Stability and bifurcation for surfaces with constant mean curvature (Q1687577)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Stability and bifurcation for surfaces with constant mean curvature |
scientific article |
Statements
Stability and bifurcation for surfaces with constant mean curvature (English)
0 references
4 January 2018
0 references
Assume that \(\Sigma\) is a two-dimensional, oriented, compact, connected smooth manifold with smooth boundary \(\partial\,\Sigma\), and \(X:\Sigma\to\mathbb R^3\) is a smooth immersion having constant mean curvature equal to \(H_0\). If \(\nu:\Sigma\to S^2\subset\mathbb R^3\) is the Gauss map of \(X\), then the volume \(V_0\) of \(X\) as the volume of the cone over \(X\) is explicitly given by \(\frac13\int_\Sigma\langle X,\nu\rangle d\,\Sigma\), where \(\langle\;,\;\rangle\) stands for the canonical inner product in \(\mathbb R^3\) and \(d\,\Sigma\) is the area element of the induced metric on \(\Sigma\). The mean curvature of an immersion \(X\) is constant if and only if \(X\) is a critical point of the area functional for all volume-preserving variations of \(X\) that fix the boundary values. For a volume-preserving, boundary preserving variation \(X_t\) of \(X\), the second variation of the area functional \(A\) is given by \(\delta^2A=-\int_\Sigma\varphi L[\varphi] d\,\Sigma\), where \(\varphi=\left\langle \frac{\partial X_t}{\partial t}\big|_{t=0},\nu\right\rangle\), \(L\) is the self-adjoint operator associated with \(\delta^2A\) defined as \(L[\varphi]=2\delta H=\Delta\varphi+\|d\nu\|^2\varphi\), \(\partial H\) is the first variation of the mean curvature, and \(\Delta\) is the Laplacian on \(\Sigma\) with the metric induced by \(X\). A constant mean curvature \(X:\Sigma\to\mathbb R^3\) is said to be stable if the second variation of the area functional is nonnegative for all volume-preserving variations that fix the boundary. \(X\) is said to be unstable if it is not stable. If \(E=\ker(L)=\{0\}\), then \(X\) admits smooth deformations of the form \(X_t=X+\varphi(t)\nu\), for some smooth \(1\)-parameter family \((-\varepsilon,\varepsilon)\ni t\mapsto\varphi(t)\) with \(\varphi(0)=0\), such that the mean curvature of \(X_t\) is constant and equal to \(H_0+t\) for all \(t\). In this paper, the authors give criteria for the existence of smooth bifurcation branches of fixed boundary constant mean curvature surfaces in \(\mathbb R^3\) and discuss stability/instability issues for the surfaces in bifurcating branches. It is shown that if \(X:\Sigma\to\mathbb R^3\) is an immersion with constant mean curvature \(H_0\), \(\dim(E)=1\), and \(\int_\Sigma ed\Sigma\neq 0\) for some \(e\in E\setminus\{0\}\), then there exists \(\varepsilon >0\) and a map \((\xi,\eta):(-\varepsilon,\varepsilon)\to E^\perp\times\mathbb R\) of class \(C^1\) such that \((\xi(0),\eta(0))=(0,H_0)\), and for all \(t\in(\varepsilon,\varepsilon)\), the map \(X_t=X+(te+\xi(t))\nu:\Sigma\to\mathbb R^3\) is an immersion having constant mean curvature equal to \(\eta(t)\) and with \(X_t|_{\partial\Sigma}=X|_{\partial\Sigma}\). This property shows that no bifurcation from the family \((X_t)\) can occur at \(t=0\). The authors are interested in determining constant mean curvature immersions that bifurcate from a given one-parameter family \((X_t)\) and satisfy a fixed boundary condition. They show that if \(L[\varphi]=-\lambda\varphi\) with \(\varphi_{|\partial\Sigma}=0\) and \(\lambda_n\) are \(n\)-th eigenvalues of \(L\), then bifurcation may occur for the family \((X_t)\) only in the case where \(\lambda_k=0\) for some \(k\geq 2\). To study the stability of constant mean curvature surfaces in a bifurcation branch it is needed to study only the case when \(\lambda_2=0\). If \(H(t)\) and \(V(t)\) are respectively the mean curvature and the volume of \(X_t\), then \(X\) is stable when \(H'(0)V'(0)\geq 0\) and \(X\) is unstable when \(H'(0)V'(0)<0\).
0 references
bifurcation
0 references
constant mean curvature surfaces
0 references
stability
0 references