Statistical Analysis of Fine Particle Resuspension from Rough Surfaces by Turbulent Flows

Abstract

Particle resuspension plays a part in indoor aerosol dynamics and has received increasing attention due to its ability to prolong human exposure to airborne particles. A stochastic model of turbulence-induced particle resuspension from rough surfaces is proposed based on the statistical nature of the process. Deposited fine (micro-or nano-size) particles are generally immersed in the viscous sublayer of the incompressible turbulent boundary layer and are subjected to aerodynamic forces that can be approximated by log-normal distributions due to penetration of turbulent inrushes and bursts into the viscous sublayer. Similarly, the adhesion force between particles and surfaces could be approximated by statistical distributions according to the statistical nature of surface roughness. Three common types of adhesion force distributions, i.e. log-normal, Weibull, and Gaussian distributions, are specifically explored. Predicted resuspension fractions versus free stream velocity are in good agreement with experimental data reported in the literature. Using the proposed stochastic model, influences of various parameters (composite Young's modulus, surface energy, adhesion force distribution, velocity distribution, fluid density, and particle diameter) on the threshold friction velocity (u * 50 ) and friction velocity divergence (Δu * ) are analysed. The information sheds light onto the controlling of the particle resuspension process. The proposed model extends the current capability of modeling particle resuspension by considering different types of adhesion force distributions.

Local experimental methodology for the study of microparticles resuspension in ventilated duct during fan acceleration

“…The static models are based on the disruption of the particle/surface static equilibrium through a static force or momentum balance (depending on the mechanism identified as being responsible for resuspension among rolling, sliding or lifting). One can cite the models developed by Goldasteh et al (2013) and You & Wan (2014, 2017. Another approach of those models is to consider a kinetic probability density function (PDF) of some parameters influencing the phenomenon.…”