The sharp Hardy uncertainty principle for Schrödinger evolutions (Q605916)

The authors study \(u\) in \(C([0, T],L^2(\mathbb{R}^n))\) satisfying the Schrödinger equation \[ \partial_t u= i(\Delta u+ V(x,t)u)\quad\text{in }\mathbb{R}^n\times [0,T].\tag{1} \] They give conditions in terms of the potential \(V\) and the behaviour of \(u(0)\) and \(u(T)\) at infinity which are sufficient in order that \(u\equiv 0\). For instance they prove that \(\equiv 0\), provided \(V\) is bounded, \(\lim_{R\to\infty}\| V\|_{L^1([0, T], L^\infty(\mathbb{R}^n\setminus B_R))}= 0\) and there exist positive \(\alpha\), \(\beta\) such that \[ \| e^{|x|^2/\beta^2} u(0)\|_{L^2(\mathbb{R}^n)}<\infty\quad\text{and}\quad \| e^{|x|^2/\alpha^2} u(T)\|_{L^2(\mathbb{R}^n)}< \infty \] and \(T/\alpha\beta> 1/4\). They show that this condition is sharp in the sense that there exist \(V\) and a nonzero function \(u\) in \(C^\infty([0,T],{\mathcal S}(\mathbb{R}^n))\) satisfying above conditions in the case \(T/\alpha\beta= 1/4\). As a byproduct they derive estimates for the Gaussian decay of solutions of (1).