Half-quantized anomalous Hall conductance in topological insulator/ferromagnet van der Waals heterostructures

This study employs first-principles calculations and tight-binding models to investigate the magnetization-induced gap, sidewall states, and half-quantized anomalous Hall conductance in three specific ferromagnet/topological insulator van der Waals heterostructures, while also analyzing the factors that may hinder exact half-quantization in realistic systems.

The Gist

The Big Picture: The "Half-Quantum" Magic Trick

Imagine you have a special kind of material called a Topological Insulator (TI). Think of this material like a chocolate bar with a hard shell and a soft, conductive center.

• The Inside (Bulk): The chocolate inside is an insulator; electricity cannot flow through it.
• The Shell (Surface): The shell is a superhighway where electrons can zip around freely without getting stuck.

In a perfect, untouched chocolate bar, this "superhighway" exists on both the top and the bottom. Because there are two highways running in opposite directions, they cancel each other out in terms of a special magnetic effect called the Hall Effect. It's like two people pushing a car with equal force from opposite sides; the car doesn't move sideways.

The Goal: The scientists wanted to create a situation where only one of these highways is active, creating a "half-quantized" effect. This is a rare and exotic state of physics that proves a deep mathematical theory about how the universe works (called the "Parity Anomaly").

The Recipe: Stacking Magnetic Sandwiches

To get this "half" effect, the researchers built a Van der Waals Heterostructure. In plain English, this is a magnetic sandwich made of ultra-thin layers of different materials stacked on top of each other.

1. The Bread (The TI): They used a thin slice of Bi₂Se₃ (Bismuth Selenide), which is the topological insulator.
2. The Topping (The Ferromagnet): They placed a single layer of a magnetic material (like CrI₃, Cr₂Ge₂Te₆, or MnBi₂Te₄) on top of the TI.

What happens when you add the magnetic topping?
Think of the magnetic layer as a force field or a one-way gate. When it touches the top surface of the TI, it breaks the "time-reversal symmetry."

• The Result: The top surface highway gets a "speed bump" (an energy gap) and stops conducting electricity in the usual way. It becomes a special, one-way street.
• The Bottom Surface: The bottom of the TI is far away from the magnet, so it remains untouched. It stays a normal, two-way highway.

The Experiment: Testing Three Different Toppings

The team tested three different magnetic "toppings" to see which one worked best:

1. Cr₂Ge₂Te₆ (CGT): A magnetic crystal.
2. MnBi₂Te₄ (MBSe): Another magnetic crystal.
3. CrI₃ (CrI): A magnetic crystal made of Chromium and Iodine.

The Findings:
In all three cases, the magnetic topping successfully created a gap on the top surface.

• The Top Surface: Became a "gapped" state that contributes exactly half of the expected quantum Hall effect ($e^2/2h$).
• The Bottom Surface: Remained "gapless" (metallic). It acts like a leaky pipe, letting some electricity flow straight through without contributing to the special magnetic effect.

The "Leaky Pipe" Problem:
In a perfect world, you want only the half-quantum effect. But because the bottom surface is still conducting, it adds a "background noise" (longitudinal resistance).

• Analogy: Imagine you are trying to measure the speed of a single runner (the top surface) in a race. But there's a whole crowd of people walking casually on the track (the bottom surface). You can still see the runner is doing something special, but the crowd makes the total picture a bit messy.
• The Good News: Even with the "crowd" on the bottom, the special "half-quantum" signal from the top is still strong enough to be detected. The math shows it is exactly half of the full quantum value, just as predicted by theory.

The Sidewalls: The "Ghost" Currents

The paper also looked at what happens at the edges of these materials (the "sidewalls").

• Normal Quantum Effect: Usually, the current flows in a tight, invisible line right along the very edge, like a train on a track.
• This "Half" Effect: The researchers found that the current on the sidewalls behaves differently. Instead of a tight line, it spreads out a bit more, decaying slowly into the material like smoke drifting away from a candle.
• Why it matters: These "smoke-like" currents are still spinning in a specific direction (chiral) and are responsible for the half-quantum effect. They are the physical proof that the "Parity Anomaly" is real.

Why Should We Care?

1. Proving a Theory: This confirms a 40-year-old prediction from quantum field theory (the Parity Anomaly) using real, tangible materials.
2. Future Electronics: This is a step toward low-power electronics. Because these currents flow without resistance (mostly), they could be used to build faster, cooler computers.
3. The "Half" is the Key: Achieving a "half-quantized" state is a stepping stone. If we can control this, we might eventually build devices that manipulate information using the geometry of space itself, rather than just moving electrons around.

Summary in One Sentence

The scientists built a magnetic sandwich where the top layer creates a special "half-speed" electrical effect, proving a deep law of physics, even though the bottom layer of the sandwich is still leaking a little bit of electricity.

Technical Summary

1. Problem Statement

The paper addresses the challenge of experimentally realizing the half-quantized anomalous Hall effect (1/2-QAHE), a condensed matter manifestation of the parity anomaly in (2+1)-dimensional quantum field theory.

• Theoretical Context: In a 3D topological insulator (TI), breaking time-reversal symmetry (TRS) on one surface opens a mass gap in the Dirac cone, theoretically yielding a half-quantized Hall conductance ($\sigma_{xy} = \pm e^2/2h$). However, due to the Nielsen-Ninomiya doubling theorem, Dirac cones in 2D appear in pairs. If both top and bottom surfaces of a TI film are magnetized, the contributions sum to an integer QAHE or an axion insulator state.
• Experimental Challenge: Isolating the half-quantized contribution requires magnetizing only one surface while keeping the other gapless. Previous experiments (e.g., Cr-doping or CrGe$_2$Te$_6$ coupling) have observed values near $0.5 e^2/h$ but with non-zero longitudinal conductivity ($\sigma_{xx}$), attributed to metallic states on the gapless bottom surface.
• Open Questions: The nature of the edge currents in these half-quantized systems is unclear. Unlike the exponentially localized edge states of integer QAHE, theory suggests half-quantized systems may exhibit power-law decaying sidewall currents. Furthermore, the impact of the metallic bottom surface on the precise quantization of the Hall conductance needs rigorous theoretical validation.

2. Methodology

The authors employed a multi-scale computational approach combining first-principles calculations with tight-binding modeling:

• Density Functional Theory (DFT):
  • Used the Vienna Ab-initio Simulation Package (VASP) with Projector Augmented Waves (PAW).
  • Exchange-correlation described by PBE-GGA with van der Waals corrections (DFT-D3 with Becke-Johnson damping).
  • Spin-orbit coupling (SOC) was included due to heavy elements.
  • Hubbard $U$ corrections (GGA+U) were applied to localized $d$-electrons in magnetic layers (Mn, Cr).
• Model Construction:
  • Substrate: A 6-quintuple-layer (6QL) Bi$_2$Se$_3$ thin film acting as the TI.
  • Ferromagnetic (FM) Overlayers: Three distinct 2D van der Waals magnets were interfaced with the top surface of the Bi$_2$Se$_3$:
    1. Cr$_2$Ge$_2$Te$_6$ (CGT)
    2. MnBi$_2$Se$_4$ (MBSe)
    3. CrI$_3$ (CrI)
• Transport & Topological Analysis:
  • Constructed maximally localized Wannier functions (MLWFs) to generate real-space tight-binding Hamiltonians.
  • Calculated Anomalous Hall Conductivity (AHC) using the intrinsic Berry curvature formalism (WannierBerri).
  • Computed Layer-Resolved Chern Numbers to spatially map topological contributions.
  • Analyzed nanoribbon geometries (20 nm width) to study sidewall state localization and decay.

3. Key Contributions & Results

A. Electronic Structure and Gap Opening

The study confirmed that all three FM/TI heterostructures successfully break TRS locally on the top surface:

• Gap Formation: A magnetic exchange gap opens at the Dirac point of the top surface states (ranging from ~45 meV for CGT to similar magnitudes for others), while the bottom surface states remain gapless and Dirac-like.
• Interface Dependence: The magnitude of the gap and the Fermi level shift depend on the specific chemical interface (e.g., Se-Te contact in CGT vs. Se-Se in MBSe). The CGT interface showed strong hybridization, while CrI$_3$ (with a larger interlayer distance) showed weaker interaction but still induced a gap.

B. Half-Quantized Anomalous Hall Conductance

• Quantization: In all three systems, when the Fermi energy ($E_F$) is positioned within the exchange gap of the top surface, the calculated AHC ($\sigma_{xy}$) is exactly half-quantized ($\approx 0.5 e^2/h$).
• Role of Bottom Surface: The gapless bottom surface provides a parallel metallic channel, resulting in non-zero longitudinal conductivity ($\sigma_{xx}$). Crucially, the authors demonstrate that this dissipative channel does not destroy the half-quantization of the transverse response; the topological contribution remains robust.
• Berry Curvature: The dominant contribution to AHC arises from sharp peaks in the Berry curvature near the $\Gamma$-point, driven by interband mixing at the gapped top surface.

C. Layer-Resolved Topology

• Spatial Localization: Layer-projected Chern number analysis reveals that the topological response is highly localized. The maximal contribution comes from the FM layer and the topmost quintuple layer of the TI, decaying rapidly into the bulk. This confirms the surface nature of the 1/2-QAHE.

D. Sidewall States in Nanoribbons

• Decay Behavior: Analysis of a 20 nm nanoribbon revealed that the sidewall states associated with the magnetized top surface are chiral and spin-polarized.
• Power-Law Decay: Unlike the exponentially localized edge states of integer QAHE, these sidewall states exhibit a slower, power-law decay into the bulk of the ribbon.
• Physical Origin: These states, though not "proper" topological edge states in the Chern insulator sense, generate the spin-polarized chiral currents responsible for the half-quantized Hall conductance.

4. Significance and Implications

• Validation of Theory: The work provides a rigorous theoretical explanation for recent experimental observations where AHC values cluster around $0.5 e^2/h$ despite the presence of finite $\sigma_{xx}$. It confirms that the half-quantization is a robust topological feature of the gapped surface, independent of the dissipative bottom channel.
• Material Design: By comparing CGT, MBSe, and CrI$_3$, the paper offers guidelines for selecting FM layers to optimize exchange gaps and interface quality in TI/FM heterostructures for future experiments.
• Fundamental Physics: The identification of power-law decaying sidewall currents offers a new perspective on the edge physics of parity-anomaly systems, distinguishing them from standard integer quantum Hall systems.
• Applications: These findings support the feasibility of using such heterostructures to probe the quantized topological magnetoelectric (TME) effect and the axion field, potentially enabling new low-power spintronic devices and topological quantum computing components.

In summary, the paper establishes that van der Waals heterostructures of 2D ferromagnets and topological insulators are viable platforms for realizing the parity anomaly state, characterized by a robust half-quantized Hall conductance coexisting with metallic transport, driven by localized sidewall currents with unique decay properties.