Fast reaction limits and convergence rate for nonlinear bulk-surface reaction-diffusion systems modeling reversible chemical reactions

1. Problem Statement

The paper investigates the fast reaction limit of a nonlinear bulk-surface reaction-diffusion system modeling a reversible chemical reaction. The physical setting involves:

• Bulk Domain ($\Omega$): A bounded domain in $\mathbb{R}^N$ containing a chemical species $U$ with concentration $u(x,t)$.
• Surface Boundary ($\Gamma = \partial\Omega$): The boundary where a second species $V$ with concentration $v(x,t)$ resides.
• Reaction: A reversible reaction occurs on the surface $\Gamma$:
  $$\alpha U \underset{k}{\overset{k}{\rightleftharpoons}} \beta V$$
  where $\alpha, \beta \geq 1$ are arbitrary stoichiometric coefficients, and $k$ is the reaction rate constant.
• Dynamics: $U$ diffuses in $\Omega$, and $V$ diffuses along $\Gamma$ (governed by the Laplace-Beltrami operator). The reaction couples the bulk and surface via a flux condition.

The governing system (with $k = 1/\epsilon$) is:
$$\begin{cases} \partial_t u - d_u \Delta u = 0, & x \in \Omega, t > 0 \\ \partial_n u = -\frac{\alpha}{\epsilon}(u^\alpha - v^\beta), & x \in \Gamma, t > 0 \\ \partial_t v - d_v \Delta_\Gamma v = \frac{\beta}{\epsilon}(u^\alpha - v^\beta), & x \in \Gamma, t > 0 \end{cases}$$
The authors aim to prove that as the reaction rate $k \to \infty$ (i.e., $\epsilon \to 0^+$), the solution $(u_\epsilon, v_\epsilon)$ converges to a limit system $(w, z)$, and to quantify the rate of this convergence.

2. Methodology

The authors employ a combination of entropy methods, a-priori estimates, and compactness arguments to handle the singular limit.

A. Existence and Uniqueness

• Approximation: To handle the nonlinearity and potential singularities, they first introduce an approximate system with a regularized reaction term $F_\delta(u, v) = \frac{u^\alpha - v^\beta}{1 + \delta|u^\alpha - v^\beta|}$.
• Galerkin Method: They prove the existence of global weak solutions for the approximate system using the Galerkin method.
• Entropy Functionals: Crucial to the analysis are specific entropy functionals ($L^p$-type and logarithmic) that are non-increasing in time. These provide uniform bounds on the solutions independent of the parameters $\epsilon$ and $\delta$.
  • $L^\infty$ Bounds: Derived from $L^p$ entropy estimates as $p \to \infty$.
  • Gradient Bounds: Derived from logarithmic entropy estimates (involving terms like $\int \frac{|\nabla u|^2}{u}$).

B. The Fast Reaction Limit ($\epsilon \to 0$)

• Product Space Approach: A major difficulty is the bulk-surface coupling, which prevents simply summing the equations (a technique used in pure bulk systems). The authors define a product space $Z = \{(\phi, \phi|_\Gamma) : \phi \in H^1(\Omega), \phi|_\Gamma \in H^1(\Gamma)\}$.
• Time Derivative Estimates: By testing the system with functions in $Z$ (specifically combining the bulk and surface equations with weights $\beta$ and $\alpha$), they eliminate the singular reaction term $\frac{1}{\epsilon}(u^\alpha - v^\beta)$. This allows them to bound the time derivative $\partial_t(u_\epsilon, v_\epsilon)$ in the dual space $L^2(0, T; Z^*)$.
• Compactness: Using the Aubin-Lions lemma on the product space, they establish strong convergence of a subsequence.
• Limit Identification: They prove that the limit satisfies $u^\alpha = v^\beta$ on the boundary, allowing the reduction of the limit system to a single parabolic equation with a nonlinear dynamical boundary condition.

C. Convergence Rate ($\alpha = \beta$)

• Regularity Assumptions: To obtain a convergence rate, the authors assume equal stoichiometric coefficients ($\alpha = \beta$) and strictly positive initial data ($u_0, v_0 \geq m_0 > 0$).
• Lower Bounds: They prove that the solutions remain strictly positive uniformly in $\epsilon$, which is essential for handling the nonlinear terms in the error analysis.
• Energy Difference: They define the error variables $U = u_\epsilon - w$ and $V = v_\epsilon - w|_\Gamma$. By analyzing the time derivative of the $L^2$-norm of the error and utilizing the mean value theorem on the reaction term, they derive a differential inequality.
• Gronwall-type Argument: The singular term provides a strong damping effect ($-\frac{1}{\epsilon}\|U-V\|^2$), which dominates the error growth, leading to the convergence rate.

3. Key Contributions and Results

1. Strong Convergence to a Limit System

The primary result is the proof that as $\epsilon \to 0$, the solution $(u_\epsilon, v_\epsilon)$ converges strongly in $L^p(0, T; L^p(\Omega) \times L^p(\Gamma))$ to a limit $(w, z)$.

• The Limit System: The limit is a heat equation in the bulk with a nonlinear dynamical boundary condition:
  $$\begin{cases} \partial_t w - d_u \Delta w = 0, & x \in \Omega \\ d_u \nabla w \cdot n = -\frac{\alpha}{\beta} \left( \partial_t (w^{\alpha/\beta}) - d_v \Delta_\Gamma (w^{\alpha/\beta}) \right), & x \in \Gamma \end{cases}$$
  This extends previous linear results ($\alpha=\beta=1$) to the fully nonlinear case with arbitrary $\alpha, \beta \geq 1$.

2. Novel Technical Approach for Coupling

The paper overcomes the difficulty of bulk-surface coupling by utilizing product spaces ($Z$) and the Aubin-Lions lemma. This approach avoids the need for the system to be linear or for the equations to be summed in a way that eliminates the reaction term directly, which was a limitation in previous works.

3. Quantitative Convergence Rate

For the specific case of equal stoichiometric coefficients ($\alpha = \beta$) and strictly positive initial data, the authors establish a convergence rate:
$$\|u_\epsilon(t) - w(t)\|_{L^2(\Omega)} + \|v_\epsilon(t) - w|_\Gamma(t)\|_{L^2(\Gamma)} \leq C \sqrt{\epsilon}$$
This $O(\sqrt{\epsilon})$ rate is derived by exploiting the strict positivity of the solutions and the specific structure of the limit system.

4. Significance

• Mathematical Rigor: The paper resolves an open problem regarding the fast reaction limit for nonlinear bulk-surface systems, which had previously only been solved for linear cases.
• General Applicability: The methodology using entropy functionals and product spaces is robust and can likely be adapted to other coupled bulk-surface problems involving singular perturbations.
• Physical Relevance: The results provide a rigorous mathematical foundation for modeling chemical reactions where the reaction rate is much faster than diffusion, a common scenario in cell biology (e.g., membrane signaling) and material science. The derived limit system offers a computationally efficient model (reducing a coupled system to a single equation with a complex boundary condition) for such fast-reaction regimes.
• Error Estimation: The derivation of the convergence rate is significant for numerical analysis, as it informs the accuracy of approximating the fast-reaction system with the limit system in simulations.