Flow and particle manipulation across flow regimes via shape modulation

Chan, Fan Kiat

Permalink

https://hdl.handle.net/2142/120500 Copy

Description

Title

Flow and particle manipulation across flow regimes via shape modulation

Author(s)

Chan, Fan Kiat

Issue Date

2023-03-27

Director of Research (if dissertation) or Advisor (if thesis)

Gazzola, Mattia

Doctoral Committee Chair(s)

Gazzola, Mattia

Committee Member(s)

• Goza, Andres
• Chamorro, Leonardo
• Hilgenfeldt, Sascha

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• computational fluid dynamics
• viscous streaming
• fog harvesting
• high performance computing
• particle manipulation

Abstract

Flow–particle manipulation has found applications in a multitude of physical scenarios, from microfluidics and healthcare to feeding in ciliated biological organisms or fog droplets collection in arid regions. A potential strategy involves, either passively or actively, shifting body shapes to systematically affect surrounding flows for desired application. In this context, the overarching goal of this thesis is to investigate how morphology and topology of a body immersed in a fluid can be modulated to enhance surrounding particle manipulation, both in active and passive contexts and across flow regimes. Towards this, we organize the research activities into three thrusts: (1) Shape-induced flow and particle manipulation at low Reynolds—Viscous streaming and rectified forces. This thrust focuses on microparticle manipulation strategies in regimes where flow and/or particle inertia is important, with an emphasis on viscous streaming. Viscous streaming is a rectified flow phenomenon that arises when an immersed body undergoes small-amplitude oscillations (compared to its size) in a viscous fluid. This flow mechanism cleanly partitions fluid into ‘sealed’ domains and thus offers various opportunities for particle manipulation. Streaming has been extensively studied and characterized theoretically, experimentally and numerically for constant curvature objects such as circular cylinders, infinite flat plates, and spheres. Beyond these uniform-curvature geometries, streaming flows, especially in 3D, involving objects of multiple curvatures and shape topologies received relatively little attention, and studies have mostly focused on the observation and description of such flows, without establishing a mechanistic connection between shape and flow reorganization. Here we begin to probe the shape–flow relation by establishing systematic guidelines for streaming flow manipulation via tools from dynamical systems theory. Building on this understanding and in collaboration with experimentalists, we develop an in vitro living system capable of endogenously generating streaming flows. We further demonstrate in silico the utility of such constructs for particle transport. Finally, in order to precisely manipulate suspended particles, it is important to accurately describe the small but finite inertial effects on them, a fundamental fluid dynamical problem that has never been solved in full generality. In this context, in collaboration with physicists, we employ a computational–theoretical synergistic approach and investigate the role of force rectification on particles (contrary to flow rectification used in streaming) for manipulation in an oscillatory flow generated by a breathing bubble. (2) Shape-induced flow and droplet manipulation at intermediate Reynolds—Fog harvesting. This thrust shifts attention to flow regimes of intermediate Reynolds number, while maintaining focus on microparticles. Here we investigate the role of morphology via the specific case of fog harvesting—an engineering problem catering to humanity’s increasing demand for clean water source as a consequence of climate change and growing populations. In the Namib desert, windborne fog is the primary source of water, and an entire ecosystem has developed around mechanisms which separate water from air. For example, tenebrionid beetles make use of their bodies to intercept droplets and redirect them for consumption. While much has been written relative to surface chemistry for driving collected droplets to the beetle’s mouth, little is known about the flow mechanism responsible for the initial collision. In this context, in collaboration with experimentalists, we first investigate the effects of geometry in a minimal setting of fog-laden flow past cylindrical targets and show how morphological features consistent with the beetle’s indeed enhance harvesting efficiency. We then perform simulations to analyze the underlying flow and its relation to droplet deposition. Using this understanding, we enable the realization of an engineered harvesting device with efficiency comparable to that of the beetle. (3) Scalable flow simulations involving immersed soft slender structures. This thrust transitions from studying the mechanics of flow–particle manipulation to enabling a computational tool for future investigations extending beyond the research conducted in this thesis. The results presented in the thrusts above were made possible thanks to a previously established robust simulation tool that employs a remeshed Vortex Method coupled with Brinkmann penalisation and projection approach. While the method excels in simulating fluid–structure interaction (FSI) problems across various flow regimes, it is limited to resolving bulk bodies that are either non-deforming or actively morphed via imposed shape motions, subsequently limiting our investigations to bulky rigid objects. In this context, a fast simulation tool capable of robustly resolving flow features in various settings and accurately capturing the two-way FSI of immersed soft and thin bodies becomes favorable. This thrust then focuses on the development of a scalable simulation software that preserves the robust features of Vortex Methods while introducing an immersed boundary forcing approach for resolving FSI involving soft slender structures. In particular, alongside the development of novel numerics towards the proposed method, we focus on the software implementation involved for efficiently mapping the numerical framework on supercomputing architectures. We further demonstrate the scalability, utility and flexibility of the developed software in a set of large-scale, high-fidelity simulations including flow past a sphere at high Reynolds and a magnetically-actuated immersed cilia carpet, thus spanning problems from bulk rigid body to dynamically deforming soft slender structures, and across flow regimes.

Graduation Semester

2023-05

Type of Resource

Thesis

Copyright and License Information

Copyright 2023 Fan Kiat CHan

Owning Collections