Minimising oil droplet size using ultrasonic emulsification

https://doi.org/10.1016/j.ultsonch.2009.02.008Get rights and content

Abstract

The efficient production of nanoemulsions, with oil droplet sizes of less than 100 nm would facilitate the inclusion of oil soluble bio-active agents into a range of water based foods. Small droplet sizes lead to transparent emulsions so that product appearance is not altered by the addition of an oil phase. In this paper, we demonstrate that it is possible to create remarkably small transparent O/W nanoemulsions with average diameters as low as 40 nm from sunflower oil. This is achieved using ultrasound or high shear homogenization and a surfactant/co-surfactant/oil system that is well optimised. The minimum droplet size of 40 nm, was only obtained when both droplet deformability (surfactant design) and the applied shear (equipment geometry) were optimal. The time required to achieve the minimum droplet size was also clearly affected by the equipment configuration. Results at atmospheric pressure fitted an expected exponential relationship with the total energy density. However, we found that this relationship changes when an overpressure of up to 400 kPa is applied to the sonication vessel, leading to more efficient emulsion production. Oil stability is unaffected by the sonication process.

Introduction

Nanoemulsions are colloidal dispersions comprising two immiscible liquids, one of which is dispersed in the other, with droplet sizes between 20 and 200 nm [1], [2]. The small droplet size implies that nanoemulsions flow easily, can be optically transparent and have unique texture/rheological properties [3], [4]. Nanoemulsions are also attractive from a product stability point of view; their very small size enhances their physical stability [1], [2]. These properties make them highly attractive for cosmetic products and in the food industry [3], [5], [6]. Specifically, nanoemulsions offer the potential to deliver high concentrations of oil soluble nutraceuticals or bio-active food supplements into a range of water based foodstuffs. If the emulsion droplet size is below the detection limit of the human eye (around 50 nm) then the emulsion can appear translucent and so the visual quality of the product is unaffected [7].

Typically, emulsions are prepared by physical shearing processes [8], [9], [10]. The ultimate size of a homogenized emulsion is determined by the balance between two opposing processes; droplet break-up and droplet re-coalescence [10]. The frequency of both processes is promoted by the intense shear that occurs within a high shear homogenizer such as the Microfluidizer™ or ultrasonic transducer. Droplet break-up occurs when the applied shear is greater than the Laplace pressure of the emulsion. In the simple case of a low oil volume fraction and negligible continuous phase viscosity, Taylor predicts that the emulsion radius (r):rγηcγ˙where γ is the interfacial tension, ηc is the continuous phase viscosity and γ˙ is the shear rate [10], [11], [12].

The surfactant plays a critical role in both droplet break-up and coalescence. The surfactant aids droplet break-up by lowering the interfacial tension which reduces the resistance to droplet deformation [10]. The most important role of the surfactant is to prevent the immediate re-coalescence of newly formed droplets by rapid adsorption to, and stabilization of, the newly formed interface. In this case having an excess of surfactant in the continuous phase capable of rapidly adsorbing to the interface is essential [10]. Invariably the requirements of both droplet break-up and coalescence dictate that small molecule surfactants are the most suited to the formation of nanoemulsions (compared to macromolecular emulsifiers) because of their greater ability to rapidly adsorb to interfaces and their much lower dynamic interfacial tensions.

The efficiency of droplet break-up within a homogenizer is also controlled by the nature and intensity of the shear. In most homogenizers droplet break-up occurs as a result of turbulent flow [13]. Droplet break-up in turbulent flow occurs via the action of viscous or inertial stress on the emulsion droplet. Which of these stresses dominate depends on the size of the droplet relative to the smallest turbulent eddy.1 The power density, (Pν, W/mL) the average energy dissipated per unit time and unit volume, is the main measure of the strength of the turbulence. The maximum sized droplet that can exist under a certain turbulent flow regime is given by [10]:dmax=CPν-2/5γ3/5ρc-1/5where C is a constant and ρc is the mass density of the continuous phase.

The average droplet diameter achieved is clearly a function of the intensity of shear as expressed through the power density (Pν), but will also depend upon the residence time within this shear field, τ (s) [13]. Karbstein and Schubert [15] argued that this dependency between residence time and average size is due to uneven power density distribution in the dispersing zone. To account for this dependency, they introduced the concept of “energy density”, which is the simple product of the power density and the residence time within the shear field, τ (s):Eν=Pντ

This parameter may also be described as the energy input per unit volume, or the specific energy input, Eν (J/mL) [15], [16], [17], [18]. The average droplet diameter can then be described by a simple exponential relationship [15]:dav=C·Eν-b

Arguably the most powerful homogenizer, the Microfluidizer has been the focus of much of the recent work on nanoemulsion formation [7], [13]. However, the Microfluidizer™ is very expensive and is prone to significant losses in efficiency due to high equipment wear rates. Ultrasonic homogenizers might prove a viable alternative to the Microfluidizer because they are easy to clean and are capable of achieving similar local power densities. In an ultrasonic system, physical shear is provided predominantly by the process of acoustic cavitation; the formation, growth and subsequent collapse of microbubbles caused by the pressure fluctuations of the acoustic wave. A collapse event causes extreme levels of highly localised turbulence – an implosion on a microscopic scale. It is the accumulation of many thousands of these miniature implosions that forms the basis of ultrasonic homogenization.

In a classical ultrasonic horn transducer, the cavitational bubble cloud and consequently the region of high intensity shear is focused in a small zone immediately adjacent to the transducer face. The spatial variation in the shear field can mean that power usage is relatively inefficient. This means that ultrasonic homogenization is currently only suitable for small batches [1], [15]. The performance of this process might be improved by increasing the homogeneity of the field as well as the average power density. In turn, these improvements may come from changes to equipment geometry [19], [20] or the use of higher amplitude sonotrodes [17].

Another approach is the use of a constant amplitude sonotrode at system pressures in excess of the ambient value. It is well known that increasing the external pressure increases the cavitation threshold within an ultrasonic field and thus fewer bubbles form. However, increasing the external pressure also increases the collapse pressure of cavitation bubbles [21], [22], [23]. This means that the collapse of the bubbles when cavitation occurs becomes stronger and more violent than when the pressure is at atmospheric conditions. As cavitation is the most important mechanism of power dissipation in a low frequency ultrasonic system, these changes in cavitational intensity can be related directly to changes in the power density.

Work by Henglein and Gutierrez [24] indicated that at low sonication amplitudes, the effect of the cavitation threshold was dominant and both the chemical yield and sonoluminescence arising from an acoustic field decreased with increasing pressure. Conversely, at higher amplitudes, the bubble collapse effects dominated and yields increased with increasing pressure. Similarly, Sauter et al. [25] find that low overpressures improve de-agglomeration of nanoparticles, whereas higher overpressures have a negative effect. Both Bondy and Sollner [23] and Behrend and Schubert [18] observe an optimum in emulsification efficiency at an absolute pressure of around two atmospheres which can again be attributed to these competing effects.

The application of ultrasound to the creation of nanoemulsions has been considered in several works, however, the droplet sizes achieved were generally above 200 nm [26], [27], [28], [29], [30]. Therefore in this paper, we re-examine the ability of ultrasonic homogenizers to produce nanoemulsions. Our approach is to enhance droplet formation by examining both terms in the Taylor Equation (Eq. (1)), surfactant and shear. On the surfactant side we examine how surfactant synergy can aid droplet deformation. On the shear size we examine how equipment geometry can be improved and whether the specific energy input of the sonifier can be enhanced through overpressure. Finally, we confirm through High Performance Liquid Chromatography (HPLC) that the structure of the triglyceride oil is not damaged through the use of such intensive shear. Overall the goal is to enhance ultrasonic homogenization to the point where it is capable of producing true nanoemulsions.

Section snippets

Materials and methods

Unless noted otherwise, emulsions were prepared with 5.6 wt% surfactant and 15 wt% of sunflower oil (Crisco, purchased from retail supermarkets) with the balance purified water (Millipore, MilliQ system). Three surfactants were trialed both individually and in combination; Tween 80, Span 80 and sodium dodecyl sulfate (SDS) (all from Sigma Aldrich). In some experiments, polyethylene glycol (PEG) of Molecular Weight 6000 (Chem Supply) and fresh canola oil (Crisco, purchased from retail

Surfactant system optimisation

The resistance of an emulsion droplet to deformation is governed by the magnitude of its Laplace pressure, which is controlled by the interfacial tension of the surfactant. It has been known for many decades that co-surfactants and surfactant mixing can significantly change the interfacial tension of oil–water mixtures as a result of surfactant synergy [32]. In this work, we examined if the addition of a co-surfactant (Span 80) altered the particle size of emulsions primarily stabilized by

Conclusion

It has been demonstrated that ultrasound is a viable method for producing transparent nanoemulsions of triglyceride oils in water with mean particle sizes down to 40 nm. The smallest droplet size that can be achieved is fundamentally a function of the emulsion composition. Thus, 40 nm droplets could be achieved by altering the oil content of a triglyceride/Tween/Span combination or similarly by altering the PEG content in a triglyceride/SDS/PEG system. Achieving this minimum droplet size requires

Acknowledgements

Financial support for this work has been provided by the University of Melbourne–CSIRO Collaborative Grant Scheme. Infrastructure support from both the Victorian Government Science Technology and Innovation Initiative and the Particulate Fluids Processing Centre, a Special Research Centre of the Australian Research Council is also gratefully acknowledged. Preliminary experiments and a literature review were completed by Bronwen Lee and Jan Zimak and their work is also acknowledged.

References (36)

Cited by (0)

View full text