International Journal of Geomechanics

Unsaturated constant water content triaxial compression tests with suction measurement, using an Imperial College tensiometer, and saturated consolidated undrained tests were conducted on reconstituted brickearth, a naturally unsaturated clayey silt from London. The results show that the saturated effective stress can be applied to the critical state line (CSL) and normalized stiffness for unsaturated brickearth but that Bishop's effective stress variable gives a slightly improved CSL. The stiffness derived from local instrumentation demonstrates that Bishop's effective stress is also beneficial for normalizing the stiffness modulus over the small-strain range (up to axial strains of about 3%).

In the process of tunnel excavation, the change of soil stress field will inevitably lead to principal stress axis rotation, which is a common but easily ignored phenomenon in shield tunneling. Therefore, it is necessary to consider the effect of principal stress rotation in the process of shield tunneling. In this paper, the stress field variation and ground surface deformation of cohesive soil are investigated by a numerical method via the strength parameter variation. A UMAT (user material) subroutine is programmed and embedding into the finite-element software to realize the strength parameter variation that is caused by the principal stress direction rotation during the whole tunneling process. The ground deformation considering the variation of the principal stress direction is compared with that when the variation of the principal stress direction is not considered. The simulation results show that the principal stress direction of the soil changes significantly during the tunnel excavation, and the principal stress direction of the soil within 1.5D (D = tunnel diameter) away from the tunnel axis and all the soil above the tunnel changes obviously. In addition, during the excavation, the rotation angle of the principal stress for each point in the soil gradually increases; in the direction perpendicular to the tunnel axis, the rotation angle of the principal stress increases first and then decreases from the tunnel axis to the edge of the model; when the direction of the principal stress changes, the strength parameters of the soil are modified, the ground deformation of which is larger than that before correction.

Deep learning (DL) algorithm bidirectional long short-term memory (BiLSTM) neural network is employed to model behaviors of the soil–structure interface in this study, as a pioneer research work to investigate the feasibility of using DL to model interface behaviors. Datasets are collected from 12 constant normal stress and 20 constant normal stiffness sand–structure interface tests. A modeling framework with the integration of BiLSTM is thereafter proposed. The results indicate that the BiLSTM-based model can accurately capture the responses of interface behaviors including volumetric dilatancy and strain hardening on the dense samples and volumetric contraction and strain softening on the loose samples, respectively. The effects of surface roughness, soil relative density, and normal stiffness on the interface behaviors are also investigated using the BiLSTM-based model. The predicted normal stress, shear stress, and normal displacement show good agreement with measured results.

Support damage is a typical hazard scenario associated with tunnel boring machine (TBM) drives in the squeezing rock condition. The stress release of the rock mass and the improvement of the support capacity are two effective methods to prevent support damage. The stress in the rock mass can be released by allowing a certain amount of overexcavation. The support capacity can be evaluated by solving the required support pressures corresponding to different critical strain levels. This paper aims to achieve a better understanding of preventing support damage for TBMs in the squeezing condition by quantitatively analyzing the overexcavation and the required support pressure. A semianalytical solution of the ground reaction curve is obtained. A simplified method of solving the support pressure with specific rock displacement is proposed. Based on the two solutions, the procedures for solving for the overexcavation and the critical support pressure are presented. Corresponding to the critical strain levels of 1%, 2.5%, 5%, and 10%, the response surfaces for critical support pressures in relation to the Geological Survey Index (GSI) and in situ stress are obtained via the nonlinear regression method. The influences of GSI, in situ stress, and the mechanical parameters of the concrete lining and shield on the size of the overexcavation are comprehensively investigated. A method to guide support design in the squeezing rock mass is proposed. Three tunnel cases are presented to demonstrate the application of the method, and three case histories that encounter support failures are discussed to illustrate the appropriate design measures through the proposed method.

This paper proposed a theoretical solution for cavity expansion in crushable soils. The constitutive relations of the crushable soils were described by the breakage mechanics model that explains the grain crushing induced grain size redistribution. The governing partial differential equations (PDEs) for the cavity expansion issue were formulated through the equations of equilibrium, constitutive relations, continuity conditions, and drainage conditions. The similarity transformation method was utilized to transform the PDEs to first-order linear ordinary differential equations, for which the numerical solutions were then obtained through the Runge–Kutta method. The effective stress, breakage, and specific volume around cylindrical and spherical cavities were given. The limit expansion pressure was particularly discussed through parametric analyses. The results showed that the normalized limit expansion pressure increases as the normalized critical comminution pressure $p_c^{\mathrm{\prime }}/p_0^{\mathrm{\prime }}$ increases when $p_c^{\mathrm{\prime }}/p_0^{\mathrm{\prime }}<10$ and tends to a constant value when $p_c^{\mathrm{\prime }}/p_0^{\mathrm{\prime }}>10$. The increase of the normalized bulk modulus $K/p_0^{\mathrm{\prime }}$ and critical state friction coefficient M led to the increase of limit expansion pressure, whereas the decrease of the ratio between bulk modulus and shear modulus δ, grading index $\vartheta$, and coupling angle ω resulted in the increase of limit expansion pressure. Moreover, the limit expansion pressure was not sensitive to the initial specific volume υ₀. The proposed solution could be used to interpret the issue of the pile end-bearing capacity in crushable soils.

This study aims to establish an equation to obtain the optimum tapering angle (α_opt) for bored tapered piles that is correlated with pile geometry and sand properties that vary with the relative density. This α_opt corresponds with the maximum axial bearing capacity when the volume of material in the tapered pile is maintained identical to the counterpart straight cylindrical pile. First, analytical formulations will be developed to estimate the axial bearing capacity of bored tapered piles that are embedded in sand. The proposed governing equations capture the shaft vertical bearing component of the tapered pile, which is unique to tapered piles and varies nonlinearly with the tapering angle (α). By differentiating the obtained bearing capacity equation for α, an α_opt is achieved. The finite element method (FEM) will be adopted to conduct the numerical modeling and to calibrate the model parameters of the proposed analytical equation, which considers soil nonlinearities and interactions between the tapered pile and the surrounding soil that is subjected to axial loading. The UBCSAND constitutive model will be used to simulate the soil response near the tapered pile and the model parameters will be calibrated against laboratory test results for sandy soils with different relative densities. However, due to the complexity of the proposed differentiation and inverse calculation, a numerical solution will be used to obtain the results. Then, the load–displacement curves of the tapered piles will be attained numerically and α_opt, which results in the maximum axial capacity of the pile, will be determined. The results exhibit good agreement between the analytically determined axial bearing capacity for the tapered pile and the corresponding numerical modeling predictions. Furthermore, a simplified empirical equation will be established to select α_opt, which could be used by practicing engineers.

Understanding the mechanical behavior of a reservoir rock within the CO₂-injection zone is essential for the safety of CO₂ geological storage (CGS). The water saturation of the reservoir rock could be changed with the injection of huge amounts of supercritical CO₂ (scCO₂). Triaxial compression tests were performed on sandstone under dry, H₂O-injected, scCO₂-injected, and CO₂–H₂O biphasic conditions at effective confining pressures ranging from 10 to 50 MPa to investigate the mechanical behavior of the sandstone within the CO₂-injection zone comprehensively. The results indicated that the strength of the sandstone was significantly reduced by H₂O, scCO₂, and CO₂­–H₂O biphasic fluids. The CO₂–H₂O biphasic fluid reduced the strength most significantly, followed by H₂O, and finally scCO₂. Water/scCO₂ monophasic and CO₂–H₂O biphasic fluids significantly affected the cohesion of the sandstone, but they did not affect the internal friction angle. Compared with the strength of the H₂O-injected sample, the strength of the CO₂–H₂O biphasic sample was reduced by −6.94%, 16.94%, and 3.28% at 10, 30, and 50 MPa, respectively. The CO₂–H₂O biphasic fluid significantly deteriorated the deformation modulus and stiffness and enhanced the ductility and compaction of the sandstone. A higher confining pressure could inhibit the enhanced cracking effect of scCO₂. The CO₂–H₂O biphasic zone within the reservoir was the most dangerous because faulting, surface subsidence, and permeability reduction of formation might occur in this zone. The results provided new information on the evolution of mechanical changes in saline aquifers injected with scCO₂ in CGS.

Discrete element method (DEM) is widely used to simulate particle crushing, and the simulation results are further utilized for devising constitutive models for soils and rocks. However, particle shape evolution during the process of crushing is often neglected for brevity. In this study, one-dimensional (1D) compression on crushable granular materials is simulated using two-dimensional (2D) DEM to identify the specific role of evolving fragment shape during particle crushing on the macroscopic response. Two different particle crushing approaches are adopted here. In the first approach, grain crushing is mimicked by substituting parent particles with smaller particles once a breakage criterion is attained. In the second approach, a bonded agglomerate consisting of smaller particles is used instead of an intact particle, which breaks once the bond strength exceeds. The simulation results show that despite an equal amount of particle crushing at any given strain level, the axial stress–strain responses are different in the two approaches. On the other hand, the evolution of contact fabric anisotropy reduces, which increases the lateral stress. Further, the effect of particle shape on the limit compression response is also studied. The study finally proposes an exponential relationship between particle breakage and particle shape indicator during the comminution process.

Operational parameters, such as multiple-cluster and well spacing, flow rates, and injection time, play a vital role in the deviation and growth of multiple-cluster hydraulic fractures during the implementation of modified zipper fracturing technology. Fundamental analysis of the dynamic zipper fracture growth because of changing flow rates, well spacing, and injection time does not appear to have been studied in the literature, although it is of critical importance for fracturing optimization. In this study, a fully coupled two-dimensional (2D) extended finite-element method (XFEM)–cohesive zone model (CZM) in combination with a phantom-node method (CPNM) was used in ABAQUS to investigate the initiation and propagation of multiple hydraulic fractures with 10 injection clusters that are distributed in two horizontal boreholes. This study aims to assist fracturing design to achieve minimal deviation in multiple fracture propagation and create deeper and much wider fractures between two horizontal boreholes of a modified zipper fracture (MZF) completion pattern. The growth of a hydraulic fracture in an XFEM region were verified by Khristianovic–Geertsma–de Klerk (KGD) zero toughness solution and good agreement with a negligible error of <2.5% was obtained. A series of simulation scenarios was carried out with various flow rates and three well spacings of 40, 60, and 80 m, over a long injection period to identify the optimum injection time in which a fracture grows straight with minimal deviation. The results show that to create smooth and deeper multiple hydraulic fractures in zipper fracturing, an increase in the flow rate is preferred, but the duration of the injection period should be optimized. The results of this study could be used to better address operational difficulties, such as multiple-fracture crossing and closure that is associated with unfavorable injection times, well spacing, and flow rates in MZF operations.

Granular materials are typically used as fills in earth structures, including foundation, embankment, and slope. The majority of these granular materials are built and maintained in unsaturated conditions. It is still an open question to develop simple and suitable models to capture the behaviors of unsaturated granular materials. This study aimed to formulate a model for describing the behaviors of granular materials under unsaturated conditions. An exponential function based on a phenomenological model was used to reproduce the asymptotic relationship between stress ratio and deviatoric strain during shearing, and a quantitative relationship considering variable void ratios was adopted to describe the changes of degree of saturation during shearing. In this new nonlinear model, nonlinear stiffness, nonlinear stress dilatancy, and critical state were introduced to enhance the model. Compared with the model for saturated soils, the new model for unsaturated soils could capture the suction effect on the behavior of stiffness and critical state. Finally, the new nonlinear model was used to simulate the behavior of saturated and unsaturated Chongqing granular fill and Beijing sand under triaxial stress conditions. The comparisons of experimental and simulated results showed that this model can consider the suction effect in reproducing the stress–strain and volumetric behavior of unsaturated granular materials.

The freezing process of salinized soil is a complex, dynamic, and interactive hydro–thermal–salt–mechanical (HTSM) coupled physical phenomenon. In many recent studies, soil was assumed to be saturated, and the theoretical models established were based on the framework of saturated soil, in which the influence of the vapor phase in freezing soil was neglected. In this paper, by considering the effect of vapor flow on heat movement and the relation between saturation and void ratio, an improved mathematical model will be established based on previous research. This improved hydro–thermal–salt–mechanical (IHTSM) model simulates the dynamic process of water migration, heat transfer, vapor flow, solute transport, and deformation. The numerical simulation implemented by the IHTSM model under the typical conditions of unidirectional freezing will be compared with previous research to verify the model's validity, and the various characteristics of the curves and their physical meaning will be analyzed by comparing them with the previous research. The dynamics of temperature, mass moisture content, displacement, salt content, volumetric vapor content and saturation degree in soil column will be discussed for the salinized soil during the freezing process. The results indicated that this improved model could provide a reference for the destruction process analysis of the harsh geological environment in cold, arid, and saline areas.

A numerical investigation of the effect of particle breakage on the one-dimensional compression behavior of a realistic granular assembly was carried out using the discrete-element method. The morphologies of sands were obtained using X-ray microcomputed tomography. Sand assemblies were generated in terms of the statistical distribution of sphericity from real sands. The results of single-particle compression tests showed that particle shape significantly affects crushing strength. Particle breakage had a significant influence on compressive deformation, as demonstrated by comparing breakable assemblies with unbreakable ones. The strong force distribution became broader, and the proportion of weak force increased with particle breakability. Looser samples underwent much more particle breakage than did denser samples. It was found that breakage dissipation was a small fraction of energy input compared with frictional dissipation. An excellent hyperbolic relationship between relative breakage and energy input was found, which was independent of initial void ratio.

The shear band formation and orientation in anisotropic granular materials have been studied. The anisotropy is described by making use of the microstructure tensor that complies with principles of continuum mechanics and expresses the anisotropy in a coordinate-invariance form. Further studies are conducted, based on bifurcation analysis, to investigate the conditions for the initiation and orientation of the shear band in anisotropic granular materials. The role of parameters describing the material anisotropy in shear band formation is examined. The main focus of this study is to relate principal values and directions of the microstructure tensor with the condition of localization of the deformation into a shear band. In particular, the influence of eigenvalues and eigenvectors of the microstructure tensor have been studied based on the classical theory of plasticity.

Based on the slip line theory and the upper-bound limit analysis theorem, a three-dimensional (3D) slip–rupture mechanism to design the shield tunnel face stability is proposed. This model is defined by using a spatial discretization technique. According to the stress arching theory, the vertical earth pressure at the top of the slip zone is calculated. The critical support tunnel face pressure is then deduced by using this additional load on the upper part of the slip damage zone. A series of finite difference numerical calculations based on the limit shear strain theory were performed to find the slip failure surface and obtain the critical support pressure. The results indicate that the soil-arching effect significantly affects the magnitude and the vertical stress distribution on front of the tunnel face. The good concordance obtained by the comparison of the results with the numerical simulations and other existing researches shows that the proposed model has a good applicability in both purely cohesive and frictional soils. At the same time, the critical tunnel face failure pattern is close to the numerical calculations one.

The instability of soil slopes is directly related to the shear parameters of the soil material and groundwater, which usually causes some uncertainty. In this study, a novel method, the element failure probability method (EFP), will be proposed to analyze the failure of soil slopes. Based on upper bound theory, finite element discretization, and stochastic programming theory, an upper bound stochastic programming model will be established by simultaneously considering the randomness of the shear parameters and groundwater level to analyze the reliability of slopes. The model will be solved using the Monte Carlo method based on the random shear parameters and groundwater levels. Finally, a formula will be derived for the EFP based on the safety factors and velocity fields of the upper bound method. The probability of a slope failure can be calculated using the safety factor, and the distribution of failure regions in space can be determined using the location information of the element. The proposed method will be validated using a classic example. This study could have theoretical value for further research that attempts to advance the application of plastic limit analysis to analyze slope reliability.

As one of the main filled materials of the Lianghekou (LHK) rockfill dam, the strength and deformation characteristics of LHK rockfill is of great interest for hydraulic and geotechnical engineers. Correct interpretation of the state-dependent behavior of LHK rockfill is necessary for the design, construction, and operation of the dam. Therefore, a series of large-scale triaxial tests on LHK rockfill with different initial states were carried out in this study, where the state-dependent strength and deformation behaviors of LHK rockfill were investigated. It was found that both the phase transformation state and peak state of LHK rockfill are dependent on the material state. To capture such state-dependent behavior, two elastoplastic models (i.e., α₁-model and α₂-model) based on different stress–fractional concepts were developed and compared. It was found that the fractional order (α₁) in α₁-model exhibits insignificant dependence on the material state of LHK rockfill, whereas the fractional order (α₂) in α₂-model is dependent on the material state. As the fractional order determines the plastic flow direction of rockfill, it should be identified carefully, otherwise, an overestimation or underestimation of the critical-state stress may occur when using α₂-model. Further validation against the large-scale triaxial data of LHK rockfill shows that without using additional plastic potentials, the two models can reasonably reproduce the state-dependent behavior of LHK rockfill. A slightly better performance on simulating the stress–strain response is observed when using α₁-model, whereas α₂-model works better when simulating the stress–dilatancy response.

This paper presents an analytical study on the roles of fluid viscosity, pH, and ionic strength in the variations of the interparticle forces and the force equilibrium of a spherical particle resting on a particle bed. The purpose of this analytical study is to provide a theoretical explanation of the observed trends of the effects of viscosity, pH, and ionic strength on the particle mobilization under laminar flow conditions in an experimental program. Viscosity affects the mobilizing hydraulic shear force, while pH and ionic strength affect the interparticle surface forces (i.e., van der Waals force and electrostatic force) between a particle and the particle bed. In this paper, the experimental program and the observations on particle mobilization under the effects of viscosity, pH, and ionic strength are first overviewed; then, the hydraulic dynamic force and interparticle surface forces in terms of van der Waals force and electrostatic force are computed. The study investigates the variations of the van der Waals force and electrostatic force with particle separation distance, pH, ionic strength, and temperature and how the variations affect the critical flow velocity, i.e., the flow velocity that induces particle mobilization. The following major conclusions are drawn from this analytical study. (1) As the pH increases and zeta potential decreases, the electrostatic force (repulsion) between a test particle and a bed of particles increases, and the test particle is more likely to mobilize; thus, the critical flow velocity decreases. (2) As the ionic strength increases, the repulsive electrostatic force generally decreases and could cause the net surface force on the test particle to change from repulsion to attraction; consequently, the test particle becomes less likely to mobilize; thus, the critical flow velocity increases. (3) The trend of electrostatic force variation with ionic strength remains the same when the fluid's zeta potential changes from −10 to −40 mV. As the zeta potential decreases from −10 to −40 mV and the pH increases, the electrostatic force can increase by one order of magnitude and becomes more significant when compared with the van der Waals force. (4) When the fluid temperature increases from 5°C to 50°C, the variation of the electrostatic force is not significant and is within one order of magnitude at various ionic strengths. However, the hydraulic shear force decreases significantly due to the significant decrease of viscosity when the fluid temperature increases from 5°C to 50°C; thus, it significantly increases the magnitude of critical velocity.

Field evaluation of the strength properties of jointed rock masses is a challenging task in geotechnical engineering. Typically, laboratory tests using small jointed specimens have difficulty determining the strength parameters of jointed rock masses due to the scale dependence of discontinuities and because the tests are expensive and time-consuming. Fast and continuous estimation of the unconfined compressive strength σ_cm of a jointed rock mass directly using drilling via a deep convolutional neural network (CNN) is a novel and practical field investigation method. This paper presents a CNN framework that includes (1) obtaining a training dataset; (2) determining the unconfined compressive strength σ_cm via a rock mass quality rating (RMQR) system; (3) training the CNN model; and (4) validating the results using tunnel engineering calculations. A comparison of the CNN predictive results with the true values suggests that the CNN makes good predictions across a wide range of unconfined compressive strengths σ_c of intact rock, especially for high RQD values. Due to the joint orientation, the unconfined compressive strength σ_cm of a jointed rock mass cannot be reliably determined using the σ_cm/σ_c ∼ RQD relation. By incorporating the physical variables of RQD and σ_c, which are known to affect the unconfined compressive strength σ_cm of a jointed rock mass, into the CNN, the proposed CNN model can provide better predictions than the regular CNN model. All the results predicted by the physics-informed CNN are within the accepted error range of 10%. This method is applied to the excavation of the Huangshan Tunnel in the Hanjiang-to-Weihe River Project of China and is verified as reliable via comparative studies with previous works. Thus, the proposed method represents fast and efficient prediction of the strength of jointed rock masses in rock engineering.

Embankment stability issues present design and research challenges to geotechnical engineers. Current design methodologies for the stability of embankments on columns are mainly evaluated under assumptions of a shear failure of columns, which may cause an overestimation of the embankment stability when a bending/tensile failure occurs. In addition, the coupling effect of the impact factors on the embankment stability is rarely considered. In this investigation, the effect of various factors, including the soil conditions, column arrangement, and embankment geometries, on the embankment stability is first numerically analyzed. Based on the generated numerical data, a multivariate adaptive regression splines model that maps the multidimensional relationship between the affecting parameters and the bending/tensile failure of columns is developed. The relative importance of the variables on the bending failure of columns is quantitatively analyzed. The accuracy and reliability of the developed model on predicting the bending/tensile failure of columns is validated with collected centrifuge test data from the literature.

Using local materials to build rockfill dam favors the natural resources, but the strength characteristics of rockfill are quite different for each project, which makes it necessary to explore the applicability of failure criteria of rockfill. In this paper, four kinds of rockfill with different lithology are collected from different projects, using a large-scale true triaxial apparatus developed independently by the Changjiang River Scientific Research Institute, a series of true triaxial tests under different intermediate principal stress Ratio b [= (σ₂−σ₃)/(σ₁−σ₃)] are carried out. The results show that: Ratio b strongly affects the strength of rockfill, especially in the range of 0 < b < 0.25, compared with conventional triaxial tests, strength parameters show a significant increase; compared with several classical failure criteria, the Lade–Duncan failure criteria established by testing on fine sand show better applicability in describing the whole strength evolution process, due to particles breakage that will attenuate the strength of rockfill during big deformations, the Lade–Duncan failure criteria can be treated as the upper envelope for all four kinds of rockfill, the harder the samples’ parent rock, the closer the strength to this upper envelope; SMP failure criteria show better applicability in the range of b < 0.25, so, when the effect of σ₂ is not so significant, these criteria could be used to make an assistant determine for strength of rockfill; a plane strain test is also carried out, and when testing stress approaches peak value, Ratio b of all four kinds rockfill are within a narrow range of 0.16 to 0.20, in the case of limited test conditions, strength parameters of rockfill under three-dimensional stress conditions can be roughly estimated by plane strain test initially, and a simplified failure criteria is proposed.

Mitigating the ground vibration initiated by the passage of trains to safeguard the surrounding structures is a challenge for geotechnical engineers. Among various existing methods, open trenches are one of the most common techniques for minimizing the transmission of such ground vibrations. Although there are several advantages to this technique in isolating ground vibrations, the main concern of such trenches is their instability under imposed dynamic loading. To overcome this issue, a new technique for isolating ground vibrations using horizontally buried hollow pipes is proposed in this study. The performance of such pipes in mitigating the ground vibrations produced by harmonic stress excitation has been investigated through finite-element modeling under plane strain conditions. These pipe assemblages have been found to be quite effective in reducing ground vibration transmission while eliminating instability problems when compared with open trench barriers. In-depth studies were carried out on the influence of various critical parameters relevant to the pipe installation, including barrier angle, depth, diameter, and different subsoil conditions on the vibration mitigation efficiency. Maximum efficiency of the new technique was achieved by optimizing the inclination angle, depth, and diameter of the pipe installation. Although the nature of the surface soil significantly influences the efficiency of this technique, ground vibration mitigation works best in soft soils. The numerical results were analyzed and interpreted to arrive at specific conclusions.

This study widely investigates the ultimate bearing capacity of a rigid footing on the free surface of sand overlying clay using the rigid-plastic finite-element method (RPFEM). Interface elements were introduced with the new constitutive equations developed by the authors to properly evaluate the interaction between the footing and the soil because these elements greatly affect the failure mechanism of the footing–soil system. Two friction conditions were employed for the footing surface, namely, the perfectly rough condition and the perfectly smooth condition. The RPFEM was extended to calculate the distribution of contact normal stress along the footing base corresponding to changes in the thickness of the sand layer. Several design charts were developed to directly determine the ultimate bearing capacity by increasing the internal friction angle, the thickness of the sand layer, and the shear strength of the clay layer. Two cases were considered for the clay layer below the sand layer, namely, a weak layer and a stiff layer. The failure mode of two-layered soils was found to change from the general shear mode to the punching shear mode for both friction conditions by a reduction in the shear strength of the clay layer. The sheared area of the ground was limited to the sand layer in the general shear mode, while the sheared area was distributed throughout the two layers in the punching shear mode. New bearing capacity formulas during the punching shear mode were proposed for the two friction conditions in a wide range of strength and geometric parameters, which were in close agreement with the experimental studies and are efficient enough to be used in practice.

Progress in the study of unsaturated soils and their application in practical geotechnics could be partially attributed to the advancement of their modeling and computational technologies. The soil–water retention curve (SWRC) and k-function are fundamental hydraulic properties of unsaturated soil that are important in the analysis of soil states. This study presents a new model to represent the hydraulic properties (SWRC and k-function) of bimodal soils using linear curve superposition. The resulting SWRC model was validated for common bimodal soils, and the k-function was then obtained. The relationship between the model and soil porosimetry was also deduced, leading to a better understanding of the physical influence of the model's parameters. In addition, based on the model, a solution was proposed to simulate the soil infiltration process involving partial differential equations and numerical methods. The model's efficiency in representing bimodal soils was proven, and the results from subsequent analyses confirmed its physical consistency. Therefore, the proposed model could address problems in real-world geotechnical practices.

This paper presents an analysis of the soil arching effect associated with piled embankments under cyclic loading by developing a series of two-dimensional trapdoor discrete numerical models. After identification of the microparameters by a biaxial compression test, the numerical models were validated by comparing the numerical model and plane laboratory test results. The contact force characteristics, deformation behaviors, and porosity distributions under various embankment heights were analyzed. The numerical results showed that cyclic loading leads to the degradation of soil arching as characterized by the collapse of load transmission paths and their vertical distribution. The variability observed in the contact fabric further confirmed that degradation in the higher embankment mainly occurred in the upper layers. Increasing embankment height decelerated the increasing contact force of the portion above the trapdoor and thus retarded the process of degradation. The load-induced time-dependent surface settlement accumulated with increasing cycle numbers and varied with embankment height. With an increasing number of cycles, the deformation area in the lower embankment cases above the trapdoor shifted from a triangular pattern to a rectangular pattern, while the higher embankment still maintained a triangular pattern. Furthermore, the effects of the localized loading area, stationary support width, and trapdoor displacement on soil arching concerning the loading efficiency and surface settlement were evaluated.

The consolidated drained triaxial shear tests have been performed in this work to investigate the shearing behavior of calcareous sands sampled from the South China Sea, with the focus on analyzing the influence of particle breakage on the materials shear strength. At approaching the failure limit state, the intense particle breakage and rearrangements prevented the shear stress from increasing further. Depending on the initial packing density, the loose sand sample exhibited the strain-hardening response, while the dense sand sample exhibited the strain-softening response with clear shear dilatancy after the peak shear strength has been reached. However, when the confining pressure increased, particle breakage occurred more thoroughly, and the sharpness of the peak stress disappeared gradually. For the series of tests, an upper limit of relative particle breakage existed, beyond which the confining pressure and relative density had little influence on the breakage of particles. The shear strength of calcareous sands was found to be determined by the combined effects of interparticle friction, sample dilatancy, and particle breakage. Under low confining pressures, the shear strength was mainly controlled by particle friction and sample dilatancy, while under high confining pressures, the effect of particle breakage was dominant. In this process, the volumetric strain evolved from dilation to contraction and the sample dilatancy angle decreased gradually. After breakage, the particle shape transformed from highly angular to subrounded.

This study investigates the compressive and tensile strength properties of cement-stabilized dredged sludge (CDS) improved by straw fiber (SF) and nano-SiO₂ (NS). The influence factors, including SF content, NS content, SF length, and curing time, were evaluated via a series of unconfined compressive strength (UCS, q_u) and splitting tensile strength (STS, q_t) tests. Furthermore, the microstructure evolutions and micromechanisms were explored by conducting scanning electron microscopy (SEM) tests. The results indicated that the inclusion of NS can obviously improve the UCS and STS of CDS, and the optimum NS content was 1.2%. Adding SF to CDS decreased its UCS but increased the STS. Compared with single SF or NS inclusion, the combination of SF and NS exhibited more advantage in improving the compressive and tensile strength behaviors. The optimum SF contents for NS-modified and SF-reinforced CDS (CNFDS) at 3, 7, 14, and 28 days were 0.5%, 0.4%, 0.3%, and 0.3%, respectively. Furthermore, 1.5% NS and 2–5 mm SF were the most effective in improving the UCS of CNFDS. The second-bearing effect occurred in the tensile stress–strain curves, in which the tensile stress evolution with strain was presented by a fitted linear equation. The optimum SF content, NS content, and SF length for CNFDS achieving the highest STS were 0.2%, 1.5%, and 5–10 mm, respectively. By fitting the relationship between UCS and STS, the relation of q_t = 0.17q_u for CNFDS was observed. The SEM analysis indicated that the interfacial friction and bonding between SF and cemented soil particles were mainly responsible for the strength properties improvement of CNFDS.

This paper presents an experimental investigation into the axial pullout behavior of buried medium-density polyethylene (MDPE) pipes under various pulling rates. Tests were conducted in a facility developed at the Memorial University of Newfoundland. During the tests, pipe elongations and axial strains were measured to examine the mobilization of axial force with leading end displacement, and circumferential strains were measured to capture the effects of pipe diameter change. Finite-element (FE) modeling of the tests was used to evaluate the interface friction angle and the effect of diameter changes on the normal stresses on the pipe surface, which influence the pullout resistance of the soil. The results indicated that the pipes’ elongation, diameter decrease, and pulling rate significantly influence the pullout forces for buried MDPE pipes subjected to axial ground movement. FE analysis with interface friction angles of 0.75ϕ, 0.86ϕ, and 0.9ϕ was found to successfully simulate the test conditions under pulling rates of 0.5, 1, and 2 mm/min, respectively. Based on the study, simplified methods were proposed to calculate the mobilized frictional length and pipe wall strain from the leading end displacement.

Water and mud inrush is a common geological disaster in the construction of karst tunnels. The loss of soil particles in filling-type karst disaster-causing structures is the key factor leading to the expansion of the water inrush channel. Considering the effects of water erosion, seepage force, and soil cohesive force, three-dimensional force analysis for soil particles on the side wall of the water inrush channel is conducted. The critical condition of incipient particle motion is established. The incipient flow velocity for sliding instability and rolling instability of the particle is deduced, respectively. The criterion of incipient particle motion is proposed. The expansion mechanism of the water inrush channel is revealed. The influencing factors of the incipient flow velocity are analyzed and the rules of particle loss are discussed. Finally, through the analysis of particle–fluid coupling and calculation of discrete element method (DEM)–computational fluid dynamics (CFD) coupling, numerical simulation for the incipient particle motion and channel expansion is implemented, and the proposed mechanism of incipient particle motion is verified. The results show that: (1) the incipient flow velocity of rolling instability is obviously greater than that of sliding instability; (2) the incipient flow velocity first decreases and then increases with the increasing particle radius and dip angle of the inclined plane; (3) the incipient flow velocity decreases linearly with the increase of hydraulic gradient and porosity; and (4) in the expansion of the water inrush channel, the incipient flow velocity of most particles is the critical flow velocity of rolling instability, while the incipient flow velocity of a few particles is the critical flow velocity of sliding instability.

Due to the weathering process, diorite-porphyrite dikes break up into a soil–rock mixture (SRM), which is crucial to the stability of dams and reservoirs that are under frequent fluctuations of water level. However, the grain transformation and mechanical behavior of SRM during weathering have rarely been studied. This study aims to investigate the effect of dominant grain size and content on the shear properties of SRM. Therefore, a sequence that simulates the change in dominant particle size and content during the weathering process will be proposed and a series of direct shear tests will be performed on the SRM samples and two-dimensional commercial discrete element method (DEM) program particle flow code (PFC2D) numerical models. The results show that the relationship between the shear stress and shear displacement of the SRM can be expressed by an exponential function. The shear strength and internal friction angle continued to decrease with weathering and the cohesion increased slightly. By comparing the particle flow code (PFC) simulation with the laboratory tests, the shear strength of the SRM was mainly affected by the force of friction between grains. The cohesion between particles was weak and was not a significant factor that affected shear strength.

Soil–rock mixture (SRM) shows complicated mechanical behaviors and hydraulic characteristics, which lead to challenging deposit landslides and debris problems in southwestern China. Previous studies have shown, based on several mechanical tests, that the strength parameters of natural SRM are sensitive to water content. Therefore, the complex mechanical characteristics during rainfall are vital when analyzing the deposit slope failure mechanism. This study aims to investigate the stability of a typical deposit slope in Yunnan, China that is subjected to different rainfall conditions given the parameter deteriorations and crack effect. A large scale mechanical laboratory is generally adopted to obtain the strength parameters of natural SRM to avoid the scale and boundary effects due to the particle size of the rock blocks inside that involves the consumption of time, materials, and money. Nonlinear empirical formulas to predict the deteriorated values for the cohesion and friction angles of natural SRM with a variety of water contents will be proposed in this study to address the shortage of laboratory tests. The accuracy of the formulas will be proved by comparing the estimated and measured values from previous studies. A series of two-dimensional (2D) numerical analyses will be conducted to study the effect of rainfall conditions and crack depths on deposit slope stability by considering the deteriorated strength parameters. The results indicate that the deposit slope stability gradually decreased during rainfall and the decreasing rate of the factor of safety increased as rainfall intensity increased. The comparisons between a slope with and without parameter deteriorations demonstrated that the weak resistance of SRM to water is a negligible factor in slope failures. The simulations of the cracking slope reveal that cracks could provide preferential flow channels for rainfall infiltration, which lead to the appearance of an increasing pore water pressure area and groundwater level fluctuations, which increases the probability of slope failure. As the crack depth increases, the influence of cracks on pore water pressure distribution and slope stability becomes more apparent. The combined effects of rainfall and crack depth on slope stability will be analyzed according to the simulation results. The results from this study could provide a reference to estimate deposit slope stability by considering parameter deteriorations that are subjected to rainfall and expand the understanding of the relationship between rainfall intensity, duration, and crack depth.

The existence of debris at the bottom of a rock socket significantly influences the mechanical behavior of a rock-socketed pile. To investigate the effects of debris on shaft friction and load transfer characteristics of a rock-socketed fiber–reinforced polymer (FRP) seawater sea–sand concrete (SSC) pile, two physical model tests will be conducted in this study. Cyclic and monotonic loadings will be applied on the top of FRP–SSC model piles along the axial direction. The vertical strains inside the model piles will be monitored by embedded fiber Bragg grating (FBG) sensors, which will be used to calculate axial stress inside the pile and shaft friction at the pile–rock interface. From the monitoring results, the detailed distribution of load transfer and shaft friction in each model pile is obtained. A comparison of the results from the two model pile tests indicates that the shaft friction and the end-bearing stress were significantly influenced by debris at the bottom of the rock socket. Under cyclic loading and static loading, the mobilization of shaft friction is more dominant when debris existed than when it does not. The end-bearing stress was reduced due to the existence of debris.

The effect of undrained shear strength profile is an important practical consideration for estimating the bearing capacity of a strip footing placed near slopes or excavations. Previous studies generally assume that the soil strength (c_u) is uniform or increases linearly with depth but does not vary in a horizontal direction. However, the horizontal variation in soil strength is also commonly found in practice, which is usually caused by various depositional and postdepositional factors. A modified formulation of shear strength profile was defined to characterize this strength variation and was embedded in upper and lower bound finite-element limit analysis (FELA). Seven principal failure modes were revealed and discussed. A comprehensive parametric study was performed to assess the bearing capacity and failure mechanism accounting for the orientation of strength increase. It was found that strength variation in horizontal direction significantly influences the bearing capacity and its associated failure mode. Some novel variation trends of the bearing capacity and the rotation of strip footings were found as well.