On Serrin's overdetermined problem in space forms (Q2420673)

From MaRDI portal
Revision as of 13:05, 14 August 2024 by Daniel (talk | contribs) (‎Created claim: Wikidata QID (P12): Q129063127, #quickstatements; #temporary_batch_1723632976074)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
scientific article
Language Label Description Also known as
English
On Serrin's overdetermined problem in space forms
scientific article

    Statements

    On Serrin's overdetermined problem in space forms (English)
    0 references
    0 references
    0 references
    6 June 2019
    0 references
    Let $\Omega$ be a connected bounded domain with $C^1$-class boundary in a complete simply conected Riemannian manifold with constant sectional curvature $K$. Then the following theorem is proved. \par Theorem 1.1. If there exists a solution $v$ to the equation \begin{align*} \Delta v+nKv=-1, \ \mathrm{on} \ \Omega, \\ v=0, |\nabla v|=c >0, \ \mathrm{on} \ \partial\Omega. \end{align*} Then $\Omega$ is a geodesic ball $B_R$ and depend only on the distance from the center of $B_R$. \par Up to homotheties, $M$ is either of $\mathbb{R}^n$($K=0$), the hyperbolic space $\mathbb{H}^n$ ($K=-1$) and the hemisphere $\mathbb{S}^n_+$ ($K=1$). They are uniformly described imposing $I\times \mathbb{S}^{n-1}$ the metric $g=dr^2+h^2g_{\mathbb{S}^{n-1}}$ Under this description, taking the center of $B_R$ the origin, $v$ is given by \[v(r)=\frac{H(R)-H(r)}{n\dot{h}(R)}, \quad H(r)=\int_o^r h(s)ds. \] Further more $c$ and $R$ are related by $c=h(R)/n\dot{h}(R)$. \par When $M=\mathbb{R}^n$, this theorem proved for more general equation $\Delta v+f(v)=0$ [\textit{J. Serrin}, Arch. Ration. Mech. Anal. 43, 304--318 (1971; Zbl 0222.31007)]. Restricted $f(v)=1$, its proof is simplified by considering $P$-function (Weinberger, H.F.: Remark on the preceding paper of [\textit{H. F. Weinberger}, Arch. Ration. Mech. Anal. 43, 319--320 (1971; Zbl 0222.31008)]. The authors remark extra term $nKv$ needed to obtain the Hessian of $v$ in the radial case to proportional to the metric. \par The proof is done by considering the $P$-function \[P(v)=|\nabla v|^2+\frac{2}{n}v+Kv^2, \] as follows: Adopting the Bochner-Weitzenböck formula and Schwartz inequality, and use Poho$\mathrm{\check{z}}$ev identity in space forms [\textit{G. Ciraolo} and \textit{L. Vezzoni}, Commun. Contemp. Math. 19, No. 5, Article ID 1750001, 11 p. (2017; Zbl 1370.35260)], $P(v)=c^2$ is derived (Lemma 2.1, and 2.3, Corollary 2.2). Applying standard maximal princip;e in the cases $K=0,-1$, and apply the fact that the first eigenvalue of the Dirchlet Laplacian on the hemisphere is $n$ and the corresponding eigenfunction is strictly positive, in the case $K=1$, to this fact (cf. Lemma 2.4), Theorem 1.1 is proved in \S2, the last Section.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    space form
    0 references
    complete simply connected Riemannian manifold with constant sectional curvature
    0 references
    Serrin's overdetermined problem
    0 references
    P-function
    0 references
    0 references
    0 references
    0 references
    0 references