient probe for detecting spin-selective coupling.

D. Impact of Spin-Mixing Disorder

For unpolarized leads, spin-mixing disorder suppresses conductance but preserves reciprocity and trivial topology (similar to $B_x$ ).
For polarized leads, spin-mixing disorder competes with the spin-selective coupling. As disorder strength increases, it mixes the counter-propagating modes, washing out the directional imbalance. This drives a crossover from a non-trivial non-Hermitian phase back to a trivial phase, effectively suppressing the skin effect.

4. Significance and Implications

Theoretical Insight: The work clarifies that non-Hermitian topology in QSH transport is governed by directional imbalances between helical edge modes, which can arise from contact engineering (spin-selectivity) or intrinsic edge-state polarization, rather than just broken TRS in the bulk.
Experimental Utility: It establishes the Non-Hermitian Skin Effect and related eigenvector diagnostics as powerful, practical tools for:
1. Estimating the spin polarization of contacts in QSH devices.
2. Probing the intrinsic spin polarization of helical edge states via orientation-dependent Zeeman responses.
3. Detecting the presence of spin-mixing mechanisms (disorder or interactions).
Unified Framework: The study provides a unified framework for analyzing spin-dependent transport in topological insulators using multi-terminal conductance matrices, bridging the gap between Hermitian topological physics and non-Hermitian phenomena.

In summary, the paper redefines the conditions for non-Hermitian topology in time-reversal-symmetric systems, showing that contact-induced spin selectivity and specific magnetic field orientations can induce robust non-Hermitian topological phases, offering new diagnostic pathways for quantum spin-Hall devices.