International Journal of Geomechanics

Particle breakage modifies the structure of rockfill material and influences its constitutive character, which is the key factor in earth-rockfill dam deformation control. However, the particle breakage laws for rockfill material under real stress paths during dam construction and impounding remain unclear. This paper investigates the particle breakage of basalt rockfill, which was used in the Gushui concrete-face rockfill dam, during drained triaxial compression under equal stress ratio paths (construction stress paths) and transitional stress paths (impounding stress paths). The results indicate that the type of stress path significantly influences rockfill particle breakage. Under the equal stress ratio path, particle breakage increases with shear strain and the growth rate decreases with the stress ratio. The relationship between the relative particle breakage index and the plastic work can be simulated using a hyperbolic formulation. Under the transitional stress path, particle breakage also increases with shear strain and the growth rate is influenced by the transitional confining pressure, the transitional stress path, and the equal stress ratio experienced before the transition. However, the relative particle breakage index cannot be reasonably simulated using the plastic work.

The design and construction of a tunnel in urban areas is a significant undertaking that requires ensuring ground movements, preventing the adverse impact on adjacent structures. However, the conventional methods are always deterministic, rather than taking the natural spatial variability of soil properties into account. In this study, the spatial variability of the soil profile was modeled using a two-dimensional random field and coupled with numerical analysis to obtain a probabilistic characterization of surface deformations. It was assumed that random fields were generated on the basis of the covariance matrix decomposition method in a manner consistent with a specified numerical mesh. A series of parametric analyses were conducted to study the effects of uncertainty due to the variability of soil Young’s modulus on ground movements. The results demonstrate the presented computational framework to be effective in the assessment of surface deformations. The soil variability exerts the influence both on the magnitude and distribution of surface deformations, which depends on the magnitude of the soil variability. Furthermore, it is concluded that negligence of the variability of soil properties in the probability analysis of the surface settlement can lead to an underestimation of the probability of failure, compared to the traditional deterministic analysis.

This paper deals with the reliability analysis for the elastic settlement of the surface strip footing resting on cohesionless soil using the concept of fuzzy set theory in conjunction with the finite-element method (FEM). Three types of soil profile (homogeneous, inhomogeneous, and layered) are considered for the analysis. The elastic modulus of soil, $E_s$, is considered as a fuzzy variable maintaining a certain membership function, i.e., triangular or trapezoidal. The Poisson’s ratio ($\upsilon$) is kept constant throughout the analysis. Spatial variability of the soil parameter is taken into consideration for both isotropic and anisotropic scale of fluctuations. Depending upon the soil type, variation of the probability of failure ($P_f$) for different coefficients of variation (COVs), different modular ratio, and different COV ratios of $E_s$ is examined. A parameter ($s_{\lim }^0$) indicating the required limiting settlement to obtain zero failure probability is proposed to account for various soil conditions, which is more significant for the design perspective. Results indicate that the failure probability ($P_f$), more precisely $s_{\lim }^0$, increases as the COV of $E_s$ increases. Moreover, the influence of the correlation length on the $s_{\lim }^0$ value is also highlighted, and it is observed that the $s_{\lim }^0$ values are more sensitive to the horizontal scale of fluctuation (${\delta }_h$) than the vertical scale of fluctuation (${\delta }_v$) for the anisotropic case.

Identification of three-dimensional rock blocks of rock masses is a key technique in rock mechanics and rock engineering. The boundary representation (BR) approach could identify generally shaped blocks, but it is not easily understood by practicing engineers as a result of its combinatorial topology theory. In this paper, a novel spatial topology identification algorithm of the three-dimensional complex block system of the rock masses is proposed. In this algorithm, first, three-dimensional joint networks are simulated using the geological statistical results. Second, the basic data structure of a complex block including the vertex, oriented edge, oriented loop, oriented face, and oriented body is defined. Last, four key techniques, including integrations of line segments, topology analysis of the oriented loops, forming the oriented faces, and topological identification of the rock blocks, are detailed. Based on the cutting algorithm, a program named Rock Block Cutting (RBC) was developed, and the simulation results of serval analytical and classical examples using RBC prove the correctness of the program. Applications of the RBC program in both the slope and cavern examples show that this spatial topology identification algorithm can efficiently calculate the convex rock blocks, concave rock blocks, and band the complex rock blocks with multiply connected characteristics, which provides the geometric model for the key block analysis and the stability calculation of the discrete block system.

This paper presents the experimental program conducted for evaluation of the rate-dependent and stress-relaxation behavior of unsaturated reconstituted London clay. A series of drained constant rate of strain (CRS) compression-relaxation tests with single-staged (SS-CRS) and multistaged (MS-CRS) loading modes was performed in an innovative CRS oedometer cell where soil suction evolutions were monitored using two high-capacity tensiometers (HCTs). Specimens were tested at two strain rates of $4.8\times {10}^{-7}$ and $2.4\times {10}^{-6}{\mathrm{s}}^{-1}$ and over a suction range of 0–1,905 kPa. The coupled and independent effects of strain rate and soil suction on one-dimensional stress–strain and stress-relaxation responses, including the effects of prerelaxation strain, stress, and strain rate under both saturated and unsaturated conditions, were evaluated. An increase in suction and strain rate resulted in an increase of the yield vertical net stress (${\sigma }_p$). Furthermore, it was observed that the rate and magnitude of the relaxed stresses increased with increases in prerelaxation strain, stress, and strain rate, and decreased with an increase in soil suction. At constant suction, an increase in the prerelaxation strain rate by a factor of five resulted in an increase of the relaxed stresses by a factor of 2.2–3.6. Moreover, the coefficient of relaxation ($R_{\alpha }$) was found to be suction dependent, falling within a range of 0.011–0.019 and 0.017–0.029, respectively, for slow and fast strain rates during MS-CRS tests. Comparing these results with the $C_{\alpha }/C_c$ ratio obtained from conventional multistage loading (MSL) oedometer test results revealed the validity of $R_{\alpha }=C_{\alpha }/C_c$ correlation for unsaturated reconstituted specimens.

When analyzing slopes stabilized by piles, the soil has been commonly considered dry or saturated, whereas soils are usually unsaturated in practice. Suction stress is caused by positive and negative pore water pressures, changing the strength properties of soils compared with soils without considering it and, consequently, leads to various stability conditions of a slope. In this study, a unified formula is presented for estimating soil-pile pressure in which the effect of suction stress is considered in variably saturated soils, which is an important extension of Ito and Matsui’s classical equation. Under such proposed formulation, an attempt has been made to develop a new method by means of the upper-bound limit analysis for analyzing the capacity of an unsaturated slope with vertical steady unsaturated flow reinforced by piles, and the most critical solutions for the optimal limiting surcharge are obtained by the optimization method. By comparing the results of the proposed method with other theoretical and experimental results in some examples, the method’s feasibility is verified. A numerical study for the parametric effects is also provided in this paper, which shows that the limiting surcharge is largely determined by soil properties, slope geometry, as well as seepage conditions. From an engineering application aspect, cost savings could be possible for the construction of slopes if a proper assessment of the effect of suction stress is completed.

Translational failure of the multilayered cover system (MLCS) is a key issue encountered in waste containment facilities. This study extends the existing force-equilibrium method for analyzing the stability of a uniform cover soil overlying the geosynthetic layer of the MLCS. In the extended analysis, the factor of safety (FoS) is determined at all the interfaces of the MLCS by considering the changes in dimensions of both active and passive wedges contributing to stability. Anticipating the possibility of multiple failure planes, it is important to evaluate the FoS along every interface. The stability of MLCS, influenced by slope configuration, material characteristics, and specific conditions experienced during its life cycle, was studied and the corresponding design curves were developed for a Resource Conservation and Recovery Act (RCRA) Subtitle C cover system. The stability of different layers under seepage induced by heavy rainfall and earthquake forces was investigated. The study indicates that the changes in passive wedge for different layers of the MLCS have a significant influence on FoS. The possibility of relatively higher compaction of the passive wedge without affecting the functionality of MLCS has been discussed for improved stability. The suggestions for modifications in material characteristics for improved stability of MLCS are reported in this study. The results from this study were used to develop a vulnerable interface diagram (VID) to provide a comprehensive understanding of the most vulnerable interface of MLCS. The extended translational stability analysis and the resulting VID find its application in multilayered soil slopes in reinforced highway/railway embankments, earthen dams, and natural slopes with distinct layers of soils and rocks.

This paper presents a laboratory experimental investigation of the behavior of coral sands. The investigation incorporated particle breakage to interpret its influence on consolidation and shear behavior, friction and dilatancy behavior, excess friction behavior, and state behavior of precrushed sands as measured by triaxial tests on a coral sand. Particle breakage impaired dilatancy, leading to a more contractive soil. It reduced friction and dilatancy angles, but increased basic friction angles. It also showed a complex influence on excess friction angles, revealing underestimation of the peak-state dilatancy angle by the excess friction angles. Particle breakage resulted in movement of the phase-transformation and peak states toward reduction in deviator stress, mean effective stress, and void ratio in the $e\text{-}\log p^{\prime }$ and $q\text{-}{p}^{\prime }$ planes. In the $e\text{-}\log p^{\prime }$ plane, it resulted in translation, rotation, or translation and rotation of the critical-state line. In the $q\text{-}{p}^{\prime }$ plane, particle breakage resulted in rotation of the critical-state line, incorporating lower-left movement of the critical states.

Does the overconsolidation ratio have any effect on the cone factor value used for determining the undrained shear strength? In order to find the answer, first the numerical modeling of a cone penetration test (CPT) in clay is performed in this study, and then the obtained results are used to achieve a reliable analytical solution for determining the undrained shear strength. In this regard, the numerical simulation is verified by some existing laboratory tests. Then the observations obtained from the comprehensive numerical analyses are investigated. Based on the numerical simulation results for the region located around the cone tip, the velocity field is formulated and the stress states are studied. According to the analytical solution, the vertical stress distribution is obtained by solving the differential equation of motion for the soil located around the cone tip. Determining the corrected cone tip resistance through integrating the vertical stresses applied to the cone tip surface, two new relationships are subsequently presented for the cone factor in terms of the initial conditions of clay. Based on the proposed solution, the effect of overconsolidation ratio on the cone factor, which was unclear as yet, is specifically determined. Finally, the cone factor relationship developed in this study is compared with the existing relationships in the literature as well as the field measurements.

When a retaining wall is built near a building or a mountain, the backfill width is often limited. It limits the use of the classical theoretical formula based on the assumption of an infinitely large half-space. Finite-element analysis shows that when the retaining wall movement is translational, one to three slip surfaces will develop from the wall heel, and eventually develop to the surface along the soil–wall or soil–rock interface. An effective model for calculating the active earth pressure of narrow granular backfill against the wall under translational mode is established by taking into account the inclination angle of the rock face and the friction angles of the soil–wall and soil–rock interfaces. The active earth pressure under the failure mechanisms of one to three slip surfaces are given by using the limit equilibrium method and the whole analysis method of sliding wedge. The coefficient of active earth pressures is as low as 0.5–0.7 times the Coulomb solution and the location ($h_0/H$) of the resultant of active earth pressures ranges from 0.34 to 0.44. The influence of different parameters on the number of slip surfaces was also studied by parametric analysis.

Real dynamic compaction (DC) sites are composed of a series of tamping locations. However, previous studies mainly focused on single-location tamping and paid little attention to the influence of adjacent tamping locations. This paper proposes a method for estimating the degree of final soil improvement with depth in the compacted area between adjacent tamping locations. Extensive numerical parametric studies were conducted to investigate the influence of each parameter on the degree of soil improvement. Based on the results obtained, dimensionless correlations between the relative density increment ratio at certain depths and combinations of soil and construction parameters were performed with the dimensional analysis method. Then, the dimensionless correlations, together with the crater depth and depth of improvement, were used to determinate a prediction curve to describe the variation in the degree of improvement with depth along the midline between adjacent tamping locations. The reliability of the proposed formula was examined by comparing its predictions with the measured results. Finally, the applicability of the proposed formula for the determination of construction parameters was illustrated using the Changi DC treatment project.

A coal storage wharf located in Zhenjiang, Jiangsu province, China, has a planned annual capacity of 15 million tons. The thick Yangtze River floodplain is rich in muddy-silty clay with high compressibility, low bearing capacity, and poor stability, which needs to be improved to meet engineering requirements. In this paper, a case study on soft clay treatment with the vacuum preloading method (VPM) and combined vacuum and surcharge preloading method (CV-SPM) is introduced and discussed. Negative pore-water pressure was induced by vacuum pressure in the VPM, which led to the seepage of water and consolidation of soils. However, the vacuum pressure could never exceed 101,325 Pa (1 atm), so the additional filling surcharge was usually adopted to raise the preloading load in the CV-SPM when the bearing capacity requirement was higher. To perform ground improvement in the field, first in situ soil samples were taken from the construction site before the treatment, and the properties of soils were investigated through laboratory tests, which were used as the reference to analyze the effect of ground improvement. Second, during the construction process vacuum pressure, ground settlement, and pore-water pressure as well as lateral displacement of soils were monitored and analyzed. Finally, vane shear tests and shallow plate bearing tests were carried out to investigate and evaluate the effect of ground improvement. Test results show that the undrained shear strength increased remarkably, and the bearing capacity reached the design requirements after around 100 days of preloading, indicating that VPM and CV-SPM are appropriate and effective for ground improvement of the thick muddy-silty clay laying in the Yangtze River floodplain.

Although commonly ignored in conventional design, cost savings could be possible if a proper assessment of the effect of matric suction in retaining structures with unsaturated backfill were completed. In this study, an analytical method was developed based on the limit equilibrium (LE) method and the assumptions of the plastic soil mass and planar slip surfaces. The proposed method can calculate the distributions of passive earth pressure, passive thrust, and their points of application under unsaturated conditions, which can consider the effect of apparent cohesion and adhesion within the framework of the suction stress-based effective stress approach. Furthermore, an unsaturated interface shear strength model was validated by existing experimental data and used in this method. The parametric studies were performed under the one-dimensional (1D) steady flow condition to investigate the effects of the soil friction angle, surcharge pressure, interface friction angle, backfill inclination angle, and seepage condition on passive earth pressure distributions and the air entry pressure parameter on the coefficient of total passive thrust and its point of action. Both the Coulomb and Rankine methods can be regarded as special cases of the proposed method under certain conditions. The passive earth pressures estimated by proposed method were then verified against the existing experimental data as well as the predictions by the extended Rankine and Coulomb theories.

In this paper, a noncoaxial plane strain soil model is developed in the framework of initial soil strength anisotropy that is described by taking the internal friction angle to be a function of principal stress orientations. The conventional Mohr-Coulomb (M-C) yield criterion is generalized to give an anisotropic yield criterion, with the curve in the deviatoric stress space forming an ellipse. Both rotational and eccentric ellipses are discussed. The formulation of noncoaxial constitutive equations is described by a general form in terms of the plastic strain rate. In this form, the plastic strain rate is divided into two parts: the conventional component that is derived from the classical plastic potential theory and the noncoaxial component that is assumed to be tangential to the yield surface. The newly proposed model is validated by the analytical calculations and discrete element modeling (DEM) simulation results in simple shear tests. Conclusions can be drawn that this model is generally capable of capturing the DEM observations of simple shear testing.