Improving the design of industrial microwave processing systems through prediction of the dielectric properties of complex multi-layered materials

Abstract

Rigorous design of industrial microwave processing systems requires in-depth knowledge of the dielectric properties of the materials to be processed. These values are not easy to measure, particularly when a material is multi-layered containing multiple phases, when one phase has a much higher loss than the other and the application is based on selective heating. This paper demonstrates the ability of the Clausius–Mossotti (CM) model to predict the dielectric constant of multi-layered materials. Furthermore, mixing rules and graphical extrapolation techniques were used to further evidence our conclusions and to estimate the loss factor. The material used for this study was vermiculite, a layered alumina-silicate mineral containing up to 10 % of an interlayer hydrated phase. It was measured at different bulk densities at two distinct microwave frequencies, namely 934 and 2143 MHz. The CM model, based on the ionic polarisability of the bulk material, gives only a prediction of the dielectric constant for experimental data with a deviation of <5 % at microwave frequencies. The complex refractive index model, Landau, Lifshitz and Looyenga, Goldschmidt, Böttcher and Bruggeman–Hanai model equations are then shown to give a strong estimation of both dielectric constant and loss factor of the solid material compared to that of the measured powder with a deviation of <1 %. Results obtained from this work provide a basis for the design of further electromagnetic processing systems for multi-layered materials consisting of both high loss and low loss components.