Influence of viscosity on acoustic streaming in sessile droplets: an experimental and a numerical study with a Streaming Source Spatial Filtering (SSSF) method

TL;DR

This study investigates how viscosity influences acoustic streaming in sessile droplets through experimental observations and a novel numerical model, revealing multiple flow regimes and the significant role of viscosity in flow dynamics.

Contribution

It introduces a spatial filtering-based numerical model for acoustic streaming, demonstrating viscosity's critical impact on flow regimes in sessile droplets.

Findings

Viscosity significantly affects flow regimes in droplets.

Four distinct flow regimes identified based on parameters.

Flow speed correlates with bulk wave attenuation as viscosity increases.

Abstract

When an acoustic wave travels in a lossy medium such as a liquid, it progressively transfers its pseudo-momentum to the fluid, which results in a steady acoustic streaming. Remarkably, the phenomenon involves a balance between sound attenuation and shear, such that viscosity vanishes in the final expression of the velocity field. For this reason, the effect of viscosity has long been ignored in acoustic streaming experiments. Here, we show experimentally that the viscosity plays a major role in cavities such as the streaming induced by surface acoustic waves in sessile droplets. We develop a numerical model based on the spatial filtering of the streaming source term to compute the induced flow motion with dramatically reduced computational requirements. We evidence that acoustic fields in droplets are a superposition of a chaotic field and a few powerful caustics. It appears that the caustics drive the flow, which allows a qualitative prediction of the flow structure. Finally, we reduce the problem to two dimensionless numbers related to the surface and bulk waves attenuation and simulate hemispherical sessile droplets resting on a lithium niobate substrate for a range of parameters. Even in such a baseline configuration, we observe at least four distinct flow regimes. For each of them, we establish a correlation of the average streaming speed in the droplet, which is increasingly dependent on the bulk wave attenuation as the viscosity increases. These correlations extend our results to a wide range of fluids and actuation frequencies.