Journal of Geotechnical and Geoenvironmental Engineering

The objectives of this study were to evaluate the effect of binder amendment on the unconfined compression strength (UCS) of mine tailings and assess applicability of using the amended materials in earthworks. Natural and synthetic (i.e., laboratory prepared) mine tailings were used to assess the effects of tailings particle size and tailings solids content on UCS. Two types of off-specification fly ashes and Types I and II portland cement were used as cementitious binders. Tailings and fly ash mixtures were prepared at 70, 80, and 90% solids content and amended with 10 or 20% binder. Unconfined compression strength tests were conducted on specimens cured for 7 days. In general, UCS increased with an increase in tailings particle size, solids content, and/or increase in ${\mathrm{CaO}\text{-}\mathrm{to}\text{-}\mathrm{SiO}}_2$ ratio of fly ash for amended tailings specimens. A multivariate regression model was developed to predict UCS of tailings amended with fly ash as a function of (1) tailings water content, (2) water-to-binder ratio, and (3) ${\mathrm{CaO}\text{-}\mathrm{to}\text{-}\mathrm{SiO}}_2$ ratio of fly ash. The model was developed using data from synthetic tailings specimens and validated via tests on natural tailings and data compiled from literature. The model is applicable for estimating the UCS of candidate fly ash amended hard rock mine tailings and low plasticity soils (i.e., liquid $\mathrm{limit}<50$). The UCS of high plasticity materials was underpredicted with the model.

A mass-conservative formulation of one-dimensional (1D) coupled large-strain consolidation and unsaturated flow using a piecewise-linear formulation is developed. The volume and water-content changes in the unsaturated zone are described using three-dimensional (3D) constitutive surfaces of void ratio and gravimetric water content as functions of net stress and matric suction, whereas the volume change in the saturated zone is described using a two-dimensional (2D) curve relating void ratio to Terzaghi effective stress. The hydraulic conductivity varies with both void ratio and suction in the unsaturated zone, and with void ratio only in the saturated zone. The associated nonlinearities are addressed by solving the flow-continuity equation for water mass conservation through a finite-difference scheme in a piecewise-linear manner. The unknowns are calculated from variables in the previous time step without requiring iteration. A novel algorithm is developed to smoothly model transition between the saturated and unsaturated zones. Four types of top boundary conditions are formulated to account for complex soil–atmosphere hydraulic interactions. The developed program is also capable of performing a quasi-unsaturated analysis, which uses the saturated soil’s constitutive relationships in the unsaturated zone, similar to the approach of several models currently used in practice. The theoretical and numerical development of the model is described in this paper, and is followed by verifications of the program’s implementation using a numerical example of saturated large-strain consolidation and an analytical solution to unsaturated flow. Validation of the quasi-unsaturated and unsaturated formulations is presented in the companion paper, using comparisons with numerical cases as well as laboratory and field data.

A piecewise-linear formulation for simulating coupled large-strain consolidation and unsaturated flow using a mass-conservative and noniterative solution was developed in a companion paper and implemented in a proposed program called UNSAT-CON. The formulation is tested in this paper using numerical cases as well as laboratory and field data. First, UNSAT-CON is tested by comparing model predictions to published predictions of a finite-element-based model that uses a quasi-unsaturated formulation to model evaporation: the same quasi-unsaturated formulation is implemented in UNSAT-CON to allow direct comparison. Excellent agreement of settlement over time for various levels of evaporation are obtained. UNSAT-CON reliably predicts the hydraulic exchange at the soil–atmosphere interface under the conditions of evaporation, decantation, or their combinations. The fully unsaturated formulation is tested on fine-grained tailings by calibration with a documented laboratory column test and verification with a 4.1-m-deep field trial. This testing exercise showed (1) three-dimensional (3D) constitutive surfaces are largely constrained when defined using conventional test data such as the saturated compressibility function and soil water characteristic curve; (2) good to excellent agreement between measurements and predictions are observed, including the developing suction, decreasing void ratio, increasing solid content, and increasing settlement rate due to evaporation effect on the near surface zone; and (3) modeling desaturation allows for a more-realistic simulation of deformation in the unsaturated zone than using formulations that do not model desaturation.

In Singapore, sudden changes of bedrock depth and highly variable weathering degrees in residual soils are commonly observed. This study employed the dispersive characteristics of Rayleigh-type surface waves from combined active-source and passive-source surface-wave techniques to (1) generate S-wave velocity ($V_s$) profiles of residual soils and (2) estimate spatial variations in the weathering degree and bedrock depth of residual soils in Singapore, ranging from completely weathered soil to weathered rock. Finally, the feasibility of detecting bedrock depth in Bukit Timah granite was examined by several approaches based on (1) preselected $V_s$, (2) normalized Rayleigh wave phase velocity ($V_R$), and (3) average $V_s$. The results of the field experiments showed that this combined noninvasive multichannel analysis of surface waves (MASW) method can be successfully used to determine the $V_s$ profile of residual soils, which includes approximate bedrock detection. The newly proposed average $V_s$-based approach using the ratios of average $V_s$ values of top soil layers (10–40 m) especially appears to be a simple, automatic, objective, and standardized method to identify depth to bedrock (within about 5-m resolution), which is confirmed by conventional boring investigations.

Deep excavations inevitably induce stress relief and soil movement, which may affect the safety and serviceability of nearby existing pile foundations. Although the effects of excavation on adjacent piles have been studied, excavation-induced pile settlement and change of shaft resistance deserve more investigation. In this paper, centrifuge simulation of the effects of multipropped excavation in-flight on vertically loaded floating piles in sand is reported. Piles were restrained in terms of rotation and lateral deflection at or above ground-surface level in three different tests. In addition, three-dimensional numerical analyses using an advanced soil model were carried out to back analyze the centrifuge tests. A pile-raft settled a similar amount due to excavation as a free-head pile and an elevated pile group. During excavation, a downward load-transfer mechanism in the piles can be identified. Shaft resistance in the upper part of the pile decreases due to the downward soil movement and the reduction of normal stress acting on the pile. In order to maintain vertical equilibrium, the pile has to settle to mobilize higher shaft resistance in the lower part of the pile and toe resistance. Lateral restraints imposed on the pile head have a significant influence on induced pile bending moment. Induced bending moment due to excavation can exceed the pile bending capacity.

Results of finite-element analyses calculating the undrained capacities of skirted circular foundations under uniaxial vertical, horizontal, and moment loading are presented. Parallel finite-element analyses using both the Tresca failure criterion and the modified cam clay (MCC) model are reported. The variations in capacity are presented as functions of embedment and soil-strength heterogeneity as a series of graphs. Closed-form formulas of uniaxial capacities are also provided. To account for the complexity in the results, a novel method of separating the contributions of soil above and beneath the skirt tip and for the constant and linearly varying components of the undrained shear-strength profile is outlined. This method can be used to estimate the undrained ultimate uniaxial capacity factors of skirted circular foundations in clay for any given embedment ratio and strength heterogeneity. Its versatility is also shown through validation against previously published embedded strip footing data. The results provide both equations for engineers to use in design and a database that can be used in future research to calculate and express the consolidated undrained capacity under combined loading.

The ultrarapid underpass (URUP) tunneling method, in which a shield machine is launched and received at the ground surface level, has been developed for tunneling in congested areas. This paper reports monitoring results of a full-scale test conducted for the first URUP tunneling project in China. In the investigation, special attention was paid to tunneling phases in which the shield machine was buried super-shallowly or partially above the ground surface. The paper focuses on the tunneling-induced soil movements and on the grouting process. Measurement results indicated that surface settlements can be represented by the commonly used Gaussian function regardless of the cover depth, and that the load-factor approach provides a fair prediction of the volume loss. A contracting transverse soil-displacement field was identified as the dominating pattern, although an expanding displacement field was observed at several locations. Pressure measurements were used to identify a time-dependent response of the grouting process.

Groups of in-line piles connected by beams are being used to contain errant vehicles for roadside safety and perimeter protection of sensitive infrastructure. The effectiveness of in-line groups of piles to safely redirect or stop approaching vehicles has not been well investigated from the geotechnical point of view. In this paper, the performance of barriers is examined through full-scale crash tests and numerical simulations. The full-scale crash tests consisted of two different configurations of piles and beams as well as two different soil types: one was in loose sand and the other in hard clay. Both barriers were subjected to vehicle impact: one by a medium-duty truck and the other by a pickup truck. The barriers successfully contained the impacting vehicles. Detailed finite-element models of the barriers and the soil were developed and combined with the vehicle models to simulate the dynamic events. Comparison between predicted and measured behavior was used to calibrate the models. Additional simulations were performed to evaluate the effect of major structural and geotechnical design parameters including soil strength, pile spacing, and number of piles on the impact response of these systems. Practical recommendations are drawn from the experimental and numerical work.

Suffusion is a selective erosion of fine particles under the effect of seepage flow within the matrix of coarser particles. This complex phenomenon appears as a combination of three processes: detachment, transport, and possible filtration of finer fraction. It can induce a change in particle size distribution, porosity, and hydraulic conductivity of the material. With the objective to characterize suffusion susceptibility, downward seepage flow tests were conducted. Four different cohesionless soils were tested under hydraulic-gradient controlled conditions or under flow-rate controlled conditions. This study shows the significant effect of hydraulic loading history on the value of critical hydraulic gradient. Moreover, the method characterizing the erosion susceptibility based on rate of erosion does not lead to a unique characterization of suffusion process for different types of hydraulic loading. The new analysis is based on energy expended by the seepage flow and the cumulative eroded dry mass. The results demonstrate that this approach is more effective to characterize suffusion susceptibility for cohesionless soils.

Large diameter, rigid monopile foundations have been extensively used in the fast-growing offshore wind energy industry over the last two decades. In view of the offshore environment, lateral response of the monopile usually governs its design. Even though several approaches have been recommended based on small-scale laboratory tests, there is no widely accepted method for the design of monopiles under lateral loading. Conversely, field testing on large-diameter prototype monopiles is normally impractical due to the high demand on the capacity of loading equipment. For these reasons a series of field lateral loading tests on reduced-scale monopiles were conducted at a dense sand test bed site. The model monopiles had similar aspect ratios of pile length to diameter to those used in the offshore wind farm projects, but were smaller in scale. Experimental $p\text{-}y$ curves were derived using the measured monopile responses of the lateral load tests, and a distinctive failure model was presented for monopiles in the overconsolidated dense sand deposit. Comparison of lateral responses of monopile between the measured and predicted by two current $p\text{-}y$ models showed the shear force at the pile tip plays an important role and should be accounted for in the design of laterally loaded rigid monopiles. Finally, a refined $p\text{-}y$ model for laterally loaded rigid monopiles in overconsolidated dense sand was recommended and calibrated.

Four centrifuge tests supplemented by calibrated numerical simulations were performed to evaluate passive wedge formation and limiting lateral pressures imposed on a large, stiff foundation during liquefaction-induced lateral spreading. In the centrifuge tests lateral pressures on the upslope and downslope faces of the foundation were measured using tactile pressure sensors. Lateral pressure distributions also were extracted from the simulations. Available methods to estimate the limiting lateral pressure distributions, including the Rankine, Broms, Japanese Road Association, and strain wedge methods were compared to the measured and simulated pressures. However, existing models did not capture the observed pressure distributions even when the size and shape of upslope passive wedge was defined directly from surface and subsurface displacement patterns in the centrifuge tests. A new semiempirical method is proposed to estimate limiting lateral pressures that accounts for the three-dimensional shape of the upslope passive wedge. This method produces an upslope limiting pressure distribution consistent with centrifuge-measured pressure distributions.

The potential for enhanced containment of two metals, potassium (K) and zinc (Zn), by soil-bentonite (SB) vertical cutoff walls comprising zeolite-amended backfills was evaluated on the basis of numerical transport simulations and previously measured adsorption data that exhibited nonlinear behavior over the range of concentrations of interest. The results indicate that the containment duration as reflected by the barrier flux breakthrough time increased for K with decreasing source concentration ($C_o$), increasing content of zeolite amendment, and increasing adsorption capacity of the backfill. The results for Zn are similar to those for K except at lower $C_o$ (i.e., 100 and $\mathrm{1,000}\mathrm{mg}/\mathrm{L}$), where better performance occurred with the unamended backfill relative to that for the zeolite-amended backfills. This ostensibly counterintuitive result for Zn was attributed to the geochemical conditions existing in the adsorption tests, whereby adsorption of Zn was suppressed due to unfavorable competition with elevated concentrations of redissolved metals, primarily ${\mathrm{Na}}^{+}$, resulting from the zeolite amendment. Overall, the results of the study indicate that containment of metals may be enhanced from years to a century or more with zeolite-amended SB cutoff walls, but the magnitude of any enhanced containment is dependent on both the adsorption capacity and the adsorption behavior of the specific metal with the specific backfill.

Numerical simulations and calibration of hypoplasticity constitutive parameters for Chicago clays are presented based on laboratory tests conducted on high-quality block samples and field tests from excavations located in the Chicago, Illinois area. The parameters for the hypoplasticity model enhanced with the intergranular strain concept are calibrated from the results of index tests, oedometer tests, and $K_0$-consolidated undrained triaxial compression and extension tests. Drained stress probes conducted in a triaxial cell after three different preshearing stress paths were used to compare the numerical results with the triaxial test results. One path was applied to study the clay stress-strain behavior at in situ conditions and the remaining two to isolate the effects of recent stress history. The results are shown in terms of the secant shear and bulk stiffness, shear and volumetric stress-strain responses at small and large strains, and stress path reversals.

In a spirit of adventure, a boring was made and soil moisture contents measured along a tension crack near the crest of an earth embankment. Moisture contents increased linearly with depth and then sharply decreased at what is interpreted to be the bottom of the crack. Desiccation does not appear to have been a factor because of construction delays due to rains as the embankment still was under construction. It is hypothesized that the crack was created by quasi-elastic rebound of the recently compacted soil directed normal to the intersecting ground surfaces. A second increase in moisture content that appears to be linear with depth was discovered as the boring had been extended all of the way through to the bottom of the embankment. This change in moisture contents suggests a possible basal crack that can be attributed to differential settlement, and is assumed to connect to the upper crack as it was encountered in the same boring. Hydraulic connection may have been impaired by pinching action related to differential settlement but still could allow some water to go through. A comparison of the two mechanisms suggests that the upper crack probably formed first. A basal crack that remains open should be relevant for earth dams.