Characterisation of dairy emulsions by NMR and rheological techniques

Abstract

Droplet size distribution strongly affects macroscopic properties of emulsions. In this work rheological and NMR (nuclear magnetic resonance) analyses were performed to evaluate material properties as well as droplet size distribution and possible flow-induced alteration. The investigated materials were commercial dairy emulsions, based on milk, vegetable fats, a mixture of sugars, emulsifiers (hydrophilic or lipophilic) and natural stabilizers (carrageenan and guar gum). Rheological properties of three different dairy emulsions were related to the droplet size distribution obtained by NMR and dynamic light scattering (DLS) techniques. NMR seems to be a particularly useful technique because no previous sample treatment is necessary, whereas DLS requires a significant emulsion dilution that could change the droplet size distribution. Obtained results evidence the effect of two relevant ingredients (surfactants and fats) on rheological properties and droplet size distribution of tested emulsions. For sample containing sunflower oil and a hydrophilic emulsifier, rheo-NMR evidenced mechanical instabilities also confirmed by rheological tests. Comparison of results about the droplet size distribution evaluated by NMR and DLS evidenced that in this case sample dilution could affect size distribution. Finally, rheo-NMR was confirmed to be a powerful instrument capable of determining the material structure and alterations induced by the flow.

Introduction

Emulsions are complex two-phase systems, made by droplets dispersed in a continuous phase. They are widely used in different industrial fields including oil drilling, transport, cosmetics, pharmaceutics and foods. Food emulsions are particularly interesting because their physical and chemical properties affect quality aspects like “texture” and shelf life in a complex way.

When designing a new product, a relationship between macroscopic parameters (related to perceived properties) and material microstructure is necessary to understand the effects of relevant ingredients (such as fats and emulsifiers) aiming to obtain a formulation with controlled characteristics avoiding a long “trial and error” approach.

Emulsion stability and “texture” (i.e. rheological properties) are probably the most important characteristics to be considered when a new product is studied. Stability affects product processing (e.g. shear induced separation during pumping), unit operation design (stirring systems, pumps, etc.) and shelf life (potential phase separation before commercial limits). Rheological properties are necessary to design unit operations properly (e.g. pumping systems) and determine the organoleptic characteristics perceived by the consumers.

Even though many data are available on model systems, few works discuss commercial products' properties, probably owing to the complexity of these systems containing a mixture of fats (or oil soluble components like fatty acids or sterols), different surface active components (such as emulsifier, proteins, phospholipids, solids) competing at interface, sugars, salts, hydrocolloids (McClements, 1999).

In this work, dairy emulsions typically used for whipped cream production, were studied aiming to describe their rheological properties and their stability as preliminary requirements for their commercial application. However, owing to the complexity of the system the proper techniques to investigate the relationship between microscopic structure and macroscopic parameters have to be selected.

Droplet size and distribution are one of the most important parameters in characterising emulsions, affecting stability, creaming-resistance, rheology, and chemical reactivity (Haque and Kinsella, 1988, Johns and Hollingsworth, 2007).

It is known (Ford, Borwankar, Martin, & Holcomb, 2000) that the effects of droplet size are mainly observed when emulsions are concentrated (volume fraction > 0.2–0.4), however, droplet size effects play a relevant role also when the concentration is relatively low. According to Pal (1996) when “soft-spheres”, i.e. deformable particles, are present, the particle size effects come into play even when the dispersion is dilute. Moreover, quite often in complex systems, the dispersing phase contains a three-dimensional network, owing to the presence of hydrocolloids and other structuring compounds, therefore particles cannot be considered as isolated hard spheres and particle–particle interactions can be very important in determining emulsion properties.

Owing to the relevance of these parameters, many techniques are described in the literature for droplet size distribution analysis such as optical microscopy, scanning electron microscopy (SEM), transmission electron microscopy (TEM), light scattering, ultrasound spectrometry, nuclear magnetic resonance (NMR) (Berger, 2000, Egelandsdal et al., 2001, Johns and Hollingsworth, 2007, Kiokias et al., 2004, McClements, 1999, Van Dalen, 2002), all of these techniques feature advantages and drawbacks (Johns and Hollingsworth, 2007, Kiokias et al., 2004, McClements, 1999).

Microscopy is quite easy to use and it is considered the traditional method, however, it is a time consuming technique, the sample preparation (e.g. spreading the emulsion across a slide or diluting to make it transparent) could alter the specimen to be analysed and only a 2D picture of the sample can be obtained (except when confocal microscopy is used).

Electron microscopy (SEM or TEM) is a powerful tool able to give information on morphology of droplets and air bubble and on the microstructure of the sample; however, the sample has to be prepared in the proper way (e.g. it has to be free of volatile components that could evaporate) therefore the major limitation of this technique is the difficulty in preparing samples without altering their structure.

Light scattering methods, based on the scattering of light by particles, is widely used for droplet sizing, however, it can be used only if emulsions are dilute (typical volume fraction lower than 1%) and transparent. Therefore emulsions often have to be diluted and stirred to homogenise the sample, and structure alterations could occur. Moreover, errors can arise due to the presence of small solid particles or droplet aggregates not distinguished by this technique. Ultrasound spectrometry can be applied to concentrate emulsions without sample pre-treatment, but errors can be produced by the presence of gas bubbles; moreover, a large number of thermo-physical properties are required to interpret the measurements.

Among the different experimental methods, NMR techniques are largely adopted to characterise emulsion systems, particularly to determine droplet size distribution as confirmed by the vast literature (Balinov et al., 1994, Denkova et al., 2004, Hollingsworth and Johns, 2003, Johns and Hollingsworth, 2007). Nuclear magnetic resonance (NMR) pulse field gradient (PFG) emulsion droplet sizing technique overcomes many problems being also readily applicable to suspended drops (Hollingsworth et al., 2004, Kiokias et al., 2004). It can be used to study opaque and concentrated systems and results are not affected by possible contaminants (e.g. gas bubble or suspended solids) (Johns & Hollingsworth, 2007). Finally, it is noteworthy that NMR techniques are “not perturbative” methods, giving information about the material structure without altering the sample. In addition, an interesting application of NMR analysis, i.e. rheo-NMR, is also able to study the material behaviour in flow conditions giving relevant information on the structural changes induced by flow.

In this work dairy emulsions, based on commercial formulations, were subjected to NMR diffusion study, aiming to determine droplet size, and to rheo-NMR analysis in order to evaluate wall slip and structure stability problems.

Rheo-NMR was used to validate data from the traditional rheometer estimating the potential presence of experimental problems not detectable directly (such as flow instabilities or phase separation).

After this validation, data from conventional rotational rheometer were compared to NMR diffusion results in order to correlate macroscopic rheological parameters to microscopic structure.

Finally, a comparison between NMR sizing and a classic technique requiring sample preparation, such as dynamic light scattering, was also carried out to evaluate whether dilution can alter results.

As a result, a study on emulsion stability and “texture” was carried out aiming to evaluate potential problems for commercial applications.