The effect of glass transition on non-enzymatic browning in dehydrated food systems

Non-enzymatic browning is an important cause of food quality loss. Maximum browning rates typically occur in the water activity (a{dollar}rmsb W){dollar} range of 0.55 to 0.75, depending on composition. At lower a{dollar}rmsb W{dollar}s the browning reaction proceeds more slowly, and this decrease in reaction rates has been attributed to diffusional limitations. Several models have been developed to describe browning kinetics as a function of temperature (T), and moisture content (M). Recent work has indicated the glass transition may be an important factor in limiting diffusion in low moisture amorphous food systems. While physical stability of foods has been shown to be influenced by the glass transition the issue of chemical stability has been less clear.; This research addresses the effect of glass transition on non-enzymatic browning in food systems. Experiments were done with models systems of the reactants xylose and lysine freeze dried in various matrices (polyvinylpyrrolidones, maltodextrins, raffinose, trehalose, sucrose, lactose, carboxymethylcellulose, and mixtures of these compounds). Three types of experiments were performed: (a) extent of browning was determined in systems stored over a range of a{dollar}rmsb W{dollar}s at 20{dollar}spcirc{dollar}C for a set period of time; (b) browning rates were determined in systems at several temperatures above and below the glass transition temperature (Tg) at a fixed moisture content, and; (c) browning rates were determined in systems where both temperature and moisture contents were variables.; The extent of browning related reactions was determined by absorbance measurements in the resolubilized samples. Tgs were determined by differential scanning calorimetry. Moisture sorption isotherms were also determined. Analysis consisted of plotting browning determinations versus a{dollar}rmsb{lcub}w{rcub}{dollar} or (T-Tg) and also by constructing Arrhenius plots. Arrhenius and WLF-based models were developed to describe and predict browning rates in some of the systems. It was concluded that Tg does affect browning, but temperature and moisture effects are also important. Adequate models were developed when any two of the three following parameters were used: T, M, and (T-Tg). The best models included all three parameters.