Preferential concentration of inertial fibres in a turbulent channel flow

Abstract

Large eddy simulation (LES) is developed to study particle preferential concentration, in which an initially uniform distribution of inertial particles spontaneously segregates into clusters in a turbulent flow, driven primarily by the small-scale turbulent fluctuations and slip velocity. Dynamic modelling of sub-grid scale effects on the LES and Langevin-type sub-grid scale modelling of particle motion ensures particle clustering is well captured, which is computational cheap when compared to fully-resolved simulations. Results shows that prediction of particle clustering near walls due to turbophoresis is strongly depended on (i) the particle inertia, where particle inertia is parameterised by its Stokes number, (ii) particle shape, parameterised by its aspect ratio, (iii) binning method and (iv) simulation time.