Modeling the Dielectric Properties of Granular Media to Determine Water Content

Dielectric measurements are elegant and reliable methods for retrieving water content distribution in soils for a range of hydrologic scales, and include time domain reflectometry, ground penetrating radar, and microwave surface measurements. The dielectric properties of soils are highly dependent on the volume fraction of water due to the large differences in dielectric constant between water (ε = 80), minerals (ε = 4-10), and air (ε = 1). Recent studies indicate, however, that the frequency-dependent dielectric behavior of some soil components, particularly clays, varies due to effects other than water content. This phenomenon, termed dielectric dispersion, occurs in the 0.1 to 6-GHz band used by most water content sensors. Some types of dispersion may arise from the microscopic arrangement and structure of the mineral particles, but a quantitative description of this mechanism is lacking. A computer simulation method was therefore developed to model the dielectric properties of water-bearing porous media in order to understand the fundamental mechanisms of dielectric dispersion, and to provide improved determination of water content using dielectric methods. The technique uses a first principles approach to predict the dielectric properties of granular materials in the radio and microwave regions, and can simulate an arbitrary 3D microstructure of up to several thousand particles. The approach models the particles as spheres, and simulates the multiple scattering of the electromagnetic fields using vector multipole expansions and iteration. To test the simulation method, the effective dielectric constants of both ordered lattices and random packings of glass particles were modeled as a function of porosity and compared to experimental data. The results show very good agreement between model and experiment. The dielectric dispersion and microscopic electric field structure were also modeled, and show that even simple systems can display complex and unexpected behavior. Future efforts will include developing models for nonspherical particles (e.g., clays and other platy minerals), soil microorganisms, and saturation processes in aggregate soils. These capabilities will provide greater scientific understanding and enhanced sensing methods for vadose zone and ground water monitoring