Performance study of acoustophoretic microfluidic silicon-glass devices by characterization of material- and geometry-dependent frequency spectra

TL;DR

This study uses numerical simulations to analyze how material properties and device geometry influence the frequency response of acoustophoretic microfluidic silicon-glass devices, focusing on optimizing particle focusing efficiency.

Contribution

It introduces a variational framework to characterize device response via frequency-dependent mechanical and electrical indicators, linking measurable electrical parameters to mechanical performance.

Findings

Mechanical indicators correlate with electrical indicators across frequencies.

Optimal driving frequencies depend on geometry and material parameters.

Electrical indicators can predict device performance effectively.

Abstract

The mechanical and electrical response of acoustophoretic microfluidic devices attached to an ac-voltage-driven piezoelectric transducer is studied by means of numerical simulations. The governing equations are formulated in a variational framework that, introducing Lagrangian and Hamiltonian densities, is used to derive the weak form for the finite element discretization of the equations and to characterize the device response in terms of frequency-dependent figures of merit or indicators. The effectiveness of the device in focusing microparticles is quantified by two mechanical indicators: the average direction of the pressure gradient and the amount of acoustic energy localized in the microchannel. Further, we derive the relations between the Lagrangian, the Hamiltonian and three electrical indicators: the resonance Q-value, the impedance and the electric power. The frequency response of the hard-to-measure mechanical indicators is correlated to that of the easy-to-measure electrical indicators, and by introducing optimality criteria, it is clarified to which extent the latter suffices to identify optimal driving frequencies as the geometric configuration and the material parameters vary.