Mathematical modelling for temperature prediction of model-food system during ohmic heating in sterilization temperature range

Abstract

The objectives of this research were to investigate factors affecting the electrical conductivity (σ) of model food system and to develop empirical models for σ and heat distribution during ohmic heating in sterilization temperature range. A static ohmic heating cell consisted of a cylindrical sample chamber made of teflon drilled with 2.65 cm inside diameter and titanium electrodes at both ends. The ohmic cell was tested with 0.1 and 0.02 M sodium chloride and 0.1 M monosodium phosphate at 25°C and it was found that the differences between the measured and the reported 0 were 6.75, 2.64 and 0.87%, respectively. The effect of voltage gradient (7-16V/cm) and frequency (50-1000 Hz.) on o of potato, blanched white radish, pork, and surimi were investigated. It was found that the σs of the solid foods were not significantly affected by voltage and frequency (p>0.05). For model liquid food system, investigated factors were voltage gradient (18.2-47.3 V/cm), frequency (50-1000 Hz.), salt concentration (0-1.5%), sugar concentration (0-15%), and potato starch concentration (0-8%). The results showed that voltage and frequency did not significantly affect the 0 of the system (p>0.05) while salt, sugar and starch concentration did (p≤0.05). Addition of salt increased the 0 of the system while sugar and starch would decrease. The empirical correlation between σ of the system (σ, S/m), temperature (T, °C), and salt (Sa), sugar (Sn), and starch (St) in %w/w was 0 = 0379+ 0.87lSa - 0.037lSt - 0.0298Su+ 0.00l64T + 0.025lSaT (R2 = 0.994). The model gave the best prediction with 3% average error. Ohmic heating of liquid—particle food mixture of selected combinations of solid and liquid with the volume traction of 0.2, 0.4, and 0.6 at 50 Hz. and 15 V/cm were investigated. The effective σ (σeff) of the mixture was linear relationship with temperature and the value depended on the o and volume fraction of solid (vf). The σeff could be possibly calculated from the σ of individual components, vf of solids and dimensions of ohmic cell based on circuit- analogy concept with 10% error. During heating, it was found that the solid could heat as fast as the liquid even though it was less conductive than the liquid. Increasing vf of solid from 0.2 to 0.6, caused the heat profile to change from particle-lagging to particle leading in most cases. The heating rates of the mixture decreased as the 0,1; decreased due to more low-conductive solids in the mixture. Mathematical model for temperature prediction of the slowest heating phase could be calculated from the 0,5; electric field strength and average thermophysical properties of the mixture. The model gave satisfactory agreement with 11% underprediction in most cases.