Integral formulas for hypersurfaces in cones and related questions
Filomena Pacella, Giulio Tralli · 2025
Read Integral formulas for hypersurfaces in cones and related questions by Filomena Pacella, Giulio Tralli (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-13137.txt.
Introduction
Filomena Pacella and Giulio Tralli study integral formulas for hypersurfaces in cones in two thousand twenty-five. The surfaces meet the cone wall at right angles. Minkowski type identities are checked in that setting. In the course of the work they correct a formula from past literature.
They then turn to graphs with constant mean curvature. A rigidity result from past work is stated in exact form and proved. Such results say that strong traits force a surface into a known shape. Last they give an integral bound for stable surfaces of that type in cones. The paper is a compact piece of pure geometry.
Catalog record
| CREATOR | Filomena Pacella, Giulio Tralli |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-13137
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Integral formulas for hypersurfaces in cones and related questions Filomena Pacella, Giulio Tralli Abstract . We discuss the validity of Minkowski integral identities for hypersurfaces inside a cone, intersecting the boundary of the cone orthogonally. In doing so we correct a formula provided in [ 3 ] . Then we study rigidity results for constant mean curvature graphs proving the precise statement of a result given in [ 9 ] and [ 10 ] . Finally we provide an integral estimate for stable constant mean curvature hypersurfaces in cones. 1. Introduction Let \omega be a smooth open domain on the unit sphere S^{N-1} , N\geq 2 , and let \Sigma_{\omega} be the open cone in \mathbb{R}^{N} with vertex at the origin O \Sigma_{\omega}=\{tx\,:\,x\in\omega,\,\,t\in(0,+\infty)\}. We consider a smooth hypersurface \Gamma\subset\Sigma_{\omega} with \partial\Gamma\subset\partial\Sigma_{\omega}\smallsetminus\{O\} , and denote by \nu=\nu_{x} the (exterior) unit normal to \Gamma for any x\in\overline{\Gamma} . Instead we denote by n=n_{x} the exterior unit normal to the cone at any point x\in\partial\Sigma_{\omega}\smallsetminus\{O\} . We focus on the case when \Gamma and \partial\Sigma_{\omega} intersect orthogonally which means that we assume (1.1) \left\langle\nu_{x},n_{x}\right\rangle=0\qquad\forall\,x\in\partial\Gamma. Let \Gamma_{\omega,R} be the piece of a sphere centered at the vertex O contained in the cone, i.e. \Gamma_{\omega,R}=\partial B_{R}\cap\Sigma_{\omega} where B_{R} is the ball of radius R>0 centered at the origin. Then the orthogonality condition ( 1.1 ) is obviously satisfied for \Gamma=\Gamma_{\omega,R} . The main aim of our paper is to discuss the validity of integral identities of Hsiung-Minkowski type for hypersurfaces \Gamma in the cone, satisfying the above orthogonality condition. It is already known that the following first Minkowski formula holds true (1.2) \int_{\Gamma}{1-H\left\langle x,\nu_{x}\right\rangle}=0, where H=H(x) is the mean curvature of \Gamma , see [ 12 , formula (4.8)] or [ 3 , Section 1] , see also [ 9 , page 858] and [ 10 , formula 2.3] . A short proof of ( 1.2 ) will be provided in Section 2 , for the reader convenience. Denoting by II^{\Sigma_{\omega}}(\cdot,\cdot) the second fundamental form of \partial\Sigma_{\omega} and by II^{\Gamma} the one of \Gamma with the associated second elementary symmetric function of its principal curvatures \sigma_{2} (see Section 2 for the definition) here we prove the following second Minkowski identity: Theorem 1.1 . Let N\geq 3 . Under the standing assumptions for \Sigma_{\omega} and \Gamma , we have (1.3) \int_{\Gamma}{H-\sigma_{2}\left\langle x,\nu\right\rangle}=\frac{-1}{(N-1)(N-2)}\int_{\partial\Gamma}{II^{\Sigma_{\omega}}\left(\nu-\left\langle x,\nu\right\rangle\frac{x}{|x|^{2}},\nu-\left\langle x,\nu\right\rangle\frac{x}{|x|^{2}}\right)\left\langle x,\nu\right\rangle}. A second Minkowski formula in cones under the orthogonality condition is claimed in [ 3 , formula (7) in Proposition 1, case k=2 ] but it is wrong since the integral in the right-hand side of ( 1.3 ) is missing. Actually this integral term is not zero in general when N\geq 3 , not even in the case when the cone is convex (in which case is non-positive). In [ 3 ] also higher order Minkowski formulas are claimed and we believe that they all would need some additional term and additional arguments. In Section 2 we provide two proofs of Theorem 1.1 , the second one follows the approach of [ 3 ] with all details there missing. The wrong statement in [ 3 , Proposition 1] induced in turn a wrong rigidity theorem in our previous paper [ 10 , Theorem 1.1] , see also [ 9 , Theorem 6.4] , about constant mean curvature surfaces (CMC in short) \Gamma which are starshaped with respect to the vertex of the (possibly non-convex) cone and intersect the cone in an orthogonal way. Thus, in Section 3 we prove the correct rigidity statement in Theorem 3.1 which needs an integral gluing condition to hold. Some connections between Theorem 3.1 and isoperimetric inequality for almost convex cones proved by Baer-Figalli in [ 1 ] will be commented. Indeed the CMC surfaces intersecting the cone orthogonally are the critical points of the relative perimeter functional, under a volume constraint, while the isoperimetric inequality aims to characterize the minimizers of the perimeters. In [ 7 ] it has been proved that, in convex cones, the only sets in \Sigma_{\omega} which minimize the relative perimeter are the spherical sectors. A similar characterization was obtained in [ 12 ] for the stable critical points of the relative perimeter, in convex cones. In Section 4 we prove an inequality, Theorem 4.1 , for stable critical points which allows to relax the convexity assumption of [ 12 ] . It is interesting that it depends on the first Neumann eigenvalue of the Laplace-Beltrami operator on \Gamma . Recently, in [ 6 ] , a connection between the stability of the spherical sector, as a critical point of the relative perimeter, and the first Neumann eigenvalue of the Laplace-Beltrami operator on \omega\subset S^{N-1} was established. It would be interesting to further intestigate the role of these Neumann eigenvalues in the study of CMC surfaces in cones. 1.1. Notations Let us further clarify here the assumptions and the notations used throughout the paper. The domain \omega on the sphere S^{N-1} is assumed to have a C^{2} -smooth boundary. Also, the embedded (N-1) -dimensional manifold \Gamma is assumed to be relatively open, C^{3} -smooth in its relative interior, and C^{2} -smooth up to its (non-empty) relative boundary \partial\Gamma contained in \partial\Sigma_{\omega}\smallsetminus\{O\} ; \Gamma will be always bounded, connected, and orientable. Under these assumptions, there is a bounded open set \Omega\subset\Sigma_{\omega} (called sector-like domain in [ 9 , 10 ] ) such that \Gamma=\partial\Omega\cap\Sigma_{\omega} : the choice of the unit normal \nu is fixed as the exterior unit normal with respect to \Omega . Having to deal with the two hypersurfaces \Gamma and \partial\Sigma_{\omega} , we shall introduce the relevant geometric objects for our analysis. We denote by \nabla the standard Levi-Civita connection in \mathbb{R}^{N} . Locally around any point in \Gamma , we can consider an orthonormal frame \{e_{1},\ldots,e_{N-1}\} , and define (with a slight abuse of notations) by \nabla^{\Gamma} the gradient with respect to the tangent directions to \Gamma that is \nabla^{\Gamma}u=\sum_{j=1}^{N-1}e_{j}(u)e_{j} . Also, for x\in\Gamma , the second fundamental form II_{x}=II^{\Gamma}_{x} of \Gamma is the bilinear symmetric form on T_{x}\Gamma\times T_{x}\Gamma which can be defined as II^{\Gamma}(e_{i},e_{j})=\left\langle\nabla_{e_{i}}\nu,e_{j}\right\rangle,\qquad\mbox{for }i,j\in\{1,\ldots,N-1\}. An important role in what follows is played by the mean curvature H of \Gamma , and (in case N\geq 3 ) by the second elementary symmetric function \sigma_{2} related to II^{\Gamma} : we recall here the definitions, namely H=\frac{1}{N-1}\sum_{j=1}^{N-1}II^{\Gamma}(e_{j},e_{j}),\quad\mbox{ and } \sigma_{2}=\frac{1}{(N-1)(N-2)}\left((N-1)^{2}H^{2}-\sum_{i,j=1}^{N-1}(II^{\Gamma}(e_{i},e_{j}))^{2}\right). The analogous definitions, starting from the tangential derivatives of the normal direction to the cone n , are adopted for the second fundamental form II^{\Sigma_{\omega}} of the cone \partial\Sigma_{\omega} . The reader has to keep in mind that the radial direction x (we decided to adopt the same notation for the position vector and the radial vector field with respect to O ) is always a tangent direction for the cone, i.e. (1.4) \left\langle x,n_{x}\right\rangle=0\qquad\mbox{ for any }x\in\partial\Sigma_{\omega}\smallsetminus\{O\}. Furthermore, the radial direction is also a direction of flatness in the sense that (1.5) II_{x}^{\Sigma_{\omega}}(x,\cdot)\equiv 0. One can easily check the previous identity by considering any tangential direction e\in T_{x}\partial\Sigma_{\omega} (for arbitrary x\in\partial\Sigma_{\omega}\smallsetminus\{O\} ) and computing II_{x}^{\Sigma_{\omega}}(x,e)=-\left\langle n_{x},\nabla_{e}x\right\rangle=-\left\langle n,e\right\rangle=0 . We say that the cone is convex if II_{x}^{\Sigma_{\omega}}\geq 0\qquad\mbox{ for any }x\in\partial\Sigma_{\omega}\smallsetminus\{O\} in the sense of bilinear forms. Finally, we fix the notations for the first (non-trivial)