Using RANS turbulence models and Lagrangian approach to predict particle deposition in turbulent channel flows

This study investigates the capability and accuracy of three Reynolds-Averaged Navier–Stokes (RANS) turbulence models, i.e. a Reynolds stress model (RSM), a RNG k-ε model, and an SST k-ω model in the prediction of particle deposition in vertical and horizontal turbulent channel flows. The particle movement was simulated using a Lagrangian-based discrete random walk (DRW) model. The performances of the three RANS turbulence models with and without near-wall turbulence corrections were evaluated. A new modification method for turbulence kinetic energy was proposed for the RNG k-ε model and the SST k-ω model. The results were compared with previous experimental data, empirical equation as well as simulation outcomes. It is found that the isotropic SST k-ω model and the RSM model can successfully predict the transition from the diffusion region to the inertia-moderated region. The RNG k-ε model with near-wall modifications can also reflect the V-shape deposition curve although without modifications it greatly over-predicts the deposition velocity and shows an almost straight deposition line. For all of the three turbulence models, application of near-wall corrections is able to improve the simulation results to different extents. ► We study particle deposition in turbulent channel flows using the Lagrangian approach. ► Deposition velocities are greatly over-predicted by k-ε or k-ω models. ► We propose a new and simple correction method for turbulent kinetic energy. ► Near-wall corrections are able to improve the deposition results by different extent.