(682a) Frequency-Dependent Streaming Flows and Motions from Acoustically Powered Microrobots

Acoustic streaming is a phenomena that is used to produce fluid flow for applications such as mixing in microfluidics and propulsion in microrobotics. Typically, acoustically responsive structures, such as bubbles and sharp edges, generate streaming by vibrating when excited with an acoustic field. To date, most work has focused on building microscale systems with a single type of acoustically responsive structure actuated at its primary resonance frequency. This begs the question of whether it is possible to integrate multiple acoustically responsive structures in a single microscale system, such that multiple patterns of streaming flow can be generated depending on the actuation frequency. The design of such systems remains unresolved due to an incomplete understanding of the relative magnitude of frequency dependent streaming flows from different acoustically responsive structures, such as bubbles and sharp edges.

In this work, we have developed a joint computational and experimental framework to characterize and predict the frequency-dependent streaming flows produced in systems with more than one type of acoustically responsive structure, each with a distinct resonance behavior. We then attempt to use the framework to translate our understanding of the frequency-dependent streaming flows to inform the design of acoustically actuatable microscale robots. Experimentally, we fabricated surface-immobilized microscale structures with protruding sharp edges or spherical cavities to hold bubbles of prescribed sizes with two-photon lithography. We then used piezoelectric transducers to acoustically excite the structures, visualized the flows using confocal microscopy, and measured the flow fields with micro-particle image velocimetry (uPIV). We compared these results with numerical predictions from a second-order perturbation theory model of streaming that couples structural oscillations to induced fluid flow.

Our characterization of the streaming flow revealed a frequency dependence that deviates from traditional eigenfrequency analyses of the bubble and sharp edge vibrations. We also found that bubbles produce a larger magnitude of fluid flow for a given applied voltage compared to sharp edges, when actuated at resonance. When actuated at sharp edge resonance frequencies (off resonance for bubbles) we found that the bubbles produce a streaming flow of similar magnitude to sharp edge flows. The inverse, however, is not true. Using these measurements, we attempted to predict and observe the frequency-dependent motion of a microparticle that has both a bubble and a sharp edge.

Ultimately, our findings provide a framework for the rational development of actuation schemes to control frequency dependent streaming flows. Primarily, our uPIV results indicate that frequency cannot be varied independent of voltage when attempting to switch between which structure is primarily responsible for producing streaming flow. Additionally, we found that while bubbles can be actuated independently of sharp edges, sharp edges cannot be actuated independently of bubbles. This has potential applicability in choosing what frequency-voltage pairs to actuate an acoustically powered biomedical microrobot and could also be used to inform the design of microfluidic devices that have engineered streaming induced mixing patterns depending on the actuation frequency.