On a Nirenberg-type problem involving the square root of the Laplacian (Q2440974)

From MaRDI portal
Revision as of 23:01, 19 March 2024 by Openalex240319060354 (talk | contribs) (Set OpenAlex properties.)
scientific article
Language Label Description Also known as
English
On a Nirenberg-type problem involving the square root of the Laplacian
scientific article

    Statements

    On a Nirenberg-type problem involving the square root of the Laplacian (English)
    0 references
    0 references
    0 references
    21 March 2014
    0 references
    The authors solve the following nonlinear equation on \(\mathbb S^2\) \[ \frac{\Gamma\Big(\sqrt{(-\Delta_g)+\frac{1}{4}}+1\Big)}{\Gamma\Big(\sqrt{(-\Delta_g)+\frac{1}{4}}\Big)}u=c_2Ku^3,\qquad u>0,\leqno {(*)} \] where \(c_2\) is an explicit constant, \(\Delta_g\) is the Laplace-Beltrami operator, \(K\) belongs to the Hölder space \(C^{1,1}(\mathbb S^2)\) and \(\Gamma\) is the gamma function. The proof is based on the study of the non-compact orbits of the gradient flow of an explicit functional \(J\) associated to \((*)\), these orbits are called \textit{critical points at infinity} of \(J\). Theorem 1.1 is the main result which states: if \(K\) satisfies \(\Delta K(y)\neq0\) when \(\nabla K(y)=0\) and the condition \(\sum_{y\in \mathcal{K}^+}(-1)^{2-\widetilde{i}(y)}\neq 1\), then \((*)\) has at least one solution, where \(\widetilde{i}(y)\) stands for the index of \(K\) at \(y\in \mathcal{K}^+:=\{y\in \mathbb S^2\text{ such that }\nabla K(y)=0,\;-\Delta_gK(y)>0\}\). The authors study the properties of \(J\), precisely, close to potential critical points at infinity, they give an expansion for both \(J\) and its gradient \(\nabla J\) (see the third section). Next, the authors give a characterization of critical points at infinity, this by building up an appropriate pseudo-gradient at infinity (see Proposition 4.1 and Proposition 4.3). The fifth section is reserved on the proof of the main theorem. A similar theorem of Theorem 1.1 is provided when \(K\in C^{1,1}(\mathbb S^n)\) with \(n\in [2,2+2\sigma)\) such that \(\sigma\in (0,1)\), see Theorem 6.1, and regarding its demonstration, the authors provide the outlines.
    0 references
    fractional Laplacian
    0 references
    critical exponent
    0 references
    critical points at infinity
    0 references
    Nirenberg problem
    0 references

    Identifiers