Modeling and optimization of acoustofluidic micro-devices

Abstract

We investigate how the combination of numerical simulation tools and optimization routines can be used to design micro-devices. Experimental devices that are designed in this way can only provide optimal performance if the simulation model, used in the optimization procedure, reflects the real device characteristics accurately. Owing to this fact, the modeling of acoustofluidic devices is summarized. The mathematical formulation of the optimization problem, the parameterization of the device design and the implementation of the optimization loop is addressed alongside with practical recommendations for the chosen genetic algorithm optimization. In order to validate the implementation, an optimized planar resonator is compared with the optimal geometry given in the literature. The optimization of a typical 3D micro-device shows that devices can be designed to generate any desired acoustic mode shape at maximum pressure amplitude. The presented automatic design approach is of great practical relevance for the development of highly optimized micro-devices and it can speed up and facilitate the design-process in the growing field of acoustofluidics.

A Numerically Efficient Damping Model for Acoustic Resonances in Microfluidic Cavities

“…As often in experiments, the device is piezo-electrically excited at a frequency that leads to a strong time-harmonic field in the fluid cavity. The 3D device model contains a silicon device body, a glass lid, a piezoelectric transducer, and a glue layer between the transducer and the silicon (Hahn et al (2014)). The device geometry and the potential energy density at 0.84 MHz are illustrated in Fig.…”

Performance Study of Acoustophoretic Microfluidic Silicon-Glass Devices by Characterization of Material- and Geometry-Dependent Frequency Spectra

“…Note that, since the objective function used in Ref. [21] depends on the particle properties, it would be better to speak about separation performance for the X particle rather than device performance. In this paper, we are interested in investigating the device performance, and thus we will introduce indicators that do not take into account the particle properties but instead exclusively pertain to the device characteristics.…”

“…Studies including the thermoviscous and transient effects inside microchannels have been reported as well [18][19][20]. The first numerical optimization studies of acoustophoretic devices have also been performed recently, illustrating a procedure to obtain optimal acoustophoretic forces by changing the geometrical parameters of the device [21]. Other studies involve numerical characterization of the acoustic pressure wave in the microchannel and subsequent computation of particle trajectories by means of numerical integration [22][23][24][25][26][27].…”

“…As an example, in Ref. [21], an objective function, i.e., Eq. (17), containing the Gorkov potential and therefore including both the device performance and the particle properties is used to optimize the performance in the space of the parameters.…”

Imaging local acoustic pressure in microchannels

“…To aid design and improve performance, physical models have been developed to describe the acoustic resonances [23][24][25][26][27][28], which predict the resonance frequency and pressure amplitude distributions well for thin chips [29].…”