Unsaturated soils play an essential role in a variety of natural earth processes and engineered earthen systems. The pore water in an unsaturated soil system forms a complex fabric consisting of saturated pockets of water under negative pressure and a network of liquid bridges formed near the particle contact points. Water influences bulk soil behavior by modifying intergranular stress through negative pressure in the saturated pores, and by providing an intergranular bonding force through the liquid bridges. The magnitude and relevance of each mechanism, however, is highly dependent on the pore water fabric, which is readily altered with changes in suction, saturation, wetting direction, external stress, and global or localized deformation. It is evident that changes to unsaturated soil microstructure under mechanical or hydraulic loading will influence macroscopic soil behavior, but the difficulties associated with its characterization have limited development of microstructure-based frameworks for predicting macro soil response. This collaborative project seeks to observe and quantify the multiphase fabric of unsaturated soils by making use of recent advances in non-destructive imaging techniques. Microfocus X-ray computed tomography will be integrated with a series of special loading stages designed to image the microstructure of unsaturated sand specimens under controlled suction and stress conditions and over a wide range of saturation and strain. Images will be analyzed to characterize salient features of the multiphase fabric, including 3D grain orientation, particle contact normals, liquid bridge configurations, and the distribution of liquid- and gas-saturated voids. Tensors describing these features will be quantified and their evolution tracked as specimens are subject to controlled changes in suction, wetting direction, compression, and shear. Grain size, density, anisotropy, suction, confining stress, and strain rate will be treated as experimental variables. Microstructural observations will be integrated into a new constitutive framework for unsaturated soil behavior that explicitly accounts for elements of the solid, liquid, and gas fabric. The research will work to resolve the links between unsaturated soil microstructure and macroscale response, and will implement them through a new constitutive platform for predicting engineering behavior. Observations of fabric evolution with hydraulic and mechanical loading will provide direct evidence to address the bottleneck issues that currently limit our predictive understanding of unsaturated soil behavior, including wetting-drying hysteresis, coupling between suction, saturation and deformation, liquid bridge rupture, dilation, and rate effects. Understanding multiphase interactions in packed particles with wetting fluids is also critical to other scientific fields that deal with physical phenomena such as filtration, drying, pharmaceutical and ceramic agglomeration, and oil recovery. Teaching and diversity will be enhanced through graduate and undergraduate student involvement and educational module development, including activities targeted specifically for women and minorities at the University of Missouri and Washington State University.
Introduction: A majority of the geotechnical engineering problems involve unsaturated soils. These include, problems associated with precipitation-induced landslides expansive or collapsing soils, near surface contaminant transport, bearing capacity, and settlement of shallow foundations. Unsaturated soil behavior is greatly influenced by the co-existence of air and water in pore spaces which result in phenomena that have significant behavioral differences from two-phase media. Therefore, a proper understanding of these processes requires considerations that must go beyond those currently available for saturated or dry soils. The presence of negative pore water in unsaturated media is one of the main causes for the significant difference between saturated and unsaturated soil behavior. This variable defined as soil suction contributes significantly to the shear strength of soil and to the factor of safety of unsaturated slopes. However, infiltration during rainfall increases the pore-water pressure in soil resulting in a decrease in the matric suction and the shear strength of the soil. A majority of current research introduces soil suction as a separate stress variable to provide a practical solution for engineering applications. Since suction is inherently coupled with parameters like saturation, wetting direction, and intergranular stress, such an approach becomes ineffective. Within the funicular and pendular regimes (saturation less than about 90%), the water assumes a complex fabric consisting of saturated pockets under negative pressure and a network of liquid bridges formed among particles near the contact points. Suction influences soil behavior in two different ways: by modifying the intergranular stress through negative pressure in the saturated portions of the pore structure, and by providing an interparticle bonding force through the liquid bridges. The magnitude and relevance of each mechanism, however, is highly dependent on the pore water fabric, which may be highly anisotropic and is readily altered with changes in suction, saturation, wetting direction, external stress, and global or localized skeleton deformation. There is an inherent nonlinear coupling between suction, saturation, and corresponding intergranular stress that is manifest as non-linearity in macroscopic stress, strength, and volume change behavior (e.g., nonlinear increase in shear strength or stiffness with increasing suction). Measurement and characterization of this liquid fabric for unsaturated soil assemblies over a range of saturation, stress, and deformation plays a pivotal role in improving our fundamental understanding of unsaturated soil behavior. However, lack of microstructural visualization techniques had hindered the consideration of liquid fabric distribution and its evolution in macro scale geotechnical formulations. Intellectual Merit Outcomes: This collaborative research between Washington State University (PI Muhunthan) and University of Wisconsin-Madison (PI Likos, initially at University of Missouri, Columbia, MO) used X-ray computed tomography to characterize the complex fabric and used the parameters to develop a microstructure based effective stress theory for unsaturated soils. The research developed a novel suction and wetting direction controlled experimental setup and integrated it within the X-ray CT scanning system to perform real time monitoring of microstructural variations in partially saturated soil samples. It also developed and implemented several digital image processing algorithms for the microstructural characterization of the liquid fabric and its distribution in unsaturated soils. Such combination of novelty in both experimental measurements and digital characterization is rare and is what made the contributions of this research to be at the heart of the breakthrough research to quantify changes in soil fabric in real time and to develop a microstructure based effective stress theory for unsaturated soils. The results of the study have been presented in a number of journal papers and at several national and international conferences. Full and periodically updated details of the study are provided on a dedicated website www.waxct.wsu.edu/NSFProjects.html Broader Impact Outcomes: The robustness of the methodology and algorithms developed goes beyond the immediate area of research to broader areas of science. The software developed has been used to characterize recycled pavement cores, clogging of pervious concrete pavements, fiber reinforced concrete pavements, and distribution of ice lens in frozen Salmons. This project provided major funding support for a graduate student Kalehiwot N. Manahiloh. It offered him the opportunity to investigate unsaturated soil behavior by integrating classical soil mechanics theories with cutting edge X-ray CT and digital image processing technologies. Manahiloh successfully finished his Ph.D. study in 2013 (Manahiloh 2013) and has joined University of Delaware as an Assistant Professor in the Department of Civil and Environmental Engineering. His exceptional dissertation work has also enabled him to continue to advance his research in several collaborative research activities at the University of Delaware. In summary, it is evident that the intrinsic merits of the contributions of the research performed coupled with its applications to broader areas of science and technology makes the research a unique one. References: Manahiloh K.N. (2013). "Microstructural analysis of unsaturated granular soils using X-ray Computed Tomography." Ph.D. Dissertation, Washington State University.
