C. G. de Kruif

Hard sphere colloidal dispersions: Viscosity as a function of shear rate and volume fraction

The viscosities of suspensions of sterically stabilized (hard) silica spheres in cyclohexane are reported as a function of shear rate (γ) and volume fraction (6×10−4<φ<0.6). The shear thinning scales according to (ηr−η∞)/(η0−η∞) =1/(1+1.31ηγa3/kT) with limiting low and high shear viscosities described up to φ∼0.35 by η0=1+5/2φ+(4±2)φ2+(42±10)φ3 , η∞=1+5/2φ+(4±2)φ2+(25±7)φ3 . At higher volume fractions the viscosity becomes more sensitive to φ and diverges at φm=0.63±0.02 (γ→0) , φm=0.70±0.02 (γ→∞) . The experimental results compare well with existing hard sphere theories and the data of Krieger (1972) for aqueous lattices. Even at the highest volume fraction neither yield stresses nor shear thickening are observed.

Polysaccharide protein interactions

Abstract The interaction between proteins and polysaccharides, as can be observed in food related systems, is systematically discussed by separating biopolymer interactions into respectively enthalpy- and entropy-dominated types. We present examples of typically enthalpy driven phase separations, such as biopolymer incompatibility, described by the classical Flory-Huggins theory. This behavior is for instance observed for the system gelatin–dextran. In mixed systems where excluded volume or depletion interaction plays an important role the phase separation is mainly driven by entropy. Here the interaction of (random coil) polysaccharides and protein (covered) particles such as casein micelles and emulsion droplets in dairy systems may lead to phase separation as well.

Hard‐sphere Colloidal Dispersions: The Scaling of Rheological Properties with Particle Size, Volume Fraction, and Shear Rate

The steady‐shear rheological properties of four submicron sterically stabilized silica dispersions differing in particle size were measured. The high‐ and low shear limiting viscosities were found to be a function of the volume fraction only and the volume fraction at which the viscosity diverges was found to be φm=0.63±0.02 in the low shear limit and φm=0.71±0.02 in the high shear limit, independent of particle size. The shear‐thinning behavior scales in the Peclet number. At higher volume fractions, the shear‐thinning transition shifts to a lower Peclet number. The time scale on which the shear‐thinning transition takes place is comparable to the time scale of short‐time self‐diffusion, that is Dsshort/a2 and relates to the structural changes in the dispersion which were observed in an earlier rheo‐optical study.

Casein micelle interactions

Native casein micelles are considered as an association colloid, sterically stabilized by a layer of κ-casein hairs. This hairy or furry layer, as it is traditionally called, is considered as a polyelectrolyte brush. Its stability or extension is related to the brush density (renneting), charge density (pH) along the chain, concentration of (divalent) salt ions and polarizability of the solvent (ethanol content). At lowered brush density, lowered charge density, lowered polarizability of the solvent and increased calcium level, the brush will more readily collapse on the surface of the micelle. It is then assumed that, as a result, the casein micelle loses its colloidal stability and flocculates, allowing brush collapse to be easily monitored. This simple unifying picture was checked by testing theoretical scaling relations, e.g. the relation between renneting level and pH. Results show that there is a quantitatively and qualitatively correct relation between the influence of the various parameters. A simple picture of the stability of casein micelles is presented which allows a good estimate of the influence of technological treatments on the stability of skim milk.

Supra-aggregates of Casein Micelles as a Prelude to Coagulation

Native casein micelles (i.e., the micelles in fresh milk) can be treated as a collection of polydisperse hard spheres. This follows from small angle neutron scattering, viscosity, diffusivity, and other measurements. Therefore, the equilibrium and transport properties of native casein micelles in an ultrafiltration permeate solvent can be described by theories developed for colloidal hard sphere dispersions because native casein micelles are sterically stabilized by a brush of k-casein (CN) molecules (and perhaps b-CN). The k-CN brush induces a very short-ranged or steep repulsion (it rises from zero to a large value) when two micelles meet each other. Electrostatic interactions are highly screened because of the high ionic strength (0.08 M) of skim milk. The colloidal properties of casein micelles change with the technological treatment applied to skim milk. This paper describes the consequences for the casein micelle properties of heating, renneting, and acidification. It appears that the properties of the micelles can be described generally by adopting the adhesive hard sphere model. In that model, the steep repulsive interaction of two micelles is preceded by a shortranged Van der Waals attraction. By relating the strength of the attraction to the degree of technological treatment (e.g., renneting time or pH changes), the colloidal properties of the micelles can be described simply by using adhesive hard sphere theory. This theory also predicts the phase behavior of such a system. For instance, it predicts correctly the gelation of casein micelles under various conditions. The adhesive hard sphere model allows a general and consistent understanding of the colloidal properties of casein micelles caused by technological treatments. The practical relevance of the models is illustrated with a few examples.

Microencapsulation of oils using whey protein/gum arabic coacervates

Microencapsulating sunflower oil, lemon and orange oil flavour was investigated using complex coacervation of whey protein/gum arabic (WP/GA). At pH 3.0–4.5, WP and GA formed electrostatic complexes that could be successfully used for microencapsulation purposes. The formation of a smooth biopolymer shell around the oil droplets was achieved at a specific pH (close to 4.0) and the payload of oil (i.e. amount of oil in the capsule) was higher than 80%. Small droplets were easier to encapsulate within a coacervate matrix than large ones, which were present in a typical shell/core structure. The stability of the emulsion made of oil droplets covered with coacervates was strongly pH-dependent. At pH 4.0, the creaming rate of the emulsion was much higher than at other pH values. This phenomenon was investigated by carrying out zeta potential measurements on the mixtures. It seemed that, at this specific pH, the zeta potential was close to zero, highlighting the presence of neutral coacervate at the oil/water interface. The influence of pH on the capsule formation was in accordance with previous results on coacervation of whey proteins and gum arabic, i.e. WP/GA coacervates were formed in the same pH window with and without oil and the pH where the encapsulation seemed to be optimum corresponded to the pH at which the coacervate was the most viscous. Finally, to illustrate the applicability of these new coacervates, the release of flavoured capsules incorporated within Gouda cheese showed that large capsules gave stronger release and the covalently cross-linked capsules showed the lowest release, probably because of a tough dense biopolymer wall which was difficult to break by chewing.

Stability of casein micelles in milk

Casein micelles in milk are proteinaceous colloidal particles and are essential for the production of flocculated and gelled products such as yogurt, cheese, and ice-cream. The colloidal stability of casein micelles is described here by a calculation of the pair potential, containing the essential contributions of brush repulsion, electrostatic repulsion, and van der Waals attraction. The parameters required are taken from the literature. The results are expressed by the second osmotic virial coefficient and are quite consistent with experimental findings. It appears that the stability is mainly attributable to a steric layer of κ-casein, which can be described as a salted polyelectrolyte brush.

Interaction of pectin and casein micelles

Abstract The influence of pectin (low methoxyl—LM, low methoxyl amidated—LMA, high methoxyl—HM) on the stability of milk was investigated using the dynamic light scattering and viscosimetric methods. At pH 6.7, a depletion flocculation of the casein micelles was observed. This mechanism involves the exclusion of the polymer pectin chains from the space between colloidal particles (casein micelles), which induces an effective attractive interaction between the colloidal particles. If the attraction is strong enough a phase separation occurs in agreement with the theoretical predictions. At pH 5.3 the pectin molecule adsorbs onto the casein micelles; at low concentrations of pectin, a bridging flocculation is observed. On increasing the pectin concentration further the casein micelles become fully coated and attraction between the particles is lowered. On increasing the pH from 5.3 to 6.7, pectin desorbs. These experiments thus show that depending on the interaction between protein and polysaccharide, different ‘instabilities’ are observed.

Substructure of bovine casein micelles by small-angle x-ray and neutron scattering

The casein micelles of cow’s milk are polydisperse, more-or-less spherical, protein particles of up to several hundred nanometer in size, containing about 7% by dry weight of calcium phosphate. Small-angle neutron scattering with contrast variation and small-angle X-ray scattering were used in critical tests of models of casein micelle substructure. An inflexion in the neutron scattering curve near Q � /0.35 nm � 1 was observed in heavy water which became a more pronounced subsidiary maximum at the match point of the protein. In water-rich buffers, where the contrast between protein and calcium phosphate is small, the inflexion was less apparent. The position of the inflexion and its variation in shape and relative importance with contrast matching are explained poorly, if at all, by the submicelle models of casein micelle substructure. However, the observations are explained by a model in which a relatively uniform protein matrix contains a disordered array of calcium phosphate ion clusters. A notable achievement of the model is the prediction of the position of the subsidiary maximum from independent measurements of the intrinsic viscosity of micelles, their mass fraction of calcium phosphate and the mass of the core of a calcium phosphate nanocluster. # 2002 Elsevier Science B.V. All rights reserved.

Effects of carrageenan type on the behaviour of carrageenan/milk mixtures

Abstract The apparent hydrodynamic diameter of casein micelles was determined as a function of hydrocolloid concentration at 60°C and during cooling to 20°C. The systems studied contained skim milk 100-fold diluted in permeate in the presence of either lambda-, iota-, kappa-carrageenan or guar gum. These measurements show that all three forms of carrageenan adsorb onto casein micelles, whereas guar gum does not. Lambda-carrageenan (which is always in the coil conformation) adsorbs at all the temperatures studied, whereas the iota and kappa forms adsorb only at temperatures below the onset of the helix–coil transition. The results agree with the idea that adsorption of carrageenan occurs only above a certain minimum charge density. Phase diagrams were established at 60°C for the three carrageenans and the rheological behaviour was followed on cooling and at 25°C. The results suggest that the lambda-carrageenan bridges the casein micelles at low carrageenan concentrations, leading to the sedimentation of carrageenan/milk mixtures at 60°C and structure formation on cooling. At 60°C, iota- and kappa-carrageenan induce depletion flocculation of casein micelles above a critical carrageenan concentration. On cooling, systems containing iota-carrageenan clearly form a network at the helix–coil transition temperature. The network is formed by mixed carrageenan/casein micelle crosslinks. In the presence of excess carrageenan, this network is reinforced by carrageenan/carrageenan crosslinks. For the kappa-carrageenan/casein micelle mixtures the existence of mixed carrageenan/casein micelle crosslinks is less obvious but seems likely.