English

Overdetermined problems with fractional Laplacian

Analysis of PDEs 2025-06-23 v4

Abstract

Let N1N\geq 1 and s(0,1)s\in (0,1). In the present work we characterize bounded open sets Ω\Omega with C2 C^2 boundary (\textit{not necessarily connected}) for which the following overdetermined problem \begin{equation*} ( -\Delta)^s u = f(u) \text{ in Ω\Omega,} \qquad u=0 \text{ in RNΩ\mathbb{R}^N\setminus \Omega,} \qquad(\partial_{\eta})_s u=Const. \text{ on Ω\partial \Omega} \end{equation*} has a nonnegative and nontrivial solution, where η\eta is the outer unit normal vectorfield along Ω\partial\Omega and for x0Ωx_0\in\partial\Omega (η)su(x0)=limt0u(x0tη(x0))ts. \left(\partial_{\eta}\right)_{s}u(x_{0})=-\lim_{t\to 0}\frac{u(x_{0}-t\eta(x_0))}{t^s}. Under mild assumptions on ff, we prove that Ω\Omega must be a ball. In the special case f1f\equiv 1, we obtain an extension of Serrin's result in 1971. The fact that Ω\Omega is not assumed to be connected is related to the nonlocal property of the fractional Laplacian. The main ingredients in our proof are maximum principles and the method of moving planes.

Keywords

Cite

@article{arxiv.1311.7549,
  title  = {Overdetermined problems with fractional Laplacian},
  author = {Mouhamed Moustapha Fall and Sven Jarohs},
  journal= {arXiv preprint arXiv:1311.7549},
  year   = {2025}
}

Comments

Added a missing assumption (1.3) in Theorem 1.1 and Theorem 1.2, which is used in the proof of Lemma 4.3

R2 v1 2026-06-22T02:17:30.345Z