The use of ultrasonics for nanoemulsion preparation

https://doi.org/10.1016/j.ifset.2007.07.005Get rights and content

Abstract

Oil-in-water emulsions are an important vehicles for the delivery of hydrophobic bioactive compounds into a range of food products. The preparation of very fine emulsions is of increasing interest to the beverage industry, as novel ingredients can be added with negligible impact to solution clarity. In the present study, both a batch and focused flow-through ultrasonic cell were utilized for emulsification with ultrasonic power generation at 20–24 kHz. Emulsions with a mean droplet size as low as 135 ± 5 nm were achieved using a mixture of flaxseed oil and water in the presence of Tween 40 surfactant. Results are comparable to those for emulsions prepared with a microfluidizer operated at 100 MPa. The key to efficient ultrasonic emulsification is to determine an optimum ultrasonic energy intensity input for these systems, as excess energy input may lead to an increase in droplet size.

Industrial relevance

The preparation of oil-in-water emulsions is a common feature of food processing operations. The use of ultrasound for this purpose can be competitive or even superior in terms of droplet size and energy efficiency when compared to classical rotor­stator dispersion. It may also be more practicable with respect to production cost, equipment contamination and aseptic processing than a microfluidisation approach. The present paper shows that ultrasound can be effective in producing nanoemulsions for use in a range of food ingredients.

Introduction

Food grade emulsions formed from oils rich in Omega-3 polyunsaturated fatty acids (PUFA), such as fish oil and flax seed oil are commercially attractive because of the potential health benefits associated with their consumption (Burdge and Calder, 2006, Kolanowski and Laufenberg, 2006). Flax seed oil has the extra benefit of a pleasant odor compared to fish oil (Burdge & Calder 2006). The preparation of such emulsions with small droplet size is of particular interest. Small droplet sizes in general leads to a creamier mouth feel and greater emulsion stability (McClements, 2004). Furthermore, reductions in oil droplet sizes below 100 nm have the potential to provide a translucent emulsion (Tadros, Izquierdo, Esquena & Solans, 2004) that can be incorporated readily into beverages and food gels without a loss of clarity. There are a number of mechanisms available for the production of such emulsions. The traditional method employed in the food industry uses valve homogenization (McClements, 2004). This process is energy intensive as only a small percentage of the applied energy is effective (Tadros et al., 2004). Since the mid 1990's a high energy emulsification device, a microfluidizer, has gained prevalence (McClements, 2004, Strawbridge et al., 1995). This technique uses particle-particle collision through a microfluidic channel architecture, rather than a straight shear field, to cause particle size reduction (McClements, 2004).

The use of low frequency ultrasound for emulsion formation is well established, at least on a laboratory scale (Richards, 1929, Abismail et al., 1999) However, most work to date has focused on the preparation of synthetic emulsions, for example for the paint industry or in the preparation of polymeric nanoparticles. The development of such techniques for the food industry is a much more recent phenomenon (Freitas et al., 2006, Jafari et al., 2006).

Ultrasonic emulsification is believed to occur through two mechanisms. Firstly, the application of an acoustic field produces interfacial waves which become unstable, eventually resulting in the eruption of the oil phase into the water medium in the form of droplets (Li & Fogler, 1978a). Secondly, the application of low frequency ultrasound causes acoustic cavitation, that is, the formation and subsequent collapse of microbubbles by the pressure fluctuations of a simple sound wave. Each bubble collapse (an implosion on a microscopic scale) event causes extreme levels of highly localised turbulence. The turbulent micro-implosions act as a very effective method of breaking up primary droplets of dispersed oil into droplets of sub-micron size (Li & Fogler, 1978b).

Studies to date comparing ultrasonic emulsification with rotor–stator dispersing have found ultrasound to be competitive or even superior in terms of droplet size and energy efficiency (Ma and Hsu, 1999, Abismail et al., 1999, Tadros et al., 2004). Microfluidization has been found to be more efficient than ultrasound, but less practicable with respect to production cost, equipment contamination and aseptic processing (Abismail et al., 1999). Comparing mechanical agitation to ultrasound at low frequency, Tadros et al. (2004) found that for a given desired diameter, the surfactant amount required was reduced, energy consumption (through heat loss) was lower and the ultrasonic emulsions were less polydisperse and more stable. It is the purpose of the present paper to further investigate the usefulness of ultrasound to generate food grade oil-in-water nanoemulsions and in particular to identify equipment-related constraints.

Section snippets

Materials and methods

All base emulsions were prepared from unrefined organically grown cold pressed flax seed oil (long chain triglyceride oil) as supplied by Stoney Creek (Victoria, Australia) and reagent grade Tween 40 (C16), as supplied from Sigma Aldrich (Sydney, Australia). Unless otherwise stated, the emulsion formulation was 15 vol.% flaxseed oil, 5.6 vol.% Tween 40 and 79.4 vol.% deionised water.

Two ultrasonic experimental set-ups were utilised. Batch experiments employed a Branson Sonifier of nominal power

Results and discussion

The size of an emulsion droplet formed by homogenization is controlled by the interplay between droplet breakup and droplet coalescence (Tadros et al., 2004, McClements, 2004). Droplet break up is controlled by the type and amount of shear applied to droplets as well as the droplets resistance to deformation (Laplace pressure) which is determined by the surfactant (Tadros et al., 2004, McClements, 2004). The rate of droplet coalescence (related to droplet stability) is determined by the ability

Residence time and input power

Batch experiments were initially conducted to examine the effect of power intensity, pre-emulsification and sonication time. As shown in Fig. 3, a total sonication time of five minutes was found to produce optimum results, with additional sonication providing no greater reduction in droplet size.

The effect of applied power was next considered. Conventionally it would be expected that the amount of shear would increase with the applied power, the emulsion particle size should then decrease with

Equipment design issues

While batch experiments are useful for laboratory studies, continuous flow equipment will almost always be required in an industrial size application. A major challenge for the introduction of ultrasonic technology is the effective design of such flow through equipment. In particular, it is essential that in such devices, all elements of the fluid experience similar levels of ultrasonic power intensity. With the present flow through cell, it was apparent that a fraction of the fluid flow

Conclusions

A range of food grade emulsions have been prepared from a flaxseed oil/water mixture. Results show that there is an optimum power input level beyond which droplet coalescence and cavitational bubble cloud formation restricts performance. Increasing residence time reduces droplet sizes to a point, but continued sonication beyond one to five minutes is ineffective. While the batch cell produces better results, continuous equipment is likely to be more viable in a commercial environment. In order

Acknowledgements

Funding for this project was partially provided by the University of Melbourne-CSIRO Collaborative Research Support Scheme. Some infrastructure funding was also provided by the Victorian government through a Science Technology and Innovation Infrastructure Grant. This financial assistance is gratefully acknowledged.

References (18)

There are more references available in the full text version of this article.

Cited by (0)

View full text