Towards harnessing hydrodynamic cavitation for producing emulsions: Breakage of an oil drop in a vortex based cavitation device

Highlights

• Investigated breakage of a single drop injected in hydrodynamic cavitation (HC) device.

• Developed computational model to simulate droplet size distribution.

• Provided systematic process for choosing parameters of the breakage model.

• Provided new insights into drop breakage in vortex based HC device.

Abstract

Emulsions are used as delivery vehicles for many active and functional chemical ingredients. There is an increasing need for developing new ways of producing emulsions on demand having desired critical quality attributes. Hydrodynamic cavitation (HC), which is a process of generation, growth and collapse of vapour cavities leading to intense localised shear and energy dissipation, offers an attractive platform for producing emulsions. In this work, we present experimental and computational results of a breakage of a single oil drop injected in a HC device. A single drop of oil was injected in a continuous stream of demineralised water flowing through a vortex based HC device. Tween 20 surfactant (2%) was used to prevent coalescence. The droplet size distribution (DSD) at the outlet of HC device was characterised using the imaging method. A detailed 3D computational model was developed to simulate turbulent cavitating flow through the considered HC device and resulting drop breakage and droplet size distribution. The developed model was used to quantitatively relate Sauter mean diameter (d₃₂) and droplet size distribution (DSD) at the outlet with the operating parameters of the HC device. A single-pass through the vortex based HC device was able to bring down droplet size from ∼10³ µm to ∼10¹ µm. The data and the model discussed in this work will provide a systematic basis for developing models, cavitation devices and operating protocols to produce emulsions with desired DSD.

Introduction

Liquid-liquid emulsions are widely used in various industry sectors such as food (cream, mayonnaise, milk products), personal care (cosmetics, lotions), healthcare (drug delivery systems) and chemicals (reactions, extraction). Droplet size distribution (DSD) is one of the key desired critical quality attributes (CQA), which subsequently affects other attributes such as rheology, appearance, stability etc. Significant work therefore has been done and is being done on developing ways of producing emulsions of desired DSD. Several different types of equipment are used for producing emulsions at industrial-scale and laboratory-scale. These include mechanically agitated vessels, rotor-stator, high-pressure homogeniser, ultra-sonication, hydrodynamic cavitation, micro-channels and membranes [1,2]. Agitated vessels and high-pressure homogeniser/ rotor-stators are mainly used in industrial production systems, whereas; membrane and ultrasonic systems are extensively used in laboratory-scale or developmental systems. Among these equipment, hydrodynamic cavitation (HC) offers an attractive alternative for producing liquid-liquid emulsions at both, the laboratory as well as industrial scales [3]. Ranade (2019) [3] has highlighted promise of harnessing hydrodynamic cavitation (HC) for producing emulsions on demand using compact HC devices.

The HC is a process of generation, growth and collapse of vapour cavities [4,5] leading to intense shear and localized high-velocity jets which can be harnessed for drop breakage. The HC devices are broadly classified into linear flow devices (like orifices or venturi) and swirling flow-based HC device (vortex diodes). The requirement of high inlet pressures (typically more than 500 kPa), the propensity to erosion and less control on resulting DSD of emulsions are some of the disadvantages of the linear flow devices in industrial-scale production [6]. The swirling flow-based HC devices either involve active rotor-stator arrangement for realising cavitation [7] or a configuration, which generates swirling flow without any moving parts [6,8]. Vortex based HC devices were shown to be quite effective in producing emulsions [3,9]. We have also recently investigated the influence of oil volume fraction (over the range of 1% to 20%) on DSD of emulsions produced in a single pass through vortex based HC device [10]. Comprehensive computational fluid dynamics (CFD) and population balance equations (PBE) based model was developed to simulate DSD generated by HC device. The breakage rate model proposed by Laakkonen et al. [11] was used and detailed sensitivity analysis on the breakage rate parameters was performed in our previous work [10]. The model was shown to be successful in describing the experimental data with the help of one adjustable parameter. In the present work, we have connected the parameters of the single drop breakage simulations with the simulations with higher volume fraction of oil [10]. The progress of drop breakage as drops move within the device was examined and discussed by analysing DSDs at different locations inside the device through CFD simulations. These results provide key insights and connect the results to operation with higher volume fraction emulsions. The fundamental understanding developed in the present study will be useful for producing liquid-liquid emulsion with hydrodynamic cavitation devices.

Many fundamental investigations on single droplet breakage have been carried out since the early time [12,13]. Despite many theoretical studies available on the distribution of daughter droplets from single droplet breakage, experimental investigations and CFD simulations of single droplet breakage in an industrially relevant equipment and resultant drop size distribution are limited [14–17]. Zaccone et al. [14] experimentally investigated the dynamics of single drop breakage in liquid-liquid stirred dispersions. They identified a breakup mechanism and developed a physically based model for the daughter drop size distribution that was originally formulated by Diemer and Olson [17]. Simon et al. [15] experimentally investigated single drop breakage using a rotating disk contactor. Based on the experiment, they discussed the possibility of droplet breakage due to different turbulence processes, i.e., breakage due to oscillation, shear stress, turbulent intensity and accelerated flow. They described the breakage frequency and daughter drop size distribution based on the correlated breakage probability using modified Weber number. Maaβ and Kraume [16] performed single drop breakage experiments in a stirred vessel to determine drop breakage rates. They used different breakage models available in the literature and analysed influence of model parameters on the DSD predictions. The study showed the importance of selecting appropriate breakage model parameters to simulate the DSD accurately. Several researchers simulated DSD and d₃₂ of emulsion in various equipment using different breakage models [10,18,19]. John et al. [18] predicted the drop size distribution of emulsion in a sonolator and discussed the role of satellite droplets on the DSD. They suggested that an understanding on the dynamics of droplet breakage in particular equipment is essential to predict the nature of DSD. No such study of breakage of a single dispersed drop in HC devices is available in the open literature. We have attempted to bridge this gap in the present work.

In this work, we have experimentally and computationally investigated breakage of a single oil drop injected in a vortex based HC device. The droplet size distribution (DSD) at the outlet of HC device was measured after injecting an oil drop in a surfactant containing demineralised water flowing through the HC device. A comprehensive model based on computational fluid dynamics (CFD) and population balance equations (PBE) was developed to simulate cavitation occurring in the considered HC device and resulting drop breakage and DSD. Several drop breakage models are available in the published studies which are typically based on the assumption of homogeneous isotropic turbulence [19]. These breakage models contain multiple parameters which may vary with considered liquid-liquid systems, design of devices and the range of turbulent energy dissipation rates [15,16,18,19]. The selection of the model parameters is important and not adequate information on this aspect is available. In this work, we have discussed influence of parameters of breakage model on simulated DSD and provided some basis for appropriate selection of these parameters. An attempt is also made to quantitatively understand drop breakage process occurring in the HC device by identifying the area where significant drop breakage occur. The observations on the breakage of injected single oil drop were related to the results obtained with continuous injection of oil droplets with higher oil volume fractions. The presented approach, model and results will provide a systematic basis for simulating drop breakage and emulsion formation in HC devices and will facilitate harnessing cavitation of producing emulsions of the desired DSD.