Journal of Geotechnical and Geoenvironmental Engineering

Centrifuge modeling of pore pressure buildup in a sand deposit as a result of shaking is evaluated by comparison with a large-scale experiment. In large-scale Test SG-1, a 5.6-m-thick, mildly sloping deposit of hydraulic fill clean Ottawa sand of $D_r=40\%$, was subjected to 5 s of low-intensity base shaking ($<0.02g$) that induced excess pore pressures short of liquefaction. Three centrifuge experiments using various soil deposits and saturation fluids were conducted and compared with the large-scale test. One of these centrifuge simulations used the same Ottawa sand and $D_r=40\%$ of the prototype, a viscous pore fluid, and dry pluviation deposition, which created a soil fabric stiffer than the prototype. The other two centrifuge simulations used silty sand saturated with water. The pore pressure buildup in one of the silty sand tests was in good agreement with the prototype, while the other two centrifuge deposits did not develop any excess pore pressure. The various responses in the four tests are explained by various levels of cyclic soil shear strain using a cyclic strain approach. The cyclic strains in the two tests with no pore pressure were smaller than the threshold strain needed to start pore pressure generation in sands ($<0.01\%$), while the cyclic strains in the two tests that built up pore pressure were greater than the threshold ($0.03\%>0.01\%$). In addition, two more experiments using Ottawa sand were also conducted: a large-scale level-ground test and a centrifuge sloping test subjected to a greater excitation. The results of the six experiments including the two additional tests are very consistent, further verifying the validity of the strain approach and threshold strain for both level-ground and mildly sloping-ground sites.

The effective design of earthquake-resistant structures and liquefaction mitigation techniques requires an improved understanding of the development and consequences of liquefaction. In this paper, the results from centrifuge experiments of structures with shallow foundations on liquefiable sand were used to evaluate the predictive capabilities of a state-of-the-practice numerical tool. Fully-coupled numerical simulations with the UBCSAND model implemented in FLAC-2D captured building settlements measured in these experiments reasonably well for one scaled input motion, mostly within factors of 0.7 and 1.8. The soil model captured the overall contribution of deviatoric displacement mechanisms and localized volumetric strains during partially drained cyclic loading. The primary limitation of the model became evident for slower rates of earthquake energy buildup, when the extent of soil softening and building displacement was overestimated by up to a factor of 4. The observations from recent case histories, the results of the experiments, and the insights gained from the numerical analyses are combined to provide guidance on the evaluation of building response on liquefiable sand and the performance of liquefaction remediation strategies.

Numerical simulations of a centrifuge test, in which dissipation patterns, lateral spreading, and shear strain localization were measured and recorded, are performed for validation of the numerical modeling approach and insight on the deformation mechanisms. The constitutive model calibration process is performed based on available laboratory data and is described in detail. The model container and construction sequence are simulated to provide good approximations of the boundary conditions and initial stress distributions. Two approaches for simulating successive shaking events are presented and compared. The key observations and mechanisms from the centrifuge test, namely (1) the dynamic response and onset of liquefaction, (2) the amount and pattern of surface deformation, (3) the patterns of pore pressure dissipation and void ratio redistribution, and (4) the difference in response between a sand profile treated and not treated with liquefaction drains, are all reasonably captured and bounded by the simulations.

Liquefiable soil-structure interaction material models are developed and implemented in the open-source finite-element modeling platform OpenSees. Inputs to the free end of the $p\text{-}y$ materials include the ground motion and mean effective stress time series from a free-field soil column. Example simulations using a single $p\text{-}y$ element attached to a soil element demonstrate key features. The models are then used to analyze centrifuge experiments of a single pile in a level liquefiable profile and a six-pile group in a sloping liquefiable profile that resulted in lateral spreading. Measured displacements and mean effective stress time series are used as inputs to isolate the response of the material models from predictive uncertainties in free-field ground motion and excess pore pressure. The predicted pile response agrees reasonably well with measurements. The cyclic mobility behavior of sand in undrained loading is shown to be an important mechanism affecting bending moments in the piles; neglecting the dilatancy component of the sand's response (i.e., ignoring the cyclic mobility behavior) can result in underprediction of the demands imposed on the piles.

This paper presents the results of laboratory investigations into the deformation of coal fouled ballast stabilized with geogrid at various degrees of fouling. A novel track process simulation apparatus was used to simulate realistic rail track conditions subjected to cyclic loading, and the void contamination index (VCI) was used to evaluate the level of ballast fouling. The experimental results show that coal fines act as a lubricant, causing grains of ballast to displace and rotate, and as a result, accelerate its deformation. However, coal fines also reduce ballast breakage because of a cushioning effect, that is, by reducing interparticle attrition. The inclusion of geogrid at the interface between the layer of ballast and subballast provides additional internal confinement and particle interlocking via geogrid apertures, which reduces deformation. A threshold value of $\mathrm{VCI}=40\%$ is proposed to assist practitioners for conducting track maintenance. If the level of fouling exceeds this threshold, the geogrid reinforcement significantly decreases its effectiveness and the fouled ballast exhibits a premature dilation. Based on the experimental results, an equation incorporating the VCI was proposed to predict the deformation of fresh and fouled ballast. This equation improves track design and assists in favorable decision support for track maintenance. Additionally, the discrete element method (DEM) was also used to capture the deformation of fouled ballast subjected to cyclic loading, whereas the DEM results were compared with experimental observations.

Sand response depends on particle morphology (size and shape). In geotechnical research and practice, size is typically assessed by sieve analysis and particle shapes are qualitatively described. Technological developments mean that digital images of sand particles can easily be obtained, enabling shape to be quantified. The complexity associated with many digital image analysis algorithms seems to have restricted their use to fundamental research studies. This study introduces a pragmatic approach for quantitative shape analysis that has the potential to be broadly adopted in geotechnical engineering research and practice. The approach generates three shape measures (convexity, sphericity, and aspect ratio) that can easily be calculated from digital images. Following an analysis of these shape measures and the imaging method used here, a database of 36 sands, including many of the sands commonly used in geotechnical research, is presented. The subjective nature of qualitative description is clear from the discrepancies that were found in published shape assessments of these sands. Convexity, sphericity, and aspect ratio data for each of the 36 sands are presented. The relevance of these parameters to geotechnical engineering is established by comparing them with widely used qualitative descriptions.

This paper presents results from field tests using explosive generated spherical stress waves to induce residual excess pore pressure and liquefaction in large saturated sand specimens. Twenty-two single spherically shaped explosive charges ranging from 0.00045 to 7.02 kg were suspended and detonated in water located over saturated sand. Little or no residual pore pressure was induced in loose, dense, and very-dense saturated specimens at peak radial particle velocity less than approximately $0.07\hspace{0.17em}\text{m/s}$ (peak shear strain less than approximately 0.005% at peak stress; late-time shear strain less than approximately 0.015% at peak displacement). Liquefaction was approached when peak radial particle velocities exceeded $0.49,0.52, \text{and} 0.71\hspace{0.17em}\text{m/s}$ (peak shear strains exceeded 0.03, 0.03, and 0.04% at peak stress; late-time shear strains exceeded 0.09, 0.09, and 0.12% at peak displacement) in the loose, dense, and very-dense specimens, respectively. Peak radial particle velocity and peak strain required to induce a given pore pressure ratio increased with increasing relative density and effective stress. Empirical relationships developed are for single charge detonations.

Following the 2008 Wenchuan, China earthquake, investigators from the Institute of Engineering Mechanics, China Earthquake Administration identified 118 sites with surface liquefaction effects. Following this reconnaissance, Chinese dynamic penetration tests (DPT) were conducted at 19 sites with liquefaction effects and 28 nearby sites without liquefaction effects. Boreholes with nearly continuous core samples were drilled near most DPT soundings revealing 1–4 m of clayey sediment overlying thick deposits of cobbley gravel. The DPT readily penetrated gravelly sediments to depths of 7–15 m. Layers that liquefied at each site are identified as the layer below the water table with the lowest average or matrix DPT resistance. DPT data are analyzed to develop probabilistic criteria for liquefaction resistance. The DPT is a robust, efficient, and effective tool for measuring penetration resistance in gravelly soils and for liquefaction hazard analyses. With further development, the DPT test could be useful for penetration and liquefaction evaluation of gravelly soils in the United States and other countries where liquefaction of gravels is an important issue.

With increased urbanization, deep foundation (bridges, signage, walls, etc.) selection is moving toward the minimization of disturbance and installation time, as well as addressing quality control and assurance issues. Unfortunately, many types of deep foundations involve noise and vibration during installation (e.g., driven piles) or integrity and reduced resistance issues (e.g., drilled shafts, both conventional and post grouted tip, continuous flight auger piles). This paper presents a new foundation type, a jetted and grouted precast pile, which uses the advantages of several proven deep foundation installation techniques. The installation of the new pile is comprised of three distinct phases: (1) pressurized water-jetting of a precast pile into the ground; (2) side grouting of the pile; and (3) tip grouting. The pile has two separate side grouting zones, each with its own grout delivery system. Each grout zone is covered with a semirigid membrane, which results in radial expansion of the soil during side grouting and horizontal orientation of the major principal stress. Small-scale testing revealed excellent bonding between the pile and the grout, as well as improved mobilized pile-soil skin and tip resistance. Both experimental and FEM modeling of the grouting and axial loading were performed on various sized jetted and grouted precast piles in cohesionless soil. On the basis of this study, a methodology that predicts expected grout pressures during grouting, unit side and tip resistance, and the load-displacement response of a pile in cohesionless soil is proposed. Further testing and research is required to validate the proposed methodology before it can be implemented in practical situations.

The strength of soil treated by cement deep mixing varies greatly, even at a single column, because of the variability of in situ soil properties, the variability of mixing effectiveness, and other factors. The variation in strength would affect the entire behavior of the cement-treated column. This paper presents a numerical experiment that investigates the effect of spatial correlation on the unconfined compressive strength of cement-treated columns. Finite element analyses incorporating random-field theory were performed to simulate the unconfined compression behavior of full-scale column samples in which the material properties vary with respect to spatial autocorrelation. The numerical results indicate that spatial autocorrelation affects the compression behavior of full-scale cement-treated columns. The overall strength of the sample in the absence of spatial autocorrelation is slightly lower than that of the uniform sample of mean strength. On the other hand, the overall strength of the sample in the presence of spatial autocorrelation is significantly lower than that of the uniform sample of mean strength. These results suggest that spatial autocorrelation should be considered when evaluating the overall strength of full-scale cement-treated columns within a relevant design procedure.

This paper presents the findings of a preliminary research program that aimed to investigate the mechanisms governing the progression of piping erosion in organic soils and attempted to use the findings to reduce the severity of piping erosion of sand. The hypothesis was that the presence of organics within mineral soils results in a reduction in piping erosion progression. Erosion behaviors of the soils were quantified using a simple erosion test with a preformed hole to simulate an initial piping channel. The research was split into five test phases. In Phase 1, the influence of grain-size distribution was eliminated to develop a preliminary understanding of the role that organic matter plays in the erosion process. The results indicated that organic matter likely contributed to a reduction in piping erosion. The second phase excluded both grain-size distribution and individual particle shape as variables in the erosion process. The results further confirmed the connection between organic matter content and erosion resistance. The third phase revealed the positive correlation between the reduced piping erosion and increased organic matter content. The fourth phase investigated the quantitative relationship between certain biologically derived substances (polysaccharides and glomalin) and piping erosion reduction. The positive correlations that were preliminarily derived from this phase indicated that these substances likely play a major role in reducing piping erosion. The final phase comprised of introducing organic matter into mineral soil and quantifying the subsequent changes in geomechanic properties. The results suggested that the introduction of organic materials into mineral soil decreases strength, increases consolidation settlement, and reduces permeability.

Ottawa sand specimens premixed with 0, 3, and 5% bentonite by dry mass of sand were tested under undrained static and cyclic loading to investigate the effects of bentonite on the static and cyclic shear strength of the sand. The results show that allowing the bentonite to hydrate within the sand pore space increases the cyclic resistance of the sand. For the same skeleton relative density and cyclic stress ratio, cyclic tests on specimens with sufficient hydration times showed a significant increase in the number of cycles required for liquefaction compared with clean sand. When the specimens were allowed an extended postconsolidation aging period, the cyclic resistance increased further. Resonant column and cyclic triaxial tests showed that this is a result of the delay in the generation of excess pore pressure in the presence of the bentonite suspension in the pore space. The improvement in cyclic behavior does not occur at the expense of the static resistance of the soil under working loads because undrained static triaxial tests on specimens with bentonite showed only a minor decrease in the small-strain internal friction angle compared with the clean sand, while the critical-state internal friction angle remained unchanged. To address the need to deliver the bentonite suspension in a sand deposit, a method for engineering the rheology of concentrated bentonite suspensions through the addition of sodium pyrophosphate (SPP) was developed. With the addition of 0.5% SPP by mass of clay, the viscosity of concentrated (10%) bentonite dispersions dropped to a value that allowed delivery of the bentonite into the sand matrix through permeation. Over time, the bentonite suspension recovered the thixotropic properties that ensured its effectiveness in mitigating liquefaction. As a result, sand specimens permeated with 10% bentonite suspensions showed a large increase in liquefaction resistance. Changes in the rheological properties of the pore fluid with time also explain the increase with aging of the cyclic resistance of the sand-bentonite specimens.

In this paper, the results of three series of drained tests carried out on sands using hollow cylinder apparatus are presented. The noncoaxiality, defined as the difference between the major principal stress direction and the corresponding principal strain increment direction, is investigated. In the first series of tests, the sand was isotropically consolidated before being sheared with the fixed principal stress direction. In the other two series of tests, the sand specimens were isotropically consolidated and then sheared by rotating the major principal stress axes while the deviator stress level was either fixed (pure stress rotation) or increased continuously (combined shear loading). The experimental results provide clear evidence for material noncoaxiality when the rotation of principal stress direction is involved. The results from these series of tests show that the degree of noncoaxiality depends on the level of deviatoric stress and the stress increment direction. It tends to decrease when the specimen is approaching a failure state. It was also observed that the effect of soil density on noncoaxiality is more significant at lower shear stress levels. Test results on two different materials, Portaway sand and Leighton Buzzard sand, were also compared with the study of the influence of material anisotropy associated with sand particle characteristics. Numerical simulations of granular materials using the discrete element method (DEM) were conducted to understand the effect of the initial material anisotropy produced during sample preparation. The DEM results were consistent with those obtained from the hollow cylinder tests.

A physics-based method is proposed for simultaneously obtaining the hydraulic conductivity, Young’s modulus, and Poisson’s ratio of soil materials using the uniaxial consolidation test. A nonlinear poroelastic model is presented, and the settlement data from consolidation tests are fitted to the model at each load step with the least-squares error method to inverse the parameters. The model results perfectly fit the experimental data in the initial load steps but slightly deviate from the data in later load steps as a result of secondary settlement and a largely increased Young’s modulus. The inversed parameters are compared with those calculated from the uniaxial consolidation test and those found in the literature. The comparison results demonstrate that the inversed parameters are reasonable. The proposed method provides both an estimation of parameters and the parameter-change information during a consolidation test. The method is simple, efficient, and versatile for obtaining poromechanical parameters with the uniaxial consolidation test. These parameters are useful for groundwater, geomechanical, and mining engineering.

Natural polymers (biopolymers) are discussed as environmentally friendly and sustainable grouting chemicals. This paper presents guidelines for selecting potentially useful biopolymers for strengthening cohesionless soil. Agar and six modified starches were identified for further study over a range of concentrations (1–4% agar and 0.5–1% starch). Experimental results demonstrated the compatibility of agar and starch. Depending on the biopolymer concentration, the unconfined compressive strength of the sand treated with agar and starch biopolymers ranged from 158 to 487 kPa. Triaxial compression tests over a range of confining pressures also indicated that the biopolymers effectively increased the cohesion intercept and stiffness of the treated sand.