Asymptotically or super linear cooperative elliptic systems in the whole space (Q365853): Difference between revisions

The authors consider the following cooperative elliptic system \[ \begin{cases} -\Delta u = \xi (x)u + f(x,u,v) & \mathrm{ for }\,\, x \in {\mathbb R}^N , \\ -\Delta v = \zeta (x)v + g(x,u,v) & \mathrm{ for } \,\, x \in {\mathbb R}^N ,\\ u \to 0, v \to 0 & \mathrm{ as } \,\, | x | \to \infty \end{cases} \tag{1} \] where \(\xi ,\zeta \in C({\mathbb R}^N,{\mathbb R}), f,g \in C({\mathbb R}^N \times {\mathbb R}^2 ,{\mathbb R})\), and there exists a function \(F\in C^1({\mathbb R}^N \times {\mathbb R}^2,{\mathbb R})\) such that \(\nabla F= (f,g)\). Here \(\nabla F\) denotes the gradient of \(F\) in \((u,v)\in {\mathbb R}^2\). Let \(L_1= -\Delta -\xi \) and \(L_2=-\Delta -\zeta \) where \(\xi ,\zeta \) are \(1\)-periodic in \(x_j\,\, (j=1,2,\ldots ,N)\). The authors assume that (\(L_1\)) \(\underline{\mu }_i :=\sup (\sigma (L_i) \cap (-\infty ,0))< 0 < \overline{\mu }_i:= \inf (\sigma (L_i) \cap (0,\infty ))\), where \(i=1,2\). (\(F_1\)) \(F\in C^1({\mathbb R}^N \times {\mathbb R}^2 ,{\mathbb R})\) is \(1\)-periodic in \(x_j\,\, (j=1,2,\ldots ,N)\), and \(\nabla F(x,U)=o(| U | )\) as \(| U | \to 0\) uniformly in \(x\). (\(F_2\)) \(F(x,U) \geq 0\) for all \(U \in {\mathbb R}^2\) and \(\frac12 (\nabla F(x,U),U) > F(x,U)\) if \(U \in {\mathbb R}^2 \setminus \{0\}\). Define \(H(x,U)= \frac12 (\nabla F(x,U),U)-F(x,U)\), \(\mu _0= \max \{ \overline{\mu }_1, \overline{\mu }_2\}\) and \(\rho _0 = \min \{ \rho _1,\rho _2\}\) where \(\rho _i= \min \{ - \underline{\mu }_i, \overline{\mu }_i\}\), \(i =1,2\). For the asymptotically linear problem, furthermore assume, (\(F_3\)) \(F(x,U) = \frac12 \eta | U | ^2 + G(x,U)\) where \(\eta \in (\mu ,+ \infty )\) for some \(\mu > \mu _0\), \(G \in C^1({\mathbb R}^N\times {\mathbb R}^2 ,{\mathbb R})\) and \(\nabla G(x,U)/| U | \to 0\) as \(| U | \to \infty \) uniformly in \(x\). (\(F_4\)) There exists \(\alpha \in (0,\rho _0/2]\) such that \[ \frac{| \nabla F(x,U)| }{| U | } \geq \frac{\rho _0}{2} - \alpha , \quad \forall U \in {\mathbb R}^2. \] Next, for the superlinear case, assume that (\(F_3'\)) \(F(x,U)/| U | ^2 \to \infty \) as \(| U | \to \infty \) uniformly in \(x\). (\(F_4'\)) \(H(x,U) \to \infty \) as \(| U | \to \infty \), and there exist \(r_0>0\) and \(\sigma > \max \{1,N/2\}\) such that \[ \frac{| \nabla F (x,U)| ^{\sigma }}{| U | ^{\sigma }}\leq C_0 H(x,U) \] if \(| U | \geq r_0\), where \(C_0 \) is a positive constant. The authors main theorem is as follows. { Theorem.} Suppose that (\(L_1\)) and (\(F_1\))-(\(F_2\)) are satisfied. If either (\(F_3\))-(\(F_4\)) or (\(F_3'\))-(\(F_4'\)) hold, then (1) has at least one ground state solution. They also give some examples.