Unveiling In-Gap States and Majorana Zero Modes in Superconductor-Topological Insulator Bilayer model

This paper theoretically investigates a superconductor-topological insulator bilayer model, demonstrating that increasing interlayer tunneling and hybridization can spatially separate Majorana zero modes from Caroli-de Gennes-Matricon states and induce distinct $p$-wave-like features, thereby providing a framework for stabilizing and identifying Majorana modes in engineered topological systems.

The Gist

The Big Picture: Hunting for "Ghost" Particles

Imagine you are trying to build a super-powerful computer (a quantum computer) that can solve problems normal computers can't. To do this, you need a very special type of building block called a Majorana Zero Mode (MZM). Think of these as "ghost particles." They are their own antiparticles, and if you can trap them, they are incredibly stable and perfect for storing information without errors.

The problem is, these ghosts are hard to find. They usually hide in materials that are very messy or hard to control. This paper proposes a new, cleaner way to catch them using a "sandwich" made of two special layers:

1. The Bottom Layer: A Topological Insulator (TI). Think of this as a material that is an insulator on the inside (electricity can't flow) but acts like a super-highway on its surface (electricity flows easily).
2. The Top Layer: A standard Superconductor (SC). This is a material that conducts electricity with zero resistance.

The goal is to make the superconductor "breathe" its special properties into the topological insulator, creating a hybrid system where these ghost particles can appear.

---

The Secret Ingredient: The "Tunnel" (Interlayer Coupling)

The researchers built a computer model of this sandwich. The most important thing they studied was how tightly the two layers are connected. They call this connection interlayer tunneling ($t_\perp$).

The Analogy: Imagine the Topological Insulator is a calm lake, and the Superconductor is a windy sky above it.

• Weak Connection: If the layers are far apart, the wind doesn't really affect the water. The water stays calm.
• Strong Connection: If the layers are very close, the wind (superconductivity) creates big waves and ripples in the water.

The paper found that as you increase this "wind" (tunneling strength), something surprising happens. The "safe zone" where the ghost particles can hide (the energy gap) doesn't stay in the center of the lake. It moves to the edges.

The Ripple Effect: Friedel Oscillations

When the safe zone moves, it changes how the "ghost particles" behave.

The Analogy: Imagine dropping a stone in a pond. You get ripples spreading out.

• In a normal superconductor, the ripples are simple and predictable.
• In this new sandwich model, because the layers are mixing so strongly, the ripples become complex and wavy. The researchers call these Friedel-like oscillations.

This means the ghost particles aren't just sitting still; they are vibrating in a specific pattern. This pattern is a "fingerprint" that proves you are dealing with this special hybrid material, not just a regular superconductor.

The Vortex: A Whirlpool Trap

To catch these ghosts, scientists often create a vortex (a whirlpool) in the superconductor by poking a hole in it (an "antidot") and spinning a magnetic field through it.

The Analogy: Think of the vortex as a whirlpool in a river.

• The Ghosts (MZMs): These are the rare, magical fish that only swim in the very center of the whirlpool. They are the ones we want for our quantum computer.
• The Noise (CdGM States): These are regular fish that also get caught in the whirlpool but swim in circles around the center. They are annoying because they look a lot like the ghosts, making it hard to tell them apart.

The Big Discovery:
The paper found that by turning up the "wind" (increasing the tunneling strength between the layers), the distance between the magical ghosts (MZMs) and the annoying regular fish (CdGM states) gets much bigger.

• Why this matters: In the past, the ghosts and the noise were so close together that it was hard to tell them apart. Now, the researchers show that by tuning the connection between the layers, we can push the noise away, leaving the ghosts isolated and easy to identify. It's like turning up the volume on the ghost's voice while turning down the noise.

The "Spin" Twist

One of the coolest findings is about spin (a quantum property like a tiny magnet).

• In a normal superconductor, the fish swim in pairs that are perfectly balanced.
• In this sandwich model, the "ghosts" and the "noise" have a weird, unbalanced spin. The paper shows that the ghosts have a specific "handedness" (like a left-handed glove) that is different from the noise. This is a clear signal that the material has developed unconventional superconductivity (acting like a p-wave superconductor, which is the "holy grail" for quantum computing).

The "Single-Layer" Shortcut

Finally, the researchers realized that instead of simulating the whole complex two-layer sandwich (which takes a lot of computer power), you can create a simplified map (a single-layer model).

• The Analogy: Instead of simulating every molecule of the ocean and the sky, you can just draw a map of the water level that predicts the waves perfectly.
• This simplified model works surprisingly well and can help other scientists quickly test ideas without needing supercomputers.

Summary: Why Should We Care?

This paper is like a user manual for catching quantum ghosts.

1. It tells us that the connection between the two layers is a "dial" we can turn.
2. Turning that dial moves the "safe zones" and creates unique wave patterns.
3. Most importantly, it shows us how to separate the real ghosts from the fake ones by adjusting that dial.

This gives experimentalists a clear recipe: Build your sandwich, tune the connection strength, and you will have a much better chance of finding the stable Majorana particles needed to build the quantum computers of the future.

Technical Summary

1. Problem Statement

The search for a robust platform for topological quantum computing remains a central challenge in condensed matter physics. While Majorana Zero Modes (MZMs) are promising candidates for topological qubits, their experimental identification is complicated by the difficulty in distinguishing them from trivial low-energy states, such as Andreev Bound States (ABS), Caroli–de Gennes–Matricon (CdGM) states, or Yu–Shiba–Rusinov (YSR) states.

Existing platforms include:

• Superconductor (SC) – 3D Topological Insulator (3DTI) interfaces.
• Spin-orbit coupled semiconductor nanowires proximitized by an SC.
• Quantum spin liquids and moiré platforms.

A specific challenge lies in the Fe(Te,Se)/Bi$_2$Te$_3$ heterostructure. While experiments suggest these systems host topological superconductivity, the nature of the superconducting state and the specific role of interlayer coupling in shaping the in-gap spectrum (particularly the separation between MZMs and CdGM modes) requires a detailed microscopic theoretical understanding. Most existing models treat the proximity effect implicitly by adding a pairing potential directly to the TI surface, neglecting the explicit physics of the superconducting layer and interlayer tunneling.

2. Methodology

The authors introduce a microscopic bilayer model to address these gaps, explicitly modeling the interface between a 3DTI surface state (SS) and a 2D $s$-wave superconductor (SC).

• Hamiltonian Construction:
  • TI Layer ($H_{TI}$): Modeled on a 2D square lattice using a tight-binding approach with Rashba spin-orbit coupling ($\lambda$) and a mass term ($m$) to isolate a single Dirac cone at the $\Gamma$-point, avoiding fermion doubling.
  • SC Layer ($H_{SC}$): Modeled as a conventional $s$-wave superconductor with on-site potential, hopping, and a pairing potential ($\Delta$). The parameters are tuned to qualitatively match the electronic structure of Fe(Te,Se).
  • Hybridization ($H_{hyb}$): The two layers are coupled via an interlayer tunneling amplitude ($t_\perp$), representing the direct hopping of electrons between the TI and SC layers.
• Formalism: The system is solved using the Bogoliubov–de Gennes (BdG) formalism, resulting in an 8-dimensional Hamiltonian per lattice site (accounting for particle/hole and TI/SC degrees of freedom).
• Geometry:
  • Momentum Space: Used to analyze band structure, spin polarization, and the formation of the proximity-induced (PrI) gap.
  • Real Space: Simulations performed on finite lattices with open boundary conditions to study chiral edge modes (CEMs).
  • Antidot/Vortex: A circular antidot (radius $R=10$) is introduced in the SC layer where the lattice sites are removed, and a magnetic flux quantum ($\Phi_0$) is threaded through the center to create a vortex. This setup traps in-gap bound states.
• Comparative Analysis: The bilayer results are compared against:
  1. A standalone $s$-wave superconductor with an antidot.
  2. An effective single-layer model where the SC layer is "integrated out" to reproduce the bilayer physics.

3. Key Contributions

1. Explicit Interlayer Coupling: Unlike previous works that use an effective pairing potential, this study explicitly includes the SC layer and the interlayer tunneling strength ($t_\perp$) as a tunable parameter.
2. Momentum-Selective Gap Shift: The authors demonstrate that increasing $t_\perp$ shifts the minimum of the proximity-induced gap away from the $\Gamma$-point to a finite momentum ring in the Brillouin zone.
3. Spatial Oscillations: They identify that this momentum shift induces Friedel-like oscillations in the wavefunctions of edge states and vortex-bound states, a feature absent in standard single-layer models.
4. MZM-CdGM Separation: The study reveals that strong hybridization ($t_\perp$) significantly enhances the energy separation between MZMs and the lowest CdGM states, even as the bulk PrI gap decreases.
5. Angular Momentum Asymmetry: The analysis uncovers distinct angular momentum symmetries and spin-momentum locking effects in the CdGM states that differ fundamentally from conventional $s$-wave vortices.

4. Key Results

A. Proximity-Induced (PrI) Gap Evolution

• Low $t_\perp$: The PrI gap opens at the $\Gamma$-point. The gap magnitude is determined by $t_\perp$ and the SC gap $\Delta_0$.
• High $t_\perp$: As $t_\perp$ increases (beyond $\approx 1.2\Delta_0$), the gap minimum shifts discontinuously to a finite momentum $|k_F|$, forming a ring-shaped contour in the Brillouin zone.
• Consequence: This shift leads to a redistribution of the density of states and alters the spatial decay of edge modes.

B. Chiral Edge Modes (CEMs)

• The system supports topologically protected edge states.
• Spatial Profile: In the strong coupling regime, the edge mode probability density exhibits oscillatory decay into the bulk, directly linked to the finite Fermi momentum ($k_F$) of the shifted gap.
• Layer Redistribution: As $t_\perp$ increases, the weight of the edge mode shifts from the TI layer toward the SC layer, and the spatial profile evolves from Lorentzian-like to Gaussian-like.

C. Vortex-Bound States (MZMs and CdGM)

• MZM Identification: Two zero-energy modes are found: one localized at the system edge and one at the vortex core. Both exhibit strong spin polarization and localization.
• CdGM Modes: Several discrete in-gap states (CdGM) appear.
  • Energy Separation: Crucially, increasing $t_\perp$ increases the energy separation between the MZM and the first CdGM state. This is vital for experimental detection, as it reduces the risk of MZMs being obscured by trivial states.
  • Spatial Structure: The MZM is sharply localized at the vortex center in the TI layer but peaks at the antidot boundary in the SC layer.
  • Angular Momentum: The wavefunctions exhibit specific angular momentum quantum numbers ($m_J$). Unlike conventional $s$-wave vortices where spin symmetry is preserved, the TI/SC hybrid shows angular momentum shifts between spin-up and spin-down components due to spin-momentum locking.

D. Comparison with Standalone $s$-wave Superconductor

• In a pure $s$-wave superconductor with an antidot, in-gap states are Andreev/CdGM states with degenerate spin and orbital symmetries.
• In the bilayer, the states exhibit emergent $p$-wave-like features, including angular momentum asymmetries and spin-polarized wavefunctions, confirming the topological nature of the hybrid system.

E. Effective Single-Layer Mapping

• The authors successfully mapped the bilayer results to an effective single-layer model by tuning the on-site energy ($\epsilon_{TI}$) and the effective pairing ($\Delta_{eff}$).
• This mapping accurately reproduces the MZM-CdGM energy spectrum, offering a computationally efficient tool for future studies, though it requires parameter tuning to account for the momentum shift of the gap minimum.

5. Significance and Implications

• Experimental Guidance: The paper provides concrete predictions for tuning the stability of MZMs in SC–3DTI heterostructures (like Fe(Te,Se)/Bi$_2$Te$_3$). Specifically, it suggests that stronger interlayer coupling is beneficial for isolating MZMs from trivial CdGM states.
• Differentiation of States: The distinct spatial oscillations and angular momentum signatures provide new "fingerprints" to experimentally distinguish true MZMs from trivial zero-energy states in Scanning Tunneling Microscopy (STM) experiments.
• Theoretical Framework: By explicitly modeling the SC layer, the work bridges the gap between abstract topological models and realistic material interfaces, highlighting the non-trivial role of the superconductor in shaping topological properties.
• Topological Proximity Effect: The results underscore the interplay between the superconducting proximity effect and the topological proximity effect, suggesting that hybridization strength is a critical control parameter for engineering topological superconductivity.

In summary, this work establishes that interlayer tunneling is not merely a coupling mechanism but a critical knob for engineering the spectral and spatial properties of topological superconductors, offering a robust pathway to stabilize and identify Majorana zero modes.