International Journal of Geomechanics

A three-dimensional numerical analysis was conducted to evaluate the performance of mechanically stabilized earth (MSE) walls with free-headed piles installed within the geosynthetic-reinforced zone subjected to lateral loading. The numerical models were first verified with the available field test results of laterally loaded piles within MSE walls. A parametric study was conducted to investigate the influence factors on both the MSE walls and the piles, including the location of piles behind the MSE wall facing, the diameter of piles, the stiffness of reinforcement layers, and the reinforcement length ratio. Load–displacement curves, wall facing displacements, and vertical profiles of lateral earth pressure distributions behind the wall facing are discussed for each influence factor at the ultimate lateral load of the pile. Numerical results show that the ultimate lateral load capacity of piles linearly decreased with the decreased pile offset distance behind the wall facing from four times to two times the pile diameter. The lateral earth pressure distribution behind the wall facing induced by the loaded pile was nonlinear with the maximum lateral pressure located within the upper part of the wall. Tensile stiffness of reinforcement layers had a significant effect on increasing pile lateral load capacities and reducing wall facing displacements. The reinforcement length ratio had limited effects on the pile and wall performance if this ratio was greater than 1.0.

The loss of effective stress due to increasing excess pore pressure that results in the upward migration of soil particles, that is, subgrade fluidization and mud pumping, has been a critical issue for railways over many years. Traditional methods such as experimental and analytical approaches can capture macroscopic quantities such as the hydraulic conductivity and critical hydraulic gradient, but they have many limitations when microscopic and localized behavior must be captured. This paper, therefore, presents a novel numerical approach where the microscopic properties of fluid and particles can be better captured when the soil is subjected to an increasing hydraulic gradient. While particle behavior is simulated using the discrete element method (DEM), the fluid dynamics can be described in greater detail using the lattice Boltzmann method (LBM). The mutual LBM-DEM interaction is carried out, so the particle and fluid variables are constantly updated. To validate this numerical method, laboratory testing on a selected subgrade soil is conducted. The results show that the numerical method can reasonably predict the coupled hydraulic and soil fluidization aspects, in relation to the experimental data. Microscopic properties such as the interstitial fluid flowing through the porous spaces of the soil are also captured well by the proposed fluid-particle coupling approach.

Soil-shrinkage curve (SSC) is an intrinsic constitutive relationship characterizing soil deformation due to variation of water content under drying. The determination of SSC has been challenging due to the difficulty of accurate measurement of volume change and the access of a broad range of water content. In this study, SSCs of three different types of soils are measured successfully by image process on photos of soil cake samples under multiple steps of drying in a humidity-controlled environmental chamber. The development of volumetric, radial, and vertical strain, and the geometry factor of the deformation are analyzed to validate homogeneous shrinkage during the entire drying process. The soil deformation is analyzed on the effect of transient drying and its dependency on water content. The sensitivity of mechanical and hydraulic properties of soil responding to time concludes that the equilibrium is not a necessity for the mechanical property measurement such as SSC. This feature implies that a transient drying process can be implemented to significantly shorten the time duration for measurement of soil shrinkage without an equilibrium condition in soil desiccation.

As most previous studies have neglected the positive influence of salinity (osmotic suction) on most coastal soils in Australia, the design of transport infrastructure involving these soils has often been overly conservative. In this study, a laboratory approach based on direct shear testing was explained to determine the stress–strain behavior of compacted coastal silty clay (CL) at different levels of osmotic suction generated by various salinity (NaCl) concentrations. A broad data set for a total of 147 direct shear tests conducted on remolded and recompacted test specimens at seven different initial matric suction conditions was analyzed to develop a semiempirical model that captures the effect of osmotic suction on the soil shear strength. The results suggested that the greater the initial matric suction, the more pronounced the role of osmotic suction. The proposed semiempirical model was governed by an electrical conductivity relationship with the osmotic suction generated by soil salinity. A new parameter χ₂ was introduced to quantify the role of soil salinity in the apparent soil shear strength corresponding to different levels of osmotic suction. When this novel relationship was coupled with the conventional matric suction theory, the overall unsaturated shear strength of a saline soil could be properly evaluated, as proven by the close proximity of the predictions to the measurements.

Particle breakage is a typical characteristic of crushable granular soil under high pressure, which has great effects on its stress–strain behaviors. The phenomenon of the critical state line (CSL) shifting downward in the compression plane caused by particle breakage was depicted by a breakage-dependent critical state plane (BCSP). Particle breakage was incorporated into a void ratio–pressure state parameter to modify Rowe’s stress–dilatancy equation, and then, the state parameter was incorporated into the bounding stress ratio and plastic modulus. Due to the impact of high pressure on particle breakage, the pressure-dependent plastic modulus parameters were introduced. A breakage-dependent bounding surface plasticity model was proposed to capture the influence of particle breakage on the state-dependent stress–strain behaviors for silica and coral sands, and the transition of complex breakage-dependent critical states resulted from the competition between the contraction due to particle breakage and the dilatancy due to particle rearrangement.

Methane hydrates are an attractive source of future clean energy and are abundant in voids of sediments, in permafrost regions, and along continental slopes. Understanding the mechanical behavior of methane hydrate-bearing sediment (MHBS) is essential for commercial gas production from hydrate reservoirs. In this paper, the effects of the intermediate principal stress are studied on the microscopic and macroscopic mechanical behavior of MHBS-filled porous media. Using discrete element method (DEM) simulations, methane hydrates are modeled as cemented agglomerates filling the pores of soil at a specific hydration level. The simulated sample is sheared under strain-controlled conditions to different values of the intermediate principal stress ratio b. The results suggest evolution rules of the principal stresses that depend on the parameter b. The friction angle increases with b, reaching a peak value at a certain value, after which it decreases. Monitoring local variables, including principal strong fabric, contact rose diagram, coordination number, and damage parameter, the effects on the macroscopic bulk behavior are demonstrated. In particular, the changes in the fabric tensor are strongly correlated with response to changes in the principal stresses, the parameter b, and the strain tensor.

Dynamic tensile mechanical characteristics of coal-rock mass under the freezing effect are an important subject that is supposed to be studied during the blasting of open-pit coal mines in Northwest China. In this study, the split Hopkinson pressure bar (SHPB) test system was used to perform dynamic Brazilian disc tests on saturated sandstone at normal temperature (25°C) and negative temperatures (−5°C, −10°C, −20°C, and −30°C). The high-speed failure process, macrofracture roughness and mesofracture morphologies of the samples were observed at different temperatures. The experimental results showed that the dynamic tensile strength of sandstone first increased and then decreased with the decrease of freezing temperature. In addition, a statistical model of the variation of sandstone tensile strength under the effect of temperature and loading rate was established. At different temperatures, all samples underwent the process of crack initiation from the center and then the main cracks penetrated the sample. At room temperature, a large number of secondary cracks formed at both ends of the samples and gradually merged with main cracks that do not occur at low temperatures. The fracture roughness of the sample at normal temperature was significantly greater than that at freezing temperatures, and there were no obvious rules of the roughness of fracture surfaces at different freezing temperatures. In the end, the change in the macroroughness of the fracture surface was explained by the brittle–ductile transition of the mesofracture morphologies of the sandstone samples at different temperatures.

In addition to various factors, the magnitude of lateral earth pressure depends on the direction and amount of the wall rotation. The current literature presents several expressions to calculate lateral earth pressure that captures the soil geometry and load characteristics. However, to the best of the author’s knowledge, there is no practical expression in the literature, in which the amount of wall rotation is incorporated. In this study, using the results of finite element (FE) models, one formula for the coefficient of active and four formulas for passive earth pressure will be presented. In each of the four passive pressure formulas, a different perspective will be adopted. Especially in two of the passive earth pressure formulas, the amount of wall rotation and soil modulus will be included as input parameters. For validation purposes, a physical retaining wall test will be modeled in the FE program and, subsequently, the test and model results will be compared.

Drainage techniques can be employed to reduce damage of pile rafts in liquefiable ground by decreasing the build-up of excess pore water pressure and maintaining soil stiffness during earthquakes. This paper aims to evaluate the impact of drainage wells on the seismic response of floating piled rafts in fully saturated ground. In practice, floating piles are commonly employed as underpinning elements beneath the raft foundations to reduce undesirable settlement of the raft. Centrifuge experiments were performed on 2 × 2 piled raft foundations with and without drainage wells in mildly sloping ground. Two types of ground involving a uniform sand layer and a uniform sand layer with a superficial gravel layer were reconstructed, whereas the piles were floating within the liquefiable sand. The models were fully instrumented and sine waves were introduced to the models' base at the 80 g centrifugal acceleration. Records of the transducers demonstrate that the drainage wells system significantly decrease the settlement of foundation, dissipation time under dynamic loading by forming closer drainage routes, and excess pore water pressure ratio. In contrary, presence of drainage systems significantly increases the rafts acceleration response, and the bending moment and the axial force of the piles. Results of the piled raft systems in the presence and absence of the drainage wells and the superficial gravel layer are compared with together and also compared with the far-field in terms of the acceleration response, permanent settlement, and excess pore water pressure.

Offshore wind energy is an attractive alternative in pursuing the nation's clean energy goals due to the significant demand for electricity in the coastal areas of the United States. Locating sites further offshore in deeper water can provide stronger, more consistent wind power resources and can mitigate aesthetic concerns. This motivates a need for improvements in the floating offshore wind turbine (FOWT) technology. As foundation costs comprise a significant fraction of the total cost for offshore wind power development, reducing the cost of the mooring system can play a significant role in making floating offshore wind economically competitive. Previous studies led to the development of a novel, efficient multiline ring anchor (MRA) system that can provide significant capital cost savings. Preliminary research shows that the MRA has a clear advantage under lateral loading by attaching wing plates to the cylindrical core of the anchor. In this study, two-dimensional finite-element (2D FE) analyses were performed to understand how wing plates affect the MRA performance under horizontal loading and provide reliable estimates of the ultimate load capacity. The results show the collapse mechanisms and bearing factors can be affected by width, the total number of wing plates, and load angles. This study also presents plastic limit analysis (PLA), based on the upper bound solution, to validate the 2D FE results by comparison and to confirm whether the postulated collapse mechanism was correct. The results obtained in the current study indicated that PLA can be a benchmark solution to evaluate the ultimate load capacity of the MRA with a satisfactory agreement with the FE-computed values.

When fine sand is treated with microbial-induced calcite precipitation (MICP), the connections of loose sand particles are reinforced by being filled with crystals of calcium carbonate (CC) caused by the presence of microorganisms. According to the research of scholars, in the process of MICP, the common anhydrous CC crystals are calcite, vaterite, and aragonite. To explore the factors that influence the penetration resistance (PR) of a biomineralization crust (BMC) and to consider the influence of sand particles and CC crystal morphology on strength, this paper reconstructs the sand particles and CC crystals by means of three-dimensional reconstruction from photographic images. Two numerical models based on a particle flow code were established using the discrete element method and were used to investigate separately (1) the unconfined compressive strength (UCS) of a biomineralized sand column; and (2) the PR of a BMC. The research shows that the UCS of an MICP sand column of calcite crystal form is 4.9 MPa, which is 2.39 times that of one of vaterite forms. It is found that the thickness of the BMC and its CaCO₃ content has a great influence on the PR of the BMC. The PR shows an increasing trend due to the increase of CC content when the penetration speed is 1.5 m/s and the coating thickness of the coating layer is 2.5 cm. The PR test value at 2.5 cm in the soil layer increases from 52.8 to 103.9 kPa when the penetration speed is 1.5 m/s, the CaCO₃ content is 12%, and the thickness of the BMC is increased from 2.5 to 4.5 cm. It is concluded that because the thickness of the BMC and the CC content increases, the PR of the test points in the soil also increases. The results of this study complement the test results of a field-scale experiment and have important value for the large-scale application of MICP technology to desert sites for desertification control.

The stability of surrounding rock for large or super-large section chambers is extremely hard to control in deep coal mines, as well as in close-distance chambers. Based on the on-site geological conditions of the super-large-section chamber group of a coal gangues separation system in Longgu Coal Mine, Heze City, Shandong, China, this paper will establish a FLAC3D numerical simulation model to research the failure evolution of the surrounding rock under different chamber spacings and thicknesses of the loosening circle. The results show that the interaction between chambers increased with the decrease of chamber spacing and the increase in the loosening circle thickness, the critical spacing (L₂) range was 15–10 m, and the critical thickness of the loosening circle was 4–5 m. Based on this, the instability mechanical models of the surrounding rock that is controlled by chamber spacing and loosening circle will be established, respectively. Then, L₂ and loosening circle radius without chain instability (R*^a) will be obtained theoretically. Finally, the theoretical verification will be realized by on-site calculations. This research revealed the chain instability mechanism of the chamber surrounding rock and provided theoretical guidance and reference for the close-distance parallel chamber with a super-large section design under deep mining conditions.

This paper presents a study on the face stability of earth pressure balance (EPB) shield tunnels in dry granular soils using the discrete element model (DEM) method to replicate realistic excavation processes of the EPB machine. Analyses were conducted at different buried depths to determine the limit face pressures and the face failure model induced by the EPB shield machine with varying open ratios and rotation speeds of the cutterhead. The failure kinematics of the soil and the soil arching effect were also determined. The results showed that both the open ratio and the rotation speed of the cutterhead have significant effects on the stability of the face in terms of the failure kinematics, limit face pressures, and the soil arching effect. The failure zone was highly dependent on the open ratio, but it depended less on the rotation speed of the cutterhead. The stability of the face is overestimated when the excavation is conducted with a cutterhead that has a large open ratio, but it is underestimated when the excavation is conducted with a cutterhead that has a smaller open ratio. Ignoring the effect of the configuration of the cutterhead and the rotation speed results in unexpected and unrealistic results.

Soil arching is relevant in various engineering applications particularly in cases where there is relative movement between soil and structure. When a portion of the soil mass yields, redistribution of stresses take place between the yielding mass and its surrounding. Several trapdoor experiments have been performed with different improvements and extensions over the past few decades to study soil arching. Numerical modeling of a trapdoor mechanism is performed in this study and the suitability of Hardening Soil constitutive model over the Mohr-Coulomb model has been studied. In the present study, extensive parametric studies were conducted to examine the extent of the zone of influence and zone of yielding. The zone of influence, beyond which there is no influence of soil arching, is bounded by straight lines to form a trapezoidal shape. The dimensions of the influence zone depend on the fill height (H) and trapdoor width (B). The two families of failure surfaces, internal and external, are described and their categorization with respect to the H/B ratio is well explained. Furthermore, the zone of yielding and the region enclosed within limits of influence and failure planes are classified as active and passive zones, considering the dilation and compression in the direction of trapdoor displacement. The equilibrium of stress transfer in soil arching that remained as an unjustified assumption until now is systematically brought out in the present study. The present study also highlights salient characteristics of the plane of equivalent settlement and critical height. The present study throws some light on the need for further understanding of arching in various field applications, such as piled embankments, tunnels, culverts, buried pipelines, excavations supported by contiguous pile walls, and slopes stabilized by a row of piles.

To reflect the contributions of different-ordered asperities to the peak shear strength, a directional uphill projection area parameter was proposed to estimate the joint roughness coefficient (JRC) for two-dimensional joint roughness profiles. Using Barton's 10 reference profiles, a series of 2D JRC estimation formulae were established by weighting the two-ordered uphill projection area parameters at different sampling intervals (SI) to consider the effects of both shear resistance and shear dilatancy. Next, the 2D uphill projection area parameter was extended to a 3D uphill projection volume parameter, which shows good agreement with the worn areas of rock joints. Based on the shear tests of 36 square sandstone joints, a new 3D JRC estimation formula was proposed, and this was successfully applied to the JRC estimation of 29 rock joints with irregular profiles. Comparison of three different calculation methods of joint length indicates that equivalent length is more suitable to estimate the JRC of irregular rock joints. This proposed method can also reflect the anisotropic characterization of a rock joint.

Reinforcement of disturbed or discontinuous rock strata by installing reinforcement bars (bolts) has always been the primary measure for stabilizing the ground. Bolts are generally installed perpendicular to a discontinuity plane to obtain the maximum resistance against shearing. However, in practice, rock bolts are installed at any suitable orientation with respect to the shearing direction. In such cases, few examples exist in the literature to quantify the shear strength of the reinforced joints as well as the efficacy of the installed bolts. This paper presents an extensive experimental work to establish the relationships between shear resistance of the reinforced joint surface and its deformation behavior with the applied normal stress and bolt orientation. Concrete jointed samples of 10 cm cube size are casted and a reinforcement bar is inserted with cement grout at an angle of −30°, 0°, and +30° with respect to vertical axis. Five different normal loads are applied with three repetitions in each case. Altogether 45 direct shear tests are conducted at a constant strain rate of 0.2 mm/min and results are obtained in terms of shear stress versus shear displacement. The results suggest improved frictional behavior of the joint with reinforcement and also confirm that normal orientation of the bolt provides the best performance in terms of peak shear strength and post-yielding behavior. A closer look into the bending of the bolt rod reveals that at certain orientations, the bolt is subjected to shear loading, behaving as a Timoshenko beam, and it provides higher resistance to shear movement. In some other orientations, the bolt rod behaves like an Euler-Bernoulli beam and, hence, shear resistance is low. A procedure to estimate bolt performance for different orientations based on normal stress and bolt orientation is presented in this paper. The results provide qualitative and quantitative insights as to why the normal orientation bolts perform better. The study also demonstrates that positive orientation angle with respect to the shearing direction can also be the second choice if normal bolts cannot be installed.

The distribution of stresses within individual granular particles subjected to one-dimensional (1D) confined compression is affected by various particle-scale properties, such as morphology, kinematics, position, crystal structure, and the interaction of the particles (i.e., the local contact network). A thorough examination of the effects of particle-scale properties on the variation of stresses within individual particles and the fracture of particles in granular materials is essential for the theoretical analysis and modeling of the engineering behavior of granular materials. In this study, three-dimensional (3D) synchrotron microcomputed tomography (SMT) and 3D X-ray diffraction microscopy (3DXRD) that are nondestructive techniques will be concurrently used to quantify the variation of internal particle stresses, as well as the particle-scale properties of a specimen composed of Ottawa sand that will be subjected to a 1D confined compression. The effects of the particle properties on the variation of the particles internal stresses and the fracture of particles will be studied in detail. The importance of the particle-scale properties based on their effect on the variation of particle stress will be ranked using a feature selection algorithm called RReliefF. The significant particle properties (kinematics, contact number, the relative orientation of the local contact network for the loading direction, and position) by the feature selection algorithm will be investigated further to uncover their influence on the internal particle stresses. In addition, the effects of the particle-scale properties on the fracture of the particles will be examined. The fracture of the sand particles proved to be a complex phenomenon that was controlled by particle-scale properties and by the interaction between the particles. The relative orientation of the fracture surface in the fractured particles was examined to assess the effect particle shape and local contact network of the particles on the fracture of the particles.

This paper proposes an analytical method based on drained solutions of cavity expansion and contraction in a unified clay and sand model to investigate tunnel–soil–pile interactions. Cavity expansion analyses are used to evaluate the effects of pile installation on ground stresses and to determine pile end-bearing capacity and the distribution of shaft friction. Cavity contraction methods were adopted to replicate the tunnel convergence–confinement response using the singularity and image method for ground loss and ovalization of a shallow tunnel in a semi-infinite medium. A 2D model was developed that evaluates changes in mean stress and specific volume during pile installation and tunnel excavation. Outcomes from the developed analytical approach are compared against data from centrifuge tests in silica sand; results demonstrate that trends in pile load capacity degradation, mobilized safety factor, and tunneling-induced pile settlement can be satisfactorily predicted for the case of a tunnel excavated beneath a pile with a constant service load. Criteria based on pile capacity, safety factor, and settlement are proposed that can be used to determine a critical tunnel volume loss or evaluate pile safety level. The paper contributes to the understanding of tunnel–soil–structure interaction mechanisms and provides an efficient means of conducting a preliminary risk assessment of tunnel–pile interaction.

Silt is widely distributed in North China Plain, especially in the Yellow River flood basin. Unfortunately, untreated silt cannot be directly used as a subgrade filler. Therefore, it is of great practical significance to explore the reinforcement of silt in the Yellow River flood basin. To solve this problem, the dynamic characteristics of silt added with different proportions of lignin were studied. The correlation among lignin content, curing time, dynamic shear modulus, and damping ratio of the improved soil was tested. The results showed that the dynamic strength and shear modulus can be effectively improved by adding lignin. The optimal lignin content for the dynamic strength and shear modulus was 5%. To understand the microscopic mechanism of lignin-treated silt, scanning electron microscopy (SEM) was performed on lignin samples. The SEM results showed that with an increase in lignin content, a certain amount of cementitious materials was formed in the soil. These cementitious materials wrap up the soil particles and connect the fine soil particles to form larger aggregates. The research results can provide a reference to further improve the physical and mechanical properties of the silt in the Yellow River flood basin, and this study is instructive.

The catastrophic failure of coal under super-critical stresses and mining-induced disturbance is becoming one of major safety risks of underground mining. Research of the failure pattern of coal under impact load is helpful for understanding its burst behavior in order to mitigate the burst risk. To investigate the fragmentation characteristic and burst behavior of coal under impact load, drop weight tests of coal samples were conducted. This paper found that coal samples have high peak stress, pulverized fragmentation, and intensive burst energy under impact load. The fragments induced by the impact load have a relatively consistent distribution mode, which can be characterized by a fractal model. The burst energy accounts for more than 99% of the impact energy input while fragmentation energy only accounts for no more than 1%.

Dynamic soil properties curves (DSPCs) [that consist of modulus reduction (G/G_max) and damping ratio (β) curves] are important input parameters for ground response analysis (GRA). However, due to the nonavailability of regional DSPCs, in most cases, regional GRA is carried out based on globally established DSPCs. It is noted that the issue of spatial variability in the previously established DSPCs has been raised in numerous studies. In addition, for most of the cases where GRA is based on previously established DSPC’s from other regions, whether their outcome for surface seismic hazard applies to the region under consideration is debatable. Although the determination of regional DSPCs based on laboratory tests can be carried out, this method is flawed due to sample disturbances and equipment compliance. Under these circumstances, inverse GRA performed on downhole seismic array records can be used to determine in situ DSPCs. However, existing inverse GRA frameworks have limitations, one of them being the estimation of DSPCs for combined layers that only exist between successive accelerometers and not for individual soil layers or type. In addition, although frameworks do exist to determine G/G_max for a combined layer, to date there is no consistent approach that can be carried out to estimate β from inverse GRA in the frequency domain. This study proposes a new framework to estimate G/G_max, β, and shear strain (γ) that follows frequency domain methodology. The proposed framework will be applied to records from a downhole array located in Lotung in the central part of Yilan County, Taiwan and IWTH27 from the Kiban Kyoshin network (KiK-net) array in Japan following which G/G_max, β, and γ for the surficial layer will be determined. It is noted that the proposed methodology could be used to determine the DSPCs for recorded ground motions that exist at the top and bottom of a layer as well as for a surface layer that is underlain by rock with ground motions available at the surface and within the rock medium.

This study will apply a newly developed coupled thermo-hydro-bio-mechanical (CTHBM) numerical model to a typical full-scale landfill cell geometry with horizontal trenches (HTs) as the leachate recirculation system (LRS) and will evaluate the effect of the leachate injection pressure (IP), the horizontal spacing of the leachate injection locations (IS), and the mode of leachate injection (IM), on some of the major long-term hydraulic (e.g., wetted area), mechanical (e.g., differential waste settlement), biochemical [e.g., methane (CH₄) gas production], and thermal (e.g., elevated temperatures) characteristics of bioreactor landfills. Based on the results from a series of numerical simulations with different IP, IS, and IM, it was determined that the variable IP had the most significant impact on the landfill performance compared with IS or IM. Higher IPs lead to a linear increase in the wetted area and a corresponding increase in the degraded waste area, thereby reducing the time taken for waste stabilization. However, an increase in IP leads to larger elevated temperature zones within the landfill and larger differential settlements on the landfill surface. Larger IS caused significantly higher differential waste settlements due to the formation of unwetted zones within the landfill cell. Increasing the duration of gaps between the intermittent injection periods significantly reduced the extent of elevated temperature zones formed within the landfill. The relative shear displacements in the geosynthetic interface in the bottom liner and final cover system were mainly influenced by the IP.

High accuracy in the simulation of the discrete-element method (DEM) depends on the proper selection of microparameters. In this study, the range of microparameters was determined through sensitivity analysis. Subsequently, four levels of orthogonal experimental tables were established and 148 sets of data were collected. In addition, five data mining methods, namely, support vector regression (SVR), nearest-neighbor regression (NNR), Bayesian ridge regression (BRR), random forest regression (RFR), and gradient tree boosting regression (GTBR), were used to establish a microparameter prediction model. The results indicate that machine learning methods have significant potential in determining the relationship between macro and microparameters of the DEM model. RFR achieved the best performance among the five models whether the input data were collected from the tests of the Brazilian tensile strength and uniaxial compression or only the uniaxial compression test. In addition, the deviation between the predicted and measured macroparameters was less than 8%. This approach allowed for more accurate modeling of complex structures in a rock under various stress conditions through DEM simulations.

This paper deals with the procedure to determine the required mechanical properties of plastic concrete used to build a new cutoff wall as a remediation measure in an earth-fill dam constructed in a narrow canyon and subjected to internal erosion. The study includes a detailed three-dimensional numerical simulation of the staged construction and reservoir impoundment of the dam before and after the cutoff wall installation to consider the influence of the past loading history and upcoming static and seismic loading on the cutoff wall behavior. The static analysis uses the hyperbolic model by considering a two-phase approach for the core elements to simulate the construction pore-pressure generation and its consolidation effect. The dynamic analysis uses a nonlinear hysteretic model to consider the shear modulus reduction and damping ratio increase with the shear strain increase. A variety of preliminary mechanical properties are selected based on a procedure discussed in the paper and assigned to the plastic concrete to assess the cutoff wall responses. This assessment not only introduces the critical zones in the wall, where they experience high shear and axial strain but also help to finalize the mechanical properties, which are low-elasticity modulus with high tensile and shear strength. However, by using the conventional mix designs, it is not possible to produce such material, according to the laboratory test data collected from the literature. Therefore, polypropylene fibers are added to the mix design to obtain the required mechanical properties, whose test results are in the paper.

To investigate its fatigue damage characteristics, natural gypsum rock was tested under multiple cyclic loading conditions. According to the fatigue test, a nonlinear viscoelastoplastic fatigue model was established by combining several fatigue elements to describe the fatigue process of gypsum rock based on the equivalent stress assumption. The results indicate that the fatigue damage behavior of gypsum rock is substantially different from that of other types of rocks because of its particular mode of damage involving crystal and tabular crystal microstructures. The fatigue characteristics of gypsum rock, such as the fatigue life and fatigue deformation evolution, were found to be sensitive to multiple cyclic loading conditions such as the stress level, cyclic frequency, and stress amplitude. The proposed nonlinear viscoelastoplastic fatigue model could predict the fatigue deformation evolution of gypsum rock well, and the fatigue model parameters varied depending on the cyclic loading conditions. When the upper limit stress was less than the fatigue damage threshold of gypsum rock, the proposed fatigue model was similar to the modified Kelvin model, which could only describe the decelerating and stationary stages of the deformation evolution. When the upper limit stress was higher, the proposed model could describe the entire fatigue deformation process of gypsum rock well, especially the acceleration stage.

This study investigated the small-strain shear modulus behavior of compacted loess under isotropic consolidation. The small-strain shear modulus was obtained from bender element tests during the isotropic consolidation. The effects of excitation frequency, stress level, and stress history on small-strain shear modulus were analyzed. The test results suggested that the excitation frequency affects the measured values of shear wave velocity. An interval of wave path length-to-wavelength ratio was proposed to select the appropriate excitation frequency. Based on the test results, the effect of the stress level and stress history on the small-strain shear modulus of normally consolidated and overconsolidated soil was examined, whereby an empirical model was proposed to characterize the variation of small-strain shear modulus. The logarithmic law relationship mitigates the limitation in existing models that the power law relationship is incapable of accurately predicting the small-strain shear modulus for cases with large overconsolidation ratios. Finally, a framework was established to account for the effects of excitation frequency, stress level, and stress history on the small-strain shear modulus of compacted loess.

In geotechnical engineering, the design of cantilever sheet pile walls is one of the most important and complex tasks due to the use of different limit equilibrium design methodologies to estimate the depth of embedment. In some countries, official codes do not fully cover cantilever sheet pile wall design, so designers tend to adopt foreign codes and manuals generally calibrated to the local conditions. In this paper, the performance of several combinations of limit equilibrium design methodologies from international codes and manuals are evaluated using a proposed multiphase Monte Carlo simulation-based probabilistic design procedure. The results show that the direct implication of the safety factors and soil/wall friction on D/H is related to a significant change of the relative degree of conservatism of codes and manual design methodologies from high to small probability levels. It is found that even when the results of several codes and manuals methodologies agree, their implementation does not produce a design with a low risk of failure, especially for those methodologies that involve the European limit state design (LSD) approach. It is evidenced that for methodologies with unfactored passive pressures, the uncertainty in the response may not be calculated in a mechanically consistent way through reliability analysis. Finally, it is determined that the effect of soil variability in the depth of embedment is not linear when an increment of uncertainty is incorporated into the design, especially for the design combination constituted by the Hansen method and the North American LSD approach.

Analytical solutions for displacement, strain, and stress in a single energy pile provide rational, mechanics-based qualitative and quantitative understanding of thermomechanical load transfer mechanism in energy piles. To this end, thermomechanically induced axial displacement, strain, and stress evolutions are derived for different tip restraints and for several different loading scenarios including net heating, net cooling, and compressive and tensile mechanical loads. In addition, the analytical solutions are successfully validated against centrifuge test results. The analytical solution for the location of a thermal null point indicates that it depends on the geometry and stiffness of the pile, stiffness of the soil, and amount of head and tip restraints. In addition, the thermal null point moves upward into the top half of the pile with increase in the ratio between the amounts of head and tip restraints. It is also found that formation of a tension zone in energy piles is likely for several different load scenarios. At a constant mechanical load, the length of a tension zone depends on the geometry of the pile, ratio of the soil and pile stiffness, and the magnitude of thermal load. An increased magnitude of a net cooling combined with a constant compressive axial force increases the length of a tension zone and magnitude of tensile stress in the pile. On the contrary, the length of a tension zone and magnitude of tensile stress decrease with the increase in magnitude of heating at a constant tensile axial force. In summary, the analytical solutions and their features presented herein provide a fundamental, rational, mechanics-based framework for advancing the understanding of a load transfer mechanism and soil structure interaction in energy piles, thus contributing directly toward their safer design and wider use, and increased sustainability of civil engineering infrastructure.

The stability of a single circular tunnel in cohesive undrained soil was investigated under plain strain conditions using a two-dimensional finite-element limit analysis (FELA) program and a shear strength reduction method (SSRM). The stability for this problem was well expressed by using Broms and Bennermarks’ original stability number (N = (σ_s + γH − σ_t)/S_u) owing to the undrained loading conditions. The variations of the factor of safety and the critical stability number were presented for a series of tunnel cover-to-diameter ratios (C/D) in both collapse and blowout scenarios. Also, the extent of surface failure for some selected depth ratios was determined in the case of tunnel collapse. The obtained results were compared and validated by using the finite-difference method, as well as other existing solutions available in the literature. The dimensionless ratios employed in this study make the design charts suitable to cover a broad range of tunnel geometries and soil parameters.

In geotechnical tests, a thin layer of lubricants is usually applied along interfaces between tested soils and the sidewall of a container to minimize interface friction (shear strength). The effectiveness of lubricants in reducing the interface friction has not yet been fully evaluated. In this paper, the effectiveness of several commonly used lubricants in reducing the sidewall friction is presented through backfilling and stress measurement tests performed in a lightly instrumented column. The interface friction angles are back calculated from the measured stresses, using an arching analytical solution. The results show that the most efficient among the tested lubricants is a combination of Teflon and silicone grease, which can reduce the interface friction angle to 16.8°, a value far from being qualified as negligible. For most cases, the application of lubricants has little or no effect in reducing the sidewall friction. For some cases, the application of a lubricant material can even increase the sidewall friction angle. To verify these somehow surprising results, direct shear tests were performed to obtain direct measurements of the interface friction angles between the tested sand and the tested lubricant materials. Very small normal stresses were applied and the sand was gently placed in the shear box to imitate the testing conditions of column filling tests. The good agreement between the directly measured and back-calculated interface friction angles indicates that the column tests and the back-calculated results are reliable. Further analysis reveals that the small cohesion of lubricant materials can become negligible under a high stress state. Under a low stress state, this small cohesion can be translated into high friction. It is concluded that the tested lubricants may be efficient under high stress conditions, but not efficient under low stress conditions. For the latter case, the consideration of zero sidewall friction between the tested soil and lubricated sidewall of the container can lead to inaccurate and even totally wrong results.

Based on the piecewise-linear consolidation model, this paper presents a consolidation model that considers the drainage conditions and non-Darcian flow of clayey soils. In the proposed model, the continuous drainage boundary is used to reflect the drainage performance of the soil's boundary, and Hansbo’s equation is applied to describe non-Darcian flow of clayey soils. The applicability of the proposed model is validated by comparing its results with existing results. Then the influences of drainage conditions, non-Darcian flow, the ratio of compression index to permeability change index. and self-weight of soil on the consolidation process are analyzed through several cases. The results indicate that both a continuous drainage boundary and Hansbo's flow can significantly slow down the consolidation process, and this effect becomes more obvious with the increase of the ratio of compression index to permeability index. In addition, without considering the influence of soil's weight, the consolidation rate and the final settlement of the soil layer will be underestimated and overestimated, respectively.