Numerical investigation of aerosol deposition on a single 2D fiber

Abstract

The aerosol filtration technique deserves extensive attention, due to its wide applications found in health, environment, and industry. The simulation of aerosol filtration over fibrous filters involves the numerical investigation of the motion and capture of particles (aerosols) under different fluid operating conditions.The capture of aerosol in fibrous filters involves the following deposition mechanisms: Impaction, interception, diffusion and gravitation. The motion of the particles is governed by the relevant forces acting on the particles. Due to the dilute concentration of the aerosol, the influence of particle motion onto the fluid and particle-particle interaction can be neglected. The particles are removed by a fibrous filter when they collide and stick to the surface of the fiber. An Euler-Lagrangian approach is engaged to simulate the particle trajectories. The air flow field (continuous phase) is simulated in Ansys CFX (Euler approach) and the particle trajectories (dispersed phase) are computed by Lagrangian simulations in MATLAB.Real fibrous filters consist of a large number of fibers of non-uniform size and random orientation and position. For simplification, a single 2D fiber is considered and placed normal to the airflow to analyse several deposition mechanisms. The efficiency at which the particles are collected by a single fiber from an aerosol stream is called single-fiber efficiency EΣ, which is highly dependent on particle size. The particle motion is also affected by diffusion, where the dispersion of particles due to Brownian motion follows a Gaussian distribution. The diffusion coefficient can be inferred from the mean distance travelled by the particle and agrees well with the theoretical diffusion coefficient calculated from the Stokes-Einstein equation. The diffusion mechanism becomes significant at particle sizes below 0.2 µm. The single-fiber efficiency is determined considering all deposition mechanisms and compared with existing theoretical models.