Maximum principles for Laplacian and fractional Laplacian with critical integrability (Q6157824)

It is well known that, given an open connected set \(\Omega\) in \(\mathbb{R}^n\) (\(n\geq 3\)) and a function \(c:\Omega \rightarrow \mathbb{R}\), the strong maximum principle for the Schrödinger operator \(-\Delta +c(x)\) holds when \(c\in L^p(\Omega)\), for some \(p>\frac{n}{2}\), and fails if \(p<\frac{n}{2}\). The first main result of this paper states that the critical integrability condition \(c\in L^{\frac{n}{2}}(\Omega)\) is not sufficient for the validity of the strong maximum principle for the operator \(-\Delta +c(x)\), even if \(\|c\|_{L^{\frac{n}{2}}(\Omega)}>0\) is small. Next, the validity of the maximum principle is investigated for the more general operator \(-\Delta +\vec{b}(x)\nabla +c(x)\) in a smooth bounded open domain \(\Omega\), where, in this case, the critical integrability condition for the coefficient \(\vec{b}(x)\) is \(\vec{b} \in L^n(\Omega)\). As a result, the authors show that the maximum principle holds when \(\vec{b} \in L^n(\Omega)\), \(c\in L^{\frac{n}{2}}(\Omega)\), and \(\|\vec{b}\|_{L^n(\Omega)}+\|c^-\|_{L^{\frac{n}{2}}(\Omega)}\) is small. As a corollary, the validity of a strong maximum principle for the operator \(-\Delta +\vec{b}(x)\nabla\), when \(\vec{b} \in L^n(\Omega)\) and \(\|\vec{b}\|_{L^n(\Omega)}\) is small, is derived. Finally, the authors investigate the maximum principle and the strong maximum principle for operators involving the fractional Laplacian \((-\Delta)^s\), \(s\in (0,1)\). In particular, they improve a known result on the validity of maximum principle for the distributional solutions \(u\) of \((-\Delta)^su\geq 0\) in \(\Omega\) by removing the lower semicontinuous assumption on \(u\), and (partially) extend the above results to the operator \((-\Delta)^s +\vec{b}(x)\nabla +c(x)\).