ABSTRACT Sorption isotherms evaluate water exchange between food and environments with different water activities (aw), providing valuable information on physical changes and susceptibility of deterioration due to lipid oxidation and microbial growth. In this context, the present study aimed to characterise the adsorption and desorption curves of extruded dry foods for dogs and cats using the Guggenheim‐Anderson‐De Boer (GAB) model, as well as to compare these results with those obtained from a simpler and faster method, the drying curve. Three foods for dogs and three for cats were evaluated. The equilibrium moisture content by isotherm method was determined at 30°C using the gravimetric method with six saturated saline solutions, covering aw from 0.11 to 0.83. The isotherm curves between aw and moisture content were plotted by GAB model. The drying curve for each food was plotted between aw and moisture using direct data obtained during the drying of samples after extrusion in an air oven at 60°C, to compare this result with the desorption curve. The parameters for each curve using the GAB model were determined using RStudio software. To compare the drying curves with the sorption isotherm, the Wilcoxon paired t ‐test was used. All pet foods exhibited a type II isotherm. The average R 2 was 0.99 for the adsorption curves, 0.87 for desorption and 0.99 for drying curves, showing a good fit of the models. The adsorption and desorption isotherms, as well as the drying curves, showed different predicted equilibrium moisture contents, with higher values predicted by the desorption method. In conclusion, the adsorption curves were the most suitable for the dry pet foods evaluated, and the drying curve could not be recommended due to the lower accuracy compared to the sorption isotherms.
Silva et al. (Tue,) studied this question.