Journal of Geotechnical and Geoenvironmental Engineering

This paper proposes a new simple approach for assessing dynamic soil–structure interaction effects on structures supported by embedded massive foundations. The classical substructure method was revisited by proposing an exact decomposition approach which allows performing the inertial interaction analysis of the superstructure without modeling foundation and soil, which are replaced by properly defined impedances. The proposed approach requires redefinition of simplified formulas for both the dynamic stiffnesses and the kinematic interaction factors, which are referred to the top of the massive foundation. Accurate expressions for the complex impedances were derived based on rigorous finite-element results. The same formulas were used to calibrate a four-spring model for the analysis of the kinematic interaction problem, resulting in different expressions for the kinematic interaction factors depending on the adopted assumptions. The proposed approach, in conjunction with the novel formulas for impedances and kinematic factors, was applied to the case of bridge piers on caisson foundations, and the results were more accurate than those of the existing simplified procedures.

Jetted and grouted precast piles (JGPPs) are a newly developed urban-friendly deep foundation technique that employs water jetting for pile penetration and pressure grouting to enhance the axial and torsional resistance of precast piles. Previous experimental studies of JGPPs in a large test chamber setup have revealed that they can withstand very high axial loads; however, validation of the same in a typical field scenario was needed. This paper presents the pioneering study of the full-scale axial response of JGPPs in cohesionless soil condition. The study showed that JGPPs exhibit high skin and tip resistances under axial top-down loading. The paper also proposes a direct method to predict the ultimate skin resistance of JGPPs using the pressuremeter test results. The experimental results were compared with the values estimated using both the proposed and existing methods.

Soil strength improvement employing microbial induced calcite precipitation (MICP) involves bio-geo-chemical interaction among bacteria (here, Sporosarcina pasteurii), sand, and cementing solution. A critical review of literature opened the gap in exploring the strength gain for a given amount of ${\mathrm{CaCO}}_3$ precipitation, investigating the influence of chemical concentration and its application methodology. Thereby, the current study aims to independently maximize the calcite precipitation efficiency and unconfined compressive strength while targeting the calcite deposition of 10%–12%. In the present study, the Taguchi method of experimental design followed by analysis of variance and analysis of mean was adopted to optimize the outcome parameters. Results depicted the extreme variation in obtained strength in a small range of calcite deposition. Further, the microscopic investigation employing micrographs and elemental mapping reveals a higher impact of ${\mathrm{CaCO}}_3$ depositional characteristics than its amount on strength gain. The current study envisages the importance of chemical application parameters to improve efficiency and economy while reducing soil contamination due to non-reacting chemicals.

Recently, permeable geotextile tubes in conjunction with prefabricated horizontal drains (PHDs) have become increasingly popular for dewatering high water content slurries or sludge. However, how to analyze the consolidation process of the sludge in the geotextile tube so as to provide a proper design and prediction becomes a technical challenge. In this paper, we have proposed a two-dimensional plain-strain consolidation model for sludge consolidation in a geotextile tube under combined fill and vacuum preloading. A semi-analytical solution was obtained and validated through experimental observations. A salient finding of this study is the identification of a critical condition at which the optimum consolidation efficiency is achieved. Consolidation efficiency decreases gradually beyond this critical condition, which arrives later as the PHD pave rate and element height to width ratio increase. Furthermore, this analytical method clearly shows how preloading affects the dewatering process and the effect of fill surcharge is more pronounced than that of vacuum preloading of the same magnitude, owing to the vacuum attenuation and leakage.

Soil sorptive potential (SSP) has recently been conceptualized as the sum of four known electromagnetic potentials in soil: cation and surface hydration, van der Waals attraction, electrical attraction, and osmosis due to electrical double layer. The SSP is most pronounced near the soil particle or intracrystalline surface and rapidly decays with increasing distance therefrom, governing the highly spatially varying characteristics of many fundamental soil properties such as pore water pressure, soil water density (SWD), and soil water phase transition. A novel framework was developed to determine the functions of SSP and SWD, directly using the experimental soil water isotherm (SWI) data with the aid of closed-form SWI and SWD models. A wide spectrum of soil types was examined to validate the proposed framework. Results indicate that the SSP in these soils can vary up to six orders of magnitude within the first three layers of adsorbed water molecules, leading to abnormally high values in both water pressure (${\sim 10}^3\mathrm{MPa}$) and SWD (${1.26\mathrm{g}/\mathrm{cm}}^3$) at the soil–water interface. The predicted SWD curves are comparable to the existing experimental SWD measurements, and the controlling parameters for the SSP calibrated by the predicted SSP curves also show good agreement with the values reported in the literature, all confirming the validity of the proposed framework. It is concluded that soil sorptive potential and soil water density functions can be reliably determined from soil water isotherm data.

A well-established framework for the design of excavation support systems (ESSs) is not available in the literature. Several guidelines that have evolved from engineering practice are conventionally used in the design of ESSs. The key indicators for judging the performance of ESSs are wall deformation, internal wall stresses, and surface soil settlement information. There is evidence from several case studies that the measured wall deformations are typically lower than the predicted values. Such behavior may be attributed to ignoring the influence of capillary suction. The key objective of the study presented in this paper is twofold; the first is highlighting limitations in the current design practice and the second is to pave the way for comprehensive design of ESSs. This is achieved by undertaking numerical analysis on a silty clay taking account of the influence of saturated and unsaturated soil conditions to investigate the performance of ESSs based on a rational design approach. In addition, the proposed approach is validated based on wall deflections information using the published results of a case history. The proposed approach is promising for implementing the mechanics of unsaturated soils in geotechnical engineering practice for the rational design of ESSs. The paper also highlights some limitations of the present study and the need for future studies to provide verifications using more case histories.

The catastrophic failure of the Córrego do Feijão Dam 1 in January 2019 has raised many questions regarding the safety of tailings storage facilities constructed by upstream methods. The stability of these structures is strongly affected by the internal stratigraphy and properties of the hydraulically deposited, segregated tailings materials (from slimes to coarse sands) and by control of seepage conditions. The authors conducted an independent review of site investigation data reported by an Expert Panel (December 2019) and evaluated the stability for a critical section of the Feijão Dam 1 using 2D numerical limit analyses under various representations of fill stratigraphy and shear strength properties. The results show that the structure has a low factor of safety, $\mathrm{FS}=1.11–1.16$ on average, based on conventional assumptions of drainage conditions for fill strata. The computed failure mechanism is strongly affected by the presence of a layer of normally consolidated, low permeability slimes near the base of the fill, and by the lack of drainage provisions within the starter dam. The analyses show a clearly defined critical mechanism of failure over the full height of the fill and suggest that the dam has smaller margins of safety than previously published limit equilibrium results. The analyses do not resolve or identify causes of the actual dam failure, which occurred more than three years after the end of construction but do suggest that more reliable methods of slope stability are needed in design.

Cyclic loading on caisson and pile anchors can be a significant design issue, particularly with respect to cumulative vertical displacements that can lead to reduction in embedment depth and load capacity of the anchor. Most existing design methodologies are based on monotonic load capacity, often accounting for potential soil strength degradation due to cyclic loading. Because the cyclic strength criterion is often based on cumulative strains, these approaches arguably implicitly consider cumulative deformations. However, cumulative plastic deformations under storm conditions should be investigated. Piles and caissons serving as anchors for floating offshore structures experience inclined cyclic loading. This paper presents (1) monotonic and cyclic loading tests in the vertical and lateral directions, and (2) a finite-element (FE) investigation of different aspect ratios ($L/D=4$, 5, and 6) of caissons subjected to different load angles ($\psi =0°$ to 45° from the horizontal). Additionally, it presents a precise method to calibrate the constitutive model (nonlinear kinematic hardening model), where the calibrated model was match to monotonic and cyclic lateral and vertical loading tests. Nonuniform load amplitudes were used in the FE investigation. The primary finding from the investigation was that the influence of the cumulative displacements increases when increasing the load amplitudes, especially for the load inclinations of 0°–20° for the lateral displacements and 20°–45° for the vertical displacements.

This paper establishes an approximate circular cone strain wedge model and proposes an analytical method that can effectively connect practice with theory for the behavior of laterally loaded flexible piles with both free-head and fixed-head conditions in various types of soil based on the real shape of the strain wedge. The analytical method of the curved strain wedge model, in which the Duncan-Chang model is used to describe the stress-strain relationship of the soil and the relationship between the shear strain $\gamma$ and the horizontal strain $\epsilon$ is improved on the basis of the Mohr circle of strain, is applicable to clay and sand as well as $c\text{-}\phi$ soil. A method of calculating the shear stress along the pile is proposed for circular piles according to the friction direction between the pile and the surrounding soil. The modulus of the soil foundation reaction is calculated along the whole pile length based on the finite-difference method, without calculating the height of the strain wedge, which can truly reflect the nonlinear deformation characteristics of the pile foundation under lateral loading. The proposed method is verified by six case studies, including five field measurement experiments for various types of soil and one finite-element analysis. Comparisons show that the agreement between the results from the curved strain wedge method and those from the field tests and the finite element analysis is generally satisfactory.

Following the 2012 Emilia-Romagna seismic sequence, widespread liquefaction of silty sands was observed, providing the opportunity to enhance our knowledge of the influence of fines content on seismic hazard and mitigation works. This paper presents the results of a thorough geotechnical investigation performed in connection with full-scale controlled blast tests in Bondeno, Italy, a small village that suffered liquefaction in 2012. Piezocone (CPTU) and seismic dilatometer (SDMT) tests were performed in natural and improved soils after rammed aggregate pier (RAP) treatment to a depth of 9.5 m to provide accurate soil characterization, evaluate liquefaction, and verify the effectiveness of the ground improvement. The combined use of piezocone (CPTU) and flat dilatometer (DMT) data provided reliable estimates of the overconsolidation ratio and at-rest earth pressure coefficient and highlighted the soil improvement in silty sands between 4 and 9 m in depth. Shear-wave velocity measurements showed a low sensitivity to RAP installation. The treatment effectiveness was also confirmed by the use of the simplified procedures for liquefaction assessment, underlining the important influence of the adopted fines profile and by the blast-induced liquefaction. CPTU and DMT parameters remained approximately unchanged between the piers after the detonation.

The hydraulic performance of geosynthetic clay liners (GCLs) comprising anionic polymer–enhanced bentonites (EBs), or EB-GCLs, based on permeation with 500 mM NaCl and 167 mM ${\mathrm{CaCl}}_2$ at an average effective stress of 27 kPa and hydraulic gradients of $\sim 300$ was evaluated. The EBs included powdered sodium bentonite (NaB) enhanced with one of seven anionic polymers, including linear, noncrosslinked poly(acrylic acid) (PA) with low, medium, and high molecular weights, sodium carboxymethylcellulose (CMC) with either low viscosity or high viscosity, a covalently crosslinked sodium polyacrylate, and an in situ polymerized bentonite polymer composite. The NaB was enhanced with polymer at mass loadings of 5%, 8%, or 10% using dry-sprinkling, dry-mixing, and/or wet-mixing methods. Hydraulic conductivity tests at a low hydraulic gradient ($\sim 35$) also were conducted to determine the effect of seepage forces. Overall, the hydraulic performance of the EB-GCLs to the NaCl solution was better than that to the ${\mathrm{CaCl}}_2$ solution. Dry sprinkling (DS) with PA generally provided the best hydraulic performance, with hydraulic conductivity values based on permeation with 500 mM NaCl and 167 mM ${\mathrm{CaCl}}_2$ of $1.3\times {10}^{-11}$ and $4.0\times {10}^{-11}\mathrm{m}/\mathrm{s}$, respectively, for the high-molecular-weight PA added at 5% via DS and $2.9\times {10}^{-11}$ and $4.7\times {10}^{-11}\mathrm{m}/\mathrm{s}$, respectively, for the medium-molecular-weight PA added at 8% via DS. The CMC viscosity grade did not affect the hydraulic conductivity of the EB-GCLs. All EB-GCLs eluted polymer during permeation, regardless of mixing method or polymer type. Polymer elution was correlated with preferential interaggregate flow paths in multiple EB-GCL specimens, indicating the importance of polymer retention in maintaining low hydraulic conductivity of EB-GCLs. Because standard termination criteria for the measurement of the hydraulic conductivity of conventional unenhanced GCLs (e.g., as per current ASTM standards) do not address the potential impact of polymer elution from EB-GCLs, caution should be exercised when adapting these criteria to measure the hydraulic conductivity of EB-GCLs such as those evaluated in this study.

Liquefaction assessment has primarily been performed using in situ penetration testing, but this practice has become problematic for gravelly soils. For example, standard penetration test (SPT)- or cone penetration test (CPT)-based correlations can become unreliable owing to interference with large gravel particles, while the Becker Penetration Test, commonly used for gravelly soil, can be relatively expensive and requires conversion to an equivalent sand blow count. As an alternative, probabilistic liquefaction-triggering curves have been developed based on shear-wave velocity ($V_s$) using gravel sites in the $M_{w}7.9$ Wenchuan earthquake. These curves have significant uncertainty because of the small data set. In this study, new probabilistic triggering curves for gravel liquefaction have been developed based on a $V_s$ data set. The data set consists of 174 data points (96 liquefaction and 78 no liquefaction) obtained from 17 earthquakes in seven countries within different geological environments. The larger data set better constrained the curves and reduced the range between the 15% and 85% probability of liquefaction curves, indicating less uncertainty. These triggering curves for gravel are shifted to the right relative to comparable curves for sand, indicating that higher $V_s$ values are necessary to preclude liquefaction. To account for the influence of the different earthquake magnitudes on liquefaction, a magnitude scaling factor (MSF) was developed specifically for gravel. This curve falls within the range of other MSF curves for sands based on $V_s$.

Layered soil conditions are common in marine sediments, and deduction of strength properties for thin interbedded layers is often challenging. Full-flow penetrometers such as T-bar are increasingly being used in offshore site investigations and show high potential for differentiating the shear strength transition between layers. This paper reports the results from large deformation finite-element analysis undertaken to provide insight into the response of a T-bar penetrometer as it continuously penetrates through stiff–soft–stiff clay layers. The numerical results are validated against data from a centrifuge model test. Trapped soil is observed for any nonsmooth interface as the T-bar penetrates from a stiff to a softer soil layer, which affects the measured resistance of the soft layer. The effect of a cavity above the advancing T-bar was also investigated. The presence of a cavity can increase the sharpness of the transition in resistance from one layer to another, which thus affects estimation of the boundary of the soft layer. The evolution of trapped soil from the top layer within the soft layer, and then its erosion in the bottom layer, were explored. The research found that the measured resistance profile in the soft layer is sensitive to the strength ratio between the top layer and soft layer, but the resistance of the bottom layer is more reliable. Based on deduced measurement parameters from the resistance profile, a new interpretation framework is proposed for interpreting the layer boundaries and undrained shear strength of the interbedded soft layer.

Pore-water pressure in soil is caused by three physically distinguishable sources: ambient (environmental) pressure, surface tension–induced capillary pressure, and the soil’s electromagnetic potential–induced adsorptive pressure. The former two form the conventional concept of pore-water pressure, which is considered a constant within a soil-water-air representative elementary volume and can be directly measured by piezometer (under saturated and compressive states) or tensiometer (under unsaturated and tensile states). The third one can be called adsorption-induced pore-water pressure and is localized within a certain distance to the particle surface of soil or intercrystalline surface of swelling clay. The adsorption-induced pore-water pressure is always compressive and dictates the water phase transition in soil by altering water’s freezing point, density, and viscosity, among other physical properties. A framework of quantifying the adsorption-induced pore-water distribution via the measured soil water isotherm is presented for any soil type under any given water content. It is demonstrated that the adsorption-induced pore-water pressure can be up to 1.6 GPa in the first few layers of hydration, but will diminish to zero at a distance equivalent to the gravimetric water content $>1\%$ for sandy soil and greater than a few percent for silty soil. In clayey soil, the adsorption-induced pore-water pressure can sustain tens of megapascals even at much farther distance, equivalent to $\sim 30\%$ water content. In expansive clay, the adsorption-induced pore-water pressure inside the crystalline lamellae can exceed 800 MPa. The soil water density functions of a silty soil and a bentonite clay predicted by the proposed framework matched well with that measured independently from the conventional consolidation testing, validating the framework to determine the spatial distribution of the adsorption-induced pore-water pressure in soil.

Expansion and contraction of an integral abutment bridge cause an abutment to move toward and away from its backfill due to seasonal temperature changes, thus causing high horizontal earth pressures behind the abutment and backfill surface settlements. This paper presents the results of four model tests conducted to simulate both translational and rotational movements of abutments and investigate horizontal earth pressures behind the abutment and the backfill surface settlements. The test results showed that the translational movement of the abutment likely prevented soil ratcheting from occurring in the upper portion of the backfill due to stress adjustments in the backfill after simulated seasonal temperature changes (i.e., causing limited earth pressure increase behind the abutment). Furthermore, the existing methods used to predict the horizontal earth pressures behind the abutment overestimated the pressures at the bottom of the abutment if the translational movement of the abutment was permitted. In addition, the backfill surface settlement near the abutment, induced by the seasonal temperature changes, was a function of footing rigidity and the displacement magnitude of the abutment and continued even after 30 cycles of abutment movement.

In geothermal engineering, the long-term thermomechanical responses of sand and clay are difficult to predict due to the lack of an essential understanding of thermally induced volume change. Recent experimental studies have measured thermal creep of dry sand, glass beads, and clays. However, this universal phenomena of thermal creep has not been properly included in constitutive models of soils. In this paper, the Tsinghua ThermoSoil (TTS) model, previously developed for clay, is extended to describe thermal and mechanical creep of sand. The elemental thermal creep and thermal cyclic behavior of Todi clay and Bangkok sand, reported under hydrostatic stress or oedometric conditions, are reproduced by the proposed formulation. These comparisons show the following key features of behavior described by the extended model (e-TTS): (1) thermal creep of clay can produce either compressive or s1welling behavior depending on the consolidation stress history, (2) the heating rate and temperature range have significant influences on thermal creep, but only small effects on swelling behavior, (3) thermal cycling induces compression of sand, which is dependent on effective stress level and relative density, and (4) the e-TTS model predicts stabilization of thermal creep strains for sands over within 50–100 thermal cycles due to equalization of elastic and locked-in hysteretic strains.

Centrifuge testing data are presented to elucidate the influence of particle size and gradation on liquefaction potential and dynamic response. The physical, index, and material properties of nine test soil mixtures, sharing a common geologic origin and ranging in $D_{50}$ from 0.18 to 2.58 mm and $C_u$ from 1.53 to 9.86, were quantified and compared to the range of values exhibited by clean sands in the literature. Each centrifuge model was subjected to 15 dynamic loading events across a range of relative density and Arias intensity levels. The high permeability of the poorly graded soils prevented flow liquefaction; however, the gap and well-graded soils generated excess pore pressures similar to clean sands even though large particles were present. Despite similar pore pressure responses, the gap and well-graded test soils exhibited lower cumulative volumetric strains than the clean sand because of enhanced dilation. The tendency for the gap and well-graded soils to dilate is theorized to stem from their enhanced packing efficiency and increased shear stiffness.

Analytical solutions are presented for the active force on a retaining wall from a single wedge of unreinforced soil, as obtained from a general Coulomb-type limit equilibrium analysis. The solutions can accommodate variable wedge geometry, pore pressure, shear-strength parameters, surcharge stress, applied loads, pseudostatic seismic coefficients, and a tension crack zone. Closed-form expressions are derived for the critical angle of the base failure plane and corresponding maximum effective normal force on the wall for a triangular soil wedge. Verification checks show exact agreement with existing analytical solutions for simplified conditions, including Coulomb and Mononobe-Okabe active earth force. A numerical example is provided to demonstrate the method for a gravity retaining wall.