Bulk-Surface Coupled PDE with an Open Boundary

1. Problem Formulation

The paper investigates a bulk-surface coupled partial differential equation (PDE) system defined on a two-dimensional domain $\Omega_1$ with boundary $\Gamma = \partial \Omega_1$. A key feature of this problem is the presence of an open boundary $S \subset \Gamma$ with two endpoints $\partial S = \{p_1, p_2\}$.

The system models time-independent coupled diffusion where:

• Bulk Domain ($\mathbb{R}^2 \setminus S$): The scalar field $u$ satisfies the Laplace equation $-\Delta u = 0$.
• Surface ($S$): A surface field $U$ is coupled to the bulk via the continuity condition $u = U$ on $S$.
• Coupling Mechanism: The surface field satisfies a surface Laplace-Beltrami equation coupled with the jump in the normal derivative of the bulk field across $S$:
  $$-\Delta_\Gamma U + [\partial_n u] = f \quad \text{on } S$$
  where $[\partial_n u]$ represents the jump in the normal derivative across the open surface.
• Boundary Conditions:
  • Neumann condition on the endpoints of $S$: $\nu \cdot \nabla_\Gamma U = g$ on $\partial S$.
  • Decay condition at infinity: $u(x) = O(|x|^{-1})$ as $|x| \to \infty$.
• Compatibility: The source terms $f$ and $g$ must satisfy a global compatibility condition to ensure the solution decays at infinity.

Significance of the Open Boundary: Unlike problems involving closed surfaces (which are well-studied), open surfaces introduce geometric singularities at the endpoints. The domain is non-Lipschitz, and solutions typically lack standard Sobolev regularity near the edges, exhibiting specific singular behaviors (e.g., $d^{-1/2}$ or $d^{3/2}$ where $d$ is the distance to the endpoint).

2. Methodology

The authors employ a multi-stage analytical and numerical approach:

A. Reformulation via Boundary Integral Equations

The problem is reformulated as an integro-differential equation on the open surface $S$.

• The bulk solution $u$ is represented as a single-layer potential $V_\Omega \psi$, where $\psi = [\partial_n u]$ is the unknown density supported on $S$.
• The system reduces to finding a pair $(U, \psi)$ satisfying:
  $$-\Delta_\Gamma U + \psi = f$$
  $$V_S \psi - U = h$$
  where $V_S$ is the restriction of the single-layer operator to $S$.

B. Existence and Uniqueness Analysis

• Functional Spaces: The analysis is conducted in Sobolev spaces tailored to open surfaces, specifically $H = H^1(S) \times H^{-1/2}_S(\Gamma)$.
• Operator Decomposition: The single-layer operator $V_S$ is decomposed into a positive, coercive part ($V_{S,0}$) and a compact smoothing part ($V_{S,1}$).
• Gårding Inequality: Using this decomposition, the authors prove that the associated bilinear form satisfies a Gårding inequality. Combined with the Fredholm alternative, this establishes the existence and uniqueness of weak solutions.

C. Regularity and Singularity Analysis (Wiener–Hopf Technique)

To address the lack of regularity at the endpoints, the authors utilize the Wiener–Hopf decomposition technique:

1. Half-Line Model: They first analyze the system on a half-line ($S = \mathbb{R}_+$) using Fourier transforms and symbol factorization.
2. Asymptotic Expansion: They derive the precise asymptotic behavior of the solution near the endpoints.
   • The density $\psi$ exhibits a singularity of order $d^{-1/2}$.
   • The surface field $U$ exhibits a singularity of order $d^{3/2}$.
3. Global Regularity: By patching local solutions using a partition of unity, they establish that the remainder of the solution (after subtracting the singular terms) possesses higher regularity ($H^{5/2+\epsilon}$ for $U$ and $H^{1/2+\epsilon}$ for $\psi$).

D. Finite Element Method (FEM)

Based on the derived singularity structure, the authors propose an enriched Finite Element Method:

• Standard FEM Limitation: Conventional piecewise-polynomial FEM suffers from reduced convergence rates (approx. $O(h^{1/2})$) because it cannot resolve the $d^{-1/2}$ singularity.
• Enriched Space: The discrete space is augmented with singular basis functions that explicitly capture the $d^{-1/2}$ and $d^{3/2}$ behaviors near the endpoints.
• Error Analysis: A rigorous a priori error analysis proves that the enriched method restores optimal convergence rates ($O(h)$ in the energy norm).

3. Key Contributions

1. First Well-Posedness Proof for Open Surfaces: This is the first work to rigorously establish the well-posedness (existence and uniqueness) of bulk-surface coupled PDEs specifically for open boundaries, a case previously unaddressed compared to closed surfaces.
2. Singularity Characterization: The paper provides a precise mathematical characterization of the edge singularities using Wiener–Hopf techniques, identifying the exact power-law decay rates ($d^{-1/2}$ and $d^{3/2}$) for the density and surface field, respectively.
3. Enriched Numerical Scheme: The development of a finite element method that incorporates these singularities. This approach overcomes the "pollution" of standard FEM near geometric singularities.
4. Rigorous Convergence Proof: The authors provide a theoretical proof showing that the enriched method achieves first-order convergence, whereas standard FEM is limited to half-order convergence.

4. Results

• Theoretical:
  • Proved that the weak solution exists and is unique in the space $H^1(S) \times H^{-1/2}_S(\Gamma)$.
  • Demonstrated that if the data is smooth, the solution admits an asymptotic expansion involving singular terms plus a smooth remainder.
  • Established that the solution is a classical solution ($C^2$ away from endpoints, $C^1$ up to endpoints) satisfying the PDE pointwise.
• Numerical:
  • Example 1 (Straight Line): For a straight line segment with constant forcing, the standard FEM showed a convergence order of $\approx 0.5$, while the enriched FEM achieved $\approx 0.9$ (approaching 1.0).
  • Example 2 (Semicircle): Similar results were observed for a curved boundary. The standard FEM exhibited spurious oscillations near the endpoints due to the unresolved singularity, whereas the enriched FEM captured the singular behavior smoothly and accurately.
  • The energy norm errors confirmed the theoretical prediction that enriching the basis functions is necessary to recover optimal convergence rates.

5. Significance

This work bridges a critical gap in the mathematical analysis of coupled bulk-surface problems, which are ubiquitous in biophysics (e.g., chemical reactions on cell membranes) and materials science (e.g., diffusion in crystalline systems with cracks or interfaces).

• Theoretical Impact: It extends the theory of boundary integral equations to non-Lipschitz domains with open boundaries, providing a framework for analyzing edge singularities in coupled systems.
• Computational Impact: It offers a practical and efficient numerical strategy. By explicitly modeling the singularity, computational resources are not wasted on refining meshes to resolve singularities that can be handled analytically. This leads to more accurate simulations of physical phenomena involving open interfaces with significantly fewer degrees of freedom compared to standard adaptive mesh refinement.