International Journal of Geomechanics

Sadov’s formula is widely used to solve the seismic wave velocity problem of single-hole blasting in calculations in blasting engineering. There is a significant difference between the calculated results for group-hole blasting and the actual situation in an open-pit mine. The modification and application of Sadov’s formula for group-hole blasting in open-pit mines is discussed in this paper. First, the insufficiencies of Sadov’s formula under the condition of group-hole blasting in an open-pit mine are analyzed. A theoretical derivation method for modifying the maximum charge coefficient is proposed. Second, blasting vibration data of the Hejing limestone mine of Guangxi Huarun Cement (Pingnan) Co. Ltd., obtained by onsite monitoring, are analyzed. The vibration data of the single-hole blasting are processed by linear superposition and fitting. The difference between the superposition results of the same single-hole vibration signal and different single-hole vibration signals and the onsite monitoring data is compared and analyzed. A modified expression of Sadov’s formula is obtained. The onsite blasting parameters of the group-hole blasting are verified. Finally, the safety distance problem of the grouting curtain in blasting construction is solved using the modified Sadov’s formula. The results show that the modification scheme of maximum charge Q in Sadov’s formula is determined through theoretical analysis and formula derivation. The peak vibration velocity of the group hole blasting suitable for the Pingnan mining area is deduced, and the correctness of the peak vibration velocity is verified.

The tunnels located in the shallow depths of loose saturated sand are significantly prone to liquefaction-induced uplift. Research works are, therefore, in progress to propose efficient techniques for mitigating uplift. In this study, 1 g physical modeling was used to assess the performance of helical piles for decreasing liquefaction-induced uplift. The effects of pile length, number of pile helixes, and the pile spacing in plan view were investigated. The uplift mechanism of the tunnel and helical pile system was also analyzed. The results demonstrate that the penetration of the helical piles into the dense layer underlying the superficial liquefiable sand has decreased tunnel uplift significantly. However, excessive close pile spacing along the tunnel resulted in shear surface interference, and the efficiency of the excessive number of helical piles decreased significantly. The detailed view of the uplift mechanism showed that utilization of the piles extended the transition phase of uplift during shaking. Helical piles can efficiently restrict the possibility of rapid uplift of the tunnel and shorten the duration of the primary uplift phase.

Undrained torsional shear tests were carried out on remolded loess specimens using a hollow cylinder apparatus to investigate the deformation behavior and noncoaxiality (where the direction of the plastic strain increment is not coaxial with the direction of the principal stress) of remodeled loess when the magnitude and direction of the principal stress change simultaneously. It can be found that the deformation behavior of tested samples is significantly influenced by the intermediate principal stress coefficient and the rotation range of the principal stress. The influences of elastic strain, rotation range of principal stress, intermediate principal stress coefficient, and cycle period on the noncoaxiality were also studied. Analysis of the test results demonstrate obvious noncoaxiality of the remolded loess. The noncoaxiality shows segmentation characteristics. The noncoaxiality will be overestimated if the elastic strain is considered and is negative when the principal stress rotates in the reverse direction. Reversal of the principal stress leads to abrupt changes in noncoaxiality. However, the noncoaxiality is similar in the process of forward rotation and reverse rotation of the principal stress. With the same cycle period, the noncoaxiality angle decreases with an increase in intermediate principal stress coefficient, but the effect of intermediate principal stress coefficient decreases with an increase in the cycle period. Increases in the cycle period increase the noncoaxiality of the remolded loess, while the influence of the rotation range of the principal stress on the noncoaxiality is not significant.

Freeze–thaw damage of rocks causes many engineering problems in cold regions. During the cyclic freeze–thaw process, the frost-heaving pressure, which is produced by 9% volumetric expansion of pore water, will drive the expansion and growth of microscopic pores. However, the change law of microscopic pore structure and how it affects the macroscopic properties of rocks under freeze–thaw are still not clear. In this study, the microscopic pore structure of five sandstones after different freeze–thaw cycles were continuously measured by mercury intrusion porosimetry (MIP). It shows that the microscopic pore structure of these sandstones has obvious fractal characteristics under freeze–thaw conditions. The fractal dimension has a gradual reduction with increasing freeze–thaw cycles. It illustrates that the distribution of the pore size is more and more uniform and that the complexity of the pore structure decreases. In addition, a bimodal exponential model can be well used to characterize the size distribution of nano- and micropores for these sandstones. The characteristic radii of nano- and micropores calculated by the bimodal exponential model have an obvious increasing trend with increasing freeze–thaw cycles. In addition, although the pore radii in these sandstones increase due to the freeze–thaw action, the volumetric fractions of nanopores and micropores are not changed remarkably. By using the least squares regression method, the porosity, P-wave velocity, and uniaxial compressive strength (UCS) are highly correlated with the fractal dimensions and characteristic pore radii. The P-wave velocity and UCS decrease with the increase of characteristic pore radius or reduction of fractal dimension. This study has filled the gap between microscopic pore structure change and deterioration of macroscopic physicomechanical properties under freeze–thaw, and it also provides a better understanding of the macro–micro frost damage mechanism for rocks.

Vertical drains improve the consolidation of soft clay, characterized by the large void ratio, high compressibility, and high water content, but the consolidation theory still needs to be improved. This paper investigated the large-strain consolidation of soft clay with vertical drains under free-strain theory, considering the elastic viscoplastic constitutive relation of soft soil, the coupled vertical–radial non-Darcian flow, and the logarithmic relationship of the permeability coefficients. Meanwhile, the nonideal cylindrical unit cell of the soil was captured by taking into account the permeability of the vertical drain decreasing with time (variable well resistance) and the smear effect. The proposed model was verified by degenerating into the cases in the extant literature and applying it to two laboratory tests using the finite-difference method. The most salient finding of the parametric analysis was that the average degree of consolidation curve might show two ups and downs due to the rheological property of soft clay. Several parameters reflecting the impacts of the smear zone and well resistance were also studied. The results demonstrated that some factors related to soil permeability and boundary conditions could enhance the phenomenon of increasing pore pressure.

Suffusion refers to the phenomenon in which fine particles in internal unstable soil are carried by seepage through skeleton pores. It is a type of internal erosion that can lead to major hazards for hydraulic and geotechnical engineering. In this study, the suffusion process is reproduced in gap-graded soil with realistically shaped coarse grains by coupling the discrete-element method (DEM) with dynamic fluid mesh (DFM), which can reproduce the pores formed by coarse particles and adapt to the deformation of the soil skeleton by changing the mesh according to the movement of coarse particles. Realistically shaped clumps are generated according to 49 Leighton Buzzard sand grains and applied to be the coarse grains in the gap-graded soil numerical model. We examine the roles of relevant parameters and compare the results with the model formed by spherical coarse grains. Two mechanisms (interlock effect and retention effect) hindering fines migration are identified at the pore-scale. The realistically shaped grains can better interlock, reducing porosity and squeezing fine particles. In addition, fines migration can be better prevented due to the roughness of the particle surface. Sensitivity analysis shows that the erosion mass increases with increasing hydraulic head, fine content, and the rising rate of the hydraulic head. Our results show that the erosion weight of the spherical particle model is much larger than that for the model of realistic grain shape; however, no evident difference in the suffusion pattern is observed between these two models.

Carbon dioxide blasting is an excellent alternative technique for blasting. Nowadays, bench carbon dioxide blasting has been gradually used in mining and rock engineering. To evaluate the essential characteristics of the ground vibration induced by bench carbon dioxide blasting, we monitored the single-hole bench carbon dioxide blasting vibration signal and proposed a novel method for determining variational modal decomposition (VMD) parameters based on the sparrow search algorithm (SSA). After the signal has been decomposed by the VMD method with the optimized initial parameters, combined with Pearson correlation coefficients (PCC), the noise and false components from the original signal can be removed. Then, the time–frequency characteristics of the monitored ground vibration were analyzed through the Hilbert transform. Latterly, based on the monitored peak particle velocity (PPV) in carbon dioxide blasting, an empirical equation for the ground vibration was established. It is indicated from the results that the SSA–VMD–PCC method can be successfully applied to the vibration signal preprocessing of carbon dioxide blasting. The vibration energy of bench carbon dioxide blasting is mainly concentrated in the range of 0–128 Hz. The attenuation law of PPVs in two monitoring directions obeys the power function, and the PPVs in the jet direction are greater than that in the vertical jet direction. Compared with the single free-surface blasting, ground vibration excited by the bench blasting with two free surfaces is weaker. Moreover, the blasting vibration empirical equation proposed by dimensional analysis can effectively describe the attenuation of the PPV in bench carbon dioxide blasting.

Rock brittleness plays an important role in rock engineering, thus it is practically valuable to assess rock brittleness accurately. Many brittleness indices have been developed, but few methods are developed for layered rocks. Based on energy evolution, the authors propose a new index for assessing brittleness of layered rocks. The new index indicates that high rock brittleness means elastic energy accumulated effectively before peak and dissipated abruptly after peak. For the sake of verifying the proposed method, a series of compression tests were performed. The results demonstrate that rock brittleness decreases gradually with increasing confining pressure. As orientation angle increases, rock brittleness weakens and then strengthens overall. The brittleness of the specimen with an orientation angle of 0° is slightly higher than that with an orientation angle of 90°. The new index was compared with three existing brittleness indices, and the results indicate that the proposed method can evaluate brittleness of layered rock effectively.

Cohesion and rolling friction play key roles in governing the behavior of soil; however, only a limited number of studies have been able to assess the simultaneous contributions of these two microparameters on the macroproperties of soil. In this respect, the innovation of the current study includes an attempt to examine the interplay of these two primary parameters on the angle of repose (AoR) based on the discrete-element method (DEM). Lifting cylinder tests on cohesive wet sand have been carried out in DEM, while the cohesion and rolling friction are captured through proposed computational models. In this paper, macroparameters, such as the geometry and developmental stages of sand piles obtained in DEM simulation, are compared with experimental data, while their microevolution is quantified in detail. The results show that a large AoR can only be obtained when the cohesive and rotational frictional forces work in tandem. Increasing the cohesion and rolling friction results in smaller contact numbers, with increasing chain-like connections between particles and larger pore spaces to account for a larger AoR. For the first time, this study distinctly identifies three major stages that contribute to the AoR, based on the development of contact numbers and the transformation of energy. Accordingly, the linkage between macroscale AoR and the microstructural coordination number is formulated with varying levels of cohesion and rolling friction. The DEM results prove that the more cohesive the particles are, the greater the delay in the dissipation of kinetic energy.

A simplified analytical method is proposed for one-dimensional unsaturated soil consolidation theory. First, in the one-dimensional compression test of unsaturated soil, the effective stress principle of saturated soil under compression is introduced to replace the constitutive relation expressed by the state of double stress in Fredlund’s theory. Based on this, the governing equations composed of two bi-variable partial differential equations are obtained. In the process of derivation, the dissipation law of pore pressure during the consolidation process is analyzed and the theoretical rationality is demonstrated. When solving the system of equations, considering that the initial excess pore water and the gas pressure are caused by instantaneous loading, Hilf’s theory is improved to calculate the change of pore pressure. This improvement is more in line with the actual situation and simplifies the approximate calculation method to obtain the analytical solution when the coupling effect of water and gas is considered. This method is verified by comparing with Terzaghi’s theory and Fredlund’s theory. It is shown that the solution of Terzaghi’s theory is a special case of this method for saturated soil, and the results of the dissipative process obtained with the method is close to Fredlund’s theory.

This paper presents a work to define four fundamental concepts, namely, buoyant force, submerged unit weight, seepage force, and critical hydraulic gradients, for saturated geomaterials including soils, rocks, and concrete under normal and high pressures using the general effective stress (GES) concept along with Terzaghi’s effective stress. In particular, four typical GES expressions are used for this purpose, and their impacts on the definition of the four concepts are compared based on available experimental evidence in the literature. The results suggest that (1) Terzaghi’s effective stress can be physically validated in the context of Archimedes’ principle for soils under normal conditions; (2) the generalized buoyant force on the unit volume of saturated geomaterials is the product of the unit weight of pore fluid and the GES coefficient tensor; and (3) the generalized seepage force theoretically acts in the direction of pore fluid flow only when the GES coefficient tensor is proportional to the permeability coefficient tensor. These four fundamental concepts have a profound significance for geotechnical applications with GES and thus merit further validation with adequate laboratory and in situ observations.

Abstract

A simplified numerical method was proposed for computing the slip-line field, as well as the associated earth-pressure coefficient, in a cohesionless backfill with an inclined surface lying behind an inclined rough wall. The potential failure zone, in either active or passive cases, was divided into the Rankine zone, which was rigorously obtained by the theory of plasticity, and the transition zone being further divided into a series of triangular slices. Within the transition zone, the theoretical relationship between the inclination of the interslice force and that of the slip surface was established by satisfying the Mohr–Coulomb failure criterion, and equations involving the forces on a typical slice were formulated in accordance with the force and moment equilibrium conditions. An iterative procedure was presented for computing the lateral forces on the wall by adjusting the inclination of the slip surface until the stress condition on the upper boundary of the transition zone was satisfied, resulting in the two families of slip lines. Several examples demonstrate the slip-line field configurations, and the computed earth-pressure coefficients were found to agree fairly well with those of other numerical methods.

Practical Applications

The determination of earth pressures and the associated critical failure surface play an important role in the design of retaining walls and other problems of geotechnical engineering. Conventional methods are frequently used for this purpose, but the errors involved are sometimes unacceptably large in cases of inclined rough walls. Hence, certain numerical methods have been employed, which, despite their sophistication, have not yet approached exact solutions. This paper presents a simplified numerical procedure for approaching theoretical slip-line fields and earth-pressure coefficients for the most frequently encountered cases of inclined rough walls with inclined cohesionless backfills. With this method, no assumptions need to be made about the shape of the slip surface or the inclinations of the forces acting on the boundaries of the triangular slices, into which the potential slip zone is divided. Furthermore, both the stress equilibrium condition and the Mohr–Coulomb criterion were completely satisfied within the failure zone and on its boundaries. An effective iterative procedure was provided for determining the slip surface location, which converged rapidly, resulting in an approximately theoretical configuration of slip-line field and associated earth-pressure coefficient values if a sufficiently large number of triangular slices were employed.

In practical engineering, slopes near reservoirs are susceptible to collapse during the rising and falling of water levels, resulting in huge financial losses. Three-dimensional (3D) upper-bound analysis is an effective vehicle to assess slope stability under water drawdown. However, in previously published 3D upper-bound analysis, the pore water pressure distribution caused by drawdown is often approximately determined by employing a pore-pressure coefficient, which is not theoretically sound and fails to give rigorous upper-bound estimations of slope stability. To overcome this shortcoming, the hydraulic head distribution of a slope subjected to drawdown was determined numerically using seepage flow calculations. The obtained hydraulic head distribution was subsequently incorporated into the 3D rotational failure mechanism of the kinematic approach of limit analysis, so as to deliver an upper-bound solution to slope safety factors. To validate the proposed approach, a case study on the Chenjiawan slope at the Three Gorges Reservoir, China, and comparisons with numerical calculations and previous studies are performed. Four different drawdown conditions are considered in this study and corresponding stability charts are provided for directly assessing the safety factors of slopes subjected to different drawdown conditions. The effects of different drawdown processes on the slope’s stability are studied, showing the unfavorable effect of the external drawdown process and the beneficial effect of the internal drawdown process.

Shallow foundations are commonly resting on partially saturated soil and are very likely to be exposed to severe variations of temperature. The degree of saturation and matric suction will be changed at the elevated temperatures, thus resulting in different hydromechanical behavior and a considerable variation in the imposed suction stress. In this study, the effect of temperature increase on the ultimate bearing capacity of strip shallow foundations resting on partially saturated clay layer subjected to vertical (V)–horizontal (H)–moment (M) combined loading is examined through a set of finite-element limit analyses (FELA) adopting lower bound theorems and second-order cone programming (SOCP). The unified effective stress for partially saturated soils is first incorporated into the soil yield function, considering the significant effect of temperature variations on suction stress by means of the well-established previously developed nonisothermal soil–water retention curve (SWRC) model. The equilibrium equations corresponding to the combined loading are also incorporated into the FELA formulations to account for the effects of inclined and eccentric loadings. Accordingly, the failure loci of obliquely and eccentrically loaded shallow foundations are presented for a wide range of temperatures. It is shown that with an increase of the temperature in the underlying unsaturated clay, the suction stress within the soil medium significantly rises, and thus, the failure wedges as well as the failure loci for both eccentric and inclined loadings substantially expand at elevated temperatures, revealing the considerably greater bearing capacity of the strip footing. The contact pressure beneath the foundation in all loading scenarios increases with an increase in the temperature of the underlying clay. This trend is much less pronounced at higher angles of load inclination, while it is barely affected by the load eccentricity. In addition, with an increase in the induced temperature and consequently the suction stress within the partially saturated clay beneath the foundation, the failure wedge expands, thereby resulting in an overall increase in the bearing capacity of the shallow foundation against vertical–horizontal–moment combined loading.

This paper develops an innovative undrained solution for cylindrical cavity expansion using a hypoplastic model that could well reflect the mechanical properties of clay, especially the structural degradation of unstable clay. Then, the constitutive relationship of the hypoplastic clay model for cavity expansion problems is established by adding the strain rate–displacement relationship in cylindrical coordinates. Therefore, simple first-order ordinary differential equations are obtained to solve the problem of cavity expansion for structured clay. This paper compares the proposed solution with the Abaqus finite-element solution and conducts an extensive analysis of the model parameters. The results prove that the solution can accurately capture the expansion responses under undrained conditions. The variations and distributions of the internal cavity pressure, excess pore pressure, and sensitivity in unstable structural clay are obtained, which show obvious trends of structural degradation. The normalized expansion pressure and excess pore pressure decrease as the value of OCR increases. The normalized cavity pressure and structural degradation radius decrease as the structural parameters increase, while the opposite change is observed in the excess pore water pressure. The results of the application analysis can be used as a reference for clarifying the variation in the bearing capacities of piles and the disturbance of surrounding soil in practical engineering.

In the field of rock mechanics, the stress–seepage–damage coupling numerical simulation of rock has always been a hot topic but a difficult problem. Based on this background, we newly derive the smoothed particle hydrodynamics (SPH) form of the seepage equation and a two-dimensional (2D) stress–seepage–damage coupling constitutive model. The proposed coupling model considers the heterogeneity of the engineering rock mass and overcomes the difficulty of the conventional SPH algorithm in stress–seepage–damage calculation. Stable one-dimensional (1D) seepage is first simulated to verify the correctness of the seepage equation and that the simulation results coincide with traditional analytical solutions. Then two triaxial compression experiments considering seepage and no-seepage conditions are simulated to show that the coupled model can well simulate the progressive rock failure process and the change in permeability. The correctness of the coupling model is verified by comparing the simulated results with experimental results. The existence of a seepage field advances the initial cracking time and weakens the peak strength of the rock. Finally, the progressive damage processes of surrounding rock excavation under unloading action are simulated, showing that stress–seepage–damage coupling model has application prospects in rock engineering. The research results may provide some references for the application of the SPH method in the stress–seepage–damage coupling simulation of rock.

Abstract

Excavation usually involves the three-dimensional behavior of the supporting wall and nearby ground as indicated by wall and ground movements. In this study, wall deflection and ground settlement behavior with excavation were investigated, focusing on the effect of excavation shape considering various excavation geometries and corner configurations. For this purpose, the three-dimensional finite-element analyses were performed for both regular and irregular excavation shapes. It was found that the wall deflection and surface settlement both increased with an increasing length-to-width ratio for the regular excavation shape. For the irregular excavation shape with a protruding corner, the corner-wall deflection and surface settlement were smaller than the longer-side wall and the regular excavation shape due to the corner stiffening effect. The location of the maximum wall deflection for the longer-side wall shifted to the longer-strut side when a corner existed. The depth of the maximum wall deflection for the longer-side wall was around 0.6 times of the excavation depth and deeper at the corner wall. The primary influence zone of ground settlement behind the wall was 1.0–1.5 times of excavation depth and larger at the corner side due to the combined wall deflection at the corner.

Practical Applications

In this study, the excavation behavior, including wall deflection and nearby ground settlement, for irregular excavation configurations was investigated. It was clearly observed that the wall deflection and ground settlement behaviors for regular and irregular excavation shapes were different. The main differences between the regular and the irregular excavation configurations were seen at the event of a protruding corner, given by the walls of W3 and W4 in the irregular excavation configuration. For the zone of the protruding corner, the depth of the maximum wall lateral deflection became deeper, and the influence range of surface settlement behind the wall was wider than that of the regular excavation configuration case at the same width and configuration. This behavior was attributed to the three-dimensional excavation effect that took place at the protruding corner. Wall deflection and surface settlement profiles were affected by both sides of the corner, and additional strains into the ground were caused. This indicates that adjacent structures near the protruding corner zone can be affected in a wider range and thus should be monitored more carefully. The results and characterized parameters obtained in this study can contribute to further enhancing excavation and wall design as well as to better establishing and networking the field monitoring system. This would be particularly important in urban areas where various types of structures and buildings are heavily located.

Jointed rock masses with different orientations and spacings are widely exposed in rock engineering applications, such as underground caverns, slopes, and tunnels, which significantly influence the rock engineering stability and mechanical behavior of the rock masses. To better evaluate the mechanical response of the jointed rock masses during the engineering excavation period, an equivalent constitutive model for the jointed rock masses was established based on the joint compliance tensor (JCT) and was adopted to analyze the excavation-induced mechanical performance of Jinping II hydropower station underground caverns. First, the calculation method of the joint compliance tensor and equivalent elastic compliance matrix of the rock masses containing multiple joint sets was established considering the deformation characteristics of the joint system comprehensively based on the Oda fracture tensor theory and linear superposition principle. Second, an equivalent constitutive model for the jointed rock masses was established considering the joint spacing and connectivity. Finally, combined with the field investigation and monitoring data, the mechanical mechanism of unloading failure of the jointed rock masses in Jinping II hydropower station underground caverns was analyzed using the proposed model and numerical simulation.

Biostimulated microbially induced carbonate precipitation (MICP) is an innovative soil improvement approach utilizing indigenous ureolytic microbes to hydrolyze urea. The objective of this study is to investigate the effect of biochar amendment on the shear strength of biocemented calcareous sand through the biostimulation approach. Biochar powder produced from softwood forestry residues with a maximum size of 0.045 mm was used. The specimens were prepared at two initial relative densities (i.e., 16.2% and 54.6%) and four biochar contents (i.e., 0%, 0.1%, 0.3%, and 0.5% by weight), and a series of direct shear tests were conducted at five normal stress levels (i.e., 50, 100, 200, 400, and 600 kPa). The cementitious content, peak shear strength characteristics, and microstructural features were determined during and at the end of the tests. The results indicate that the amendment of biochar in biocemented sand could increase its overall cementitious content. It is also found that whether biochar amendment could enhance shear strength depends on the biochar content, cementitious content, and initial relative density. The higher initial relative density (54.6%) and moderate biochar content (0.3%) are favorable conditions for greater sand shear strength enhancement provided by biochar. Microscopically, the biochar powders are embedded within biocementation. Mechanistically the effects of biochar on the shear strength of biocemented sands are (1) increasing cementitious content, (2) creating weak points inside biocementation, and (3) reducing interparticle friction.

Urban underground comprehensive pipe galleries (referred to as UCPGs) often have large rectangular cross sections and are now widely used as underground tunnel spaces for accommodating various engineering pipelines used for power, signal communication, gas, oil, heating, water supply, and drainage channels in China. The force–displacement relation for a pipe–soil interaction is very important since it can be used to evaluate the pipe behavior under various conditions. Conventional theoretical analysis methods are usually proposed for circular cross-sectional pipes, while few methods have been presented for UCPGs with large rectangular cross sections. This paper studies the two-dimensional rectangular pipe–soil interaction problem. The limit resistances of underground rectangular pipe galleries (URPGs) in undrained clay under purely horizontal and vertical loads are first investigated through two-dimensional finite-element limit analysis (FELA). Subsequently, the failure envelope of the URPGs under relative pipe–soil movement along various directions is investigated. Parametric analyses are carried out to derive the empirical equations of the limit resistance and failure envelope of the URPGs. The proposed empirical equations are verified by comparing the theoretical predictions with the FELA results and other researchers’ studies, and they provide a basis for constructing the p–y relation of the pipe–soil interaction.

A narrow backfill zone is formed when retaining walls are built near existing stabilized structures (e.g., rock faces). In such circumstances, the classical passive earth pressure coefficient is no longer applicable, and a correction factor is required for the design. This paper aims to develop analytical solutions for estimating the passive earth pressure problem of narrow cohesive-frictional backfills behind retaining walls. The novel arched differential element method considers both effects of the horizontal shear stress in backfills and the soil arching, and it is employed to estimate the passive earth pressure distribution along with wall depth. The solutions are compared against those published experimental data, analytical approaches, and finite-element limit analysis solutions. The factors influencing the distribution of passive earth pressure are also undertaken using a series of parametric studies. To implement the derived solutions into a routine design, a modified practical design equation is presented following the standard Coulomb’s solutions. This work provides a theoretical guideline for the initial design of retaining walls with narrow soils, and it should be of great interest to practitioners.

Abstract

This paper investigates the critical stability of purely cohesive soil (ϕ = 0) slopes under seismic conditions using a novel concept of the critical slope surface (CSS). A theoretical model is devised on the basis of a unified framework of the method of stress characteristics and the pseudostatic approach. In the presence of pseudostatic seismic forces, a system of differential equations governing the equilibrium stress field is combined with the Mohr–Coulomb yield criterion. Subsequently, the curvilinear CSS derived at the state of critical equilibrium with a factor of safety of 1.0 is endorsed to operate as an autoguided measure for slope stability. Unlike the traditional limit equilibrium and limit analysis methods, the present method does not involve any preordained failure mechanism in the analysis. The CSSs obtained from the present analysis mostly anticipate steeper profiles with lesser height than the traditional linear slopes reported in the literature. In addition, the curvilinear CSS exhibits a certain morphological resemblance with naturally occurring soil slopes. A finite-element simulation of the CSS further establishes the versatility of the proposed concept to design an equivalent linear profile with a prescribed height. A set of design charts based on the CSS concept is provided to mandate a quick and inexpensive stability analysis in practice.

Practical Applications

In slope stability analysis, the optimal inclination of a linear slope often becomes a prerequisite to ensure a sufficient margin of safety by providing a flat slope geometry compared with that required for critical stability with a factor of safety of 1.0. The present critical slope surface-based stability concept ensures critical stability at minimal computational costs. The critical slope surface appears to be a boundary between the zone of stability and instability. Hence, for site-specific soil properties and seismic conditions, flat slope profiles compared with the derived critical slope surface ensure a stable configuration and vice versa. Therefore, the guidelines for designing proposed linear slopes can be orientated in advance to prevent an imminent collapse. In the case of existing slopes, the necessity for a remedial measure to maintain critical stability can also be assessed a priori using the current concept.

In this paper, the peripheral stability of a horseshoe-shaped tunnel in undrained clay with linearly increasing and directionally dependent strength is studied using the lower-bound theorem of plastic analysis coupled with finite-element formulation and second-order cone programming. The present analysis computes required supporting pressure for a tunnel constructed in isotropic/anisotropic normally consolidated (NC)/lightly overconsolidated (LOC) clay with linearly varying undrained strength across the depth, and isotropic/anisotropic heavily overconsolidated (HOC) clay whose undrained strength remains constant throughout the depth. Dimensionless stability charts have been produced for expressing the undrained stability of horseshoe tunnels in terms of anisotropic and heterogeneous parameters, tunnel cover depth, and normalized overburden pressure. The combined influence of these parameters on the peripheral stability of tunnels and associated failure mechanisms have also been examined and discussed. Further, a few analyses were done to examine the advantage of a horseshoe-shaped cross section in comparison with a circular cross section.

Because of its simplicity, cost-effectiveness, and high efficiency, bolt-anchoring technology has been widely used in various fields such as tunnel and slope engineering. In this study, the mechanical behavior and damage pattern of bolted joints are investigated by a direct shear test of bolted joints, considering different bolt numbers, anchoring angles, and arrangements. The results showed that there are five distinct stages in the shear stress–displacement curve of the bolted joint, namely, the compaction deformation stage, the elastic deformation stage, the separation and drop stage, the yield damage stage, and the residual deformation stage. An increase in the bolt number increased the elastic modulus and shear strength of the bolted joint, although the yield deformation of the bolted mass first increased and then decreased. An increase in the anchoring angle caused the shear strength of the bolted joint to first increase and then decrease. The optimal anchoring angle was approximately 45°, and the safety factor was higher under large anchoring angles than under small anchoring angles. In particular, multiple rows of widely spaced bolts arranged perpendicular to the shear direction had significantly higher shear strength than a single row of narrowly spaced bolts placed parallel to the shear direction. These findings offer a guideline for the design of bolted rock mass in engineering projects.

According to the excavation methods of tunneling, a damaged zone may be developed in the rock mass around the tunnel, the mechanical properties of which can be significantly altered. This issue is of great importance in the analyses of both the short-term and the long-term performance of tunnels. In this paper, a closed-form solution is presented to find the long-term-induced stresses and the convergence of the tunnel considering the altered zone development. To account for the time-dependency effect, the Burgers viscoelastic rheological model is assigned to altered and unaltered zones. The proposed method is applicable for the case in which the tunnel excavation is halted or for the case in which the tunnel is continuously excavated. The predictions made by the proposed analytical method are compared with those made by using FLAC^3D software. Then, a parametric study is performed to investigate the effects of different key parameters associated with the research aim. Finally, the numerical simulation is extended so that the combining Burgers substance and the Mohr–Coulomb failure criterion (CVISC) model, i.e., a viscoplastic rheological model, is adopted to the medium. The results indicate a remarkable difference between the predictions of tunnel convergence by the Burgers and the CVISC models, in very severe and extreme squeezing conditions.

This study attempts to evaluate the effects of two important structural design parameters on the dynamic behavior of back-to-back mechanically stabilized earth walls (BBMSEWs). For this purpose, a series of shake table tests were carried out on seven BBMSEW models with different overlap lengths of reinforcements (L_R) and arrangements of connecting two opposing walls to each other. Based on the results of the particle image velocimetry (PIV), a combination of a concave surface and an inclined plane was observed as the failure surface in models and it was concluded that the complete connection of two opposing walls to each other could prevent the formation of a failure wedge in BBMSEWs. Findings indicated that although separating two opposing walls from each other and reducing the overlap L_R increase lateral deformations, these could be two effective solutions for reducing the lateral earth pressure in BBMSEWs. On the other hand, comparisons have showed that Seed-Whitman and Mononobe-Okabe methods are conservative for calculating dynamic lateral pressures (σ_AE) in BBMSEWs at HPGA < 0.6g (where HPGA is horizontal peak ground acceleration) and that this conservatism fades gradually by increasing HPGA, and finally at HPGA ≥ 0.9g, the obtained values of σ_AE becomes almost greater than those predicted by these methods.

The mechanical behavior of the interfaces between fine sea sand and offshore structure is regarded as a quite important issue that influences the foundation-bearing capacity and structural safety. In this work, by employing a modified shear apparatus with a ring-transparent sample container, a series of interface shear tests were conducted to shed light on the effects of the applied normal stress, initial relative density, and the roughness of the steel surface. Based on image analysis, the horizontal deformation at various vertical positions within the sand specimen was captured. In addition, the postshearing particle-size distributions of sand specimens were measured and compared. From the acquired test results, it was demonstrated that for the dense and medium dense sand–steel interfaces, strain softening is a typical shear behavior, regardless of the roughness of the steel surface. The mobilized shear stress of the interface with a smooth steel surface reaches the peak strength earlier than that of the rough one and then rapidly declines to residual strength within a short softening stage. The shearing zone in sand specimens can be divided into a large deformation zone and a small deformation zone. The shear deformation that occurs within the sand specimens sheared against the smooth steel surface is negligible and concentrated very close to the interface. However, for rough surface and dense sand, the interlocking between sand and steel forces the deformation of sand specimens to arise in greater scope and magnitude, accompanied by significant particle crushing.

A box culvert is a tunnel under roadways or railways that transports open drains or streams. The geotechnical design of such a structure involves accurate evaluation of the earth pressure on the structure. This short technical note pertains to findings of stability factors in tables and charts as an aid to civil engineers in designing unlined square box culverts. A conventional equation that is similar to Terzaghi's bearing capacity factors (N_c, N_s, and N_γ) is adopted to examine the use of stability factors (F_c, F_s, and F_γ) to estimate the critical support pressures at collapse. Rigorous upper-bound and lower-bound solutions of the stability factors are computed using the adaptive finite-element limit analysis. To validate the present bound solutions, numerical results are compared with those published previously using other methods. Design tables, figures, and examples are presented for practical uses.

Abstract

Compositional characteristic in loess is a key factor controlling its collapsibility. Clay cementation between particles plays an important role in loess collapse. In this study, the type, content, properties, and occurrence mode of clay minerals in Malan loess were fully investigated by X-ray diffraction and scanning electron microscopy combined with energy-dispersive spectroscopy. The thickness of clay-bound water was determined by isothermal adsorption experiments and the pycnometer method. The results indicated that interstratified illite–montmorillonite and illite are two primary clay minerals in the studied loess. In particular, the illite has experienced serious degradation with low crystallinity and considerable swelling layers. Clay minerals, mixed with skeleton particles of various sizes, mostly exist in the form of aggregates and cementation between particles. The relation between the adsorbed water content of clay minerals and relative humidity (RH) was obtained; when the RH was 99%, the water content reached 5.74%, and the corresponding thickness of bound water was 10.51 Å, which is equivalent to the thickness of nearly 3.8 water molecules and as thick as montmorillonite itself, which demonstrates the high hydration swelling potential of clay minerals in the studied loess. Therefore, the hydration swelling of clay cementation was illustrated to be the primary trigger of loess collapse. This was further confirmed by single-oedometer-collapse tests using different wetting solutions with different polarities. This study highlights the important role of clay minerals in loess collapse, which contributes to further understanding of the collapse mechanism.

Practical Applications

Loess collapse is a focus issue that needs to be considered in engineering construction and disaster prevention. A full interpretation of the collapse mechanism can help us to take effective measures to reduce the collapse amount or probability of collapse. In the study, hydration swelling of clay cementation is an important factor causing a decrease in strength between particles, based on a detailed and thorough investigation of clay minerals such as properties and hydration reactions; therefore, loess soil can be effectively improved by mixing with some materials or solutions that can restrain hydration swelling of clay cementation from reducing collapse amount. In addition, loess collapsibility may be quantitatively evaluated more accurately using mathematical methods by establishing a systematic database that contains information on clay minerals and other factors such as microstructural features in different loess regions.

Most slope collapses have three-dimensional (3D) characteristics due to the uniform soil. The traditional plane-strain analyses neglect the 3D effects and could yield an overestimation or underestimation of the slope stability. The present study analyzes the stability of 3D slopes characterized by spatially variable undrained shear strength. A 3D slope reliability analysis approach that can fully consider the spatial variability of soil properties is proposed based on the plastic limit analysis upper-bound theorem. Within the framework of Monte Carlo simulation, the effects of strength gradient, the coefficient of variation, and the scale of fluctuation of soil undrained shear strength on the slope reliability index are examined. The length of the 3D failure surface is highlighted as an essential characteristic in 3D slope analysis. The slope failure mechanisms are classified, and the occurrence probability of each type is quantified. The obtained results indicate that the degree of nonstationarity and the coefficient of variation of undrained shear strength have significant impacts on the slope failure width, and the variation of failure mechanism highly depends on the combined effect of slope inclination and soil strength. The findings could guide slope reinforcement and facilitate a more efficient and accurate risk evaluation of slopes.

This paper presents a probabilistic study of anchor–soil bond stress performance in a post-grouted anchor wall. The variability of geotechnical shear strength parameters will be described by a probability density function (PDF) and as a linear and squared shear strength function with depth based on widely available standard penetration test (SPT) data on two geologies (mudflows and residual soils). Monte Carlo simulation was carried out to evaluate the mean and standard deviation of ultimate anchor strength. The study allows the factor of safety (FS) to be determined to achieve a target probability of failure (p_f). The first analyses consist of a probabilistic analysis without considering the shear strength varying with depth (SSVD), while the second one includes SSVD. Probabilistic analyses (without SSVD) show that a reduction of the p_f depends only on the design FS. The probabilistic SSVD analyses showed that the N-value at the surface (N₀), and the function gradient (ρ) had an important influence on the anchor p_f. It is shown that the use of an overall FS can be considered as a high-risk design for soils with a high N₀ variability and low ρ. Therefore, the target p_f shows that the top anchors are the most critical, so higher FS values are required (FS > 2.5), especially for mudflow soils.