Journal of Geotechnical and Geoenvironmental Engineering

The influence of discrete and randomly distributed fibers (DRDF) on the improvement of the resistance to cracking of clay-based landfill covers was studied by conducting a series of centrifuge model tests. A motor-based differential settlement simulator (MDSS) was used to induce differential settlements to the soil layers during centrifuge tests performed at $40g$ using the 4.5 m radius beam centrifuge at Indian Institute of Technology Bombay. A digital image cross-correlation (DIC) technique was used to capture deformations of the model barriers and to obtain strain field distributions. Thickness of the unreinforced barriers (URSB) and fiber-reinforced barriers (FRSB) were varied as 15 mm (similar to 0.6 m in the field) and 25 mm (1.0 m). Fiber lengths were varied as 30, 60, and 90 mm. The strain at water breakthrough of 15 mm (0.6 m) and 25 mm (1 m) FRSB were respectively 1.9 and 2.2 times higher compared to identical URSBs. A relatively crack-free surface was observed for FRSB with longer fibers even after subjected to differential settlement of 1 m.

This study presents reliability analyses of soil nail walls (SNWs) against global and sliding stability failures in layered ground. The load and resistance factor design (LRFD) has been studied and advocated for SNW design over the past few years. However, the majority of those studies were exclusively focused on SNWs in single-layer ground and few have investigated the case of layered ground. In this paper, the reliability is evaluated using the first-order reliability method (FORM) for SNWs that are designed based on traditional working stress design (WSD) methods. Based on analyses of a great variety of combinations of two soils, a stiff clay layer and a dense sand layer, reliability indices calibrated from traditional WSD exceed 3.0 for global stability designs but may be lower than 3.0 for sliding stability designs. This study also reveals that the sequence and the thickness ratio of layers are two important factors for the reliability of both global and sliding stability modes. SNWs in layered ground usually have a higher reliability of global stability than SNWs in single-layer ground. The influences of model uncertainty and soil spatial variability on the reliability are also investigated. These findings will help the future development of LRFD for SNWs.

Specific surface area (SSA) is a fundamental property of soil. A new methodology for SSA determination of both expansive and nonexpansive soil based on interpretation of the soil water retention curve (SWRC) is proposed and evaluated. The SWRC-based methodology relies on the Khorshidi-Lu water sorption model, where hydration of expansive soils is interpreted to occur on three distinct surfaces and in sequential order: exchangeable cations, intercrystalline surfaces, and external particle surfaces. Hydration at these different sites is responsible for the commonly observed hysteresis in specific moisture capacity (SMC) between adsorption and desorption paths. Hydration mechanisms for nonexpansive soils, where intercrystalline surfaces do not exist, are shown to be well captured by interpreting the Brunauer-Emmett-Teller (BET) multilayer sorption theory if the effect of cation hydration on the SWRC is considered. A suite of materials ranging from silty to clayey soils is used to demonstrate the validity of the SWRC-based methodology. A comparison of SSA values obtained using the SWRC-based methodology, two existing water-vapor-based methods, and the ethylene glycol monoethyl ether (EGME) method confirms the validity and the accuracy of the SWRC-based methodology. Agreement between SSA values calculated based on the SWRC approach and the EGME method for nonexpansive soils and expansive soils as estimated by root-square-mean error (RSME) is $\text{9\hspace{0.17em}\hspace{0.17em}}{\mathrm{m}}^2/\mathrm{g}$ and $80{\mathrm{m}}^2/\mathrm{g}$, respectively. A limitation of the EGME method for the SSA determination of expansive soils is also identified.

This paper summarizes results of laboratory investigations performed to quantify the influence of diagenesis (or aging processes) on the peak shear strength and dilatancy of an uncemented Pleistocene age sand sampled near Charleston, South Carolina. Drained triaxial compression tests were performed on high-quality intact specimens retrieved using the in situ freezing and sampling method, and on remolded specimens prepared with matching densities. The stress-strain behavior of intact specimens is accompanied by dilation and a peak shear value, whereas remolded specimens generally contracted throughout shearing. An age-dilatancy term is added to a dilatancy index equation to account for the difference between intact and remolded peak friction angles. The resulting equation suggests that dilatancies due to age and density are suppressed with increasing confining pressure. A profile of peak friction angle with depth is established from the results and compared with estimates from empirical relationships.

In zoned dams, some types of materials, when located upstream of a damaged core, may impose high hydraulic losses capable of limiting the flows converging toward the flaw in the core. In this paper, the factors influencing flow limiting by an upstream material are addressed experimentally, using the flow limitation erosion test (FLET). The results of eighteen FLETs, on five coarse, broadly graded soils as upstream material, are presented. The laboratory testing shows that the flow-limiting action is strongly dependent on the grading and the compaction characteristics of the upstream soil. The critical parameters are the fines and the gravel content, the nature of the fines, and the compaction water content. Practical rules for the preliminary estimation of the likelihood of flow limiting by an upstream coarse material being effective in stopping the progression of erosion in zoned dams are suggested, based on the results of the experimental study.

The load transfer method is an efficient and practical tool for the analysis of single piles and combined systems such as pile-raft foundations and rigid inclusions. The existing transfer curves rely either on the measurement of the pressuremeter modulus, which is not widespread at the international level, or they have been developed only for specific pile and ground types. In order to develop new transfer curves valid for all pile and ground types without the need of pressuremeter tests, a database of 50 instrumented pile load tests was examined. The existing curves were first compared to the measurements at the pile shaft and at the tip. Then the parameters of the most appropriate curve types, the cubic root and the hyperbolic curve, were calibrated in order to best match the measurements. A single set of parameters for each proposed curve type is suitable enough for most pile and ground types. The good agreement of the proposed curves was confirmed by applying them to assess the overall load-settlement curve of 72 other pile load tests.

A constant-volume direct simple shear (DSS) apparatus was used to characterize the liquefaction resistance of cemented paste backfill (CPB) prepared using unclassified fine-grained tailings. The effects of curing age, cement content, and initial void ratio (also expressed through initial solids content of the mixture) on the resistance to liquefaction were investigated. Higher cement contents, longer curing periods, or higher initial solids contents were found to increase CPB liquefaction resistance in a consistent manner. The commonly used “100-kPa unconfined compressive strength (UCS) rule of thumb” for CPB liquefaction susceptibility, as originally adopted from lightly cemented sands, was found to be conservative under a large earthquake-induced cyclic stress ratio (CSR). For two different mixes, a UCS of approximately 70 kPa was found adequate to resist liquefaction under a CSR typically generated by a maximum ground acceleration of 0.3 g. For the samples tested, a UCS of 70 kPa was found to correspond to a shear wave velocity of $220\mathrm{m}/\mathrm{s}$, which may be a more convenient criterion for assessing liquefaction resistance, given the potential for in situ measurement.

Microbially induced calcite precipitation (MICP) is a sustainable biological ground improvement technique that is capable of altering and improving soil mechanical and geotechnical engineering properties. In this paper, laboratory column studies were used to examine the effects of some key environmental parameters on ureolytic MICP mediated soils, including the impact of urease concentrations, temperature, rainwater flushing, oil contamination, and freeze–thaw cycling. The results indicate that an effective crystal precipitation pattern can be obtained at low urease activity and ambient temperature, resulting in high improvement in soil unconfined compressive strength (UCS). The microstructural images of such crystals showed agglomerated large clusters filling the gaps between the soil grains, leading to effective crystals formation. The rainwater flushing was detrimental to the biocementation process. The results also indicate that traditional MICP treatment by the two-phase injection method did not succeed in treatment of oil-contaminated soils, and the proposed premixing of bioflocs with soil can significantly improve UCS and stiffness of oil-contaminated soils. Finally, MICP-treated soils showed a high durability to the freeze–thaw erosion, which is attributed to the interparticle contact points and bridging of crystals formation.

Cavity expansion theory plays an important role in many geotechnical engineering problems, including the cone penetration test (CPT). One of the challenges of interpreting CPT data is the delineation of interfaces between soil layers and the identification of distinct thin layers, a process which relies on an in-depth understanding of the relationship between penetrometer readings and soil properties. In this paper, analytical cavity expansion solutions in two concentric regions of soil are applied to the interpretation of CPT data, with a specific focus on the layered effects during penetration. The solutions provide a large-strain analysis of cavity expansion in two concentric regions for dilatant elastic-perfectly plastic material. The analysis of CPT data in two-layered soils highlights the effect of respective soil properties (strength, stiffness) on CPT measurements within the influence zones around the two-soil interface. Results show good comparisons with numerical results and elastic solutions. A simple superposition method of the two-layered analytical approach is applied to the analysis of penetration in multilayered soils. A good comparison with field data and numerical results is obtained. It is illustrated that the proposed parameters effectively capture the influence of respective soil properties in the thin-layer analysis. It is also shown that results based on this analysis have better agreement with numerical results compared with elastic solutions.

This paper presents the results of free vibration tests carried out at different load levels on a system of three near-shore steel pipe piles vibro-driven into soft marine clay. Piles are arranged in an L-shaped horizontal layout and are free at the head. The instrumentation consists of strain gauges placed at selected levels along the shaft of the loaded pile and accelerometers at the head of the receiver piles. The aim of this experiment is to analyze the dynamic soil–water–pile and pile-to-pile interaction and to investigate the development of nonlinearities at increasing load levels. The results of experimental modal analyses, in terms of natural frequencies and damping ratios of the system, are presented and the complex dynamic behavior of the vibrating soil–water–pile system and the pile-to-pile interaction are discussed. For a better reliability assessment of the system response in the range of linear behavior, the results of free vibration tests at the lowest level of the applied loads are compared with those obtained from impact load tests.

Nonlinear dynamic numerical simulations of infinite slopes are used to examine the effects of spatial variability in penetration resistances on liquefaction-induced settlement and lateral spreading for gently sloping ground and to develop guidance on selecting representative properties for uniform models. Simulation results for models having uniform properties (uniform models) are compared to results for models having spatially correlated Gaussian random field property distributions (stochastic models). Results are presented for sets of stochastic model realizations and a set of ground motions scaled to a range of peak ground accelerations. Computed responses and ground deformations (settlements and lateral displacements) for these uniform and stochastic models are used to identify representative properties (as a percentile of the stochastic distributions) for which the uniform models produce reasonable agreement with the median of the stochastic model responses. The effects of the scales of fluctuation, nonliquefiable crust layer thickness, sand layer thickness, ground motion characteristics, sand density, and other factors are evaluated. Implications for engineering practice are discussed.

In this paper, a new procedure for rapidly determining the hydraulic properties of unsaturated soils is developed. The proposed procedure is used to characterize the evolution of the instant degree of saturation in an unsaturated soil during a multistep outflow experiment. By introducing the concept of capillary relaxation, a theoretical model is developed and adopted to describe the dynamic hydraulic properties of unsaturated soils under constant-volume conditions. A series of nonequilibrium multistep outflow experiments are performed on various types of unsaturated soils. It is shown that the evolution of the instant degree of saturation or the outflow mass generally follows an exponential law. Both experimental results and theoretical simulations illustrate that flow instability and water blockage can occur in the soil during multistep outflow experiments. The theoretical predictions are compared to the experimental results, showing that the proposed procedure can be effectively used to determine the hydraulic properties of unsaturated soils. Because the outflow test is performed under nonequilibrium flow conditions, the experimental time for measuring the hydraulic properties is much shorter than those needed for conventional equilibrium procedures.

A numerical solution has been developed for large-strain consolidation incorporating non-Darcian (nonlinear) radial flow with varying compressibility and permeability coefficients. The solution can accommodate both conventional fill surcharge as well as vacuum preloading. The smear effect caused by mandrel-driven vertical drains is also captured in the analysis. The proposed model is verified by comparing it with FEM simulation, existing laboratory data, other existing theoretical solutions, and its advantage of capturing the multiple factors influencing radial drainage and consolidation is demonstrated. The effects of non-Darcian flow are found to be significant for obtaining an accurate solution, unlike numerous past solutions that are based on linear Darcy flow. The salient finding of this study is that the conventional small-strain theory can overestimate the rate of consolidation with radial drainage, especially for highly compressible soils such as estuarine clays under substantial preloading pressures. It is also found that a considerable difference (larger than 5%) between large-strain and small-strain solutions inevitably occurs once the vertical strain exceeds approximately 15%, which can be regarded as a threshold beyond which the large-strain analysis becomes increasingly important. The proposed model is applied to a case study at Ballina Bypass (NSW, Australia), where prefabricated vertical drains have been installed in soft estuarine clay subjected to a combination of fill surcharge and vacuum preloading.

Meyerhof analysis of bearing capacity for shallow foundations placed atop the crest of slopes, as standard in the literature, is applied using a set of modified factors based on footing location, soil strength, and footing size. These factors, however, are restricted to soils that are purely cohesionless or cohesive. Accounting for soils with both $c^{\prime }$ and ${\varphi }^{\prime }$ is important when constructing foundations in proximity or adjacent to slopes comprised of native soils. This technical note uses an upper-bound limit state plasticity failure discretization scheme, known as Discontinuity Layout Optimization, to present revised bearing capacity coefficients for foundations adjacent to frictional-cohesive slopes by presenting revised factors for footings not adjacent, but in proximity of slopes. A set of reduction coefficients that can be directly applied to the classical bearing capacity formulation is presented for ease of application.

Modulus reduction and damping values are commonly plotted against cyclic shear strain amplitude (${\gamma }_c$), and the resulting curves are known to depend on mean effective stress ($p^{\prime }$), plasticity characteristics, strain rate, and number of loading cycles. The dependence on $p^{\prime }$ is potentially problematic for undrained effective stress analysis where excess pore pressure may develop during loading. This paper presents a new concept in which normalized modulus reduction ($G/G_{\text{max}}$) and damping ($D$) values are plotted against stress ratio ($\eta$) rather than ${\gamma }_c$. Relations developed for sand, clay, and peat are found to be essentially pressure-independent when $G/G_{\text{max}}$ and $D-D_{\text{min}}$ are plotted versus $\eta$,whereas all three are pressure-dependent when plotted versus ${\gamma }_c$. This finding is potentially useful for undrained effective stress analysis where $p^{\prime }$ may change during loading, and provides a new approach for interpreting laboratory tests in future development of $G/G_{\text{max}}$ and $D$ curves.