Disorder-induced crossover from phase-averaging to mode-mixing regimes in magnetic domain walls of a second-order topological insulator

This paper investigates electronic transport across magnetic domain walls in 3D second-order topological insulators under Anderson disorder, revealing a disorder-induced crossover from a phase-averaging regime to a mode-mixing regime characterized by distinct two-step plateaus in conductance fluctuations and Fano factors.

The Gist

Imagine you have a tiny, magical highway made of a special material called a Second-Order Topological Insulator. In this material, electricity doesn't flow through the middle like water in a pipe; instead, it travels strictly along the sharp edges and corners of the wire, like cars driving on a very specific, invisible track.

Now, imagine there is a "traffic jam" in the middle of this highway called a Magnetic Domain Wall. This is a place where the magnetic direction flips. In a perfect, clean world, this wall creates a special loop: two lanes of traffic run side-by-side around the wall, forming a perfect circle.

The Perfect Race (The Clean Limit)

In a perfectly clean system, these two lanes act like a race track. The cars (electrons) take both paths at the same time. Because they are waves, they interfere with each other. If you change the magnetic field (like changing the wind), the race results change in a smooth, predictable, wavy pattern. It's like a perfectly synchronized dance.

Enter the Chaos: Disorder

Now, imagine we start throwing obstacles onto the track. Maybe we drop some pebbles, or the road gets bumpy. This is what physicists call disorder.

The paper investigates what happens to our electrical traffic when the road gets messy. Surprisingly, the average amount of electricity that gets through settles down to a very specific, "half-way" value (50% of the maximum). But the way the traffic behaves changes dramatically depending on how messy the road gets.

The authors discovered that as the road gets messier, the system goes through two distinct phases, like two different types of traffic jams:

Phase 1: The "Confused Runner" (Phase-Averaging Regime)

The Analogy: Imagine two runners on a track. The road is a bit bumpy, so one runner gets delayed by a pebble, and the other gets delayed by a puddle. They are still running in their own lanes, but they arrive at the finish line at slightly different times.

• What happens: The "bumpiness" (disorder) scrambles the timing (phase) of the runners. Sometimes they arrive together, sometimes apart.
• The Result: Because the timing is random, the "wavy" dance pattern disappears. Instead, the traffic flow averages out to that steady 50%.
• The Signature: If you look closely at the traffic statistics, you see a specific "U-shape" pattern. It's like saying, "Most of the time, the traffic is either very fast or very slow, but rarely in the middle." This is the Phase-Averaging Regime.

Phase 2: The "Crowded Marketplace" (Mode-Mixing Regime)

The Analogy: Now, imagine the road gets so messy that the lanes disappear entirely. The runners can't stay in their lanes anymore. They start bumping into each other, swapping places, and running in all directions. The two separate lanes have merged into one giant, chaotic crowd.

• What happens: The electrons aren't just delayed; they are completely mixed up. The "lane" structure is destroyed.
• The Result: The traffic flow is still 50% on average, but the fluctuations (the ups and downs) change. The "U-shape" disappears and becomes a flat line. It's like saying, "Anything can happen; the traffic is completely random." This is the Mode-Mixing Regime.

How Do We Know the Difference?

The paper is brilliant because it shows that just measuring the average traffic (conductance) isn't enough to tell these two phases apart—they both look like 50%. You need to look at the noise or the fluctuations.

The authors found two "fingerprints" to tell the difference:

1. The Fluctuation Plateau: In the "Confused Runner" phase, the traffic noise settles at a specific level (about 0.35). In the "Crowded Marketplace" phase, it drops to a different level (about 0.29).
2. The Fano Factor: This is a fancy way of measuring how "bumpy" the current is. It jumps from 1/4 in the first phase to 1/3 in the second.

Why Does This Matter?

Think of this like a new way to tune a radio.

• Old way: You just turn the knob to get a signal.
• New way: By intentionally adding "noise" (disorder) to the material, we can switch the device from one operating mode to another.

This discovery suggests that engineers could design future electronic devices (like super-fast switches or sensors) that use controlled disorder as a feature, not a bug. By tweaking how "messy" the material is, we can control how electricity flows, turning a device from a "phase-averaging" mode to a "mode-mixing" mode.

In short: The paper shows that when you mess up a perfect quantum highway, the traffic doesn't just get worse; it changes its personality. First, it gets confused about timing, and then, if you mess it up enough, it becomes a chaotic crowd. And we now have the tools to tell the difference.

Technical Summary

1. Problem Statement

The paper investigates electronic transport across a magnetic domain wall (DW) in a three-dimensional (3D) second-order topological insulator (SOTI) subject to Anderson disorder.

• Context: In clean 3D SOTIs, magnetic DWs host four co-propagating one-dimensional (1D) topological edge states (TESs) that form a closed loop, acting as an effective Aharonov-Bohm (AB) interferometer. This results in sinusoidal conductance oscillations dependent on magnetic flux.
• The Challenge: While disorder is known to suppress AB oscillations and induce a half-quantized conductance plateau ($0.5 e^2/h$), the underlying physical mechanisms driving this phenomenon in different disorder regimes have often been treated separately. Specifically, the transition between Phase-Averaging Regime (PAR) (where disorder randomizes phases but preserves channel integrity) and Mode-Mixing Regime (MMR) (where disorder causes complete scattering and mixing between channels) has not been unified within a single platform.
• Goal: To identify and characterize the disorder-induced crossover from PAR to MMR in a 3D SOTI DW, distinguishing these regimes using higher-order conductance statistics beyond the average conductance.

2. Methodology

The authors employed a combination of tight-binding modeling, non-equilibrium Green's function (NEGF) transport calculations, and analytical theoretical derivations.

• Model: A 4-band tight-binding Hamiltonian on a cubic lattice was used to model a 3D SOTI nanowire with a magnetic DW.
  • System: A nanowire with diagonal magnetization ($\pm \mathbf{n}_{110}$) creates a surface gap, leaving 1D topological hinge states (THSs). The magnetic DW reverses the magnetization, creating four TESs that connect the THSs into an AB interferometer.
  • Disorder: Anderson disorder (random on-site potential $U_i$) was introduced in the central scattering region.
• Simulation:
  • Quantum transport was calculated using the NEGF method to determine transmission coefficients and conductance ($G$).
  • Large-scale numerical simulations (averaging over thousands of disorder configurations) were performed to compute ensemble-averaged conductance ($\langle G \rangle$), conductance fluctuations ($\sigma_G$), probability distributions of conductance ($P_G$), and local current density profiles.
  • The Fano factor ($F$) associated with shot noise was calculated to probe higher-order cumulants.
• Theory:
  • PAR Theory: Developed a framework assuming random dynamical phase accumulation along the 1D paths while maintaining unidirectional propagation.
  • MMR Theory: Applied Random Matrix Theory (RMT) for quantum chaotic cavities with broken time-reversal symmetry, assuming complete mixing of transport channels.

3. Key Contributions and Results

A. Discovery of a Two-Step Crossover

The study reveals a distinct disorder-induced crossover characterized by two specific plateaus in the conductance fluctuation ($\sigma_G$), despite the average conductance ($\langle G \rangle$) remaining at the half-quantized value of $0.5 e^2/h$ across all regimes.

1. Phase-Averaging Regime (PAR) - Moderate Disorder:
   • Fluctuation: $\sigma_G$ saturates at a plateau of $\approx 0.35 e^2/h$ (theoretical value $\sqrt{2}/4$).
   • Mechanism: Disorder randomizes the dynamical phase difference ($\Delta \phi$) between the two arms of the interferometer. The TESs remain topologically protected and unidirectional.
   • Statistics: The conductance distribution follows a U-shaped Beta distribution ($B(0.5, 0.5)$).
   • Current Profile: Current remains confined to the hinges/DW interface with a "hollow" spatial profile.
2. Mode-Mixing Regime (MMR) - Strong Disorder:
   • Fluctuation: $\sigma_G$ collapses and settles into a second plateau at $\approx 0.29 e^2/h$ (theoretical value $1/\sqrt{12}$).
   • Mechanism: Strong disorder destroys the 1D nature of the TESs, causing scattering between the arms and into the 3D bulk (complete mode mixing).
   • Statistics: The conductance distribution becomes uniform over $[0, 1]$.
   • Current Profile: Current spreads into the bulk, forming a "diffusive cloud."

B. Theoretical Validation

• Analytical Derivations: The authors derived exact analytical expressions for $\langle G \rangle$ and $\sigma_G$ in the PAR limit, showing perfect agreement with numerical data.
• Fano Factor: The Fano factor ($F$) was computed as a robust experimental signature:
  • PAR: $F = 1/4$.
  • MMR: $F = 1/3$.
  • The evolution of $F$ from $1/4$ to $1/3$ provides a clear metric for identifying the crossover.

C. Spatial Dependence of Disorder

The study investigated the effect of disorder location (bulk vs. surface vs. hinge):

• Bulk and Surface Disorder: Both induce the full crossover from PAR to MMR.
• Hinge Disorder: Only induces the PAR (single plateau at $\sigma_G \approx 0.35$). The 1D nature of the hinge states is preserved even at strong disorder because the scattering is confined to the hinges, preventing the inter-arm mixing required for MMR.

4. Significance

• Unified Framework: This work provides the first unified platform to study and distinguish between phase-averaging and mode-mixing mechanisms within a single topological system.
• Beyond Average Conductance: It demonstrates that the average half-quantized conductance plateau is insufficient to characterize transport regimes. Instead, second-order cumulants (conductance fluctuations and the Fano factor) are essential "fingerprints" for distinguishing between topological protection (PAR) and chaotic mixing (MMR).
• Disorder Engineering: The results suggest that disorder can be engineered (e.g., by controlling the spatial distribution of impurities) to tune electronic transport between distinct regimes, offering new avenues for designing DW-based topological and spintronic devices.
• Experimental Feasibility: The identification of specific values for $\sigma_G$ ($0.35$ vs. $0.29$) and $F$ ($1/4$ vs. $1/3$) provides concrete, measurable targets for experimental verification in 3D SOTI nanowires (e.g., doped Bi$_2$Se$_3$ or Bi$_2$Te$_3$).

In summary, the paper elucidates how increasing disorder transforms a coherent topological interferometer into a chaotic diffusive system, using precise statistical metrics to map the transition from phase-randomization to mode-mixing.