On the pure polar flow within two concentric spheres (Q1384942)

This paper deals with the problem of steady axisymmetric flow of a Newtonian incompressible fluid in a concentric spherical gap. The driving force acting on the flow is a pressure difference between the North and South Pole. The differential equations which govern the flow are the Navier-Stokes (NS) equations and the continuity equation. The authors assume purely meridional flow (i.e., vanishing radial velocity component, too) in a spherical coordinate system \(R, \Theta, \varphi\) for all Reynolds numbers, \(0\leq \text{Re}\leq\infty\). However, it is easy to show that a solution of the NS equations under all these assumptions, and which would satisfy the normal velocity component boundary condition, does not exist, except for the two trivial and unreal limiting cases: no-flow \((\text{Re}=0)\) and potential flow \((\text{Re}=\infty)\). Nevertheless, the authors offer an ``approximate solution'' of the NS equations for all Reynolds numbers which only linearly combines the \(R\)-dependent differential equations for the flow at \(\text{Re}=0\) and the potential flow into one third-order nonlinear ODE, and use this as a ``substitute'' for an exact differential expression which is derived from the full NS equations and which has no solution for arbitrary Re. Usually, an approximate solution of a differential equation is understood as a procedure which leads arbitrarily close to the exact solution with increasing number of iterations. This is not the case when using this approach. Furthermore, it is obvious that by this approximation an uncontrolled violation is done of the assumption that the trajectories in the meridional plane (including the gap walls) are concentric circles, and, consequently, that the radial velocity component vanishes everywhere in the flow domain. A backward control under permission of an error-velocity component, which would show how good (or how bad) the NS equations are satisfied in the entire flow domain at this approximation, if possible at all, is not presented. Therefore, this reviewer do not quite understand the mathematical and physical validity of this ``approximate solution''.