Predictive modeling of dielectric heating in porous foods is challenging due to their nature as multiphase materials. To explore the relationship between the topological structure of multiphase foods and the accuracy of dielectric mixture models, the degree of anisotropy of two cooked rice samples with 26 and 32 % porosity was determined, and their dielectric properties were estimated using the Lichtenecker (LK), Landau-LifshitzLooyenga (LLL), and Complex Refractive Index Mixture (CRIM) equations. These properties were used in a predictive finite-element model for reheating an apparent homogeneous rice sample on a flatbed microwave (MW) for 120 s. The results were compared with experimental data and a validated two-element model. Unlike LK and LLL equations, the CRIM equation predicted heat accumulation towards the edges of the container at the two values of porosity ratio evaluated, in accordance with the experimental results and the isotropic nature of the sample. The simulated temperature distributions suggest that the three evaluated equations could predict the MW heating behavior of rice to some extent, but that in order to obtain more accurate results, it could be useful to obtain an empirical topology-related parameter specific for this sample. These results can provide insight on the relationship between the topology of the porous structure in the sample and the adequacy of different dielectric mixture models.

来源平台：JOURNAL OF FOOD ENGINEERING