🔢 mathematics

Mobility-Gap Robustness and Středa-Hall Separation in Disordered Axion Pumping

This paper demonstrates that the noncommutative second Chern number of a $(3+1)$ D disordered axion pump remains quantized well into the mobility-gap regime beyond global spectral gap closure, whereas the Středa-Hall response rapidly fails, thereby establishing a robust distinction between true bulk topological protection and finite-size artifacts.

Original authors: Bryan Leung, Emil Prodan

Published 2026-09-24

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Bryan Leung, Emil Prodan

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 quiet, invisible world of solid materials, electrons do not always behave like tiny, independent particles bouncing around a room. Under the right conditions, they move in coordinated, collective patterns that give the material a hidden, rigid structure. This structure is not made of atoms, but of mathematical rules that dictate how electricity flows. For decades, physicists have known that if a material is perfectly ordered, these rules create a "spectral gap," a forbidden zone of energy where no electrons can exist. This gap acts like a safety fence, protecting the material's special electrical properties from being scrambled by minor imperfections. However, real materials are never perfect; they are always filled with disorder, like a room where the furniture has been slightly knocked askew. The big question has been whether these delicate mathematical protections can survive when the safety fence itself disappears, leaving the electrons to wander through a chaotic, disordered landscape.

The specific puzzle this research tackles involves a theoretical machine called an "axion pump." Imagine a material that is slowly and smoothly changing its internal state over time, cycling through different configurations in a loop. In a perfect world, this cycle would act like a pump, moving a precise, quantized amount of electrical charge from one side of the material to the other, a process that is robust and unchangeable. This phenomenon is the dynamic version of a static "axion insulator," a material known for its strange ability to convert electric fields into magnetic ones. While scientists have studied these materials in their static, frozen states, they have never been sure what happens when you set them in motion and then introduce the inevitable disorder found in real-world samples. Does the pump keep working, delivering its exact charge, or does the chaos break the mechanism?

A team of researchers set out to answer this by running massive, high-resolution computer simulations of a three-dimensional axion pump. They built a digital model of a material made of stacked layers, similar to a deck of cards, where the connections between the layers change rhythmically as time passes. To mimic the messy reality of the physical world, they introduced random disorder into the model, essentially shaking the deck of cards and scrambling the energy levels of the electrons. They then tracked two specific things as they increased the amount of disorder: the total amount of charge pumped by the cycle, and a related measurement that describes how the surface of the material reacts to magnetic fields.

The results revealed a surprising and distinct separation between these two behaviors. As the researchers cranked up the disorder, they first reached a point where the energy gap that usually protects the system completely vanished. In a one-dimensional version of this problem, this would have been the end of the line; the pump would have stopped working, and the precise charge delivery would have broken down immediately. But in this three-dimensional system, something different happened. Even after the energy gap closed, the total charge pumped by the cycle remained locked at its precise, quantized value. The pump continued to work perfectly, delivering the exact same amount of charge, even though the "safety fence" was gone.

However, the story was not the same for the second measurement. The response of the material's surface to magnetic fields, which is usually tied to the pumping action, began to drift away from its expected value the moment the energy gap closed. This divergence acted as a crucial control, proving that the system was not just a computer glitch or a trick of the simulation. It showed that the core property of the pump—the ability to move charge—was being protected by a different kind of stability. In physics, this is called a "mobility gap." While the energy gap was gone, the electrons in the bulk of the material were still stuck in place, unable to move freely through the disorder. This lack of movement was enough to preserve the pump's precision, even as the surface behavior began to unravel.

The researchers mapped out exactly where this protection ended. They found that the pump remained robust up to a specific level of disorder, which they calculated to be around 18 units of randomness. Beyond this point, the electrons finally broke free from their localized spots, the pump lost its precision, and the system became a normal, disordered insulator. This critical threshold was determined by the weakest link in the cycle, a specific moment in the pump's rhythm where the electrons were most likely to escape their confinement. The study confirmed that the pump's success relied on the electrons being trapped in place, not on the existence of a clean energy gap.

This finding is significant because it challenges the long-held assumption that a clean energy gap is strictly necessary for these topological effects to survive. The research suggests that in three dimensions, the sheer inability of electrons to move through a disordered medium is enough to protect the material's most fundamental electrical signatures. It draws a sharp line between the behavior of one-dimensional systems, where disorder quickly destroys the effect, and three-dimensional systems, where the electrons can remain localized and the pump can keep working. The work provides strong numerical evidence that the quantized magnetoelectric effects predicted for these materials are far more robust than previously thought, surviving well into the chaotic regime where the energy gap no longer exists.

Technical Summary: Mobility-Gap Robustness and Stˇreda-Hall Separation in Disordered Axion Pumping

Problem Statement
The paper addresses a fundamental question in the theory of topological phases: whether the quantization of a dynamically pumped Chern number persists in the presence of strong disorder after the global spectral gap has closed. While static noncommutative Chern numbers are known to be protected by mobility gaps (regions where states remain localized despite the absence of a spectral gap), it remains unclear if this protection extends to adiabatic cycles, specifically axion pumps. Previous work on one-dimensional (1D) disordered pumps indicates that quantization breaks down immediately upon spectral gap closure, even if instantaneous states remain localized. The authors investigate whether the distinct behavior of three-dimensional (3D) systems, where mobility edges separate localized and extended states, allows for the robustness of the second Chern number ( $Ch_2$ ) in a (3+1)D axion pump beyond the spectral-gap regime.

Methodology
The authors employ large-scale real-space computations to evaluate the noncommutative second Chern number and the generalized Stˇreda-Hall response under strong disorder.

Model: A (3+1)D axion pump model consisting of a stack of bilayer Haldane sheets along the $z$ -axis with $\tau$ -dependent interlayer coupling. Anderson disorder is introduced via on-site potentials drawn uniformly from $[-W/2, W/2]$ .
Topological Invariants:
- The second Chern number ( $Ch_2$ ) is calculated using the Kernel Polynomial Method (KPM) with Chebyshev expansions of the Fermi projector $P$ . The trace per unit volume is approximated via disorder averaging over local matrix elements.
- The Stˇreda-Hall response ( $\partial_\phi Ch_{0,3}$ ), representing the change in the pumped weak Chern invariant per unit transverse flux, is computed by inserting a uniform flux via Peierls phases.
Scaling and Analysis:
- System sizes up to $L=60$ ( $N \approx 864,000$ orbitals) and KPM orders up to $M=4096$ are used to ensure convergence.
- Level statistics (adjacent-gap ratio $\langle r \rangle$ ) are analyzed to identify Anderson transitions and locate mobility edges.
- Finite-size scaling and extrapolation in $1/M$ are performed to distinguish thermodynamic behavior from finite-size artifacts.

Key Contributions and Results

Separation of $Ch_2$ and Stˇreda-Hall Response: The study demonstrates a clear separation between the bulk invariant $Ch_2$ and the Stˇreda-Hall response $\partial_\phi Ch_{0,3}$ after the global spectral gap closes at $W_g \approx 14.5$ .
- $Ch_2$ remains quantized near $-1$ well into the mobility-gap regime, converging toward $-1$ up to disorder strengths of $W \lesssim 16.5$ .
- In contrast, $\partial_\phi Ch_{0,3}$ departs rapidly from the quantized plateau immediately after $W_g$ , serving as an internal control that distinguishes bulk mobility-gap protection from the breakdown of the gap-dependent Stˇreda identity.
Identification of Critical Disorder and Mobility Bottleneck:
- Level statistics identify a 3D unitary Anderson transition at each pump slice.
- The critical disorder strength for the cycle is determined by the "mobility-edge bottleneck" at $\tau = 0$ , where delocalization occurs first. The critical disorder is estimated at $W_c \approx 18$ .
- The topological correction length $\xi_{top}$ , which measures the scale over which finite systems feel the approaching mobility edge, grows rapidly as $W$ approaches $W_c$ , exceeding the maximum system size ( $L=60$ ) near $W \approx 17$ .
Dimensional Contrast: The results provide numerical evidence that the (3+1)D axion pump behaves fundamentally differently from the (1+1)D disordered pump. While the 1D case shows immediate breakdown of quantization upon gap closure, the 3D system retains quantization in the mobility-gap regime, provided the disorder does not exceed the critical threshold set by the mobility bottleneck.

Significance and Claims
The paper claims to provide direct numerical evidence for the mobility-gap robustness of axion pumping in (3+1)D systems. This finding suggests that the quantized magnetoelectric coupling in 3D symmetry-protected topological insulators may remain robust beyond the spectral-gap regime, a long-standing open question.

The authors emphasize that their results do not constitute a rigorous mathematical proof of quantization in the mobility gap for this specific dynamical system. They note that while the link between $Ch_2$ and the index of a Fredholm operator exists for static systems, establishing this link for the axion pump requires proving that the derivatives of the Fermi projector belong to a Sobolev space—a condition that may require new techniques due to the absence of disorder in the temporal ( $\tau$ ) direction.

Ultimately, the work highlights that the breakdown of the generalized Stˇreda relation (indicated by the divergence of $\partial_\phi Ch_{0,3}$ ) does not necessarily imply the loss of bulk topological quantization ( $Ch_2$ ) in the presence of a mobility gap, distinguishing the fate of dynamically pumped topological invariants from their static counterparts in lower dimensions.

Technical Summary: Mobility-Gap Robustness and Stˇreda-Hall Separation in Disordered Axion Pumping

More like this