Acta Geotechnica最新文献

The mechanical properties of clay minerals are largely dependent upon the chemical compositions and the mesoscale fabrics of the constituent particles. This paper describes results of a series of mesoscale molecular dynamics simulations of the hydrostatic compression and shear strain behavior for initially randomly oriented assemblies of 10³ illite primary particles. The particles are simulated as rigid-body ellipsoids that interact through the single-site, Gay–Berne potential function. This corresponds to a coarse-grained model based on prior atomistic scale computation of the potential of mean force for water-mediated interactions between pairs of particles through free energy perturbation method. We investigate the mesoscale fabrics of the NPT-equilibrated assemblies for confining pressures ranging from 1.0 to 125 atm, including path dependence associated with unloading and reloading. We analyze and quantify the geometric arrangement including particle orientation, specific surface area, properties of particle stacks/aggregates, and interstack pair correlation functions. The compression of each particle assembly is associated with large irrecoverable changes in void ratio, while unloading and reloading involves much smaller, largely recoverable volumetric strains. The results are qualitatively similar to macroscopic compression behavior reported in laboratory tests. We simulate the uniaxial and shear behavior at each of the equilibrated pressure states through a series of strain-controlled steps, allowing full relaxation of the virial stresses computed at each step. The simulations investigate directional and path dependence of the shear behavior for strain deviations up to 0.2%. The results show the onset on nonlinear stiffness properties at strain levels (sim)0.01% and hysteretic behavior upon unloading and reloading. Small-strain stiffness properties of the particle assemblies are qualitatively in good agreement with quasi-static, elastic stiffness properties reported for illitic clays.

Anhydritic claystones are widely distributed in the Gypsum Keuper formation. Their swelling is associated with the chemical process of anhydrite to gypsum transformation and has caused extensive damages in tunnels. Even though this problem has attracted great scientific interest, an adequate mathematical description of the swelling of anhydritic rocks is still missing. The present paper contributes towards closing this gap by formulating a coupled chemo-mechanical constitutive model, which considers anhydritic rock as an elastoplastic porous medium according to the principle of effective stresses, with a Mohr–Coulomb yield criterion, a non-associated flow rule and an additional, chemically induced strain component. The volumetric chemical strain is equal to the sum of the changes of the volume of the solids and of the pore volume. The change of the volume of the solids depends on the stoichiometry of the chemical reaction and is proportional to the mass of the transformed anhydrite. The pore volume may increase or decrease during the anhydrite to gypsum transformation, depending on how gypsum grows. The pore volume increases if the gypsum crystals crack and expand the matrix, and decreases if the gypsum crystals precipitate within the available pore space. The proposed model considers experimental results according to which the higher the stresses and porosity, the lower the increase in pore volume. In addition, the model assumes that the chemical strains are coaxial with the principal stresses and that the volumetric chemical strain in each principal direction is inversely proportional to the corresponding principal stress. The model is calibrated with results of tests on artificial anhydrite-kaolin specimens and achieves a very high correlation degree (R² = 0.92).

The efficiency of alkali-activated ground granulated blast furnace slag in stabilizing dredged sediments with high water contents is suboptimal because the activators become diluted. To improve stabilization efficiency, additives such as nano-CaCO₃ are proposed. However, some of the proposed additives may not be practical owing to their high costs. This study experimentally investigates the addition of Na₂CO₃ for the stabilization of dredged sediment with high water contents (i.e., 100%) using Ca(OH)₂-activated slag. Experimental results show the optimal content of Na₂CO₃ to obtain the highest 28-day unconfined compressive strength of stabilized sediments is 0.2% gravimetrically. Below the optimal content, the strength increases with Na₂CO₃ content. Above the optimal content, a decrease in strength is observed. By examining the reaction products and microstructure of the stabilized dredged sediments, it is observed that the coupling mechanism of cation exchange and calcite precipitation promotes the development of finer capillary pores, leading to a reduction in interpore connectivity and lower structural heterogeneity of the fine capillary pores. Experimental evidence from this study broadens the practical applications of sustainable soil stabilization using additives.

Accurate estimation of soil properties is crucial for reliability-based design in engineering practices. Conventional empirical equations and prevalent data-driven models rarely consider uncertainty quantification in both measurement and modelling processes. This study tailors three uncertainty quantification methods including Bayesian learning, Markov chain Monte Carlo and ensemble learning into data-driven modelling, in which support vector regression is selected as the baseline algorithm. The compression index of clay is adopted as an example for model training and testing. In this context, Bayesian learning and Markov chain quantify uncertainty by considering the distribution of function and hyper-parameters, respectively, while different sampled data are employed to explore model uncertainty. These models are evaluated in terms of accuracy, reliability and cost-effectiveness and also compared with Gaussian process regression, etc. The results reveal that based on built-in structural risk minimization, sparse solution and uncertainty quantification, developed models can capture more accurate and reliable correlations from actual measured data over other methods. Their practicability and generalization ability are also verified on a new creep index database. The proposed probabilistic methods are also compiled into a user-friendly platform, showing a significant potential to enrich the data-driven modelling framework and be applied in other geotechnical properties.

Granular flow is a typical process that occurs in sediment disasters, including rockfalls, avalanches and landslides, etc. The runout distance in granular flow is closely associated with the ultimate impact range of these sediment disasters. However, this factor is often highly sensitive to various physical parameters and exhibits significant randomness. Hence the study of granular flow is crucial to elucidating the mechanism of such disasters and even to disaster prevention and mitigation. In recent years, a numerical simulation called discrete element method (DEM) that simulates at the particle level has been widely used in this field. Based on the above situation, this study aimed to capture the critical DEM input parameter combinations for risk assessment in a four-dimensional parameter space considering the particle size distribution. XGBoost feature importance is employed to decide the search priority, and its results indicate that the friction angle with bottom surface (FABS) and coefficient of restitution (COR) are the key parameters. The two key parameter spaces were then comprehensively explored using Gaussian process regression response surfaces. The correlation between the FABS and runout distance appeared as a convex function. The COR exhibited diverse degrees of approximately linear correlation with the runout distance throughout the granular flow. The particle size distribution indirectly led to inconsistencies between the bidisperse flow and other granular flows in the influence mechanisms of the key parameters. By clarifying this effect, we efficiently identified two critical parameter combinations for granular flow DEM simulation.

In cold regions, the frozen soil–rock mixture (FSRM) is subjected to cyclic loading coupled with freeze–thaw cycles due to seismic loading and ambient temperature changes. In this study, in order to investigate the dynamic mechanical response of FSRM, a series of cyclic cryo-triaxial tests were performed at a temperature of −10 °C on FRSM with different coarse-grained contents under different loading conditions after freeze–thaw cycles. The experimental results show that the coarse-grained contents and freeze–thaw cycles have a significant influence on the deformation properties of FSRM under cyclic loading. Correspondingly, a novel binary-medium-based multiscale constitutive model is firstly proposed to describe the dynamic elastoplastic deformation of FSRM based on the coupling theoretical framework of breakage mechanics for geomaterials and homogenization theory. Considering the multiscale heterogeneities, ice-cementation differences, and the breakage process of FSRM under external loading, the relationship between the microscale compositions, the mesoscale deformation mechanism (including cementation breakage and frictional sliding), and the macroscopic mechanical response of the frozen soil is first established by two steps of homogenization on the proposed model. Meanwhile, a mixed hardening rule that combines the isotropic hardening rule and kinematic hardening is employed to properly evaluate the cyclic plastic behavior of FSRM. Finally, comparisons between the predicted results and experimental results show that the proposed multiscale model can simultaneously capture the main feature of stress–strain (nonlinearity, hysteresis, and plastic strain accumulation) and volumetric strain (contraction and dilatancy) of the studied material under cyclic loading.

This paper deeply couples the exponential-type nonlinear strain softening with the anisotropic method of microstructure tensor combined stress invariants, proposing an effective strength formula that reflects the anisotropy evolution of soil. Furthermore, an expression for the anisotropy ratio k of strength as an equivalent plastic strain-related variable is derived. For natural clay, this evolution of strength anisotropy is incorporated into the Mohr–Coulomb-matched Drucker–Prager (MC-matched DP) yield criterion within the Cosserat continuum framework, resulting in a more refined soil constitutive model. The main strength parameters required for this model can be conveniently obtained based on conventional soil tests, and the model functionality can be degraded through parameter adjustments. The detailed procedure of stress updating algorithm and the elastoplastic tangent modulus matrix are provided for the constitutive integration. Through the finite element implementation, the superiority of the model is demonstrated compared with existing literature. Also, a biaxial compression example is systematically analyzed to prove that the model can effectively reflect the sensitivity of soil to loading direction. Moreover, the evolution of the shear band morphology, particle rotation in the shear band, and the anisotropy degree presented by the model are consistent with previous experimental studies and discrete element method (DEM)-related literature results. Furthermore, the proposed model effectively addresses numerical convergence issues and mesh size dependence usually encountered in classical models during the simulation of strain localization occurred in the soil.

A new simplified method of analysis is proposed for predicting the final steady state behaviour of sloping capillary barriers subjected to continuous rain of constant intensity. In contrast to an existing simplified method, the proposed new method assumes approximate final steady state suction profiles on vertical cross-sections of the finer layer that are appropriate for sloping capillary barriers, with flow parallel to the slope in the lower part of the finer layer. Numerical validation, performed by hydraulic FE modelling, shows that, in all cases studied, the final steady state profiles of suction, degree of saturation and horizontal seepage velocity predicted by the new simplified method are excellent matches to the corresponding results from FE simulations. As a consequence, values of water storage capacity and water transfer capacity are accurately predicted in all cases, together with the final steady state variation of water stored with horizontal coordinate. A parametric study shows the influence of key variables (slope angle, material of finer layer, thickness of finer layer and rainfall intensity) on water storage capacity and water transfer capacity of sloping capillary barriers.

The layered distribution of hydrates significantly influences the mechanical properties of hydrate-bearing sediments (HBS). A comprehensive understanding of the mechanical and deformation behaviors of layered HBS is essential for the safe and effective exploitation of hydrates. In this study, marine clay from the South China Sea and quartz sand were used to simulate hydrate-bearing clayey-silty sediments, and layered hydrate-bearing clayey-silty sediments (LHBCSS) were prepared. A series of consolidated-drained triaxial tests were conducted, and the results were compared with those from homogeneous hydrate-bearing clayey-silty sediments (HHBCSS) to analyze the differences in mechanical properties and deformation characteristics. The shear strength and deformation behavior of LHBCSS were further investigated. The results show that the layered distribution of hydrates reduces the initial stiffness and strength of HBS, while promoting strain hardening in the specimens. The failure strength of LHBCSS is significantly influenced by the effective confining pressure, with the clay content having no obvious effect. However, the clay content is negatively correlated with the secant modulus (E₅₀). The internal friction angle of the LHBCSS is higher than that of the HHBCSS, and the cohesion of the LHBCSS gradually increases with the clay content. The layered hydrate distribution causes the volumetric strain of HBS to favor shear contraction. The maximum shear dilatation rate of LHBCSS is notably lower than that of HHBCSS, and the clay content has a minimal effect on the critical stress ratio of LHBCSS. The layered distribution of hydrates alters the stress behavior between the upper and lower layers of the specimen, with the low hydrate saturation layer having a greater influence on the overall strength and deformation characteristics of the HBS.

In recent years, structuralized cementation method has become a novel and promising method for reinforcement of coarse-grained soil slope, which has been proved by practical application. Nevertheless, structuralized cemented slopes with gradient material properties cannot be analyzed using the current methods. The structuralized cemented slope is divided into pure soil zone, variation zone, and solidification zone. The constitutive models of cemented soil in the three zones are proposed, respectively. Slip surfaces of structuralized cemented slopes are all assumed as circular arc shape. It is proved that the displacement compatibility rule is valid for the structuralized cemented slope. A new simple approach is proposed through extension of simplified Bishop slice approach for analyzing structuralized cemented slope stability degree with different cement distribution under various conditions. Our proposed approach involves few parameters examined by element tests. Through analyzing structuralized cemented slopes based on vertical loading and excavation conditions, our proposed method is validated, showing agreement with centrifuge model test analysis. Application of this approach to an actual slope based on excavation conditions reveals that an expanded solidification zone enhances slope safety and reduces shear deformation. Stability remains relatively constant once the solidification zone reaches a certain size. This result underscores the practical value of the proposed method in predicting and optimizing the stability of structuralized cemented slopes.