Sharp uncertainty principle inequality for solenoidal fields (Q2692135)

From author's abstract: ``This paper solves the \(L^2\) version of Maz'ya's open problem (see Zbl 1398.35019) on the sharp uncertainty principle inequality \[ \int\limits_{\mathbb{R}^N}|\nabla\boldsymbol{u}|^2dx\int\limits_{\mathbb{R}^N}|\boldsymbol{u}|^2|\boldsymbol{x}|^2 dx\geq C_N\bigg(\int\limits_{\mathbb{R}^N}|\boldsymbol{u}|^2dx\bigg)^2 \] for solenoidal (namely divergence-free) vector fields \(\boldsymbol{u}=\boldsymbol{u}(\boldsymbol{x})\) on \(\mathbb{R}^N\). The best value of the constant for \(N\geq 3\) turns out to be \(C_N=\frac{1}{4}\left(\sqrt{N^2-4(N-3)}+2\right)^2\) which exceeds the original value \(N^2/4\) for unconstrained fields. Moreover, we show the attainability of \(C_N\) and specify the profiles of the extremal solenoidal fields: for \(N\geq 4\), the extremals are proportional to a poloidal field that is axisymmetric and unique up to the axis of symmetry; for \(N=3\), there exist extremal toroidal fields, in addition to extremal poloidal fields; for \(N=2\), the extremal fields are all toroidal.'' The proof is based on the decomposition \(\boldsymbol{u}=\boldsymbol{u}_P+\boldsymbol{u}_T\) of solenoidal fields \(\boldsymbol{u}\), where \(\boldsymbol{u}_P\) is a poloidal field and \(\boldsymbol{u}_T\) is a toroidal field. This decomposition allows to separate the problem into finding best constants for \(\boldsymbol{u}=\boldsymbol{u}_T\) and \(\boldsymbol{u}=\boldsymbol{u}_P\). The computation of the best constant for a toroidal field is carried out by reducing it to the problem of finding the best constant for scalar fields with zero spherical mean. For \(\boldsymbol{u}=\boldsymbol{u}_P\) the estimation of the best constant can be reduced to an one-dimensional minimization problem and this minimization problem is solved by the author using a Laguerre polynomial approach.