Journal of Geotechnical and Geoenvironmental Engineering

This paper presents a three-dimensional (3D) finite-element analysis of a bridge abutment and foundation system subjected to the kinematic demands of lateral spreading. The selected bridge system corresponds to the Mataquito River bridge in Chile. The 2010 Maule earthquake caused extensive lateral spreading on both river banks at this site. Observations suggest that 3D effects may have contributed to reducing the structural demands placed on the abutment and foundations during lateral spreading. Several 3D models representing the existing site geometry and a widened approach embankment are developed and used to estimate and compare bending demands in the shaft foundations supporting the bridge abutment. A simplified design method that accounts for pile pinning effects is also assessed as a possible tool to address this problem.

The three-dimensional failure criterion for soils was expanded to include concrete and used for rocks. The failure criterion is expressed in terms of the first and the third invariants of the stress tensor and it requires two parameters for characterization of the strength of soils and three parameters for concrete and rocks. For soils, one parameter is used to describe the opening angle of the failure surface, i.e., similar to the friction angle, and the second parameter is employed to describe the curvature of the failure surface in meridian planes that contain the hydrostatic axis. The third parameter, used for concrete and rocks, is slightly higher than the uniaxial tensile strength and is used to capture the cohesion and tensile strength. Typically, the results of three triaxial compression tests and the tensile strength from a Brazilian test or from an estimate are required for the determination of these three parameters. Careful inspection of parameters, determined from numerous high-quality tests on soils, concrete, and rocks presented in the literature, revealed that two of the parameters relate to each other. Therefore, it is possible to obtain the two parameters from the results of a single triaxial compression test. The relation between the two parameters is demonstrated, and it is shown how the results from a single test are used to determine the parameters to describe the three-dimensional failure surface for soils, concrete, and rocks.

An experimental program of large-scale direct shear tests has indicated that shear displacement of a high-density polyethylene (HDPE) geomembrane (GM) over a gravelly compacted clay liner (CCL) under moderate to high normal stress conditions can cause much greater damage to the geomembrane than static pressure alone. Essentially, no damage was observed at low normal stress. The greatest geomembrane damage occurred at high normal stress (1,658 kPa) and yielded an average of $169\hspace{0.17em}\text{holes}/{\text{m}}^2$, with a maximum hole size of 23 mm. Size, angularity, and hardness of the gravel particles are also important factors with regard to potential damage. Geomembrane damage was greatly reduced when a needle-punched geosynthetic clay liner (GCL) was placed between the geomembrane and gravelly CCL, including one test conducted at very high normal stress (4,145 kPa). The findings suggest that gravelly soils should be viewed with caution for the construction of GM/CCL composite liners for landfill bottom liner systems and other moderate- to high-stress applications. If there is a reasonable expectation for GM/CCL interface shear displacement, project-specific direct shear tests should be conducted to determine the potential for shear-induced geomembrane damage. Recommendations are provided for the performance of such tests and for design options when damage mitigation is necessary.

An experimental program of large-scale direct shear tests has indicated that shear displacement of a gravel drainage layer and nonwoven geotextile protection layer over a high-density polyethylene (HDPE) geomembrane under moderate to high normal stress conditions can cause much greater damage to the geomembrane than static pressure alone. Essentially, no damage was observed at low normal stress. The greatest damage occurred at high normal stress (1,389 kPa) using a lightweight geotextile ($335\hspace{0.17em}\text{g}/{\text{m}}^2$) and yielded an average of $31\hspace{0.17em}\text{holes}/{\text{m}}^2$, with a maximum hole size of 29 mm. Surprisingly, geomembrane damage measured using a lightweight geotextile was greater than that measured using no geotextile due to a change in failure surface location. For the same conditions, shear-induced damage was slightly less for a geomembrane placed on a compacted sand subgrade than on a compacted clay subgrade. Interface shear strength increased significantly with decreasing geotextile mass/area due to greater out-of-plane deformation of the geomembrane. The findings suggest that the placement of a gravel drainage layer on top of a HDPE geomembrane, even with a protection nonwoven geotextile, should be viewed with caution for landfill bottom liner systems and other moderate- to high-stress applications. If there is a reasonable expectation for interface shear displacement, project-specific direct shear tests should be conducted to determine the potential for shear-induced geomembrane damage. Recommendations are provided for the performance of such tests and for design options when damage mitigation is necessary.

A numerical model, called CS3, is presented for one-dimensional, large strain consolidation of layered soils. The algorithm accounts for vertical strain, soil self-weight, conventional constitutive relationships, changing material properties during consolidation, unload/reload, time-dependent loading and boundary conditions, an externally applied hydraulic gradient, and multiple soil layers with different material properties. CS3 can accommodate equilibrium and nonequilibrium profiles for the initial void ratio as well as variable profiles for preconsolidation stress and applied stress increment. Verification checks show excellent agreement with available analytical and numerical solutions. Several numeric examples are used to illustrate the capabilities of CS3 and highlight errors that may occur when multilayer systems are modeled as a single layer with average properties. Finally, settlement estimates obtained using CS3 are in good agreement with field measurements for the Gloucester test fill.

Although the behavior of shallow foundations under vertical load combinations has been the subject of numerous studies, the response of shallow foundations subjected to combined horizontal and torsional loading has received considerably less attention. New offshore applications of shallow foundations for liquid natural gas facilities and other subsea structures have underscored the importance of the behavior of shallow embedded foundations subjected to combined in-plane translation and torsion. This study investigates the undrained bearing capacity of rectangular and square shallow foundations under eccentric horizontal loads through comparisons of various limit-equilibrium and plastic-limit analysis solutions to three-dimensional finite-element (FE) solutions. In general, the plastic-limit approach considered in this paper agrees well with the FE solutions, although it has some tendency to overpredict capacity at greater embedment depths. The studies revealed a general insensitivity in the shape of the yield envelope to variations in embedment depth, which permits a simplified analysis suitable for first-order estimates of load capacity. The variables considered in this study include footing aspect ratio, embedment depth, and load direction in addition to eccentricity.

Laboratory fall cone penetration is measured as a function of saturation for four sands using equipment more commonly used to quantify plasticity and undrained strength of fine-grained soils. Relations between penetration depth and saturation display nonmonotonic behavior characteristic of the more general behavior of unsaturated sands. Penetration-saturation ($P\text{-}S$) relations are interpreted in three behavioral regimes: $P$ is largest for dry sand, decreases sharply with increasing $S$ up to residual saturation, continues to decrease slightly and approximately linearly between residual saturation and $S$ of approximately 50 to 90%, and then increases at full saturation to a penetration value less than that under dry conditions. Observations are partially attributed to suction induced by a dilative soil rupture zone and are interpreted using an effective stress framework that links soil-water retention curves and suction-stress characteristic curves for the sands.

Under some circumstances, leaving a composite geomembrane/geosynthetic clay liner (GCL) exposed to solar radiation in the field has been shown to cause shrinkage of the underlying GCL. Recent field studies have shown that leaving a composite liner exposed can also lead to erosion of bentonite from the GCL due to downslope moisture migration. This paper reports an experimental technique that reproduced similar erosion in the laboratory on a typical landfill side slope of $3H:1V$. The test method simulates the features that occur with the erosion of bentonite caused by downslope migration of evaporative water in the field. The laboratory tests demonstrate that erosion features can be present but may not be visible unless the appropriate back lighting is used. Erosion features measuring over 25 mm in width were produced. The test method simulated the features that were observed with bentonite erosion in the field and has the potential for use in examining other factors that may affect this type of erosion. The test method developed can be used for examining the response of GCLs when part of a composite liner that will be left exposed. The findings from this laboratory study provide additional motivation for timely covering of composite liner systems as recommended by GCL manufacturers.

The influence of curing temperature on the strength development of concrete, mortar, or cement-stabilized granular soils has been widely reported in the literature. However, there are fewer studies focusing on cement-stabilized clay. The key difference is the presence of pozzolanic reactions in cement-stabilized clay, which give rise to more complex strength-development behaviors than those of concrete, mortar, or cement-stabilized granular soils. This paper aims to clarify this difference using extensive data collected as part of a land reclamation project in Singapore. Five different mixes of cement-stabilized Singapore marine clay are cured under different temperatures. Their strength development with time is studied by testing the specimens under unconfined compression at various curing ages (up to 1 year). The experimental results show that a higher curing temperature gives rise to not only higher short-term strengths but also higher long-term/ultimate strengths. Based on the experimental results, a model is proposed that combines the maturity theory commonly used for concrete or mortar with a proposed temperature-enhanced strength factor, which takes into account the effect of curing temperature on pozzolanic reactions occurring in cement-stabilized clay. This proposed model is validated by another independent Singapore data set and data sets collected from the literature.

Deformation of bridge approach embankments caused by earthquake-induced liquefaction in the underlying soils can impose large force and displacement demands on the piled abutments. An equivalent static analysis (ESA) procedure that accounts for the beneficial coupling of the embankment displacements with the mobilized bridge/pile restraining forces is evaluated against the results of three centrifuge tests involving three nonpiled embankments and three piled embankments. The centrifuge tests, which showed that the restraining forces from the different pile configurations reduced the embankment displacements by 20–80%, relative to those of the nonpiled embankments, are briefly summarized. The ESA procedure is described and applied to the centrifuge tests using different sets of input parameters to isolate the questions of how well liquefaction-induced displacements are estimated for embankments without piles, and how well the relative pile pinning effects are estimated. The results of nonlinear deformation analyses (NDAs) of the same centrifuge tests are compared with those from the ESAs and are used to illustrate important differences in their capabilities. The relative merits of the ESAs and NDAs for evaluating the performance of pile abutments affected by liquefaction are discussed.

This paper analyzes the validity of the yield surface and flow rule of several isotropic and anisotropic constitutive models. These aspects are studied based on drained triaxial tests, with a number of different stress paths in a computer-controlled stress path cell. The Cam clay, modified Cam clay, and $p\text{-}q\text{-}\theta$ models are used to study the isotropic behavior, whereas the anisotropic conditions are analyzed using the Melanie, MIT-E3, and S-CLAY1 models. Both under the isotropic and anisotropic conditions, the results indicate that the surface described by the undrained effective stress path is suitable for simulating the yield and plastic potential surfaces. In terms of isotropic behavior, the $p\text{-}q\text{-}\theta$ model is the most suitable for predicting the yield, whereas the plastic potential function is closely simulated by the yield function of the modified Cam clay model. For the anisotropic models studied, the MIT-E3 model provides a better approximation of the anisotropic behavior of the Santa Clara clay.

The pressure–void ratio relationship of many structured soils in the virgin compression range is highly nonlinear. The initial part of the compression curve is characterized by the breakdown of structures, whereas the behavior in the postdestructuration range is influenced by soil mineralogy. A robust pressure–void ratio relationship should include parameters that account for the distinct mechanisms that control the behavior in the destructuration and postdestructuration ranges. A new pressure–void ratio relationship based on a modified secant compression index is proposed. It is shown that the variation of the proposed secant compression index with a logarithm of pressure can be approximated by a hyperbolic form with two parameters. The new relationship has been verified in a wide range of naturally structured soils. Parametric studies conducted show that one parameter controls the compression behavior within the stress range where the destructuration is dominant, and the other parameter controls the behavior beyond it where the structure is destroyed and mineralogy controls.