Positive solutions for singular elliptic equations with critical Hardy-Sobolev exponent (Q651980)

The authors investigate the following semilinear elliptic equations \[ \begin{aligned} -\Delta u + \lambda \frac{u}{| x -a | ^2} -\gamma \frac{u}{| x | ^2} = & \frac{Q(x)}{| x | ^s } | u | ^{2^*(s)-2}u + g(x,u) ,\quad u>0 \,\, \mathrm{in}\,\,\, \Omega ,\\ \frac{\partial u}{\partial \nu }+ \alpha (x) u=& 0\,\, \mathrm{on}\,\,\, \partial \Omega \end{aligned} \tag{1} \] where \(\Omega \subset {\mathbb R}^N\) \((N\geq 3)\) is a bounded domain with smooth boundary \(\partial \Omega \), \(0\) and \(a(\neq 0) \) belong to \(\partial \Omega \), \(\lambda >0\), \(0<\gamma < \overline{\gamma }:=(N-2)^2/4, 2^*(s)= 2(N-s)/(N-2)\) \((0<s<2)\) is the critical Hardy-Sobolev exponent. \(\nu \) denotes the unit outward normal vector to \(\partial \Omega \), \(Q(x)\) is a positive continuous function on \(\overline{\Omega }\) and \(0\leq \alpha (x) \in L^{\infty }(\partial \Omega )\). The authors assume that the nonlinear term \(g(x,u)\) satisfies \((g_1)\) there exists a function \(a\in L^{\infty }(\Omega ), a(x) \leq 0 \) such that \(\displaystyle{\lim _{t \to 0^+}\frac{g(x,t)}{t}= a(x) }\) uniformly for \( x \in \overline{\Omega }\), \((g_2)\) \(\displaystyle{ \lim _{t \to + \infty } \frac{g(x,t)}{t^{2^*(s)-1}}=0}\) uniformly for \(x \in \overline{\Omega }\), \((g_3)\) \(\displaystyle{ \frac12 g(x,t)t -G(x,t)\geq 0}\) where \(G(x,t)=\int _0^tg(x,s)ds\). Then they get the main theorem. \textbf{Theorem 1.1} Assume that \(N\geq 5\), \(0<\gamma < \overline{\gamma }-1\), \(H(0)>0\) where \(H(x) \) denotes the mean curvature at \(x\). If \(| Q(x)-Q(0)| =o(| x | )\) for \(x\) close to \(0\), \((g_1), (g_2) \) and \((g_3)\) hold, \(\| \alpha \| _{\infty }\) small enough and \(\gamma < \gamma _0\), then the problem (1) admits a positive solution. Theorem 1.1 generalizes the work of \textit{Y.-Y. Shang} and \textit{C.-L. Tang} [Nonlinear Anal., Theory Methods Appl. 70, No. 3, A, 1302--1320 (2009; Zbl 1160.35422)] and \textit{D. Cao} and \textit{J. Chabrowski} [Rev. Mat. Complut. 20, No. 2, 309--338 (2007; Zbl 1132.35307)]. The proof is based on using variational methods. In order to overcome the lack of compact embedding, the mountain-pass and the concentration-compactness principles are used.