A classification of solutions of a fourth order semi-linear elliptic equation in \(\mathbb{R}^n\). (Q2360448)

In this paper authors use moving planes technique to prove symmetry of solutions of problem \[ \Delta^2u=u^p_+\quad\text{in}\;\mathbb{R}^n,\quad u^p_+\in L^1(\mathbb{R}^n),\tag{1.1} \] where \(\Delta^2\) denotes the biharmonic operator in \(\mathbb{R}^n\), \(u_+=\max\{u,0\}\), \(1<p\leq n/(n-4)\) with \(n\geq 5\). Authors firstly describe the asymptotic behavior \(u,\Delta u\) near infinity by the following integral representation results. Theorem 1.1. Let \(n\geq 5\), \(p>1\) and \(u\in C^4(\mathbb{R}^n)\) be a solution of (1.1). Then the following statements hold. i) The function \(-\Delta u\) can be represented by \[ -\Delta u(x)=\frac{1}{n(n-2)\omega_n}\int_{\mathbb{R}^n}\frac{u^p_+(y)}{|x-y|^{n-2}}\, dy+c_1\,,\quad x\in\mathbb{R}^n,\tag{1.2} \] for some constant \(c_1\geq 0\). ii) After an orthogonal transformation, \(u\) can be represented by \[ u(x)=\frac{1}{2(n-4)n(n-2)\omega_n}\int_{\mathbb{R}^n}\frac{u^p_+(y)}{|x-y|^{n-4}}\,dy -\sum^n_{j=1}a_j(x_j-x^0_j)^2+c_0,\tag{1.3} \] \(x\in\mathbb{R}^n\), \(\omega_n\) is the volume of the unit sphere of \(\mathbb{R}^n\), \(a_j\geq 0\), \(c_0\) are constants and \(x^0=(x^0_1,\dots,x^0_n)\in\mathbb{R}^n\). The main result of the first part is the following theorem. Theorem 1.2. Let \(n\geq 5\) and \(u\in C^4(\mathbb{R}^n)\) be a solution of (1.1). Then for \(1<p\leq n/(n-4)\) the following statements hold. i) \(u\) is bounded from above. ii) If either \(a_j>0\) \(\forall j\) or \(c_0\not=0\), then \(u_+\) has a compact support (\(a_j\) and \(c_0\) are defined in Theorem 1.1). Moreover, \(u\) is radially symmetric about some point \(x_0\) in \(\mathbb{R}^n\) and \(\frac{\partial u}{\partial r}<0\), where \(r=|x-x_0|\). To verify that \(u\) is bounded from above authors use blow-up technique. The second part of this paper deals with some \(\sup+\inf\) type inequality for higher order elliptic equation, that is \[ (-\Delta)^mu=u^p_+\quad\text{in}\;\Omega,\; u\in C^{2m}(\Omega),\; m\geq 1,\tag{1.4} \] where \(\Omega\) is a bounded domain in \(\mathbb{R}^n\), \(n\geq 2m+1\), and \(1<p<(n+2m)/(n-2m)\). Authors remove the boundedness integral condition on the solution to (1.4) and obtain Theorem 1.3. Let \(p\in(1,(n+2m)/(n-2m))\) and \(K\subset\subset\Omega\) a compact subset. Then there exist \(C_1\) and \(C_2(K)\) depending only on \(K\) two positive constants such that \[ \sup\limits_{K}u+C_1\inf\limits_{\Omega}u\leq C_2, \] for any solution \(u\) to (1.4).