🔬 optics

Classification of Fragile Topology Enabled by Matrix Homotopy

This paper introduces a Z2 energy-resolved topological marker based on matrix homotopy and position-space descriptions to classify fragile topology in finite, disordered, and aperiodic systems, demonstrating its robustness and applicability in models like twisted bilayer graphene and photonic crystals.

Original authors: Ki Young Lee, Stephan Wong, Sachin Vaidya, Terry A. Loring, Alexander Cerjan

Published 2026-10-14

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Ki Young Lee, Stephan Wong, Sachin Vaidya, Terry A. Loring, Alexander Cerjan

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 materials science, certain patterns in how electrons move through a solid can create states of matter that are remarkably robust. For decades, scientists have classified these states based on "strong" topology. Imagine a knot tied in a string; no matter how you wiggle the string, the knot remains unless you cut the string. Similarly, in these strong topological materials, special conducting states appear at the edges or surfaces, protected by the material's internal symmetry. These edge states are immune to small imperfections or disorder, making them highly desirable for future technologies like quantum computing. However, nature has revealed a more subtle class of materials where the "knot" is not quite as secure. These are known as fragile topological phases. In these systems, the special topological features can be undone or "trivialized" simply by adding a layer of ordinary, non-topological material. It is as if the knot could be untied not by cutting the string, but by weaving in a loose, extra strand of thread. This fragility makes them difficult to identify, especially in real-world materials that are finite in size, disordered, or lack the perfect, repeating patterns found in ideal crystals.

A team of researchers has now developed a new way to find and classify these fragile phases, even in messy or imperfect materials. Instead of looking at the material's behavior in the abstract realm of momentum, which requires perfect order, they created a method that works directly with the material's physical layout in space. By translating the physical symmetries that protect these phases into a specific mathematical language involving matrices, they constructed a new tool called a topological marker. This marker acts like a detector that can scan a finite piece of material, whether it is disordered or aperiodic, and determine if it possesses fragile topology. The method provides a clear "yes" or "no" answer, represented by a simple value of either positive or negative one, and it also offers a quantitative measure of how robust that topology is against disturbances.

The researchers tested this new framework on two very different systems to prove its versatility. First, they applied it to a model of twisted bilayer graphene, a material made of two layers of carbon atoms stacked at a slight angle. In this system, they introduced disorder by randomly varying the way electrons hop between atoms, simulating the imperfections found in real samples. Their simulations showed that the fragile topological phase could survive even under strong disorder. Surprisingly, they also discovered a phenomenon where increasing the disorder strength actually caused the material to switch back into a fragile topological phase after it had become trivial. This "re-entrant" behavior, where disorder restores the topology, would have been nearly impossible to predict using traditional methods that rely on perfect, repeating patterns.

To show that their approach works beyond atomic lattices, the team also applied it to a photonic crystal, a structure designed to control light. This system was a finite block of material surrounded by air, meaning it did not have a clean, global energy gap separating its states. Traditional methods often fail in such gapless environments, but the new position-space marker successfully identified the fragile topology within a specific frequency range. The results confirmed that the topological protection remained strong even without a full spectral gap, and the localized states associated with the topology were found to persist at the boundaries of the structure.

The significance of this work lies in its ability to move beyond the idealized world of infinite, perfect crystals. By focusing on the physical position of atoms and the symmetries of the system's operators, the researchers have provided a tool that can characterize topology in the complex, disordered, and finite systems that experimentalists actually build. This opens the door to identifying fragile topological phases in a wider variety of materials, including twisted graphene, metamaterials, and photonic devices, potentially guiding the design of new technologies that rely on these subtle but powerful quantum effects. The method suggests that fragile topology is not just a theoretical curiosity restricted to perfect models, but a real, observable property that can persist and even re-emerge in the messy reality of the physical world.

Problem Statement
The classification of topological phases of matter has traditionally relied on momentum-space descriptions, such as vector bundles and K-theory, which assign topological indices to band gaps. While "strong" topological phases (e.g., Chern insulators) exhibit robust bulk-boundary correspondence and stability against the addition of trivial bands, a newer class known as "fragile topology" has emerged. Fragile topology, notably identified in the moiré flat bands of small-angle twisted bilayer graphene (TBG), is characterized by topological features that can be trivialized by the addition of trivial atomic bands.

Existing methods for classifying fragile topology, such as Wilson loops and symmetry indicators, are primarily formulated in momentum space. These approaches face significant limitations when applied to experimentally relevant scenarios involving finite-size systems, disorder, aperiodic structures, and heterostructures lacking a global bulk spectral gap. Consequently, there is a need for a classification scheme that operates directly in position space, capable of handling disordered and finite systems while providing a quantitative measure of topological robustness.

Methodology
The authors develop a position-space framework for classifying fragile topology rooted in matrix homotopy. The core of their approach involves translating the physical symmetries protecting fragile phases into mathematical matrix symmetries of the system's Hamiltonian ( $H$ ) and position operators ( $X, Y$ ).

Symmetry Translation: For a system with $C_2T$ symmetry (twofold rotation combined with time-reversal), the authors define a transpose-like matrix operation $\rho$ such that $M^\rho = (C_2T)^{-1} M^\dagger (C_2T)$ . In the physical basis, this operation relates to the system's symmetries as $X^\rho = -X$ , $Y^\rho = -Y$ , and $H^\rho = H$ .
Basis Transformation: To utilize standard matrix homotopy results, the authors introduce a unitary transformation $W = \frac{1}{\sqrt{2}}(C_2 + i\mathbb{1})$ . This transforms the operators into a new basis where the $\rho$ operation becomes the standard matrix transpose ( $\top$ ). In this basis, the transformed position operators $WXW^\dagger$ and $WY W^\dagger$ become Hermitian skew-symmetric, while $WHW^\dagger$ becomes Hermitian symmetric.
Spectral Localizer Construction: The authors construct a spectral localizer $L(x,y,E)$ by tensoring these transformed operators with Pauli matrices:
$L(x,y,E) = \kappa(WXW^\dagger - x\mathbb{1}) \otimes \sigma_x + \kappa(WY W^\dagger - y\mathbb{1}) \otimes \sigma_z + (WHW^\dagger - E\mathbb{1}) \otimes \sigma_y$
Here, $\kappa$ is a scaling coefficient balancing the spectral weight of position and energy.
Topological Marker: At the center of rotation ( $x=y=0$ ), the localizer $L(0,0,E)$ is skew-symmetric. The authors define an energy-resolved $\mathbb{Z}_2$ invariant $\zeta_E$ as the sign of the Pfaffian of this matrix:
$\zeta_E = \text{sign}[\text{Pf}(L(0,0,E))]$
This invariant distinguishes systems based on which atomic limit they can be continuously connected to without closing the local gap.
Quantitative Robustness: The local gap $\mu(x,y,E)$ , defined as the minimum absolute value of the spectrum of $L(x,y,E)$ , serves as a quantitative measure of topological protection. The invariant $\zeta_E$ is guaranteed to remain stable under perturbations $\delta H$ as long as $\|\delta H\| < \mu(0,0,E)$ .

Key Results
The framework was applied to two distinct systems to demonstrate its efficacy:

Twisted Bilayer Graphene (TBG):
- Using a four-band tight-binding model of TBG, the authors confirmed that the energy-resolved marker $\zeta_E$ correctly identifies the fragile topological phase within the bulk band gap.
- The method successfully tracked phase transitions induced by varying the inter-layer hopping phase $\phi$ , identifying gap closings at $\phi = \pi/3$ and $2\pi/3$ .
- Crucially, the framework was applied to disordered TBG models where hopping phases were randomly assigned. The results demonstrated that fragile topology persists even under strong disorder. Notably, the simulations revealed a disorder-induced re-entrant topological phase transition, where the system transitions from a fragile phase to a trivial phase and back to a fragile phase as disorder strength increases. This phenomenon is difficult to capture with momentum-space methods.
Photonic Crystals (PhC):
- The authors applied the method to a 2D photonic crystal heterostructure (6x6 unit cells surrounded by air) designed to possess $C_2T$ symmetry.
- Unlike the infinite bulk system, the finite heterostructure lacks a global bulk spectral gap due to the surrounding air continuum. Despite the absence of a spectral gap in the density of states, the frequency-resolved marker $\zeta_\omega$ successfully identified a range of frequencies exhibiting fragile topology.
- The local gap remained relatively large within this frequency range, indicating topological robustness, and the local density of states (LDoS) showed localization at the boundaries of the PhC.

Significance and Claims
The paper claims to provide an effective tool for elucidating aspects of fragile topology that are inaccessible to conventional momentum-space classifications. By relying on position-space descriptions and matrix homotopy, the proposed scheme:

Enables the direct classification of finite, disordered, and aperiodic materials.
Provides a quantitative measure of topological robustness (via the local gap) that remains valid under finite system sizes, disorder, and environmental perturbations.
Reveals complex phenomena such as disorder-induced re-entrant transitions, which challenge the stability assumptions of fragile phases.
Offers a pathway to identify and characterize fragile topology in systems lacking a bulk spectral gap, such as heterostructures and continuum models.

The authors suggest that this framework offers guidance for potential applications across various experimental platforms, including topological photonics, correlated phases in materials, and metamaterials, particularly where disorder or aperiodicity is inherent.

More like this