Effect of ingredient interactions and heat treatment on the structure and stability of dairy based oil-in-water emulsions : a thesis presented in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Food Technology at Massey University, Palmerston North, New Zealand
Oil-in-water emulsions are an important basis of many pharmaceutical, nutraceutical and food products such as mayonnaise, dressings, coffee creamers, chocolate, infant formulae, parenteral emulsions and enteral medical beverages. The effects of protein adsorption, addition of other food components (i.e. polysaccharides, carbohydrates and minerals), processing conditions (i.e. high pressure, heat and shear) and environmental conditions (i.e. pH, ionic strength and solvent quality) on the physicochemical properties of protein-stabilized oil-in-water emulsions (i.e. heat stability, creaming stability, viscosity and flow behavior) have been covered in past studies. In addition, there is increasing information on the heat stability and the viscosity of emulsions containing high concentrations of protein. However, there is very limited understanding of the inter-particle interactions in high solids emulsion systems containing high amounts of protein, oil or carbohydrate. A lack of studies that have investigated the types of interfacial proteins and non-adsorbed proteins, the aggregation state of the proteins, the particle size distributions of protein particles and oil droplets and the presence of a third component, especially under high temperature heating (i.e. 120-140 °C) conditions, has limited the understanding of how processing and ingredients influence the properties of proteins and, in turn, the physical stability and the rheological properties of protein-stabilized emulsions.
To address this issue, the research described here was aimed at determining the importance of the mechanisms of instability, such as depletion flocculation, bridging flocculation, aggregation, creaming and coalescence, in high solids emulsion systems and at understanding the colloidal forces that drive specific material/rheological properties and the storage stability of complex high solids systems. Five main techniques were used to characterize the structure-material properties of protein-stabilized emulsions: light scattering particle sizing, Turbiscan, rheology, polyacrylamide gel electrophoresis and confocal microscopy. The behaviors of interfacial proteins, non-adsorbed proteins and protein-coated oil droplets before and after heat treatment, in the presence of a third food
ingredient (polysaccharides or sugars) and with pre-homogenization protein preheat treatment were investigated.
As a prelude to addressing the mechanisms of instability of emulsions containing high concentrations of protein, before and after heat treatment (120 °C for 10 min), a stability map approach was developed. The model emulsions were successfully categorized into four model systems based on droplet-protein, protein-protein and droplet-droplet interactions with respect to the roles of the types of adsorbed and non-adsorbed proteins. Emulsions stabilized by whey proteins or containing high whey protein concentrations were shown to be highly sensitive to calcium content and heat treatment. Non-adsorbed whey proteins were involved not only in the heat-induced destabilization of whey-protein-stabilized emulsions but also when the interfacial layer was replaced by sodium caseinate (NaCas). In this case, NaCas-stabilized droplets were incorporated in the gel network, probably through heat-exchanged interfacial interaction. Gels were formed after heat treatment when NaCas was added to whey protein emulsions; this gel formation was dependent on the casein to whey protein ratio. Replacement of the continuous phase intact whey proteins with hydrolyzed whey proteins improved the heat stability of the emulsions. With one notable exception, surface-active fractions from hydrolyzed whey proteins were seen to promote droplet coalescence in Tween-20-stabilized emulsions at high heating temperatures. Model emulsions containing relatively high concentrations of micellar or non-micellar caseins were found to be heat stable. However, rapid enhanced creaming, which was associated with the depletion flocculation, occurred in these emulsions. The non-adsorbed protein concentration, the aggregation state and the polydispersity of the protein determined the extent of depletion flocculation.
Other studies on the effect of the depletion interaction potential and the continuous phase viscosity on the creaming stability of weakly flocculated NaCas-stabilized emulsions showed that an increase in the non-adsorbed protein concentration promoted a stronger depletion interaction potential, which could change emulsion destabilization to restabilization because of the formation of a space-filling droplet network. At higher NaCas concentrations (= 6% w/w), there was a significant delay in the formation of the space-filling droplet network, which was attributed to the particularly small diffusive
motion of flocculated droplet clusters and their inability to redisperse under a strong depletion interaction potential. Although the addition of maltodextrin or xanthan gum influenced the creaming behavior, neither additive could be considered to be an inert thickening agent. The reduced network formation caused by the addition of maltodextrin was not an effect of viscosity, but rather a breakdown of the caseinate nano-particles that induced the formation of the network. Once the maltodextrin concentration exceeded 15% w/w, the droplets were entirely non-interacting and would presumably cream at a rate predicted by modified Stokes’ law for concentrated emulsion (Ø > 0.2) (Tadros, 2004). Adding xanthan gum to the NaCas-stabilized emulsions resulted in local domains that were rich in either xanthan gum or NaCas and emulsion. The effect of these local domains was that the droplets experienced a stronger depletion interaction potential and the high viscosity of the combined xanthan gum and emulsion phases resulted in uniform phase separation across the emulsion rather than gravimetric separation.
Depending on the heating duration, heat treatment (120 °C) of NaCas-stabilized emulsions (2-8% w/w, protein) resulted in a weakened depletion interaction potential and a more rapid formation of the space-filling droplet network. Heat treatment of NaCas-stabilized oil droplets had relatively limited influence on the phase separation kinetics. Heat-induced degradation of non-adsorbed NaCas molecules was the dominant factor determining the creaming stability of NaCas-stabilized emulsions. The particle size, viscosity and molecular weight of continuous phase NaCas decreased as the heating time increased. A deviation of the size of the NaCas nano-particles from the optimum value (~ 20 nm in diameter) resulted in changes in the strength of the depletion interaction potential and the range of attraction. A decrease in the continuous phase viscosity also contributed to rapid formation of the droplet network at higher NaCas concentrations (> 6% w/w).
The droplet break-up and the heat stability of milk protein concentrate (MPC)-stabilized emulsions were determined to understand the impact of carbohydrate type and concentration during emulsification and subsequent heat treatment. The addition of different carbohydrates (up to 30% w/w) slightly increased the droplet diameter to similar extents. However, the addition of 30% w/w maltodextrin significantly decreased the
droplet diameter, which was attributed to the marked decrease in the dispersed/continuous phase viscosity ratio. Generally, added carbohydrates reduced the heat stability maximum in heat coagulation time-pH profiles of MPC-stabilized emulsions. The pH at the heat stability maximum was shifted towards more acidic values by an increased concentration of glucose, maltose, sucrose or trehalose but towards more alkaline values by an increased concentration of maltodextrin. The extent of destabilization also varied between carbohydrates, with trehalose being particularly effective in retaining the original heat stability of the MPC-stabilized emulsion. Reducing carbohydrates (glucose, maltose and maltodextrin) decreased the heat stability maximum more considerably than non-reducing carbohydrates (sucrose and trehalose).
Studies on the effect of pre-homogenization heat treatment on the physicochemical, microstructural and rheological properties of protein-stabilized emulsions showed that preheating the MPC and the whey protein concentrate denatured the whey proteins, leading to a reduction in subsequent heat-induced interactions between adsorbed proteins, between non-adsorbed proteins and between adsorbed and non-adsorbed proteins. This was attributed to fewer reactive groups being available for secondary heat-induced interactions after pre-homogenization heating. Extensive heat-induced aggregation during post-homogenization heat treatment resulted in delayed creaming because of the high viscosity at low shear rates. The presence of non-micellar caseins, the homogenization pressure and the order of heat treatment of the proteins were shown to significantly influence the physical stability and the rheological properties of protein-stabilized emulsions.
Overall, the results of this research have advanced our understanding of how the protein concentration, the aggregation state of caseins, the heat-induced physicochemical changes of proteins and the attractive force impact on the physical stability, rheological properties and microstructures of model protein-stabilized emulsions. This information may have important implications for developing tailor-made milk protein ingredients that allow controlled functionalities in the processing and storage of dairy emulsions and for making strategic plans to control/manipulate the properties of high solids colloid systems.