Aplicação de um planejamento simplex-centroid para modelagem da temperatura de fusão de misturas lipídicas para indicadores de tempo-temperatura

Abstract

The temperature control of food source products demand very carefully, because in situations where this variable is not properly controlled, can be the multiplication of pathogenic and spoilage microorganisms, which depreciate the sensory and microbiological physical and chemical characteristics of the theirs. In that sensors during the study to monitor the cooling temperature is seen as important. Oils and fats have a complex array of fatty acids due to these characteristics, the melting point of these materials have a wide distribution. In this way it becomes feasible mix of planning study, working with the variable melting point. One type simplex-centroid experimental design was used to determine the fusion point of stearic acid mixtures, lard and peanut oil. The experimental data were obtained by the capillary tube method. Was possible to obtain a model for predicting the melting behavior of the lipid mixture, using linear, quadratic, models special cubic and cubic complete, all of which had to be significant to describe the results obtained (p <0.05). The full cubic model showed the best results, with regression coefficient (R2) equal to 0.998343 and adjustment coefficient (R2ajustado) to 0.997515. Was detected lack of model fit indicating that despite describe adequately the experimental data there may be influence of components used in the mixture, which have complex composition.