Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Civil and Environmental Engineering

First Advisor

Mandar M. Dewoolkar

Second Advisor

Yves Dubief


Liquefied soils inflict significant risks to the natural and built infrastructure. Liquefaction induced flow failures often have disastrous consequences. Seismic design of earth structures often requires estimates of the shear strength of liquefied soils, S_u (liq). Common practices for estimating S_u (liq) relies upon empirical correlations obtained from back analysis of flow failure case histories, but there is significant scatter in these correlations. Alternate approaches to determine S_u (liq) and its relation to shear rate are relatively still in their infancy but include a handful of attempts employing laboratory, physical modeling experiments and computational fluid dynamics modeling. The research work in this dissertation examines S_u (liq), how the shearing rate may influence S_u (liq), and flow liquefaction through novel experimental work and computational modeling.

The experimental work involved design and development of a unique cyclic triaxial testing equipment integrated with a motor assembly to pull a thin plate from the bottom to top of a liquefied sand specimen. The coupon was designed to incorporate a miniature pressure transducer inside it to measure the pore pressure on the shearing plane. Experiments were performed by applying cyclic loading to the saturated sand specimens until the excess pore pressure reached a set value to trigger liquefaction, and subsequently the coupon was pulled at different speeds. Some experiments included thin layers of colored sand placed within the triaxial specimen to measure the shear band thicknesses and assess any influence of shear rate on the shear band thicknesses. These modified cyclic triaxial test results were supplemented with centrifuge experiments where the same coupon was pulled through saturated sand level ground centrifuge models liquefied in flight using a shake table. These experimental results were analyzed to develop a shear stress versus shear rate relation of the tested sand, which showed that the Bingham plastic model captures the developed shear stress-shear strain relationship.

The numerical work involved developing computational fluid dynamics (CFD)-based models and simulating the experimental coupon movement through liquefied sand modeled as a Bingham plastic fluid. Both two-dimensional (2D) and three-dimensional (3D) models were developed, but only the 3D models captured the significant impacts of the finite length of the dragged object in the third dimension. Therefore, 3D models were adopted for further simulations with two approaches. In the first approach, the coupon was modeled to be stationary within a uniform flow field in steady state condition with the boundaries set far from the coupon. In the second approach, the coupon moved through the stationary fluid within the model resembling the experimental set up including the distances to the specimen boundaries. These simulations allowed quantification of the boundary effects, which were determined to be relatively insignificant. A multi-objective optimization problem was developed to minimize the difference between the experimental and computational drag and solved using a coupled optimization-CFD framework to determine the optimum Bingham model parameters using the simpler steady state model. CFD simulations were also used to model the shear band formation around the coupon in experiments with colored sand. The simulation results predicted slightly higher shear band thicknesses but captured the oval shape of the shear band as well as the general trend of increasing shear band thickness with increasing shear rate very well for the range of coupon speeds investigated.



Number of Pages

459 p.

Available for download on Thursday, April 23, 2026