Micro-Scale Dynamics of Aerosol Resuspension: A Resolved CFD Model for Near-Surface Particle Detachment

Aerosol dispersion and deposition mechanisms have been extensively studied; however, resuspension has not enjoyed as much attention and remains a challenging topic in particle transport modelling. Despite its critical role across diverse sectors; panning nuclear safety, healthcare, transport, agriculture, and industrial processing; the complex near-surface physics driving this phenomenon are often oversimplified. Resuspension arises from the coupling between particle, surface, flow and environmental characteristics, making it strongly scenario-dependent. The underlying broad-ranging physics and the coupled interactions result in numerous disparate and competing forces that either enable or resist displacement. While several models of mechanistic, probabilistic and semi-empirical nature have been developed, they often rely on broad assumptions which can lead to observed discrepancies between predicted and observed behaviour. Although current models may be sufficient for bulk population estimates, a significant research gap remains for high-consequence scenarios for which physical accuracy is critical – a gap this study aims to bridge.

This research advances the study of aerosol dynamics by developing a finely resolved Computational Fluid Dynamics (CFD) model that comprehensively captures fundamental resuspension physics at the single-particle level. We focus specifically on reproducing real atmospheric conditions in the near-wall region, investigating how boundary layer aerodynamics interact with surface geometry and adhesion to induce lift-off. The model was implemented in Siemens’ Simcenter StarCCM+ v2502, making use of overset meshing and dynamic fluid-body interaction solvers to resolve six-degrees-of-freedom (6 DOF), three-dimensional particle motion in turbulent flow. Key features include the resolution of unsteady flow structures around the particle, detailed contact mechanics, and a custom adhesive force implementation where van der Waals (vdW) forces play a significant role. To accommodate complex real-world scenarios, the model supports variable particle sizes and shapes, surface topologies and roughness, offering high sensitivity particularly in the Stokes regime (Rep ≤ 1), examples of which are shown in Figure 1. Significant challenges addressed included accurately resolving particle-surface contacts at necessary spatiotemporal scales and implementing adhesion to support general particle-surface combinations, all while maintaining computational efficiency.
Case studies examine detachment from flat and corrugated surfaces and investigate critical velocities required to lift off particles of varying shapes. An example case study is shown in Figure 2. Results highlight the competition between aerodynamic torque and adhesive moments, distinguishing between sliding and rocking mechanisms. Further validation work is presented against experimental trajectory and detachment rates data and will be benchmarked against established Eulerian resuspension models and in-house wind tunnel experiments. Ultimately, improving the capabilities for predicting particle and powder resuspension addresses critical health and safety concerns as it helps better anticipate hazards associated with particle exposure during its transport, thus supporting better risk mitigation strategies for harmful aerosols in the environment.