Mobility and oxidative stability in plasticised food matrices the role of water

Abstract

The importance of water in food structure and stability is well known, and the role of water as a plasticiser for biopolymers has been extensively studied during the last 25 years. Recently, understanding of the mechanisms of water plasticisation of glassy carbohydrate matrices at the molecular level has increased and its relevance to the rate of mass transfer has been emphasized. There appears to be a lack of such studies with food proteins, although the water factor is similarly recognised for example in protein-based edible films. Furthermore, food stability is currently evaluated on the basis of water activity and the physical state of the matrix. Therefore, it would be important to consider whether the major food polymers, starch-based carbohydrates and proteins, really act similarly with respect to water. In the present work, the role of water in system stability and biopolymer interactions was studied in two different systems: cast films with and without plasticiser and spray-dried particles with a dispersed lipid phase. Plasticisation of amylose films by glycerol and water was studied by proton NMR relaxometry. In glassy amylose the proximity of Tg did not strongly affect the amylose mobility. The second moment M2, which is a measure of strong dipolar interactions and decreases with increasing distance between the protons contributing to it, decreased slightly with increasing water content. It was thus suggested that glassy state swelling occurred in amylose film. Swelling behaviour is probably important when mass transfer in the matrix is considered. In rubbery state, high concentration (30%) of glycerol increased the mobility of amylose despite the phase separation that occurs in these systems already at much lower plasticiser content. The data on mobility of plasticised amylose was combined with results presented earlier on oxygen permeability of these films. Although increasing mobility generally resulted in increased permeability, conditions were found in which the plasticiser induced segmental motions in amorphous amylose without appreciable loss in oxygen barrier properties. In powder particles, the stability of embedded lipid phase was studied in traditional carbohydrate carriers, i.e. hydrolysed (maltodextrin with gum Arabic as an emulsifier) and modified (octenyl succinate derivative) starches, and in whey protein isolate. Powders with oils rich in volatile flavour or in polyunsaturated fatty acids were prepared by spray-drying and characterised by laser diffraction and scanning electron microscopy for oil distribution and by differential scanning calorimetry for their glass transition temperatures (Tg). The powders were stored under controlled conditions, and the effect of relative humidity on the rate of oxidation was studied by following the increase in peroxide value during storage. Formation of hydroperoxides is linked with oxygen transfer in the system, as it is the oxygen consuming step of the reaction. In the case of powders with embedded oil rich in volatiles, the release of limonene and carvone was studied as a function of time and temperature. The starting hypothesis of this work was that a higher water vapour sorption at higher humidity would increase oxygen permeation in the matrices and lead to an increased rate of oxidation. This was in fact found to be the case in carbohydrate matrices during storage at 20°C, at which temperature the rate of oxidation in matrices was higher at RH 54% than at RH 11%. An opposite behaviour was found for bulk oil, suggesting that the effect of water in matrix-dispersed oil was due to matrix properties. At elevated temperatures, a difference was found between hydrolysed and modified starches. The stability of oil in modified starch still correlated with the proximity of Tg, whereas the hydrolysed starch completely lost its barrier properties at 50°C, which could not be explained by the Tg of the matrix. When volatiles release was studied at elevated temperature (70°C), little release from dry matrices was found. Intense release was found in the proximity of the glass transition temperature in all the systems. In whey protein isolate matrix, oxidation of matrix-embedded oil was retarded compared to that of bulk oil at all humidities, but followed almost the same pattern as bulk oil with respect to humidity. The rate of oxidation was high at low humidities (RH 0% and RH 11%), was retarded at intermediate humidities (RH 50% and RH 75%) and again increased at high humidity (RH 90%), at which caking of the powder was observed as an indication of physical instability. Thus, it appeared that water did not have a similar role in the matrix formed of globular proteins as it had in a glassy carbohydrate matrix. The importance of storage conditions and matrix properties for relating oxygen transfer to the rate of oxidation was demonstrated. Furthermore, it was proposed that the high solubility of the volatiles in non-volatile triglyceride phase was the reason for the retention of limonene and carvone in the matrices at elevated temperatures.