Semilinear degenerate elliptic boundary value problems via Morse theory (Q2339657)

In this paper, the author studies the following semilinear homogeneous elliptic boundary value problem: \[ \left\{\begin{aligned} & Au=\lambda u-g(u)\qquad \text{in}\,\, \Omega\\ & Bu= a(x')\frac{\partial u}{\partial \nu}+b(x')u=0\quad \text{on}\,\, \partial\Omega \end{aligned}\right.\leqno{(P)} \] where \(\Omega\) is a bounded domain of \(\mathbb R^N, N\geq 2\) with smooth boundary \(\partial \Omega.\) \(A\) is a second-order elliptic differential operator with real coefficients given by \[ Au= -\sum_{i=1}^N \frac{\partial }{\partial x_i}\left( \sum_{j=1}^Na^{ij}(x) \frac{\partial u}{\partial x_j}\right)+c(x)u, \] with real coefficients satisfying (1) \(\; a^{ij}\in C^\infty(\overline\Omega)\) and \(a^{ij}(x)=a^{ji}(x)\) for all \(x\in \overline\Omega\) and \(1 \leq i, j \leq N,\) and there exists a positive constant \(a_0\) such that \[ \sum_{i,j=1}^Na^{ij}(x)\xi_i\xi_j\geq a_0|\xi|^2\;\;\text{for all}\; (x, \xi)\in \overline\Omega \times \mathbb R^N. \] (2) \(\; c\in C^\infty(\overline\Omega)\) and \(c(x)\geq 0\) in \(\Omega.\) (3) \(\; a\in C^\infty(\partial\Omega)\) and \(a(x')\geq 0\) on \(\partial\Omega.\) (4) \(\; b\in C^\infty(\partial\Omega)\) and \(b(x')\geq 0\) on \(\partial\Omega.\) (5) \(\displaystyle \partial / \partial\nu= \sum_{i,j=1}^Na^{ij}(x')n_j \partial/ \partial x_i\) is the conormal derivative associated with the operator \(A,\) where \(n=(n_1, n_2,\dots, n_N)\) is the unit exterior normal to the boundary \(\partial\Omega.\) (H.1) \(\; a(x')+b(x')>0\; \text{on}\; \partial\Omega.\) (H.2) \(\; b(x') \not\equiv 0\; \text{on}\; \partial\Omega.\) Regarding the nonlinear term \(g: \mathbb R\rightarrow\mathbb R\), it is assumed to satisfy the following two assumptions (A) \(\; g\in C^1(\mathbb R)\) and \(g(0)=g'(0)=0.\) (B) \(\; \displaystyle\lim_{s\rightarrow \pm \infty}\frac{g(s)}{s}=+\infty.\) Based on Morse theory on Hilbert spaces and using Ljusternik-Schnirelman theory of critical points, the author proves the existence of multiple non-trivial solutions to the problem \((P)\).