Predicting the engineering and transport properties of soils using fractal equivalent circuit model: Laboratory experiments

Abstract

Frequency-dependent electrical measurements of soils contain useful information about their texture and structure that can be linked to their engineering and transport properties. We performed frequency-dependent electrical measurements on 29 natural soils with wide variability in physical and textural properties in a laboratory environment at a constant stress level and in the frequency range of 0.01 Hz–10 kHz. The engineering and hydraulic properties of these soils, that is, the hydraulic conductivity [Formula: see text], void ratio [Formula: see text], fines content [Formula: see text], intergranular void ratio [Formula: see text] and the dry density [Formula: see text] are concurrently measured. The electrical behaviors of the soils are modeled with an equivalent circuit model, which are described by six circuit parameters. Relationships between the circuit parameters and the soil properties (geotechnical engineering and hydraulic) are investigated. Crossplots of frequency exponent [Formula: see text] and resistivity [Formula: see text] and that of [Formula: see text] and grain percent resistivity [Formula: see text] clusters soils with high and low values of hydraulic conductivity, whereas crossplots of relaxation time [Formula: see text] and [Formula: see text] clusters soils with high and low intergranular void ratio. Regression models are developed using the parameters [Formula: see text] and [Formula: see text] to predict the hydraulic conductivity with [Formula: see text]; [Formula: see text] and [Formula: see text] to predict the intergranular void ratio with [Formula: see text] and [Formula: see text] and [Formula: see text] to predict the dry density with [Formula: see text].