Acta Geotechnica最新文献

Offshore structures exposed to prolonged cyclic loading necessitate a precise assessment of the soil–structure interface mechanical behavior. This study, building on cyclic shear tests at silty sand–steel and gravelly soil–structure interfaces, offers a comparative analysis of strength characteristic, deformation response, and physical-state evolution. Drawing from those, a well-verified mechanism-based interface modeling is formulated. The findings indicate that the shear strength at both interfaces can be uniformly characterized by the Mohr–Coulomb failure criterion. The interface elastic behavior, plastic shear strain, and compression-induced plastic volumetric strain can be effectively modeled within a unified framework. The shear-induced volumetric strain involves the cumulative and cyclic components, while the cumulative volumetric strain at both interfaces accumulates rapidly at the onset, with its rate diminishing as the cycles progress. Regarding cyclic volumetric strain, double-phase transformation points per cycle were observed at the silty sand–steel interface, whereas only single at the gravelly soil–structure interface, highlighting the influence of interface contact properties. Employing parameters calibrated from the interface shear test under constant normal load condition and confining compression test, the proposed model effectively simulates the interface strength and deformation responses under both constant normal stiffness and constant volume conditions, demonstrating the independence of interface mechanical modeling parameters from boundary conditions.

A generalized constitutive model is presented in the framework of the multilaminate method in this paper. This model does not need to explicitly define the yield and plastic potential surfaces. In this framework, the constitutive relations between stresses and strains in the form of volumetric and deviatoric components are defined on several planes in various directions, called microplane. A new volumetric–deviatoric stress space is defined on the microplanes. The new constitutive model uses a new stress space in which stress–strain relations are defined in a scalar form instead of a tensor form, and this potentially enhances the computational efficiency of the model. The proposed model captures induced anisotropy, including principal stress axes rotation effects, through microplane-specific strain histories and hardening laws. While inherent anisotropy is not addressed here, the framework allows its incorporation via directional microplane properties without changes in formulation. The results predicted by the new constitutive model for several types of clay and sand in drained and undrained conditions have been validated with experimental data. The new constitutive model has been implemented in OpenSees software, and the Mahabad earth dam is analyzed by it in a steady-state condition which demonstrates the practical application of the model.

Engineering waste mud, primarily produced during the construction of bored piles and slurry shield tunneling, represents a challenging type of construction waste to treat. Additionally, the disposal of industrial by-products also presents significant challenges. In this study, a sustainable solution is proposed by utilizing industrial by-product lignocellulosic fibers as reinforcement materials in geotechnical engineering, complemented by an eco-friendly hydrophobic polymer to treat the waste mud. The impact of additive content on mechanical properties was assessed through unconfined compressive strength (UCS) tests. The reinforcement mechanism was elucidated through microstructural observation tests, including scanning electron microscopy (SEM), X-ray diffraction (XRD), and mercury intrusion porosimetry (MIP). The results show that composite soil additive effectively improves the UCS of the engineering waste mud. The strength of the reinforced mud samples increases with the additive content and curing age, and the optimum dosages were found to be 4% lignocellulosic fiber and 4% hydrophobic polymer, enhancing the 28-day UCS of the treated mud by 347.9% relative to untreated mud. MIP and SEM results suggest that the strength improvement can be attributed to a reduction in total volume of pores and the reinforcing and toughening effects of lignocellulosic fiber. The strength gains over time are primarily attributed to the hydrophobic polymer rather than lignocellulosic fiber. In conclusion, the waste mud reinforced with by-product lignocellulosic fiber and hydrophobic polymer represents a win–win solution that simultaneously improves soil strength and recycles industrial waste.

This paper assesses the coupled effects of particle shape and water saturation on the mechanical performance of granular materials subjected to impact loading. Dynamic compression experiments at high strain rate (750 s⁻¹–950 s⁻¹) are performed on dry and wet glass particles with a porosity of 0.48 and different initial aggregate shape via the split Hopkinson bar device. Particle size and shape parameters are quantitatively characterized using high-resolution micro-CT imaging with a 6 μm spatial resolution. Multi-scale experimental evidence demonstrates that particle shape has a more dominant influence on the mechanical behaviours of granular media than the saturation condition. Specifically, the fragmentation and subsequent particle arrangement are notably shape dependent, with rounded particles tending to produce more angular fragments and irregular particles generating more spherical fragments. Additionally, while particle saturation primarily affects the extent of breakage rather than altering the fundamental mechanisms of breakage, it significantly enhances the degree of fragmentation by reducing inter-particle friction and modifying stress distributions. The study further reveals that the particle breakage index, which quantifies the extent of fragmentation, shows a stronger correlation with its characteristic particle size under impact conditions. Particle melting has been observed in both dry and saturated conditions. The coevolving trends of particle size and shape reveal two mechanisms driving changes in particle shape: fracturing-activated and melting-activated processes. These findings highlight the complex interplay between particle shape and moisture content in governing granular material behaviours, emphasizing the need for detailed understanding of these factors to accurately predict and optimize material performance in various engineering applications.

Railway embankments must be stable during earthquakes to prevent track deformation and minimize accident risks. The vertical component of near-fault pulse-like ground motions can significantly increase the risk of damage to railway embankments. Therefore, accurately assessing the effects of near-fault vertical ground motions (VGMs) on the seismic demand of railway embankments is crucial to ensuring their structural safety. This study addresses this problem from a probabilistic perspective, developing a probabilistic seismic demand model for railway embankments under pulse-like horizontal ground motions (HGMs) and VGMs to identify the optimal intensity measures (IMs). The amplification coefficient, denoted as β, quantified the increased seismic demand of railway embankments under the combined effects of near-fault pulse-like HGMs and VGMs, as opposed to that under pulse-like HGMs only. Analysis of near-fault ground motions indicated that the vertical-to-horizontal peak acceleration ratio increased as source distance decreased, averaging 0.88 and exceeding the empirical 2/3 ratio. The velocity spectral intensity (VSI) of the HGMs was identified as the optimal IM. Statistical analysis revealed average values of β ranging from 1.1 to 3.6 under various ground motions. β followed a normal distribution; its mean and standard deviation decreased as the VSI increased and increased in proportion to the VSI ratio of vertical-to-horizontal ground motions (V/H). A mathematical model for β, incorporating multilevel seismic safety reserve levels of 50%, 84%, and 98%, was established using VSI and V/H to assess the impact of random VGM on railway embankment, ensuring cost-effective yet reliable seismic resilience for railway infrastructure.

Determining the stress state in natural soil layers is essential for accurately assessing foundation bearing capacity, predicting soil deformation and optimizing construction plans. However, few efforts have been made so far to investigate the anisotropy of horizontal earth pressure through in situ tests. In this study, both K₀ stepped blade test (KSB) in different directions and pressuremeter test (PMT) are carried out at an excavated pit to explore the anisotropy of horizontal earth pressure in natural granite residual soil (GRS) deposits. Additionally, scanning electron microscopy (SEM) test is performed to analyze the orientational arrangement of soil particles of GRS. Finally, considering the excavation-induced unloading effect, the unloading response of horizontal earth pressure is conducted through a series of laboratory consolidation experiments. The results of KSB tests indicate a significant anisotropy of horizontal earth pressure, with the maximum horizontal earth pressure arising at the directions of 112.5°, as verified by conducting dilatometer test (DMT) in different directions. In addition, a tensor ({varvec{varPhi}}_{ij}) is introduced to quantify the anisotropy of horizontal earth pressure. From the microscopic aspect, the SEM test reveals a significant orientational arrangement of soil particles, which is coincident with the direction of maximum horizontal earth pressure measured by KSB test. According to the results of consolidation experiments, the horizontal earth pressure considering the unloading effect is more consistent with that measured in the in situ tests, which is greater than in the loading stage, indicating that the unloading effect leads to a significant rise in the horizontal earth stress in the GRS. This study could provide more insights into the horizontal earth pressure in the GRS, and a reference for determining the stress state in natural soil deposits.

The environmental vibration generated by train operations in multiple existing tunnels closely intersecting are complex. Therefore, 3D full-scale numerical models for dynamic analysis were established based on an ongoing tunnel construction in Chongqing, China. The dynamic response measured in actual engineering verified the rationality of the numerical model. Moreover, the dynamic interaction between six adjacent tunnels within a volume of 2700 cubic meters was analyzed. On this basis, the impact of different operating conditions of trains in existing tunnels on the new tunnel was discussed. The results show that there is a sudden increase in acceleration at about 1.0 s in the longitudinal middle section of the new tunnel. This is because all the trains in operation are roughly at the intersection with the plane of the new tunnel when trains start running 1.0 s. Additionally, the impact of the number of trains running in the existing tunnel on the dynamic response of the new tunnel may not increase linearly, but exponentially. It is worth noting that the reinforcement of the cross sections near multiple tunnel intersections should focus on the top of the section, while that farther away should focus on the section bottom. The research results provide a theoretical basis for the environmental vibration assessment of complex and multiple intersecting tunnels in modern cities. Based on the varying dynamic responses at different positions of the newly built tunnel section, reasonable layout of vibration reduction facilities can enhance passenger comfort.

The effectiveness of using transparent sand materials for simulating the soil behavior under the dynamic loading condition is not clear. In this study, the horizontal pulling tests with iron pipes embedded in were performed on the saturated transparent sand and natural quartz sand to compare the apparent viscosity and pore water pressure of both materials during dynamic tests. The test results show that saturated transparent sand exhibits a similar dynamic characteristics to the natural sand in the liquefied state, evidenced by both showing the non-Newtonian fluid properties of “shear thinning”. The apparent viscosity of transparent sand decreases by 70.2 and 58.9% during and after shaking, respectively, while that of quartz sand decreases by 71.9 and 61.8%, respectively, as the pulling speed increases from 2.2 to 10.6 mm/s. During shaking with the pipe pulling rate of 7.2 mm/s and at the lateral displacement to pipe diameter ratio of δ/D = 4, the apparent viscosity of transparent sand is only 8.8% higher than that of quartz sand. Through visual analysis, transparent sand could effectively exhibit the sand particles’ movement and shear zone evolution during the liquefaction process confirming its capability of simulating the natural sand’s flow behavior under the liquefaction phase. This would eventually provide a powerful visualization methodology to better understand the dynamic behavior of liquefied sand.

Many geological disasters in loess areas are related to soil wetting. Previous studies of loess wetting behavior focused mainly on zero vertical net stress or constant vertical net stress, while few studies have been conducted under confined conditions (constant volume). The stress regime and microstructure evolution characteristic of loess subjecting confined wetting remain unclear. This study investigates the intact loess of two burial depths (denoted as HFT10 m and HFT30 m) subjected to controlled-suction wetting under both confined and vertical unconfined conditions. The loess microstructure evolution characteristics before and after loading-confined and loading–unloading-unconfined wetting were evaluated by mercury intrusion porosimetry (MIP) tests. Under vertical unconfined wetting, the results indicated that all specimens exhibited wetting-induced expansion, with denser specimens showing higher swelling strain. Under confined wetting, the vertical stress decreased gradually, and the extent of stress reduction intensified as specimen strain increased. The HFT10 m sample showed a larger wetting-induced stress reduction than the HFT30 m samples. The soil water retention curves (SWRCs) at the degree of saturation (S_r)-void ratio (e)-suction (s) space between confined and unconfined was different, this may be due to the different microstructure evolution characteristic of unconfined wetting and confined wetting. The paths (suction against vertical net stress) were within the newly defined loading-collapse (LC) curve. The experimental results of this study highlight that the vertical stress in the soil beneath the bearing platform decreases upon wetting, which may result in an increase in the axial stress of the pile, even if the loess does not undergo collapse.

Biomineralization technology, exemplified by processes such as microbially induced calcium carbonate precipitation, offers significant advantages in terms of reinforcing sand to improve their strength and stability. However, systematic theoretical models for predicting the mechanical behavior of biotreated sand remain scarce. This study proposes a novel theoretical model based on granular thermodynamics, providing accurate predictions and insights into underlying mechanisms. The model reveals that compressive and shear stiffness anisotropy are linked to stress state and deviatoric elastic potential energy, respectively, while biotreatment reduces anisotropy and enhances stiffness and stability. By using the determinant of the elastic potential energy Hessian matrix, the limit state surface reflecting loading direction dependency is constructed, revealing the reinforcing effect of biotreatment in the sand granular system. The hardening index describes strength evolution, with elastic strain accumulation dominating early shear and plastic flow influencing later stages. Cementation degradation and hardening index affect granular temperature and energy dissipation, with higher cementation levels and hardening index reducing granular fluctuations and increasing dissipation. This theoretical model provides theoretical support for predicting and understanding the behavior of biotreated sand.