Effect of pressure on heat and mass transfer in starch-based food systems
A two-dimensional finite element formulation was developed to solve Luikov's system of nonlinear partial differential equations for heat, mass and pressure transfer. The finite element results were verified by comparing the model results with exact solutions and analytical results available in the literature. The finite element model was then used to conduct numerical investigation of the effect of pressure gradient on moisture transfer during moisture adsorption and desorption processes in various biological materials. The finite element moisture predictions obtained from Luikov's three-way coupled heat, mass and pressure transfer model (denoted as the HMP model) showed considerable difference with those obtained from Luikov's two-way coupled heat and mass transfer model (denoted as the HM model) in which the pressure gradient is assumed zero. Experiments were conducted to measure temperature and moisture variation during drying of starch and starch/fructose samples. Drying tests revealed that the samples with high initial moisture content had faster moisture desorption rates while the temperature variations were not significantly affected by the initial moisture content of the samples. The presence of fructose in the starch/fructose samples decreased the moisture desorption rate and enhanced the temperature transfer during the drying process. An exhaustive parameter estimation method was applied in conjunction with the finite element model and the drying data to estimate the moisture filtration coefficient, which has not been determined previously for starch based materials, and to estimate the heat and mass transfer coefficients as they vary with drying conditions. The finite element model was finally applied to study the temperature, moisture and pressure transfer in starch and starch/fructose samples. The model predictions were compared with the experimental results to verify the models. Both temperature and moisture predictions from the HMP model showed good agreement with experimental results. The HM model predicted temperature fairly well, but under-predicted the moisture transfer rates at later drying stages due to the fact that the pressure gradients were neglected.