Defect-induced displacement of topological surface state in quantum magnet MnBi$_2$Te$_4$

This study demonstrates that high concentrations of intrinsic antisite defects in the topological magnet MnBi$_2$Te$_4$ displace the topological surface states deep into the crystal bulk, thereby suppressing the surface gap and resolving the long-standing discrepancy between experimental observations and theoretical predictions.

The Gist

Imagine you have a very special, high-tech floor made of a material called MnBi₂Te₄. This floor is designed to be a "super-highway" for electrons, allowing them to zip along the surface without any traffic jams or resistance. In the world of physics, this is called a topological surface state.

Scientists hoped this floor would be perfect for building future quantum computers. However, they noticed a problem: the "magic" of this highway wasn't working as well as the blueprints (theoretical predictions) said it should. The energy gap that was supposed to protect the electrons was disappearing.

The Mystery: Where did the magic go?

For a long time, scientists were puzzled. They knew the material had some "imperfections" (defects), but they didn't know exactly how these tiny flaws were ruining the show.

Think of the material like a multi-story apartment building.

• The Surface (1st Floor): This is where the electrons are supposed to live and travel.
• The Layers Below: The building has layers with different "magnetic personalities." The first layer wants the electrons to go one way, but the second layer wants them to go the opposite way.

The Culprit: The "Swap" Defects

The paper reveals that the problem is caused by tiny "swap" mistakes inside the crystal. Imagine the atoms in the building are like tenants. Sometimes, a Manganese (Mn) tenant accidentally swaps places with a Bismuth (Bi) tenant.

When these two swap places (creating what scientists call an antisite defect), they act like a heavy anchor or a sinkhole.

The Big Discovery: The "Subsurface" Shift

Here is the clever part of the discovery, explained with an analogy:

Imagine you are trying to listen to a band playing on the roof of a building (the surface state).

1. In a perfect building: The band is right on the roof, loud and clear. You can hear them perfectly.
2. In a building with many "swaps": The heavy anchors (defects) pull the band down into the basement (deeper into the crystal).

The paper shows that when there are too many of these "swap" defects, they physically push the electron highway away from the surface and deeper into the crystal.

Why did the experiments look confusing?

This explains why scientists were getting mixed results:

• The Microscope (STM): This tool is like a person standing right on the roof, looking down with a magnifying glass. When the band was pushed down into the basement by the defects, the person on the roof couldn't see or hear them anymore. They thought the band had disappeared.
• The Satellite (ARPES): This tool is like a satellite in space looking at the whole building. It can "see" through the roof and detect the band even when it's hiding in the basement. So, the satellite still saw the band, but the person on the roof didn't.

The "Magnetic Clash"

Why does pushing the band down matter?
Remember the building has layers with opposite magnetic personalities?

• On the roof (Layer 1): The magnetic field is strong and helps create the "magic gap."
• In the basement (Layer 2): The magnetic field is the opposite.

When the defects push the electron highway down to the basement, it gets caught in the middle of these two opposing magnetic fields. The fields cancel each other out, and the "magic gap" closes. The highway loses its special protection.

The Solution

The paper concludes that to fix this, we need to stop the "swaps."

• Fewer Defects: If we can grow the crystal with fewer mistakes (fewer swapped tenants), the electron highway stays on the roof where it belongs.
• More Space: If the defects are far apart, they don't pull the highway down as hard. But if they are crowded together, they pull it deep into the basement.

In Summary

This paper solved a mystery by realizing that defects in the material act like invisible elevators, dragging the special electron highway from the surface down into the crystal's interior. Once down there, the highway gets stuck in a magnetic tug-of-war and stops working. To build better quantum materials, scientists now know they must keep the crystal surface as clean and defect-free as possible to keep the "highway" on the roof.

Technical Summary

1. Problem Statement

The stoichiometric magnetic topological insulator MnBi$_2$Te$_4$ (MBT) is a prime candidate for realizing the Quantum Anomalous Hall (QAH) effect and Axion insulator states due to its intrinsic A-type antiferromagnetic (AFM) order. Theoretically, the magnetic exchange interaction should open a gap at the Dirac point of the topological surface state (TSS).

However, experimental observations have been inconsistent:

• Some studies report a gapped surface state, while others observe a vanishing gap.
• The QAH effect is typically only observed at temperatures significantly lower than the Néel temperature ($T_N \approx 25$ K) and under applied magnetic fields.
• The Core Conflict: While native antisite defects (specifically Mn$_{Bi}$ and Bi$_{Mn}$) are known to exist, their exact mechanism for suppressing the surface gap has been debated. Recent theoretical work suggested that co-antisite defects might push the TSS deeper into the bulk, but this had not been experimentally validated.

2. Methodology

The authors employed a multi-modal approach combining experimental characterization with theoretical modeling to resolve the discrepancy between local surface probes and global electronic structure:

• Samples: Two MBT single crystals with varying defect densities were analyzed:
  • Sample A: Lower Mn content, lower Mn$_{Bi}$ antisite concentration (~3.8%).
  • Sample B: Higher Mn content, higher Mn$_{Bi}$ antisite concentration (~4.6–8.7% in specific regions).
• Scanning Tunneling Microscopy/Spectroscopy (STM/STS):
  • Used to identify specific defect types (Bi$_{Te}$, Mn$_{Bi}$, Bi$_{Mn}$) based on topographic signatures.
  • Performed tunneling spectroscopy ($dI/dV$) and Quasi-Particle Interference (QPI) mapping to probe the local density of states (LDOS) and surface band dispersion.
• Angle-Resolved Photoemission Spectroscopy (ARPES):
  • Used to measure the global electronic band structure and surface state dispersion, which has a larger probing depth (nanometers) compared to STM (atomic scale).
• Density Functional Theory (DFT):
  • Calculated the electronic structure of MBT slabs with varying co-antisite defect configurations (nearest-neighbor pairs vs. distant pairs).
  • Simulated STM topographies and charge density distributions to understand the vertical displacement of the TSS.

3. Key Contributions & Results

A. Discrepancy Between STM and ARPES

• STM Observations: In Sample A (lower defect density), STM clearly resolved the Dirac cone and linear dispersion, though no clear gap was observed. In Sample B (high defect density), the Dirac dispersion was almost completely absent in STM; the $dI/dV$ spectra showed a constant, slightly elevated intensity within the bulk gap, suggesting the surface state was "masked."
• ARPES Observations: Crucially, ARPES measurements on both samples revealed a pristine, clear Dirac cone with no observable surface gap, regardless of the defect density.
• Resolution: The authors reconciled this by noting the differing surface sensitivities. STM probes the topmost atomic layer, while ARPES probes deeper. The absence of the TSS signature in STM for Sample B, despite its presence in ARPES, implies the TSS has been physically displaced away from the surface.

B. Defect-Specific Effects (Local Scale)

• Mn$_{Bi}$ Defects: These defects (Mn atoms occupying Bi sites) cause a significant suppression of the $dI/dV$ intensity at the Dirac point. When Mn$_{Bi}$ defects cluster, the surface state signature vanishes.
• Bi$_{Mn}$ Defects: These defects (Bi atoms occupying Mn sites) actually enhance the $dI/dV$ intensity of the surface state.
• Bi$_{Te}$ Defects: Show negligible effect on the tunneling intensity.
• Mechanism: The suppression in Sample B is driven by the high density of Mn$_{Bi}$ defects and their proximity to Bi$_{Mn}$ defects (co-antisites).

C. DFT Validation: Vertical Displacement

• Charge Density Shift: DFT calculations revealed that in the presence of co-antisite defects (Mn$_{Bi}$ + Bi$_{Mn}$), the charge density of the surface state (both valence and conduction bands) shifts downward by 1–2 Å from the topmost septuple layer (SL) into the second SL.
• Gap Suppression: The topological surface state interacts with the magnetic moments of the second SL. Due to the A-type AFM order, the second layer has magnetization opposite to the first. As the TSS moves deeper, it experiences this opposing magnetization, which cancels out the exchange gap, effectively closing it.
• Defect Density Dependence: The effect is strongly dependent on the distance between co-antisites. When defects are closer (higher density), the vertical shift is more pronounced, and the gap suppression is more severe.

4. Significance and Implications

1. Mechanism Clarification: The study definitively proves that the "missing" surface gap in MBT is not due to the defects simply scattering electrons or doping the system, but due to a geometric displacement of the topological state into the bulk.
2. Depth as a Critical Parameter: It establishes that the vertical position of the TSS relative to the magnetic layers is a critical parameter for the magnitude of the exchange gap.
3. Guidance for Material Synthesis: To achieve robust QAH effects and large Dirac gaps in the AFM phase, future efforts must focus on minimizing native antisite defect densities, specifically Mn$_{Bi}$ and Bi$_{Mn}$ pairs.
4. Defect Engineering Strategy: The authors suggest an alternative pathway: if defects cannot be eliminated, defect engineering could be used to induce a ferromagnetic (FM) layer-by-layer coupling. In an FM configuration, even if the TSS is pushed deeper, it would interact with parallel magnetic moments, potentially preserving or even enhancing the surface gap.

Conclusion

The paper resolves a long-standing debate in the field of magnetic topological insulators by demonstrating that intrinsic antisite defects in MnBi$_2$Te$_4$ induce a vertical displacement of the topological surface state. This displacement moves the state into a region of opposing magnetization, suppressing the exchange gap. This finding provides a clear roadmap for optimizing MBT crystals for quantum transport applications by controlling defect density and potentially manipulating magnetic coupling.