This study investigates cavitating flows within a silicon-glass microfluidic venturi reactor using an integrated multi-modal approach that combines cavitation-induced noise analysis and high-speed imaging. Acoustic signal processing, primarily via power spectral density (PSD) and time-resolved frequency analysis (using wavelet transform), was employed to characterize cavitating flows across a wide range of upstream pressures (0.76 to 4.52 MPa). To the best knowledge of the authors, this is the first study on the direct correlation of time-resolved cavitating flow morphology captured via high-speed imaging of arrival, growth, and shedding of cavities with their acoustic signatures in both PSD diagrams and wavelet spectrograms in micro scale. By mapping specific PSD peaks and time-localized wavelet bursts to distinct flow regimes, a unified framework spanning inception through fully developed cavitation was constructed for a micro-venturi reactor. At lower pressures (0.76 MPa), acoustic emissions in the 2–8 kHz range corresponded to cavitation inception with low cavitation intensity. As the upstream pressure was increased, a shift toward higher frequencies (up to and beyond 18 kHz) and a broadening of the spectral distribution revealed a transition to more intense sheet and cloud cavitation regimes. These acoustic signatures were confirmed by visual observations, which captured the evolution of distinct cavitating flow patterns. The findings highlight the potential of acoustic noise measurements as a real-time diagnostic tool for optimizing microscale hydrodynamic cavitation processes, with high potential for their implementation in applications such as wastewater treatment, energy harvesting, biomedical engineering, and integrated lab-on-a-chip platforms.
Tebyani et al. (Fri,) studied this question.