🔬 mesoscale physics

Winding-control mechanism of non-Hermitian systems

This paper introduces a winding-control mechanism using conditional boundary conditions to selectively collapse specific periodic boundary spectra onto their open boundary counterparts based on winding numbers, thereby revealing a composite reconstruction of the Brillouin and generalized Brillouin zones and establishing a link between residual imaginary velocity and non-Hermitian topological transitions.

Original authors: Yongxu Fu, Yi Zhang

Published 2026-05-18

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Yongxu Fu, Yi Zhang

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 quantum system not as a solid block of matter, but as a busy highway for tiny particles. In the strange world of "non-Hermitian" physics (which describes systems that exchange energy with their surroundings, like light in a laser or sound in a crystal), these highways behave very differently depending on how you close the loop.

Here is the core idea of the paper, broken down into simple concepts:

1. The Two Types of Highways: The Loop vs. The Dead End

Usually, physicists study these systems in two ways:

The Loop (Periodic Boundary Conditions): Imagine the highway is a giant circle. If a car drives off the right edge, it instantly reappears on the left. In this "loop" world, the traffic flows in a specific direction, creating a swirling pattern. The paper calls this a spectral loop.
The Dead End (Open Boundary Conditions): Now, imagine cutting the circle open. The highway has a start and a finish. If a car hits the end, it stops or piles up. In non-Hermitian systems, this causes a "traffic jam" where almost all the cars (particles) crash into one specific wall. This is known as the Skin Effect.

2. The Problem: You Can't Pick and Choose

In the past, if you wanted the "loop" behavior, you had to close the system. If you wanted the "dead end" behavior, you had to open it. You couldn't have a system where part of the highway was a loop and another part was a dead end. The whole system had to be one or the other.

3. The Solution: The "Winding Control" Switch

The authors of this paper discovered a way to act like a traffic engineer with a magical switch. They found that the "swirling" nature of the loop (called the winding number) is tied to a hidden "imaginary speed" of the particles.

They introduced a new type of boundary condition called Conditional Boundary Conditions (CBCs). Think of this as installing a one-way gate at the end of the highway.

The Right-Permissible Gate: This gate lets cars exit to the right but blocks them from entering from the left.
The Left-Permissible Gate: This gate lets cars exit to the left but blocks them from entering from the right.

4. How the Magic Works: Sorting the Traffic

Here is the clever part: The system naturally sorts itself based on which way the "swirl" is going.

If a section of the highway has a "swirl" that wants to go right, the Right-Permissible gate lets it keep flowing in a loop. It stays a loop.
If a section of the highway has a "swirl" that wants to go left, the Right-Permissible gate blocks it. That section is forced to stop, turning into a dead end (a skin effect).

The Result: You can now take a single system and split it in half. One half behaves like a swirling loop, and the other half behaves like a crashing pile-up at the wall. You are effectively collapsing specific parts of the loop into dead ends while leaving the rest intact.

5. The "Bloch Points": The Border Crossings

Where the loop turns into a dead end, there is a specific meeting point the authors call a Bloch point. Imagine this as a border crossing between two countries. On one side, the traffic flows in a circle; on the other, it piles up against the wall. The paper shows that these points are the precise boundaries where the behavior changes.

6. Stretching and Shifting the Map

The authors also showed they can use mathematical "stretching" (similarity transformations) to move these border crossings around.

Imagine the highway map is printed on a rubber sheet. By stretching the sheet, you can move the "loop" section and the "dead end" section closer together or further apart, or even change where the border crossing happens, without changing the fundamental rules of the road.

Summary Analogy

Think of a river that naturally flows in a giant circle.

Old Way: You could either let the river flow in a circle, or you could dam it up so the water piles up at one end.
New Way (This Paper): You build a special, smart dam that only stops the water if it's trying to flow in a specific direction.
- If the water tries to flow clockwise, the dam opens, and it keeps circling.
- If the water tries to flow counter-clockwise, the dam closes, and the water piles up at the wall.

This allows the scientists to create a river that is half-circle and half-pile-up simultaneously, controlled entirely by the direction the water wants to go. This gives them a powerful new tool to design and control how energy and particles move in these exotic quantum systems.

Technical Summary: Winding-Control Mechanism of Non-Hermitian Systems

Problem Statement
Non-Hermitian quantum systems exhibit unique spectral, topological, and boundary-sensitive features, most notably the non-Hermitian skin effect (NHSE), where extensive eigenstates accumulate at boundaries under open boundary conditions (OBCs) despite being extended under periodic boundary conditions (PBCs). This sensitivity challenges conventional bulk-boundary correspondences, necessitating the generalized Brillouin zone (GBZ) formalism to characterize OBC spectra. While the relationship between spectral windings, the GBZ, and boundary conditions is established, the underlying mechanism for targeted control of these boundary sensitivities—particularly in systems with multiple PBC spectral loops—remains obscure. Existing methods lack a systematic framework to selectively collapse specific segments of PBC spectra onto their OBC counterparts based on topological invariants.

Methodology
The authors propose a "winding-control mechanism" grounded in the interplay between spectral winding numbers, imaginary velocity, and boundary conditions. The core methodology involves:

Conditional Boundary Conditions (CBCs): The authors introduce CBCs with exclusive hopping (e.g., Right-Permissible or Left-Permissible) that selectively suppress hopping across boundaries in one direction. This acts as a filter for the spectral winding number ( $W_s$ ).
Spectral-Winding-Velocity Relations: The mechanism relies on the connection between the spectral winding number $W_s$ and the residual imaginary velocity $\text{Im}(\bar{v})$ of the Fermi sea. Under PBCs, a non-zero $W_s$ corresponds to a non-zero $\text{Im}(\bar{v})$ , allowing for worldline winding. Under OBCs, $W_s$ and $\text{Im}(\bar{v})$ must vanish. CBCs are shown to act as transition boundaries that force specific spectral segments to adhere to either PBC or OBC constraints based on their winding sign.
Composite Reconstruction: The approach involves a composite reconstruction of the Brillouin zone (BZ) and GBZ. "Bloch points," defined as intersections between the PBC BZ and OBC GBZ, serve as junctions where the system switches between PBC and OBC spectral segments.
Holomorphic Mappings and Similarity Transformations: To extend control beyond binary selection, the authors incorporate similarity transformations and general holomorphic mappings. These transformations modify the relative geometry of the BZ and GBZ, allowing for the manipulation of Bloch point locations and the creation of new spectral configurations.
Model Verification: The theory is demonstrated numerically using various 1D non-Hermitian lattice models, including a paradigmatic single-band model, the Hatano-Nelson (HN) model, and multi-loop systems.

Key Contributions and Results

Selective Spectral Collapse: The paper demonstrates that CBCs can selectively collapse specific pieces of a PBC spectrum (characterized by a specific winding number $W_s$ ) onto their OBC counterparts. Segments with $W_s$ compatible with the CBC remain as PBC loops, while incompatible segments collapse into OBC arcs with $W_s = 0$ .
Hybrid Eigenstate Behavior: The resulting eigenstates exhibit nontrivial localization properties determined by the hybrid spectral structure. Eigenstates corresponding to collapsed OBC segments display skin effects (localized at boundaries), while those corresponding to preserved PBC segments remain extended. This behavior cannot be understood by analyzing the BZ or GBZ in isolation.
Bloch Points as Junctions: The study identifies Bloch points as critical boundaries separating domains of opposite winding numbers. These points facilitate smooth transitions between PBC and OBC spectral segments and determine the localization properties of the eigenstates.
Enhanced Control via Mappings: By applying similarity transformations (e.g., scaling the BZ radius) and holomorphic mappings (e.g., translating the BZ), the authors show that the location of Bloch points and the topology of the spectral loops can be systematically engineered. This allows for the design of complex spectral configurations, including those with multiple loops and disconnected GBZs.
Generalizability: The mechanism is validated across single-band models with multiple spectral loops and multi-band systems. The authors note that for multi-band systems with $Z_2$ skin effects (where net winding is zero), the control applies on a band-resolved basis.

Significance and Claims
The paper claims to provide a fundamental theoretical framework for controlling non-Hermitian topology and spectra. By establishing a direct link between the imaginary velocity of the Fermi sea, spectral windings, and boundary conditions, the authors offer a "winding-control" theory that enriches the understanding of non-Hermitian physics across spectrum, topology, and bulk-boundary correspondence.

The authors emphasize that this mechanism provides a versatile and systematic approach to manipulating non-Hermitian spectral structures, moving beyond the binary distinction of PBC vs. OBC to a hybrid regime. They propose that this formalism could be extended to interacting many-body systems, where imaginary velocity and worldline winding remain physical quantities. Furthermore, the paper suggests a potential experimental realization in acoustic crystals using directional amplifiers to tailor boundary hoppings, though this is presented as a proposal for future investigation rather than a completed experimental result. The work aims to bridge the gap between theoretical spectral topology and practical control over non-Hermitian boundary phenomena.