Geometric-type Sobolev inequalities and applications to the regularity of minimizers (Q1932408): Difference between revisions
From MaRDI portal
Changed an Item |
Normalize DOI. |
||
Property / DOI | |||
Property / DOI: 10.1016/j.jfa.2012.10.012 / rank | |||
Property / DOI | |||
Property / DOI: 10.1016/J.JFA.2012.10.012 / rank | |||
Normal rank |
Latest revision as of 13:36, 16 December 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Geometric-type Sobolev inequalities and applications to the regularity of minimizers |
scientific article |
Statements
Geometric-type Sobolev inequalities and applications to the regularity of minimizers (English)
0 references
18 January 2013
0 references
The authors prove the following geometric-type Sobolev and Morrey's inequalities. Let \(\Omega\) be a smooth bounded domain of \(\mathbb{R}^{n}\) with \(n\geq2\). Then for every \(p\geq1\) and \(r\in\{0\}\cup[1,\infty)\) there exists a constant \(C=C(n,p,r)>0\) such that the following inequalities hold for all functions \(u\in C_{0}^{\infty}(\Omega)\) (i.e.\(,\) \(u\in C^{\infty }(\Omega)\) and \(u=0\) on \(\partial\Omega)\): (a) \(\left\| u\right\| _{L^{\infty}(\Omega)}\leq C\left| \Omega\right| ^{\frac{p(1+r)-n}{np} }\left\| \left| H_{u}\right| ^{r}\left| \nabla u\right| \right\| _{L^{p}(\Omega\cap\left\{ \left| \nabla u\right| >0\right\} )}\) if \(1+r\leq n<p(1+r)\); (b) \(\left\| u\right\| _{L^{p_{r}^{\ast}}(\Omega)}\leq C\left\| \left| H_{u}\right| ^{r}\left| \nabla u\right| \right\| _{L^{p}(\Omega\cap\left\{ \left| \nabla u\right| >0\right\} )}\) if \(n>p(1+r)\). The case where \(p>1\) and \(n=p(1+r)\) is also considered. Here \(H_{u}=-\frac{1}{n-1} \text{div}\left( \frac{\nabla u}{\left| \nabla u\right| }\right) \) denotes the mean curvature of the level sets of \(\left| u\right| ,\) and the critical exponent \(p_{r}^{\ast}\) is defined by \(1/p_{r}^{\ast}=1/p-(1+r)/n\). This extends previous work due to Trudinger in the case of mean convex functions. As an application new \(L^{q}\) and \(W^{1,q}\) a priori estimates for local minimizers (and more generally, for semi-stable solutions) of reaction-diffusion problems are derived. These estimates lead to an \(L^{2n/(n-4)}(\Omega)\) bound for the extremal solution of \(-\Delta u=\lambda f(u)\) when \(n\geq5\) and the domain is convex. As the authors noticed, the extremal solutions are known to be bounded in convex domains if \(n\leq4,\) and their boundedness is expected (but still unknown) for \(n\leq9.\)
0 references
Sobolev inequalities
0 references
semilinear equations
0 references
regularity of stable solutions
0 references
mean curvature of level sets
0 references