Experimental and numerical study of airflow pattern and particle dispersion in a vertical ventilation duct

A numerical simulation has been conducted to investigate the behavior of airflow and the dispersion of size-dependent particles in the ventilation duct. To examine the prediction accuracy for our numerical simulation, we compared computational fluid dynamics (CFD) results with particle image velocimetry (PIV) measurement data of airflow in an isothermal duct model. The low Reynolds number type k-e{open} turbulent model was selected for further calculation because its predictions were the most accurate in this study. A size-dependent particle transport approach, combining the Eulerian approach with low Reynolds k-e{open} turbulent model and Lagrangian particle trajectory tracking, was adopted to investigate the airflow pattern and particle removal efficiency in the ventilation duct. Two scenarios, one with and one without a baffle, were also compared to determine the effects on the particle transport mechanisms. We compared the effectiveness of particle removal as by using the simulated results, which showed that the fraction of coarse particles retained were higher than the fraction of fine particles retained. We found that a vertical duct with a baffle in a perfect sink condition on the wall surfaces increased particle deposition and decreased particle escape. Because of the two opposing particle removal mechanisms, i.e., deposition and escape, the impact of thermophoresis on particle retaining behavior was not as significant as that of airflow velocities. © 2012 Elsevier Ltd.