The dielectric (DE) properties, specifically the DE constant (ε′) and loss factor (ε′′), were measured for vacuum-dried and freeze-dried potato samples at a microwave frequency of 2.45 GHz over a range of different moisture contents (MCs) using a DE probe and also a 2-probe electrochemical impedance spectroscopy (EIS). Third-order polynomial models (ε′ = f1(MC); and ε′′ = f2(MC)) at room temperature were developed for regression analysis. Additionally, at various temperatures (T), biphasic 3rd-order polynomial models (ε′ = f1(MC, T); and ε′′ = f2(MC, T)) were obtained to determine ε′ and ε′′ as a function of MC and T using measured data. The vacuum-dried potato sample showed a good fitness of ε′ and ε′′ (R2 = 0.95 and 0.96, respectively) to the regression model with the range of MCs from 18% to 80% (w/w), while the freeze-dried potato sample showed a good fitness of ε′ and ε′′ to the 1st-phase regression model with MC < 50% w/w (R2 = 0.95 and 0.96, respectively) and the 2nd-phase regression model with MC > 50% w/w (R2 = 0.94 to 0.96). EIS measurements were also used to obtain correlation impedances for ε′ and ε′′ determined by the DE probe method. The resulted regression analysis meets the demands for simple, rapid, and accurate assessment for transient values of ε′ and ε′′ of food products during dehydration/drying processes. The EIS method was verified to be a successful alternative to direct measurements of ε′ and ε′′.
Moisture content and temperature are crucial factors strongly influencing temperature changes in chipping potatoes under microwave heating or drying due to potential shift in dielectric constant (ε′) and loss factor (ε′′). This study provides comprehensive regression models developed based on destructive (typical microwave probe) and nondestructive (electrochemical impedance spectroscopy, EIS) measurement techniques. The models are capable to predict the unknown ε′ and ε′′ values using a wide range of known moisture and temperature values of potatoes.