Graduation Semester and Year




Document Type


Degree Name

Doctor of Philosophy in Civil Engineering


Civil Engineering

First Advisor

Anand Puppala


Predicting the volume change behavior of expansive soils has been a major challenge to the engineers in the past few years, due to the intricate couple effects caused by different soil composition parameters toward swelling. Hence, the behavior of an expansive soil is better understood with its inherent composition parameters. For this research, eight natural expansive clays have been sampled from different active swell zones across The United States. Basic soil classification was performed on these soil to identify the high plasticity and low plasticity clays. The swell behavior of these clays are studied using different testing methods and initial dry density conditions. A novel 3-D swell test apparatus was introduced to study the volumetric strains of these soils under confining pressures. These test results are utilized for the validation of swell prediction models. Different soil composition parameters like clay mineralogy, soil-suction relationship and pore distribution are studied using advanced techniques. These techniques include the determination of clay mineralogy from chemical properties of soil, soil water relationship (i.e. soil water characteristic curve) from pressure plate and filter paper techniques and pore distribution studies from mercury intrusion porosimetry technique and X-ray computed tomography tests. Once the soil composition parameters are determined, analysis of test results was performed to identify the variation of composition parameters with soil type. Based on the soil composition test results three soil swell prediction models have been formulated. The first model is based on diffused double layer effect of clay minerals present in the soil. The second model termed as MHC is based on two swell governing parameters whichare Mehcanical hydro and chemical behavior which constitutes the effect of soil matric suction, void ratio change and clay mineralogy. The final model was based on the ratio of total surface area measured from clay mineralogy and mercury intrusion porosimetry. All the models showed good predictions with the measured test results.


Civil and Environmental Engineering | Civil Engineering | Engineering


Degree granted by The University of Texas at Arlington