Numerical Solution of the Bardeen-Cooper-Schrieffer Equation for Unconventional Superconductors

This paper investigates the analytical properties and presents an efficient Galerkin-based numerical solution using B-splines for the Bardeen-Cooper-Schrieffer equation describing unconventional superconductors with long-range power-law interactions on a $d$-dimensional lattice, with specific results demonstrated for a nodal superconductor on a two-dimensional square lattice.

The Gist

Imagine a bustling city where the citizens are electrons. In a normal city (a regular metal), these citizens bump into each other and the buildings, creating traffic jams and friction. This friction is what we call electrical resistance.

But in a special kind of city called a superconductor, something magical happens. Below a certain temperature, the citizens pair up into dance partners (called Cooper pairs) and glide through the streets without bumping into anything. The friction vanishes, and electricity flows perfectly.

This paper is about figuring out the "dance rules" for a very specific, weird type of superconductor. Here is the story in simple terms:

1. The Old Rules vs. The New Mystery

For a long time, scientists knew the rules for how these electron pairs formed. They thought the pairs were held together by the "vibrations of the floor" (phonons). It was like the floor wiggling to help two people hold hands.

But then, scientists discovered unconventional superconductors (like high-temperature ones). These don't seem to follow the old floor-wiggling rules. The authors of this paper propose a new idea: maybe the electrons are holding hands because they are talking to each other from far away.

Imagine if you could feel a gentle tug on your hand from a friend standing on the other side of the city, not just the person standing next to you. This is a long-range interaction. The paper tries to solve the math puzzle of how these long-distance tugs create superconductivity.

2. The Mathematical "Recipe" (The BCS Equation)

The authors use a famous recipe called the Bardeen-Cooper-Schrieffer (BCS) equation. Think of this equation as a complex cooking recipe that tells you exactly how much "glue" (the superconducting gap) is needed to keep the electron pairs stuck together.

However, this recipe is tricky because:

• It's a loop: The amount of glue depends on how many pairs you have, but the number of pairs depends on how much glue you have. You have to solve a puzzle where the answer changes the question.
• The "Long-Distance" Ingredient: The recipe includes a special ingredient called the Epstein zeta function.
  • The Analogy: Imagine trying to calculate the total noise in a room where everyone is whispering. If everyone whispers from right next to you, it's easy. But if people are whispering from every corner of the universe, and the sound gets louder the closer they are (a "power-law" singularity), it becomes a nightmare to calculate. The Epstein zeta function is the mathematical tool that handles this "noise from everywhere" efficiently.

3. The "Nodal" Problem (The Zero Points)

In some of these superconductors, the "glue" isn't the same everywhere. It's like a blanket that has holes in it.

• In a standard superconductor, the glue is strong everywhere (like a solid blanket).
• In these unconventional ones, the glue strength drops to zero at specific points. These are called nodes.
• The Analogy: Imagine a dance floor where the music stops completely at four specific spots. At those spots, the dancers (electrons) stop dancing in pairs. This creates a "singularity" or a sharp edge in the math, which is very hard for computers to handle without crashing or getting the answer wrong.

4. How They Solved It (The B-Spline Magic)

The authors needed a way to solve this messy, looping recipe with the "far-away noise" and the "holes in the blanket."

They used a method called Galerkin with B-splines.

• The Analogy: Imagine you are trying to draw a perfect, smooth curve on a piece of paper, but the paper is made of tiny, rigid Lego blocks. If you just use big blocks, the curve looks jagged. If you use tiny blocks, it's smoother but takes forever.
• B-splines are like "smart, flexible Lego blocks." They can bend and stretch to fit the curve perfectly, even around the tricky "holes" (the nodes) and the "far-away noise" (the Epstein zeta function).
• The authors showed that by using these smart blocks, they could turn the impossible math problem into a giant, but solvable, spreadsheet (a matrix) that computers can crunch very quickly.

5. The Result: The "D-Wave" Dance

When they ran their simulation on a computer, they found a specific pattern.

• The S-Wave: A standard superconductor is like a round, uniform circle of dancers.
• The D-Wave: Their solution looked like a four-leaf clover or a cross. The "glue" is strong in two opposite directions and weak (or zero) in the other two.
• This matches what scientists see in real-world unconventional superconductors (like certain ceramics). The "clover" shape has four points where the glue vanishes (the nodes), exactly where the math predicted the "holes" would be.

Why Does This Matter?

This paper is a bridge between pure math and real-world physics.

1. It proves the math works: It shows that even with weird long-range forces and tricky "holes" in the data, we can calculate exactly how these superconductors behave.
2. It helps build the future: Understanding these "D-wave" superconductors is crucial for building quantum computers. These computers need to be super stable, and knowing exactly how the electron pairs dance (even when they have holes in their blanket) helps engineers design better, more powerful machines.

In a nutshell: The authors built a super-smart mathematical "net" (using B-splines) to catch the behavior of electrons that are holding hands from far away, successfully mapping out the complex, hole-filled dance floor of a futuristic superconductor.

Technical Summary

1. Problem Statement

The paper addresses the mathematical and numerical challenges associated with solving the Bardeen–Cooper–Schrieffer (BCS) gap equation for unconventional superconductors. Unlike conventional superconductors where interactions are short-range and phonon-mediated, this work focuses on systems with long-range power-law electron-electron interactions.

Key characteristics of the problem include:

• Nonlinearity: The equation is a nonlinear convolution integral equation for the complex matrix-valued superconducting gap function $F$.
• Long-range Interactions: The interaction kernel involves the Epstein zeta function ($Z_{\Lambda, \nu}$), which exhibits a power-law singularity at zero momentum. This function arises from the Fourier transform of the power-law interaction on a $d$-dimensional lattice.
• Symmetry Constraints: The solution must satisfy fermionic anticommutation rules. For triplet pairing ($k=2$), the gap matrix must satisfy the antisymmetry constraint $F^\top(-x) = -F(x)$.
• Nodal Superconductivity: The nonlinearity involves a term $G[F]$ containing an inverse square root. On the Fermi surface (where the dispersion relation $\xi(x)=0$), if the gap $F(x)$ also vanishes, the integrand becomes non-smooth or discontinuous. These "nodal" points create significant numerical difficulties, particularly when convolved with the singular Epstein zeta function.

2. Methodology

The authors propose a Galerkin-Petrov method using periodic B-splines as basis functions to solve the equation efficiently.

Theoretical Framework

• Operator Formulation: The BCS equation is treated as a nonlinear operator equation involving a convolution with a kernel composed of an on-site interaction ($C_1$) and a long-range interaction ($C_2 Z_{\Lambda, \nu}$).
• Sobolev Spaces: The problem is analyzed within the context of periodic Sobolev spaces ($H^s$) on the torus $T^d$. The long-range interaction operator is identified as a pseudo-differential operator of order $-\nu$.
• Analytical Properties: The authors establish the existence of solutions for scalar and matrix cases. They show that for constant kernels, the problem reduces to finding fixed points of a specific function $\varphi(s)$, proving the existence of trivial ($F=0$) and non-trivial solutions.

Numerical Scheme

• Basis Functions: The method utilizes multidimensional, one-periodic B-splines of degree $\mu$. These functions are defined recursively and possess high regularity ($C^{\mu-1}$).
• Fourier Representation: A crucial aspect of the method is the use of the Fourier coefficients of the B-splines. Since the interaction kernel is a convolution operator, its symbol is diagonal in the Fourier domain.
• Circulant Structure: By using B-splines as both trial and test functions, the resulting Galerkin matrices (mass matrix and stiffness matrix) become circulant (or block-circulant in $d>1$).
  • This structure allows for the use of Fast Fourier Transforms (FFT) to perform matrix-vector multiplications and solve linear systems with almost optimal computational complexity ($O(N \log N)$).
• Handling Singularities: The method combines the B-spline discretization with an analytic evaluation of the convolution. This allows the algorithm to handle the singularities arising from the Epstein zeta function and the nodal points (where $F(x)=0$ and $\xi(x)=0$) without requiring ad-hoc regularization or extremely fine meshing near the nodes.

3. Key Contributions

1. Efficient Numerical Solver: Development of a high-performance Galerkin scheme using periodic B-splines that exploits the circulant structure of the system matrices, making it feasible to solve high-dimensional BCS equations with long-range interactions.
2. Analytical Treatment of Singularities: The paper provides a rigorous framework for handling the non-smooth behavior of the gap function at nodal points and the singularity of the Epstein zeta function within the convolution integral.
3. Implementation of Epstein Zeta Function: The work leverages the efficient numerical evaluation of the Epstein zeta function (via the EpsteinLib C-library) to model long-range power-law interactions accurately in momentum space.
4. Extension to Matrix-Valued Gaps: The formulation handles both scalar (singlet) and matrix-valued (triplet) gap functions, incorporating necessary symmetry constraints.

4. Numerical Results

The authors present numerical results for a nodal superconductor on a two-dimensional square lattice ($d=2$).

• Parameters: $C_1 = 0.75$, $C_2 = 0.7$, and interaction exponent $\nu = 2.01$.
• Grid: A $1024 \times 1024$ grid using periodic cubic B-splines ($\mu=3$).
• Solution Type: The computed solution is identified as a $d$-wave superconductor.
  • The gap function $f(x)$ exhibits the symmetry $f(x) \sim \cos(2\pi x_1) - \cos(2\pi x_2)$.
  • The solution changes sign upon exchanging coordinates and possesses zeros (nodes) at specific points (e.g., $x = (1/4, 1/4)^T$).
• Performance: The algorithm successfully resolves the discontinuities at the nodal points where the gap vanishes and the dispersion relation is zero, demonstrating the robustness of the B-spline convolution approach.

5. Significance

• Bridging Physics and Numerics: The paper addresses a gap in the literature where mathematical and numerical treatments of unconventional superconductivity are scarce compared to physical studies.
• Quantum Computing Applications: By modeling long-range interactions that can lead to topological superconducting states, the work supports the development of materials relevant to quantum computing.
• Computational Efficiency: The use of circulant matrices and FFT-based solvers makes the simulation of large-scale, high-resolution systems (like the $1024^2$ grid used) computationally tractable, which is essential for studying complex phenomena like nodal superconductivity.
• Generalizability: While focused on a square lattice, the theoretical framework is designed to generalize to arbitrary lattices and dimensions, providing a versatile tool for future research in condensed matter physics.