\(L^p\)-theory for second-order elliptic operators with unbounded coefficients in an endpoint class (Q2249908)

The paper considers second-order elliptic operators of the form \[ Au(x) = -\text{div}(a(x)\nabla u(x))+ F(x)\cdot \nabla u(x)+V(x)u(x),\quad x\in \mathbb R^N, \] where \(N\in \mathbb N\), \(a(x)\) is positive-definite, \(a\) and \(F\) are continuously differentiable and \(V\) are locally \(L^\infty\). The functions \(a\), \(F\) and \(V\) are unbounded, but satisfy an estimate of the form \[ |a(x)| \leq \alpha \left( 1+V-\frac{\text{div}\, F}{p}\right)^{1-2/r} |x|^2 (\log|x|)^2, \quad |x|>R, \] \[ |F(x)| \leq \alpha \left( 1+V-\frac{\text{div}\, F}{p}\right)^{1-2/r} |x|(\log|x|), \quad |x|>R, \] where \(r\in [2,\infty)\). In [the author, ibid. 12, No. 4, 957--971 (2012; Zbl 1277.47060)], the operator \(A\) was studied for \(r<\infty\) and in the present article the critical case \(r=\infty\) is studied. Moreover, \(m\)-accretivity, essential \(m\)-accretivity and \(m\)-sectoriality of the minimal and maximal realization of \(A\) in \(L^p(\mathbb R^N)\), \(1<p<\infty\), are discussed. Further, an \(L^p\)-generalization of Kato's self-adjointness problem is discussed and the obtained results are applied to the operators \(-\Delta + |x|^{\beta-1} x\cdot \nabla +c|x|^\gamma\) and \(-\Delta - |x|^{\beta-1} x\cdot \nabla +c|x|^\gamma\).