Protein-Stabilized Emulsions and Whipped Emulsions: Aggregation and Rheological Aspects

By exploiting the combined gelling and stabilizing properties of the milk protein casein, creamy foam structures can be made by whipping air into a matrix of flocculated protein-coated emulsion droplets. Acidified sodium caseinate-stabilized emulsions based on liquid triglyceride oil give rise to elastic foams of low rigidity and high apparent fracture strain. Replacing all-liquid droplets with all-solid emulsion droplets (crystalline n-eicosane) produces a brittle foam of shear modulus similar to that of traditional whipped dairy cream. Addition of emulsifier (LACTEM) affects interdroplet interactions during whipping, leading to a fracture strain characteristic of traditional whipped cream, even in systems with a high proportion of all-liquid droplets. INTRODUCTION A common procedure for producing an aerated dairy system is through the shearinduced destabilization of an emulsion during whipping. In generic terms, this multiphase system can be regarded as a particle-stabilized foam. The emulsified dairy fat is semi-crystalline, and the gas bubbles incorporated during whipping are stabilized by aggregated fat globules. This type of shear-induced aggregation is known as ‘clumping’ or partial coalescence. It is caused by the crystalline fat from one milk fat globule rupturing the adsorbed layer of another under the influence of the externally applied shear forces. On shearing emulsions in the absence of air under conditions of gradually increasing applied stress, the onset of partial coalescence is associated with a dramatic jump in viscosity and substantial fat particle structuring. Traditional whipped cream can be made by whipping dairy cream (~ 35% fat) at 5 °C for 2–3 minutes in a kitchen mixer at moderately high speed until it develops characteristic ‘stiff peaks’. Fully whipped cream is a soft elastic solid that can support its own weight (shear modulus ~ 5 kPa) and is easy to ‘shape’. Additionally, it has a brittle (‘short’) rheology, with a low yield strain (< 0.1%), and it flows easily in the mouth with a smooth creamy texture. The main physical destabilization mechanism is syneresis. This typically leads to extensive separation of serum after a few hours. Question 1: Is it possible to prepare an aerated emulsion possessing the texture of traditional whipped cream, but without the structure being stabilized by clumping of partially crystalline fat droplets? To try to answer this question, we have been investigating the stabilization of model aerated systems via bridging aggregation of protein-coated emulsion droplets containing a completely liquid triglyceride oil. The underlying concept of the method is that the lowering of the pH towards the isoelectric point of the protein (sodium caseinate) leads to less steric and electrostatic stabilization, enhanced protein–protein interactions, and Protein-Stabilized Emulsions and Whipped Emulsions: Aggregation and Rheological Aspects Eric Dickinson*, Brent S. Murray and Kirsty E. Allen Procter Department of Food Science, University of Leeds, Leeds LS2 9JT, UK formation of an emulsion gel structure. Using this concept, we can make an aerated caseinate-stabilized emulsion (pH ~ 5, gas content similar to whipped cream) based on a liquid oil (groundnut oil) having good stability with respect to serum separation. However, the appearance and texture more resemble a gelled dessert (i.e., a mousse) rather than traditional whipped cream. That is, the model system is characterized by a polymer gel-like rheology, i.e., with rubberlike elasticity, a relatively low shear modulus, and a high fracture strain. Question 2: Is the difference in rheology between traditional whipped cream and our aerated acidified caseinate-based emulsion mainly due to the liquid-like character of our emulsion droplets. In experiments designed to answer this question, we replaced the all-liquid droplets (groundnut oil) with all-solid droplets (n-eicosane). This generated a considerably more rigid and brittle foam, i.e., more similar to whipped dairy cream. Moreover, unlike the partially coalesced globules of whipped cream, the aerated n-eicosane emulsions contained the stabilizing globules as clearly well separate entities, as observed by scanning electron microscopy (SEM). Therefore it is suggested that the solid-like internal character of the dispersed droplets, accompanied by the presence of less ‘soft’ droplet–droplet interactions, is essential for the support of ‘whipped-cream-like’ mechanical stresses within the network of droplets in the aerated acidified caseinatestabilized emulsion. This report focuses on the influence of a low-molecular-weight oil-soluble surfactant on the properties of these aerated acidified protein-stabilized emulsions. The emulsifier is called LACTEM (lactic acid esters of monoglycerides) which is commonly used in the food industry as a commercial ‘whipping aid’ for dairy and non-dairy creams. The objective is to explore the feasibility of using emulsifier addition to assist in the preparation of aerated acidified caseinate-stabilized emulsions containing a high proportion of all-liquid droplets and possessing a ‘short’, rigid, whipped-creamlike texture. Full details concerning the materials and methodology used in this research have been described in full elsewhere. Emphasis here is on the key results and conclusions. BASIC EMULSION PROPERTIES Samples of the oil-in-water emulsions (2 wt% caseinate, 30 vol% groundnut oil or neicosane) were prepared with different LACTEM contents (0.25, 0.5, 0.75 and 1.0 wt%) by high-pressure homogenization. To ensure complete droplet crystallization, the emulsions containing n-eicosane (melting point 37 °C) were prepared at 60 °C and subsequently rapidly cooled to 5 °C. The mean droplet diameter of the freshly prepared emulsion containing 0.25 wt% LACTEM was d32 = 0.4 μm ± 0.05 μm as determined by static multi-angle light scattering (Malvern Mastersizer). This was the same d32 value as found for equivalent emulsions without added emulsifier. For higher concentrations of LACTEM, the droplet-size distributions showed increasing evidence of flocculation, probably due to protein bridging; these flocs could be redispersed with excess sodium dodecyl sulfate (SDS). Groundnut oil emulsions with ≥ 1 wt% LACTEM were visibly unstable, as were n-eicosane emulsions containing ≥ 0.75 wt% LACTEM. This was manifest as ‘oiling off’ in the groundnut oil emulsions, and a clumped ‘cottage-cheese-like’ texture in the n-eicosane emulsions. EMULSION ACIDIFICATION We first consider the change in rheology of the emulsion systems during acidification in the absence of air incorporation. The emulsion samples were slowly acidified by hydrolysis of glucono-δ-lactone (GDL). The acidulant level was such as to reduce the pH from 7 to 5 in around 2 hours (see Fig. 1) and to give a ‘final’ value of pH ≈ 4 after around 6 hours. Figure 1. Change in pH of emulsions as a function of acidification time. The labels show different times at which whipping was begun to produce the overrun values shown in Fig. 3. Figure 2. Time-dependent storage modulus G’ (25 °C, 1 Hz) of caseinate-stabilized emulsions (30 vol% groundnut oil, 2 wt % protein) acidified with 0.6 wt% GDL and with various LACTEM concentrations: 0 wt% (thick solid line); 0.25 wt% (thin dashed line); 0.5 wt% (thin solid line); 1 wt% (thick dashed line). Oscillatory viscoelasticity measurements were carried out in the concentric cylinder cell of the Bohlin CVO Rheometer as a function of the time following addition of GDL. Fig. 2 indicates the time-dependent storage modulus G’ (at 1 Hz) of groundnut oil emulsions (2 wt% sodium caseinate, 30 vol% oil) containing 0, 0.25, 0.5 or 1.0 wt% LACTEM. With higher emulsifier content, there is a slight increase in gelation time and also a moderate decrease in the developing gel rigidity (after 3–4 h). This behaviour is similar to that previously reported for the effect of addition of a low concentration of the oil-soluble emulsifier, Span 20 (sorbitan monolaurate), on the acid-induced gelation of a sodium caseinate-stabilized emulsion. The reduction in modulus on addition of surfactant may be attributed to competitive protein displacement from the oil–water interface leading to a change in the droplet character from active to inactive filler particles. OVERRUN OF AERATED EMULSIONS According to convention, the ‘overrun’ is defined as the gas-to-liquid volume ratio expressed as a percentage. Fig. 3 shows the overrun as a function of whipping time for a sodium caseinate-stabilized emulsion (30 vol% groundnut oil, 2 wt% protein) which was gradually acidified using GDL (0.6 wt%). Whipping was done with a standard kitchen whisk beater used under controlled conditions. The overrun reaches a maximum at pH ∼ 5.1–5.2, which is roughly the stage when the foam no longer drips from the whisk beater. Figure 3. Overrun against time for whipping of sodium caseinate-stabilized emulsion (30 vol% groundnut oil, 2 wt% protein, 0.6 wt% GDL) started at different stages of acidification: A, , pH = 5.50; B, , pH = 5.40; C, Δ, pH = 5.30; D, ♦, pH = 5.20; E, , pH = 5.10. The labels A–E correspond to the same acidification times as indicated in Fig. 1. time (min) A