Relationships between moisture content and dielectric constant for subgrade and subbase of pavement courses

Abstract

Soil is used as bedding or support for all types of heavy structures, including roads, highways, and structural foundations. The strength parameter, California Bearing Ratio (CBR) is a vital consideration during the design stage of pavement and other superstructure. Moisture content in soil also plays an important role for values of CBR. In conventional methods, soil samples are taken from the field to the laboratory and moisture content is determined by standard oven drying method. These methods are laborious, time-consuming, and require excessive care and accuracy. In the presence of water, the soil shows strong dielectric performance. Microwave frequency can exploit the large contrast between the dielectric constant of free water and soil. The dielectric constant of the soil is directly correlated to soil moisture content. The dielectric constant of the soil varies with proportion of water substrate present in it. This research focused on using microwave techniques by considering frequency domain to determine moisture content of subgrade and subbase. Three samples were tested in this research: soil subgrade, sand subbase and blended subbase. Several laboratory tests such as determination of moisture content by conventional method, grain size analysis and Modified Proctor test were performed. The dielectric data measurements were taken for the three samples at operating frequency range from 1 MHz to 20 GHz. The values of dielectric constant, loss factor and conductivity of the samples were determined independently. The relation between dielectric constant, E'r and moisture content was established in polynomial formulation in this research. The dielectric constant of soil subgrade rose rapidly with the increase of moisture content. However for sand subbase, the dielectric constant rose slowly with the increase of moisture content. The conductivity of soil sample increased more rapidly with moisture changes than sand sample. It was observed that the soil sample was highly conductive while the sand sample had poor conductivity. This study was the first of its kind to apply microwave techniques for considering frequency domain in determining moisture content of pavement layers.