Particle resuspension in a fully developed turbulent square duct flow is simulated using one-way coupled large eddy simulation coupled with a Lagrangian particle tracking technique for a range of bulk Reynolds numbers (36.5 k, 83 k and 250 k) and four particle sizes ranging from 5 to 500 μm (St = 0.01–2415) considered. Results obtained for the single-phase flow show good agreement with experimental data. Predictions of the time-dependent particle-laden flows demonstrate that the secondary flow mainly dominates particle resuspension in the regions near the center and sidewalls of the duct. It is found that particle resuspension decreases with particle size. The smaller particles tend to be more prone to resuspension, and are resuspended for a longer duration than larger particles. The mean particle resuspension velocity is found to increase with the duct height. In addition, particle resuspension in the vertical direction increases with Reynolds number while the effect of particle size on particle resuspension decreases. The resuspension rate in the spanwise direction fluctuates more as the Reynolds number increases. It is also found that the average particle resuspension rate in the lower half of the duct is always close to 0.5, and is independent of time, particle size and Reynolds number. Based on a dynamic analysis, the drag force is found to dominate the resuspension of small particles, while the lift force tends to dominate particle resuspension with increasing particle size. For low Reynolds number (36.5 k and 83 k) flows, the drag force plays an important role in the upper regions of the lower half of the duct, but the lift force dominates particle behavior in the lower regions. It can be concluded that the effects of duct height on particle behavior decline significantly with Reynolds number.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)