Effects of Oil-Dispersed Phase Composition and Selected Polysaccharides on the Physical Properties and Stability of Soybean-Palm Kernel Olein Blend Oil-In-Water Emulsions Model System
Ibrahim, Nor Hayati (2008) Effects of Oil-Dispersed Phase Composition and Selected Polysaccharides on the Physical Properties and Stability of Soybean-Palm Kernel Olein Blend Oil-In-Water Emulsions Model System. PhD thesis, Universiti Putra Malaysia.
An oil-in-water (O/W) emulsion is a system that consists of oil droplets dispersed in an aqueous continuous phase. Soybean oil (SBO) is commonly being used as oildispersed phases in many O/W-based food products. The products sometimes exhibit poor physical properties and stability against temperature fluctuations that can be attributed to a high unsaturation degree of SBO. These problems can be reduced by blending of SBO with more saturated oils such as palm kernel olein (PKO). However, crystallizing tendency of PKO at low storage temperatures may lead to partial droplet coalescence, causing destabilization of the emulsion. The use of certain polysaccharides however can reduce this problem and indirectly improve the overall emulsion properties. The objective of this research was to investigate the effects of oil-dispersed phase composition and selected polysaccharides on the physical properties and stability of SBO:PKO blend O/W emulsions model system. In the first stage of this study, the effect of palm kernel olein (PKO) incorporation on physical properties and stability of O/W emulsions was investigated. Soybean oil and blends of SBO:PKO at 10-40% PKO levels were used as dispersed phases (70% volume fraction) of egg yolk-stabilized O/W emulsions. The use of PKO caused a significant (p < 0.05) increase in droplet size but a significant (p < 0.05) decrease in rheological properties of the freshly prepared emulsions. With 10-30% PKO replacements, the emulsions were stable after storage at 25°C, most probably promoted by a significant content of C8-C12 fatty acids in PKO. With 30 and 40% PKO replacements, the emulsions were unstable after storage at 5°C due to high solid fat content (14-20%) which caused a severe partial coalescence. This was mainly evidence by increases in droplet equivalent surface mean diameter from 3.65-3.70 μm to 7.80-8.97 μm and decreases in emulsion yield stress from 1.72-1.82 Pa to 0.27-0.30 Pa after 30 days of storage. Throughout the storage, peroxide and anisidine values were found to be lower in the emulsions with PKO incorporated than in the emulsion with fully SBO. Under an accelerated oxidation condition (60°C, 12 days), a calibration model based on a Fourier-transform infrared spectral region (1800-1480 cm-1) was developed to predict the peroxide value in oxidized emulsions over the range of 6-45 meq/kg. In the second stage, physical properties and stability of emulsions as affected by the presence of individually 0.5% (wt/wt) xanthan gum (XG), carboxymethyl cellulose (CMC), guar gum (GG) and locust bean gum (LBG) were evaluated. A blend of SBO:PKO at 30% PKO level was used as a dispersed phase (40% volume fraction) of the emulsions. The microstructure of stored (5°C) XG emulsion showed the presence of partially coalesced droplets, explaining a large increase in its droplet size and the presence of ‘free oil’ after centrifugation at 3,500 rpm. However, partially coalesced droplets were not observed in stored CMC, GG and LBG emulsions and no ‘free oil’ could be separated under centrifugation force. The results support the ability of these polysaccharides in reducing partial coalescence by acting as a protective coating for oil droplets. Blends of XG, CMC and LBG were also prepared according to an augmented simplex-centroid mixture design with 10 points to investigate interaction effects of these polysaccharides on the emulsion rheological properties. The strongest synergistic effect was shown by ternary blends of XG:CMC:LBG at approximately 33-67% XG levels. Yield stress, apparent viscosity, elastic modulus and loss tangent responses were successfully fitted with a special quartic model (R2 > 0.89). Hence, the mixture design with regression modelling approach was shown to be a valuable tool in better elucidating and predicting the interaction effects beyond the two-component blends.
Repository Staff Only: Edit item detail