Transport and Deposition of Fine Particulates in Turbulence: Numerical Modeling and Uncertainty Quantification

Abstract

In recent years, public awareness of the impact caused by micron-sized particles such as infectious aerosols or dust has increased drastically, ranging from severe public health concerns to various environmental issues. In addition, airborne dust and volcanic ash ingested by aircraft engines compromise the durability, performance, and safety of engine turbine components. The transport and deposition of fine (<≈O(10 μm) particulates in turbulence (e.g., dust or powder) is largely controlled by cohesive forces such as electrostatics and van der Waals. Due to their small size and cohesive nature, tracking individual particles in turbulence is challenging, and is further complicated by significant uncertainties in material properties. Although computational methods with varying levels of complexity have been developed over past decades, accurate predictive models of cohesive particle transport and deposition do not yet exist for large-scale simulations.

The main objective of this work is to develop a numerical framework tailored for resolving cohesive particle interactions in turbulence. Efficient algorithms are developed to optimally resolve particle contact forces in a direct numerical simulation (DNS) framework. The framework is then used to study the effect of electrostatics on particle transport in turbulence. It is found that the short-range electric potential plays a key role in particle clustering even in dilute suspensions. A follow-up study of charged aerosols in ionized air identifies a feedback mechanism capable of generating atmospheric turbulence via an electrohydrodynamic body force.

Turbulence-induced breakup of an aggregate of solid particles subject to van der Waals is also investigated. A phenomenological model of the breakup process is developed that acts as a granular counterpart to the Taylor analogy breakup (TAB) model commonly used for droplet breakup. Such a model is capable of predicting the onset of aggregate breakup in the absence of a resolved turbulent flow field. Finally, particle deposition in a turbulent pipe flow is studied in the presence of van der Waals and electrostatics. The sensitivity of deposition rate to uncertainties in cohesive forces is efficiently quantified using a multi-fidelity framework. Deposition is found more sensitive to electrostatics than van der Waals across all particle sizes and exhibits largest uncertainty for mid-sized particles.