The interaction of vortices with fixed boundaries is an important problem in many areas of practical interest. During the approach of the vortices to the wall, vorticity is created at the wall, and then mixing and stirring of primary and secondary vorticity lead to different scenarios that depend on the Re number. This problem is relevant to bursting in turbulent boundary layers, which is responsible for turbulence production. In the near-wall region of turbulent flows, vortex stretching plays a fundamental role. As a prelude to attempting the solution by direct simulation of 3-D Navier–Stokes equations, it is convenient to describe the mechanism associated with 2-D or axisymmetric flows. In the present study, axisymmetric vortex rings and vortex dipoles approaching a nonslip flat wall have been considered. In both cases, experimental results based on flow visualization are available.1,2 The purpose of the present paper is to do a numerical simulation with the same conditions used in the real experiments and follow the flow evolution for a much longer time than has been possible in such experiments. Both for axisymmetric and 2-D cases the numerical simulation shows, as revealed by the experiments, that, at the initial stage, very thin layers of vorticity of sign opposite to that of the primary vortex are generated at the wall. This mechanism is almost independent of Re. This thin and intense layer immediately becomes unstable and rolls up creating a secondary vortex. This vortex interacts with the primary vortex and different flow structures are generated, depending on Re. The location where these structures form depends, in large measure, on whether vortex stretching is present or not. Particularly relevant to the bursting phenomenon is the case at intermediate Re (e.g., Re=1600). Here, a new dipole is created near the wall after multiple rebounding and pairing of secondary vortices. This new structure has sufficient strength to move itself away from the wall. We conclude that mixing and stirring of vorticity are the relevant mechanisms influencing the whole process in a two-dimensional simulation. This ejection of a new structure was previously noticed in the interaction of vortex rings with a wall. From flow visualization, Walker et al., at Re=Γ/ν>3500, observed the occurrence of azimuthal instabilities and, immediately after that, they noticed a new vortex ring was rapidly ejected away from the wall.2 From these observations, those authors argued that the ejection was due to the azimuthal instabilities. In the present paper, axisymmetric calculations have been done in the same range of numbers used in the experiment of Walker et al.2 From the time evolution of the vorticity distribution, we argue that vortex pairing is the main mechanism that contributes to the generation of the new ring rather than azimuthal instabilities. The numerical simulations clearly show that this new ring is created above the primary ring and not at its interior.
Read full abstract