Purely toroidal solutions of the kinetic dynamo equation (Q1364600)

The author considers the induction equation for the solenoidal magnetic field \(B: \mathbb{R}^3\times(0, T)\to \mathbb{R}^3\): \[ {\partial\over\partial t}B=-\text{curl}(\lambda \text{ curl } B)+ \text{curl}(v\times B)\quad\text{in}\quad\Omega\times(0, T), \] \(\Omega\subset \mathbb{R}^3\) being the (spherically symmetric) fluid region. In the exterior non-conducting region \(B\) is required to be a vacuum field, i.e. \(\text{curl} B=0\) in \( (\mathbb{R}^3\setminus\Omega)\times(0, T)\), and across \(\partial\Omega\) there are matching conditions for \(B\), \(\lambda>0\) is the magnetic diffusivity, and \(v\) denotes the velocity field corresponding to the field motion. To study the above problem the decomposition of the solenoidal field \(B\) into a poloidal component \(P\) and a toroidal one \(T\) \[ B=P+T=\text{curl curl}(\varphi\text{ id})+\text{curl}(\psi\text{ id}), \] is of extreme importance. The author proves that a purely toroidal magnetic field \(B=\text{curl}(\psi\text{ id})\) indeed can occur as solution in nontrivial configurations of the problem under consideration, i.e. for nontrivial, physically reasonable fields \(v\). The open problem concerning (global existence and) decay of \(\psi\) for \(t\to\infty\) is discussed briefly.