Abstract

Laminar flow is dominant in the microfluidic device, and hence it is difficult to improve the mixing performance efficiently. To solve this problem, a fast and homogenized mixing application through microstreaming generated by ultrasonic resonance in fluid has been proposed. In this article, the acoustic streaming patterns in two-dimensional microchannel are investigated based on the Reynolds stress method (RSM) and limiting velocity method (LVM). It is found that the classical Rayleigh streaming pattern can be well described by both methods, but the LVM adopts uniform meshes with fewer elements for more efficient computation. Therefore, the LVM is more suitable to explore acoustic streaming and microfluidic mixing performance in three-dimensional models. Results also show that the similar sound pressure distribution and different acoustic streaming patterns are found in microchannels with different aspect ratio. Four symmetric vortices are formed in transducer-plane streaming model driven by ultrasonic standing waves at h/w = 1/20, while eight symmetrical acoustic vortices in Rayleigh-like acoustic streaming model at h/w = 1/3. When the inlet velocity is large, the mixing effect of the fluids is not significant. With the decrease of inlet velocity, the mixing efficiency is increased gradually. In addition, the mixing performance of the two models is close to each other at the same inlet velocity. The solution flow rate in the Rayleigh-like microchannel is seven times larger than that of the transducer-plane streaming model. Therefore, the disturbance intensity in the Rayleigh-like acoustic streaming model is stronger.

摘要

层流在微流体装置中占主导地位，因此很难有效地提高混合性能。为了解决这个问题，已经提出了一种通过流体中超声波共振产生的微流进行快速均匀混合的应用。在本文中，基于雷诺应力法（RSM）和极限速度法（LVM）研究了二维微通道中的声流模式。发现经典的瑞利流模式可以用两种方法很好地描述，但 LVM 采用具有更少元素的均匀网格来提高计算效率。因此，LVM 更适合探索三维模型中的声流和微流体混合性能。结果还表明，在不同纵横比的微通道中发现了相似的声压分布和不同的声流模式。在超声波驻波驱动的换能器平面流模型中形成四个对称涡流h / w  = 1/20，而类瑞利声流模型中的八个对称声涡流在h / w  = 1/3 处。当入口速度较大时，流体的混合效果不显着。随着入口速度的降低，混合效率逐渐提高。此外，在相同的入口速度下，两种模型的混合性能彼此接近。类瑞利微通道中的溶液流速是换能器平面流模型的七倍。因此，类瑞利声流模型中的扰动强度更强。