Heat and mass transport during microwave heating of mashed potato in domestic oven-model development, validation, and sensitivity analysis

A 3-dimensional finite-element model coupling electromagnetics and heat and mass transfer was developed to understand the interactions between the microwaves and fresh mashed potato in a 500 mL tray. The model was validated by performing heating of mashed potato from 25 °C on a rotating turntable in a microwave oven, rated at 1200 W, for 3 min. The simulated spatial temperature profiles on the top and bottom layer of the mashed potato showed similar hot and cold spots when compared to the thermal images acquired by an infrared camera. Transient temperature profiles at 6 locations collected by fiber-optic sensors showed good agreement with predicted results, with the root mean square error ranging from 1.6 to 11.7 °C. The predicted total moisture loss matched well with the observed result. Several input parameters, such as the evaporation rate constant, the intrinsic permeability of water and gas, and the diffusion coefficient of water and gas, are not readily available for mashed potato, and they cannot be easily measured experimentally. Reported values for raw potato were used as baseline values. A sensitivity analysis of these input parameters on the temperature profiles and the total moisture loss was evaluated by changing the baseline values to their 10% and 1000%. The sensitivity analysis showed that the gas diffusion coefficient, intrinsic water permeability, and the evaporation rate constant greatly influenced the predicted temperature and total moisture loss, while the intrinsic gas permeability and the water diffusion coefficient had little influence. Practical Application: This model can be used by the food product developers to understand microwave heating of food products spatially and temporally. This tool will allow food product developers to design food package systems that would heat more uniformly in various microwave ovens. The sensitivity analysis of this study will help us determine the most significant parameters that need to be measured accurately for reliable model prediction.