Optimization of combined microwave-hot air roasting of malt in terms of energy consumption and neo-formed contaminants content and simulation of the temperature distribution in malt during the microwave heating

Abstract

This research consists of three parts. The first part aimed to determine color development and kinetics of Neo-Formed Contaminants (NFCs) formation during combined microwave-hot air roasting of malt. Malts were roasted at different specific microwave powers (2.50, 2.75, and 3.00 W.g⁻¹) for 3.4 min followed by hot air roasting at 200 °C for 0, 37.5, 75, 112.50, and 150 min. Changes in color represented by color difference (ΔE) and formation of NFCs including hydroxymethylfurfural (HMF), furfural, furan, and acrylamide were analyzed by using fluorometric and physico-chemical method. The browning (ΔE) of malts increased logarithmically with roasting time. Furan determined by the fluorometric method increased linearly with time which indicated that the formation of furan followed a zero order kinetic model. The changes in HMF, furfural, and acrylamide with roasting time could not be fitted with any kinetic model. The second part aimed to optimize the process of combined microwave-hot air roasting of malt in terms of energy consumption and NFCs formation. Malts were roasted by combined microwave-hot air at various specific microwave powers (SP = 2.50 to 3.00 W.g⁻¹), microwave heating times (tmw = 3.3 to 3.5 min), oven temperatures (Toven = 180 to 220 °C), and oven heating times (toven = 60 to 150 min). The response variables; color, energy consumption by microwave (Emw) and oven (Eoven), total energy consumption (Etotal), and quantity of HMF, furfural, furan, and acrylamide were determined. Comparing with conventional processes (Toven = 180 °C, toven = 2.5, 3.5, and 5 hours for coffee, chocolate and black malts, respectively) combined microwave-hot air reduced Etotal by 40, 26, and 26% for coffee, chocolate and black malt, respectively, and reduced HMF, furfural, furan, and acrylamide contents in black malt by 40%, 18%, 23% and 95%, respectively. The proposed quadratic models of color, Emw, Eoven, Etotal, HMF and furfural contents were experimentally validated. The experiment was carried out using the following parameters: SP = 2.50 W.g⁻¹ for 3.48 min, Toven = 215 °C for 136 min. Percentage difference (%D) values between the predicted values and the experimental ones were less than 20%. In the last part, microwave heating of malts in a beaker (7.4 cm diameter and 9.5 cm height) was simulated using finite element method. The model was experimentally validated by heating malts in a lab scale microwave oven operated at 2,450 MHz. The average value of difference %Davg between the measured and the simulated temperature at the center of the bed was +10%. Heat transfer to a 75-g malt bed subjected to various microwave powers (2.50, 2.75, and 3.00 W.g⁻¹) for 120 seconds and malt beds (75, 100, and 125 g) subjected to microwave heating at 100 W for 120 seconds was simulated. The simulation results in all cases showed that absorbed microwave energy concentrated near the center of a malt bed. Increasing microwave power from 2.50 W.g⁻¹ to 3.00 W.g⁻¹ resulted in an increase in absorbed power from 2.0⁶x10⁶ W.m-3 to 2.48x10⁶ W.m-3, whereas increasing malt weight from 75 to 125 g resulted in a reduction in maximum temperature of malt bed from 288.17 °C to 53.11 °C. HMF and acrylamide contents were predicted from formation kinetic model and temperature of malt bed. HMF and acrylamide content at the hot spot in a 75-malt subjected to microwave heating at 2.50, 2.75, and 3.00 W.g⁻¹ increased as a function of time and temperature.