Loading...
Understanding cavitation-induced droplet breakage: modelling, experimentation and simulations
Citations
Altmetric:
Date
2026-02
Abstract
Liquid–liquid emulsions are central to numerous industrial applications in healthcare, food and nutrition, personal care, agrochemicals, and home care. Traditional emulsification methods, such as high-pressure homogenisers, rotor–stator mixers, and colloid mills are energy-intensive and face scalability challenges. Hydrodynamic cavitation (HC) has recently emerged as a promising alternative, offering enhanced energy efficiency and operational versatility. Despite its growing adoption, predictive models for droplet breakage in cavitation-based emulsification remain limited. Existing frameworks are often computationally expensive, designed for shear-dominated systems, and fail to capture cavitation-specific physics. This thesis addresses these gaps by developing new experimental, computational, and theoretical approaches to describe and predict droplet breakage under cavitation.
Two major contributions are made. First, a simplified drop breakage model is introduced for HC-driven emulsification. This model approximates droplet size distributions (DSDs) as three representative populations, each characterised by a mean diameter, thereby reducing computational complexity. Breakage due to both shear and cavitation is accounted for through the energy dissipation rates, with cavitation dissipation (𝜀𝑐𝑎𝑣) expressed as a factor (𝛼) of the average shear dissipation (𝜀̅). The model was applied to oil-in-water emulsions produced by three fluidic devices, hydrodynamic cavitatied based vortex diode (VD), helical coil, and fluidic oscillator (FO), spanning a wide range of operating conditions. Experimental measurements of DSD evolution confirmed the predictive accuracy of the model, which successfully reproduced Sauter mean diameter (𝑑₃₂) trends and droplet breakage efficiency across devices. Comparative analysis revealed that VD exhibited the highest drop breakage efficiency, FO was optimal for low energy inputs (<20 J/kg), HC performed best in the intermediate range (20−100 J/kg), and VD was preferable for higher energy consumption regimes (> 100 J/kg).
Second, the thesis presents the first cavitation-specific drop breakage frequency model, bridging microscale cavity–drop interactions with device-scale predictions. Direct numerical simulations (DNS) employing a volume-of-fluid (VOF) multiphase model were conducted to study droplet fragmentation under symmetric cavity collapse. Parametric analysis of interfacial tension (𝜎), viscosity ratio (𝜆) which is dispersed phase over the continuous phase viscosity, droplet-to-cavity size ratio (𝛽) , and driving pressure (Δ𝑃) revealed complex deformation and fragmentation mechanisms, dominated by vortex-induced breakup and secondary droplet formation. Results showed that 𝜀𝑐𝑎𝑣 Increased with 𝛽 , 𝜎 , and Δ𝑃 , but decreased with 𝜆. A quantitative correlation was derived between 𝜀𝑐𝑎𝑣, key parameters, and dimensionless groups (Weber and Ohnesorge numbers), identifying Δ𝑃 and 𝜎 as dominant factors. Complementary high-speed imaging experiments of cloud cavitation, generated using both ultrasonic horns (acoustic cavitation, AC) and snapping shrimp mechanisms (HC analogue), confirmed that cavitation collapse drives bi-modal droplet size distributions, with smaller droplets from cavity collapse and larger droplets from cavitation-induced shear. Comparison of AC and HC revealed higher breakage efficiency (larger b, smaller D90) in HC systems.
By integrating Computational Fluid Dynamics (CFD) insights with experimental data, this thesis develops a closure modelling framework that embeds microscale cavitation physics (obtained via simulations) into device-scale descriptions of droplet breakup. This work provides new physical insight into cavitation-assisted droplet breakup by explicitly linking resolved cavity collapse dynamics to reduced-order breakup and size-distribution models. Unlike existing approaches that rely on empirical correlations or turbulence-only closures, the models developed in this thesis incorporate cavitation-induced energy dissipation in a physically informed yet computationally efficient manner, enabling accurate prediction of the Sauter mean diameter, 𝑑32, without the computational burden associated with fully resolved simulations. Validation against experimental data for a hydrodynamic cavitation device demonstrates strong agreement, with coefficients of determination exceeding 𝑅2>0.9across the investigated operating conditions. Relative to existing CFD-based and population balance approaches, the proposed models reduce computational cost by one to two orders of magnitude while maintaining good predictive accuracy.
The key contributions of this work include: (i) a computationally efficient simplified droplet breakup model, (ii) the first cavitation-specific breakup frequency formulation, and (iii) a closure strategy for incorporating microscale cavitation dynamics into device-scale models. High-resolution axisymmetric simulations provide new physical insight into single and interacting cavity collapse and their interaction with droplets, establishing a direct physical link between collapse-induced energy dissipation and breakup behaviour. The limitations of the present work include the use of axisymmetric simulations, simplified representations of cavitation-induced energy dissipation, and phenomenological breakup kernels. These assumptions define the scope of applicability of the proposed models and motivate future extensions to fully three-dimensional simulations and bubble-resolved cavitation modelling. Overall, this work advances the mechanistic understanding of cavitation-assisted droplet breakup and provides practical modelling tools for device optimisation and scale-up, with direct implications for industrial efficiency and product quality.
Supervisor
Description
Peer-reviewed
Publisher
University of Limerick
Citation
Collections
Files
ULRR Identifiers
Funding code
Funding Information
Sustainable Development Goals
External Link
License
Attribution-NonCommercial-ShareAlike 4.0 International