Microbubble streaming flows for non-invasive particle manipulation and liquid mixing

Liquid transport at small length scales plays an increasingly important role in many emerging and inter-disciplinary fields, such as micro/nano manufacturing processes, biomedical engineering, lab-on-a-chip diagnostic technology, micro fuel cell, and bio-fluidics. As the length scale shrinks, the synergy of micro-hydrodynamics with various external force fields (capillary forces, acoustic forces, electric and magnetic fields, and optical forces) has emerged as an exciting inter-disciplinary field. In this dissertation, we study an emerging actuating mechanism -- microbubble steady streaming flows -- by emphasizing both fundamental understanding and the design and applications of bubble-based microfluidic devices. Under periodic acoustical driving, microbubbles with radius 20 -- 100µm absorbed or attached to a solid wall initiate steady streaming flows around the bubbles. These flows are driven by the Reynolds stresses in the boundary layer, a consequence of the nonlinearity of the Navier-Stokes equations. In the first part of the dissertation, we focus on manipulating micron-sized objects with microbubble streaming flows. We observe that in microbubble streaming flows, micro-particles (radius 1--5µm exhibit size-dependent behaviors: particles of different sizes follow different characteristic trajectories. Superimposing bubble streaming flow and a Poiseuille flow shapes the flow into regions of closed streamlines and open streamlines. The combined flow fields allow selective trapping of particles by size and subsequent releasing of the trapped particles. We explain these mechanisms and exploit them as a novel and general concept of manipulating microparticles. By integrating acoustically driven bubbles as active elements in various microfluidic devices, we further demonstrate specific applications including switching, sorting, focusing, and pre-concentrating of micro-particles. Although microbubble streaming flows have received increasing attention in microfluidics and have been used widely in recent years, a fundamental understanding of microbubble streaming flow lags behind the experimental progress. In the second part, we study the frequency dependence of bubble streaming flows, and the correlation between the bubble dynamics and the streaming flow patterns. In contrast to steady streaming due to simple harmonic oscillations of a solid object, microbubbles exhibit frequency dependent and more complex shape modes. As the streaming flow patterns are caused by oscillations of microbubbles in contact with walls of the set-up, an understanding of the bubble dynamics is crucial. In this part of the study, we aim to bridge the gap between the physical understanding and the experimental observations of this complex phenomenon. With high-speed imaging, we experimentally characterize the oscillation modes and the frequency response spectrum of such bubbles, driven by a pressure variation resulting from ultrasound in the range of 1kHz≲f≲100kHz. We find that (i) the appearance of streaming flow patterns is governed by the relative amplitudes of bubble surface modes (normalized by the volume response), (ii) distinct, robust resonance patterns occur independent of details of the set-up, and (iii) the experimental results compare well with the prediction of our asymptotic theory. With the bubble dynamics known from both experiment measurement and theoretical prediction, we also perform the calculation of streaming flows from the bubble dynamics using the first principle approach. The fundamental understanding of frequency dependent streaming flows in turn can guide the design of practical microfluidic applications, such as various strategies of effective mixing on the micron scale. In the final part of this dissertation, we investigate two general classes of mixing strategies utilizing microbubble streaming flows: (a) modulating the acoustic driving pattern, such as the duty cycle and driving frequency, and (b) controlling the arrangement of microbubbles, such as the number, position, and orientation of the microbubbles. More specifically, modulating duty cycling will change the steady streaming flow directly by breaking it into unsteady flows, and thus achieve more effective mixing. Modulating driving frequency f can alter the directions of the resulting far field streaming flow. Strategically switching between drastically different flow patterns leads to improved mixing. On the other hand, when using multiple bubbles as actuating elements, the distance, position, and arrangement will affect the interaction between the streaming flows caused by each individual bubbles. Steady three dimensional flow can thus be achieved through proper arrangement and positioning of the microbubbles. Finally, we show that combining strategies of (a) and (b) can yield even better mixing. Through these strategies, we also demonstrate the flexibility of using microbubble streaming flows as both active and passive micro-mixers