STIRRING AND MIXING BY GRID-GENERATED TURBULENCE IN THE PRESENCE OF A MEAN SCALAR GRADIENT

Abstract

The stirring and mixing of a passive scalar by gridgenerated turbulence in the presence of a mean scalar gradient is studied in three dimensions by DNS (Direct Numerical Simulation). Using top-end high fidelity computer simulations, we calculate and compare the effects of various fractal and regular grids on scalar transfer and turbulent diffusion efficiencies. We demonstrate the existence of a new mechanism present in turbulent flows generated by multiscale/fractal objects and which has its origin in the multiscale/fractal space-scale structure of such turbulent flow generators. As a result of this space-scale unfolding (SSU) mechanism, fractal grids can enhance scalar transfer and turbulent diffusion by one order of magnitude while at the same time reduce pressure drop by half. The presence of this SSU mechanism when turbulence is generated by fractal grids means that the spatial distribution of length-scales unfolds onto the streamwise extent of the flow and gives rise to a variety of wake-meeting distances downstream. This SSU mechanism must be playing a decisive role in environmental, atmospheric, ocean and river transport processes wherever turbulence originates from multiscale/fractal objects such as trees, forests, mountains, rocky river beds and coral reefs. It also ushers in the new concept of fractal design of turbulence which may hold the power of setting entirely new mixing and cooling industrial standards.