Journal of Geotechnical and Geoenvironmental Engineering

This paper reports the development of a particle breakage model within a finite-element (FE) framework based on the modified Mohr–Coulomb model (MMC). A modified form of the strength-dilatancy relationship is used to incorporate additional energy consumed due to particle breakage. A new particle breakage index is introduced and constitutively linked to the mobilized friction angle. The model is verified against carefully conducted drained triaxial tests on medium-dense calcareous sand obtained from offshore Australia as well as other published data. The model is implemented in the FE program ABAQUS for both small strain and large deformation FE analyses in practical field problems. Several important microscopic observations can be made from the proposed model that are not limited to identifying soil regions where particle breakage is maximized. Due to its inherent simplicity, the newly developed model can provide reasonable predictions for soil-structure interaction problems in calcareous sands.

Current design methods for geosynthetic-reinforced soil (GRS) walls in a tiered configuration mostly focus on wall stability by determining required tensile strength and layout of reinforcement. A finite difference numerical model was firstly verified with the available field measured results of a single GRS wall and model test results of a two-tiered GRS wall in the literature, and then used to analyze lateral displacements of multitiered GRS walls with modular concrete block facing. A parametric study was conducted to evaluate the effect of backfill properties (friction angle and cohesion), elastic modulus of foundation soil, reinforcement parameters (stiffness, spacing, and length), and tiered wall geometry (offset distance, number of tiers, and height ratio of adjacent tiers), on facing lateral displacements of multitiered GRS walls. The numerical results showed that an increase of the shear strength of the backfill by its friction angle or cohesion reduced the wall lateral displacement. An increase in the reinforcement length of the upper tier in the two-tiered wall from 0.35 times to 0.60 times the total wall height resulted in approximately 21.3% and 34.7% reduction in the maximum lateral displacements for lower and upper tiers, respectively. The reinforcement stiffness and spacing had a significant impact on facing lateral displacements whereas the ratio of reinforcement stiffness to spacing had a negligible influence. An increase of the offset distance or a reduction of the number of tiers remarkably reduced wall facing lateral displacements. An analytical solution was developed for estimating the lateral displacement of a two-tiered GRS wall and compared well with the numerical results.

Liquefaction in soils has predominantly been associated with saturated soils. Under seismic activity, unsaturated soils are may also be susceptible to liquefaction. The implications of neglecting unsaturated soils that are close to saturation as the foremost rule for assessment of liquefaction can be risky and catastrophic. Hence, in this study, a cyclic double-walled triaxial device capable of applying suction via axis-translation technique was modified to conduct cyclic triaxial tests on suction-equilibrated specimens at a low suction state. In unsaturated soils, the presence of highly compressible air within the soil specimens may prevent the effective confining pressure from reaching zero, which is typically considered as initiation of liquefaction. Therefore, in this study, liquefaction was considered to occur when 5% double-amplitude strain was attained in the soil specimen. A series of suction-equilibrated cyclic triaxial test results on unsaturated soils were analyzed to study and verify the possibility of liquefaction in unsaturated soils. Additional suction-controlled monotonic triaxial tests were performed to compare the stress paths of the cyclic triaxial tests with their respective critical state lines to indirectly assess the hypothesis of liquefaction mitigation using an induced partial-saturation technique. It was observed that liquefaction was possible and detected in unsaturated cohesionless soils having relative density of 50% and a degree of saturation of above 70%. However, with desaturation the resistance to liquefaction in soils increased exponentially. This demonstrated the potential for induced partial saturation to mitigate liquefaction in moderately compacted soils whose degree of saturation is reduced to less than 70%.

This study explores the mechanical behavior of glass bead pairs cemented by microbial induced calcite precipitation (MICP) when subjected to tensile or shear loading. The mineral precipitation habit and contact area are also examined using X-ray computed tomography (X-ray CT). Examination of the failure surfaces reveals three distinctive failure modes: debonding failure at the precipitate-grain interface, internal failure within the precipitate, and mixed failure. The internal failure mode appears dominant when the calcite content (CC) of the bonded glass bead pair is greater than 17%, and it results in the smallest strengths: $\sim 8\mathrm{kPa}$ in tension and $\sim 7\mathrm{kPa}$ in shear. When CC is less than 17%, the debonding failure mode is mostly observed, and the debonding failure leads to the greatest strengths: $\sim 35\mathrm{kPa}$ in tension and $\sim 13\mathrm{kPa}$ in shear. The mixed failure mode occurs when $11\%<\mathrm{CC}<20\%$, partly overlapping with the other two modes. The average tensile strength is greater than the average shear strength in all modes. The X-ray CT images demonstrate that the deposition of calcium carbonate first begins by coating the grain surface, and later shifts toward preferential precipitation at the grain contacts as the CC increases. Therefore, the relationship between contact radius and CC is bounded by the grain-coating and meniscus-filling models when CC < 20%; however, at a greater CC this relationship is bounded by the meniscus-filling and flat torus-filling models. This study presents unprecedented grain-scale mechanical responses associated with MICP-treated granular materials, which can be further extended to advance the understanding of the interplay between grain-scale cementation and the mechanical response of MICP-treated specimens, as well as the simulation of cemented soil behavior using discrete element modeling.

There are clear advantages in the establishment of reliable, direct cone penetration test (CPT) based methods for assessment of the axial capacity of driven piles. These advantages motivated the formation of a joint industry project (JIP) under the management of the Norwegian Geotechnical Institute (NGI), which initially led to the creation of a unified database of high-quality pile load tests in sand and clay. The unified database has the consensus approval of representatives of the profession and personnel in multiple companies from the offshore energy sector. This paper presents a component of the research from the second phase of the JIP, which had the objective of developing a new CPT-based method for driven piles in clay to unify several CPT-based methods that are in use today. First, a rational basis for the CPT-based formulation is described, using trends from instrumented pile tests; the description facilitates an understanding of the approach and illustrates its empirical nature and limitations. The unified database was used to calibrate the formulation and it led to good predictions for an independent database of pile load tests and for measured distributions of shaft friction.

In this study, the suction induced by stress release was investigated by measuring the suction of instantaneously unloaded MX80 bentonite specimens after wetting in an oedometer either under different vertical stresses or under a constant-volume condition. Swell-consolidation tests were also performed, and the results were compared with those of specimens wetted under different vertical stresses or a constant-volume condition. The microstructure of the unloaded specimens was investigated using mercury intrusion porosimetry (MIP). The water content and void ratio as well as the degree of saturation of specimens were also determined. Results showed that in the case of limited water exchange between macropores and micropores during stress release, the change in suction was equal to that of stress ($-\mathrm{d}s/\mathrm{d}{\sigma }_v=1$). By contrast, with water exchange between macropores and micropores, soil swelling occurred, and the variation of suction was higher than that of stress ($-\mathrm{d}s/\mathrm{d}{\sigma }_v>1$). A method was proposed allowing such water exchange between macropores and micropores to be determined ($-\mathrm{d}s/\mathrm{d}{\sigma }_{v-\text{modified}}=1$, where ${\sigma }_{v-\text{modified}}$ is the modified stress considering water exchange), evidencing the macropore drying due to stress release. Comparison of results from MX80 bentonite and other clayey soils showed that when the nonclay (grains such as quartz, calcite, and so on) fraction (NCF) was lower than 35% by mass, NCF had an insignificant effect on $-\mathrm{d}s/\mathrm{d}{\sigma }_{v-\text{modified}}$ (equal to 1). By contrast, in the case of higher NCF ($>35\%$), $-\mathrm{d}s/\mathrm{d}{\sigma }_{v-\text{modified}}$ became lower than 1, and the higher the NCF, the lower the $-\mathrm{d}s/\mathrm{d}{\sigma }_{v-\text{modified}}$ value. These findings provide useful information for better understanding the mechanism involved at the microstructure level in the stress/suction coupling during stress release in clayey soils.

Over the past decade, several researchers have demonstrated that microbially induced carbonate precipitation (MICP) has the potential to improve soil behavior in the laboratory setting. In this study, MICP was implemented at a sandy slope field site to enhance erosion resistance and surficial soil strength. Three application systems—surface spraying, prefabricated vertical drains (PVDs), and shallow trenches—were compared. Improvement of the treated soil was assessed using dynamic cone penetration, impinging jet, and pocket penetrometer tests and was monitored for 331 days. Results indicated that MICP is an effective soil improvement method for surficial and deeper applications. Penetration index values improved up to 73% and 55% at the surface and a depth of 30 cm, respectively. Critical shear stress and coefficient of erodibility values exhibited significant improvements. The surface spraying method is preferred for the treatment of large surficial areas, whereas the PVD method demonstrated deep soil improvement potential. The shallow trenches resulted in significant surficial improvements, however, in a highly localized manner. Post-treatment monitoring indicated no significant degradation of the treated areas with time and after major storm events (e.g., Hurricane Dorian). Based on the field results, a sensitivity analysis was performed to address the applicability of future MICP-field implementations in various soil types.