Surface-localized topological superconductivity in nodal-loop materials: BdG analysis

This paper theoretically demonstrates that in nodal-line semimetals like Pd-doped CaAgP, superconductivity driven by drumhead surface states naturally favors a surface-localized chiral $p$-wave pairing symmetry, which strongly enhances the order parameter at the outermost layers while gapping out the surface states.

The Gist

The Big Picture: A "Drumhead" on a Trampoline

Imagine a massive, 3D trampoline made of a special material called a nodal-line semimetal. In the middle of this trampoline, the fabric is perfectly flat and touches the ground in a giant, continuous circle. This circle is called a "nodal loop."

Now, imagine you are standing on the edge of this trampoline. Because of the physics of this special material, the edge doesn't just hang there; it creates a unique, flat platform right above the ground. The scientists call this a "drumhead surface state."

Think of the drumhead like a flat, calm pond sitting on top of a busy, rushing river (the bulk of the material). Everyone in the river is moving fast, but on this flat pond, everything is still and crowded. Because so many "people" (electrons) are crowded onto this tiny, flat surface, they start to interact with each other very intensely.

The Experiment: What Happens When They Get Cold?

The researchers wanted to know: If we cool this material down until it becomes a superconductor (a material where electricity flows with zero resistance), what kind of "dance" will the electrons on this flat drumhead do?

In superconductors, electrons pair up and dance together. There are many ways they can dance:

1. The "Chiral P-Wave" Dance: A fast, spinning, swirling dance (like a tornado).
2. The "D-Wave" Dance: A more complex, figure-eight style dance.

The team used a computer to simulate a slice of this material (a "slab") and asked: Which dance will the electrons on the surface choose?

The Results: The Surface Loves the Spin

Here is what they found, broken down simply:

1. The Surface is the VIP Zone
When the material gets cold, the superconductivity doesn't happen everywhere. It happens almost exclusively on the very outer edges (the "drumhead"). The middle of the material stays normal and doesn't superconduct at all. It's like a party where only the people on the roof are dancing, while everyone in the basement is just standing around.

2. The "Swirling" Dance Wins Easily
When the researchers let the electrons choose their dance partner:

• The Chiral P-Wave (Swirl): The electrons on the surface went crazy for this one. They paired up immediately and formed a strong, stable superconducting state. The "dance" was so strong it was confined to just the top 2 or 3 layers of the material.
• The D-Wave (Figure-Eight): The electrons tried this dance, but it was a disaster. The pairing was incredibly weak—more than 10 times weaker than the swirling dance. It was practically non-existent.

3. The "Flat Pond" Gets a Hole Punched in It
In the normal state, the "drumhead" surface had a sharp peak of energy right at zero (like a flat line). When the superconductivity kicked in, this flat line didn't just disappear; it split into two distinct peaks.

• Analogy: Imagine a calm, flat lake (the drumhead). When the superconductivity starts, it's like a giant wave crashing through the middle, creating two high walls of water on either side and a gap in the center. This "gap" is the superconducting energy gap. The fact that the flat lake turned into two walls proves the superconductivity is working perfectly on the surface.

Why Does This Matter?

This study is like a detective solving a mystery about a specific material called CaAgP (specifically, when doped with Palladium).

• The Mystery: Scientists know this material has a nodal loop and surface states. They also know it becomes a superconductor, but they weren't sure how the electrons were pairing up.
• The Clue: This paper says, "If you have a nodal-loop material with these flat surface states, the electrons are naturally biased to do the Chiral P-Wave (swirling) dance."
• The Prediction: If you look at real-world experiments on this material, you should see signs of this swirling, surface-only superconductivity.

The Takeaway

Think of the nodal-line material as a stage. The drumhead states are the spotlight on the center of the stage. The paper shows that when the "lights go down" (cooling to superconductivity), the actors on the spotlight don't just stand there; they spontaneously start a very specific, swirling dance (Chiral P-Wave) that is so strong it ignores the rest of the stage.

This gives scientists a clear roadmap: If you see a material with these "drumhead" surface states, you can bet your bottom dollar that the superconductivity happening there is likely this special, swirling type.

Technical Summary

1. Problem Statement

Topological nodal-line semimetals (NLSMs) host bulk nodal loops in momentum space, which project onto the surface Brillouin zone to form nearly flat "drumhead" surface states. These states possess a strongly enhanced density of states (DOS) near the Fermi level, making them highly susceptible to interaction-driven instabilities such as superconductivity.

• Context: Recent experiments on Pd-doped CaAgP suggest unconventional, surface-dominated superconductivity with broken time-reversal symmetry.
• Core Question: When superconductivity emerges primarily from these drumhead surface states in a nodal-line material, which pairing symmetry is favored? Specifically, how does the presence of drumhead states influence the competition between chiral $p$-wave (odd-parity) and $d_{x^2-y^2}$-wave (even-parity) pairing symmetries?
• Gap in Literature: While theoretical proposals exist for topological crystalline superconductivity in these systems, a microscopic, self-consistent analysis of how surface states bias the pairing symmetry competition in a realistic slab geometry has been lacking.

2. Methodology

The authors employ a layer-resolved Bogoliubov–de Gennes (BdG) approach combined with a minimal tight-binding model to study a slab geometry.

• Model Hamiltonian:
  • A cubic lattice model describing a nodal-line semimetal (motivated by CaAgP) with negligible spin-orbit coupling (spin treated as a trivial degeneracy).
  • The normal-state Hamiltonian ($H_0$) includes nearest-neighbor hopping and a mass term ($M$) controlling band inversion.
  • Geometry: A slab of $N=20$ layers with open boundary conditions (OBC) in the $z$-direction and periodic boundary conditions in the $xy$-plane. This setup reproduces bulk nodal loops in the interior and drumhead surface states at the boundaries.
• Superconducting Interaction:
  • An attractive local interaction ($V > 0$) projected onto specific pairing channels.
  • The pair potential is factorized as $\Delta_n(\mathbf{k}_\parallel) = \Delta_{0,n} \phi(\mathbf{k}_\parallel)$, where $\Delta_{0,n}$ is the layer-dependent amplitude and $\phi(\mathbf{k}_\parallel)$ is the form factor.
• Pairing Channels Investigated:
  1. Chiral $p$-wave: Odd-parity, form factor $\phi(\mathbf{k}_\parallel) = \sin k_x + i \sin k_y$.
  2. $d_{x^2-y^2}$-wave: Even-parity, form factor $\phi(\mathbf{k}_\parallel) = \cos k_x - \cos k_y$.
• Self-Consistent Calculation:
  • The BdG Hamiltonian is constructed and diagonalized to obtain quasiparticle energies and eigenvectors.
  • The gap amplitudes $\Delta_{0,n}$ are determined self-consistently using the gap equation at $T=0$ until convergence ($|\Delta_{0,n}|/t < 10^{-5}$).
  • Spectral properties are analyzed via the layer-resolved local density of states (LDOS) and quasiparticle dispersion.

3. Key Contributions

• Microscopic Bias Mechanism: The study provides a microscopic demonstration that drumhead surface states inherently bias the system toward chiral $p$-wave pairing over $d$-wave pairing.
• Spatial Localization: It establishes that superconductivity driven by these surface states is not a bulk phenomenon but is strongly confined to the outermost layers (within 2–3 layers), leaving the bulk interior essentially normal.
• Spectral Signature: The paper identifies specific spectral signatures (splitting of the zero-energy peak into coherence peaks) that directly link the opening of the superconducting gap to the gapping out of the drumhead band.

4. Key Results

A. Normal State Verification

• The slab model successfully reproduces the bulk nodal loop and the associated drumhead surface band.
• LDOS Analysis: The central layers ($n=10$) exhibit a bulk-like DOS, while the surface layer ($n=1$) shows a sharp, pronounced zero-energy peak corresponding to the flat drumhead band.

B. Pairing Symmetry Competition

• Chiral $p$-wave:
  • The order parameter $|\Delta_{0,n}|$ is strongly enhanced at the outermost layers ($n=1, 20$).
  • It decays rapidly within a few layers, becoming negligible in the central region.
  • The amplitude is orders of magnitude larger than the $d$-wave case.
• $d_{x^2-y^2}$-wave:
  • The order parameter is negligibly small (more than an order of magnitude smaller than the $p$-wave) across all layers.
  • The interaction strength $V$ is insufficient to stabilize a significant $d$-wave gap in the presence of the drumhead states.
• Conclusion: The drumhead states naturally favor the odd-parity chiral $p$-wave channel.

C. Spectral Properties of the Chiral $p$-wave State

• Quasiparticle Dispersion: The flat drumhead band at $E=0$ is pushed away from the Fermi level, opening a superconducting gap over most of the projected nodal loop region ($-\frac{2\pi}{3} \lesssim k_x \lesssim \frac{2\pi}{3}$). Gapless excitations remain only at isolated points (e.g., $k_x=0$).
• Surface LDOS:
  • The sharp zero-energy peak observed in the normal state is split into two coherence peaks symmetric around the Fermi energy.
  • The energy separation between these peaks directly measures the induced superconducting gap.
  • This confirms that the drumhead band is efficiently gapped out by the surface-localized $p$-wave pairing.

D. Sensitivity to Chemical Potential

• The surface-localized superconductivity is highly sensitive to the chemical potential ($\mu$). When $\mu$ is shifted away from zero (the bottom of the drumhead band), the order parameter amplitude drops sharply, indicating that the phenomenon relies on the specific filling of the surface states.

5. Significance

• Theoretical Insight: The work clarifies the mechanism by which topological surface states select pairing symmetries, suggesting that surface-localized chiral superconductivity is a natural outcome in nodal-line materials with drumhead states.
• Experimental Guidance: The results offer a qualitative framework for interpreting recent experiments on Pd-doped CaAgP. The observed unconventional, surface-dominated superconductivity with broken time-reversal symmetry is consistent with the predicted surface-localized chiral $p$-wave state.
• Methodological Value: The layer-resolved BdG approach demonstrates a robust method for distinguishing between bulk and surface-driven superconductivity in topological materials, highlighting the importance of spatially resolved order parameters in such systems.