Serrin-Type Results in Overdetermined PDEs

• Serrin-type results are symmetry and rigidity theorems in elliptic PDEs that force domains to be balls with radially symmetric solutions.
• They use the method of moving planes, maximum principles, and Hopf’s lemma to handle both scalar and system overdetermined boundary conditions.
• These theorems have practical implications in shape optimization and free-boundary problems by linking boundary data with explicit domain geometry.

A Serrin-type result refers to a class of rigidity and symmetry theorems for overdetermined boundary value problems (BVPs) in partial differential equations (PDEs), where the domain and solution are determined uniquely, often as a ball and a radial function, by the imposition of both Dirichlet and suitably chosen Neumann (or more general) boundary conditions. These results generalize and extend the foundational theorem established by J. Serrin in 1971 for the torsion problem and form a cornerstone of modern elliptic PDE, geometric analysis, and shape optimization.

1. Classical Serrin Problem and Overdetermined Systems

The original Serrin problem considers the torsion equation for a scalar function $u$ in a bounded domain $\Omega \subset \mathbb{R}^n$: $$-\Delta u = 1 \quad \text{in } \Omega, \qquad u = 0, \quad \partial_\nu u = \text{constant} \quad \text{on } \partial \Omega.$$ Serrin’s theorem establishes that under these conditions, $\Omega$ must be a ball and $u$ is radial and decreasing away from the center. The crux of the proof is the method of moving planes, exploiting maximum principles and Hopf-type boundary point lemmas.

The work of Celentano, Nitsch, and Trombetti extends this paradigm to systems of semilinear elliptic equations. Let $m \ge 1$ and consider

$$-\Delta u_i = f_i(u_i) \quad \text{in } \Omega, \qquad u_i=0 \ \text{on } \partial\Omega, \quad i=1,\ldots, m,$$

with the overdetermined boundary constraint

$$F(\partial_\nu u_1, ..., \partial_\nu u_m) = c \quad \text{on } \partial\Omega,$$

where $F$ is $C^1$, nondecreasing in each argument, strictly increasing in at least one, and each $\Omega \subset \mathbb{R}^n$0 with $\Omega \subset \mathbb{R}^n$1 locally Lipschitz, $\Omega \subset \mathbb{R}^n$2 nondecreasing. Their main result is:

Theorem (Celentano–Nitsch–Trombetti).

If the above system admits a positive classical solution in a smooth, bounded, connected domain, then $\Omega \subset \mathbb{R}^n$3 must be a ball and each $\Omega \subset \mathbb{R}^n$4 is radial and strictly decreasing in radius (Celentano et al., 2024).

2. Structural Hypotheses and Nonlinear Features

A critical aspect of modern Serrin-type results is the flexibility in the nonlinearities allowed:

• Each $\Omega \subset \mathbb{R}^n$5 splits into a locally Lipschitz part and a nondecreasing part, accommodating both nonlinear source terms and reaction phenomena.
• The coupling of Dirichlet and generalized Neumann data is only through a scalar boundary function $\Omega \subset \mathbb{R}^n$6, which can be nonlinear and need not be strictly increasing in all directions—strict monotonicity in a single component suffices.

The symmetry conclusion leverages:

• The strong maximum principle for systems,
• The Gidas–Ni–Nirenberg result on monotonicity and symmetry for semilinear PDE,
• A refined use of the Hopf boundary lemma that accounts for the coupling structure in $\Omega \subset \mathbb{R}^n$7.

This overall framework extends rigidity far beyond the classical case, accommodating systems, nonlinearity, and specialized symmetric configurations dictated purely by boundary data (Celentano et al., 2024).

3. Symmetry via the Method of Moving Planes

The moving-planes method is central. For the elliptic system, the proof strategy is:

• In each direction, consider reflection about a hyperplane $\Omega \subset \mathbb{R}^n$8, comparing $\Omega \subset \mathbb{R}^n$9 and its reflection $$-\Delta u = 1 \quad \text{in } \Omega, \qquad u = 0, \quad \partial_\nu u = \text{constant} \quad \text{on } \partial \Omega.$$0.
• At the "critical" position, cases of tangential and orthogonal touching are analyzed.
  • If the reflected solution fails to coincide with the original (strict inequality), Hopf’s lemma leads to a contradiction with the strict monotonicity of $$-\Delta u = 1 \quad \text{in } \Omega, \qquad u = 0, \quad \partial_\nu u = \text{constant} \quad \text{on } \partial \Omega.$$1.
  • If equality is achieved, symmetry is established by unique continuation.
• By iterating over all directions, the domain is shown to be a ball, and solutions to be radially symmetric.

This method flexibly adapts to multiple equations coupled only on the boundary and incorporates nonlinear Neumann-type functionals, as in the generalized framework introduced by Celentano et al. (Celentano et al., 2024).

4. Applications: Symmetry in Shape Optimization

Serrin-type results have powerful implications in variational problems with geometric constraints. Consider

$$-\Delta u = 1 \quad \text{in } \Omega, \qquad u = 0, \quad \partial_\nu u = \text{constant} \quad \text{on } \partial \Omega.$$2

where $$-\Delta u = 1 \quad \text{in } \Omega, \qquad u = 0, \quad \partial_\nu u = \text{constant} \quad \text{on } \partial \Omega.$$3 is torsional rigidity (integral of the unique Dirichlet solution to $$-\Delta u = 1 \quad \text{in } \Omega, \qquad u = 0, \quad \partial_\nu u = \text{constant} \quad \text{on } \partial \Omega.$$4), and $$-\Delta u = 1 \quad \text{in } \Omega, \qquad u = 0, \quad \partial_\nu u = \text{constant} \quad \text{on } \partial \Omega.$$5 is the first Dirichlet eigenvalue. Under a volume constraint, the Euler–Lagrange equation for critical domains leads to an overdetermined system: $$-\Delta u = 1 \quad \text{in } \Omega, \qquad u = 0, \quad \partial_\nu u = \text{constant} \quad \text{on } \partial \Omega.$$6 where $$-\Delta u = 1 \quad \text{in } \Omega, \qquad u = 0, \quad \partial_\nu u = \text{constant} \quad \text{on } \partial \Omega.$$7 and $$-\Delta u = 1 \quad \text{in } \Omega, \qquad u = 0, \quad \partial_\nu u = \text{constant} \quad \text{on } \partial \Omega.$$8 solve $$-\Delta u = 1 \quad \text{in } \Omega, \qquad u = 0, \quad \partial_\nu u = \text{constant} \quad \text{on } \partial \Omega.$$9, $\Omega$0 under Dirichlet conditions (Celentano et al., 2024).

This precisely matches the multi-component, boundary-coupled system in the main theorem and implies that any domain critical for such a functional under a volume constraint must be a ball.

5. Generalization and Rigidity in Related Models

Modern extensions, inspired by the same principles, include:

• More general elliptic operators (quasilinear, fully nonlinear),
• Problems in spaces of varying curvature (Riemannian, warped products),
• Couplings via nonlinear, possibly degenerate boundary data,
• Overdetermined problems on domains with less regularity and even in geometric measure theory.

A key unifying principle is that posing Dirichlet and suitably coupled Neumann (or generalized) boundary data on a PDE system singles out the Euclidean ball as the unique admissible domain, leading to so-called "ball-rigidity." Variations of the moving-planes method, maximum principles, P-function arguments, and boundary analysis remain at the heart of all such results (Celentano et al., 2024).

6. Methodological and Theoretical Significance

Serrin-type theorems provide:

• Central tools for proving uniqueness and structure in overdetermined PDE and variational problems,
• Deep connections between boundary data, domain geometry, and solution symmetry,
• Methodological templates adaptable to nonlinear, system, and geometric PDE settings.

They serve not only as classification results but as guiding principles in shape optimization, geometric analysis, free-boundary problems, and the study of rigidity phenomena in elliptic PDEs (Celentano et al., 2024).

• V. Celentano, C. Nitsch, C. Trombetti, "On a Serrin-type overdetermined problem" (Celentano et al., 2024)