Journal of Geotechnical and Geoenvironmental Engineering

In this study, the impacts to soil thermal properties during and after biocementation via microbially induced calcite precipitation (MICP) method on silty silica sand specimens with varying fines content (0%, 5%, and 15%) were investigated. Firstly, calcium conversion was measured after each pulse; then, the MICP-treated specimens were tested for cementation uniformity. The evolution of thermal conductivity of silty soils with the MICP treatment was assessed using a thermocouple probe. The results show that thermal conductivity of silty saturated sands increased by 17% for specimens treated to 9.7% ${\mathrm{CaCO}}_3$. The improvement in thermal conductivity was attributed to the formation of calcium carbonate bridges binding the soil grains together. The results suggested that the thermal conductivity of silty soil depends on water content, the number of treatment pulses, and the treatment uniformity through the soil specimen. Presence of fines content in the soil was found to play an important role in the distribution and uniformity of biocementation through the soil specimen. However, no statistically significant difference in the thermal conductivity values of MICP-treated specimens with different fines content was observed ($p>0.05$). The average calcium carbonate content ranged between 10.7% and 7.2% for the soils with 0% and 15% fines content, respectively. The findings of this research could be used to improve the efficiency of geothermal boreholes and other energy geostructures using MICP by improving thermal conductivity of dry and partially saturated soil.

Quantifying the earth pressure acting on retaining structures is essential for a number of engineering applications, including the assessment of damaged structures. In response to the widespread corrosion found in many cantilever retaining walls in Switzerland, this study investigates the evolution of the earth pressure acting on a retaining wall rotating about its base: the dominant deformation mode for such structures. In a controlled setup, a scaled cantilever retaining wall is first backfilled and then subjected to rotation about its base. An extensive experimental program was conducted to evaluate the effects of different soil properties, packing densities, and initial soil stresses on the earth pressure during the life of a wall. Accurate measurements allow the stick-slip behavior typical of granular soils to be observed. An analysis of this behavior suggests that its impact on the practical safety assessment of retaining walls is negligible. Tests performed under 3D conditions show how local damage to a section of the wall results in a lower active earth pressure value on that section due to stress redistribution, which causes an increase in pressure on adjacent sections. It is also observed how the initial earth pressure acting on retaining walls after backfilling results from a perturbed stress state due to wall deflection and wall friction. Its distribution is bilinear for uncompacted soil and nonlinear for compacted soil. Loosely packed soil exerts higher pressures than densely packed soil. The unloading process shows the expected behavior of loose and dense soil: it is slower and monotonic in loose soil and faster in dense soil, where the initial unloading is followed by reloading due to softening in the backfill. The experimental results are available in a public repository and can serve as a resource for the development and validation of improved verification procedures.

The load-carrying capacity of rock-socketed piles depends on the shaft resistance at the pile-rock interface which is governed by the interactions between the pile, rock, and soft interface materials (smear or infill). Despite recent advancements in 3D experimental pile testing of smeared interfaces, the knowledge in this field still remains quite limited primarily due to the experimental limitations, and high costs and fabrication challenges associated with extensive laboratory tests. This study presents a comprehensive numerical investigation to simulate 3D rock-socketed piles with smear using the discrete element method (DEM). By correlating the observations with the experimental x-ray CT images, the interactions between the pile, rock and smear are investigated for various smear area proportions in terms of the micromechanics at the asperity level. The movements of smear and rock debris in the socket and their influence on the various asperity failure modes in relation to the shaft resistance development are evaluated by monitoring the force and damage evolution in DEM. Finally, the calibrated DEM model is utilized to examine the influence of the various smear fabric parameters on the shaft response. Based on the observations, it was inferred that the shear failure mode primarily dictates the interface damage evolution, with a significant contribution from mixed-mode failure involving both tensile and shear damages. Insights from the results indicated that the effect of smear placement on the load-bearing capacity of the pile is found to be minimal compared to the other smear fabric parameters (thickness and area proportion). Moreover, the critical smear thickness to asperity height ratio was determined to be 1.75 beyond which the smear controls the load-bearing capacity of the pile. The discussions presented in this study provide a novel understanding into the smear fabric effect and act as a foundation for further research aimed at improving the reliability and efficiency of rock-socketed pile construction in soft rocks.

Modeling of the soil–structure interface has been a critical issue in geotechnical engineering. Numerous studies have simulated complex soil–structure interface behaviors. These models usually are assessed by direct comparisons between the simulations and experiments. However, little work has been done to compare the specific interface behaviors simulated by different interface models. This paper evaluated some frequently recognized interface behaviors for six different interface models. These models either were adopted from the existing literature or modified from the existing soil models, including the exponential model, hyperbolic model, hypoplastic model, MCC model, SANISAND model, and SIMSAND model. Global comparisons and effects of the soil density, normal stiffness, and shearing rate were investigated to evaluate the interface models based on Fontainebleau sand–steel interface experiments and kaolin clay–steel interface experiments. The limitations and advantages of different models under different conditions were discussed.

Laboratory and field tests were performed on sandy soils from six Pleistocene-age sites in the South Carolina coastal region to investigate the age-related resistance to liquefaction. Stress-controlled cyclic triaxial tests were used to determine the cyclic strength of soils with geologic ages ranging from approximately 59,000 to 1,200,000 years. Three sites have evidence of liquefaction in the form of sand blows that are 467 to 4,185 years old as determined from ${}^{14}\mathrm{C}$ dating of embedded organic material. The other three sites show no indications of liquefaction. Cyclic stress ratios ranging from 0.095 to 0.225 were applied to undisturbed and reconstituted soil specimens that were consolidated to an effective stress equal to 100 kPa. Soil specimen liquefaction was defined to occur when the excess pore pressure was equal to the confining effective stress. Estimates of the at-rest earth pressure coefficient were determined using measurements from the flat plate dilatometer and the cone penetrometer and were applied to the laboratory cyclic stress ratio occurring at the 15th loading cycle to determine the laboratory–field equivalent cyclic resistance ratio. The age-dependent liquefaction resistance was determined using additional data from the inner coastal plain of South Carolina and assessing the cyclic resistance ratios and their associated ${\mathrm{K}}_{\mathrm{DR}}$ ratios relative to the base data and applying one of the more recently developed liquefaction triggering model. It was found that the development of the aging factor should be independent of the liquefaction triggering model. Subsequently, the aging factor is developed using an offset that is constrained at 20 years and a ${\mathrm{K}}_{\mathrm{DR}}=1.0$, and was found to range from 1.00 at 20 years to 1.45 at 1.0 Ma for the original deposition ages of the soils and 1.00 at 20 years to 1.51 at 1.0 Ma for the data set consisting of the last disturbance and original deposition ages of the soils.

Soil physical properties derived from phase relationships such as void ratio, degree of saturation, water content, and unit weight conventionally are formulated and calculated under the implicit assumption that the volumes of the three soil phases (i.e., solid, water, and air) are independent and unaffected by interactions among phases. However, significant variations do occur, including variation in the unit weight of soil water resulting from adsorptive soil–water interactions and variation in void volume resulting from bulk swelling or shrinking with changes in moisture. The importance of accounting for soil–water interactions in calculating soil physical properties was examined by introducing a new variable-volume soil phase diagram, and was illustrated using experimental data from consolidation and shrinkage tests for silts, nonswelling clays, and swelling clays. Errors in calculating void ratio and soil unit weight in saturated consolidation tests using conventional phase relationships can be significant—as high as 48% relative error in void ratio and as high as 16% relative error in soil unit weight. Errors in calculating moisture ratio and saturation in shrinkage tests can be as high as 53% relative error in moisture ratio and 42% in saturation. The occurrence of maximum error is highly correlated with soil water content corresponding to the maximum amount of adsorbed water. Liquid limit (LL) and specific surface area (SSA) were identified as index properties that can be used to account for volume variations in phase calculations. Examples are provided to illustrate the implementation of the proposed variable-volume soil phase formulation using LL or SSA in practical problems. Analysis showed that for fine-grained soils, it is practically necessary to use general variable-volume phase relationships to define basic soil properties in lieu of the conventional phase relationships.

In this study, a series of suction-control oedometer tests were conducted on unsaturated and saturated MX80 bentonite pellet/powder mixtures. The mixture samples were imposed with various suctions from 0 to 113 MPa under a low vertical stress (0.1 MPa). Then, the mixture samples were loaded in steps with controlled suction until reaching three different target stresses ${\sigma }_v$ (0.1, 3.2, and 12.8 MPa) and instantaneously unloaded. The microstructure of unloaded samples was observed with mercury intrusion porosimetry (MIP) and microcomputed tomography ($\mathrm{\mu CT}$), together with the determination of suction, water content, void ratio, and saturation degree. Results showed that the water-retention behavior of bentonite pellet/powder mixture was rather characterized by pellets instead of powder. Due to the stress release, the measured suction after unloading $s_{me}$ was larger than the imposed suction $s_{im}$. The suction difference $\mathrm{\Delta }s$ between $s_{me}$ and $s_{im}$ depended on both suction and load. Under a given ${\sigma }_v$, the suction effect was characterized by an increase of $\mathrm{\Delta }s$ when $s_{im}$ decreased from 113 to 4.2 MPa, but by a noticeable decrease of $\mathrm{\Delta }s$ when $s_{im}$ further decreased to zero. This was found to be related to the changes of the slopes of soil-water retention curves (SWRCs): the slope was decreasing with suction increases after 4.2 MPa suction, but was increasing with suction increases before 4.2 MPa suction. As far as the stress effect was concerned, a larger $\mathrm{\Delta }s$ was identified at a higher ${\sigma }_v$ for a given $s_{im}$. The narrowing difference between $\mathrm{\Delta }s$ and the reduction of mean stress of pellet $\mathrm{\Delta }{p}^{\mathrm{pellet}}$ with increasing load was because of the decrease of larger pores due to loading. The total mixture void ratio $e_T^{\mathrm{mixture}}$ was highly dependent on suction and vertical stress, whereas the total pellet void ratio $e_T^{\mathrm{pellet}}$ was only dependent on suction in the stress range considered.

Ground reaction curves (GRCs) are increasingly used to simulate the progressive development of soil arching resulting from differential movement in soil. In this study, 20 single, double, and multiple trapdoor tests were conducted under a simultaneous displacement mode (i.e., trapdoors settle simultaneously), while 8 double trapdoor tests were carried out under a successive displacement mode (i.e., trapdoors settle successively) to examine the evolution of soil arching including the parameters related to the GRCs. This study identified and analyzed the key parameters of the GRCs. For the trapdoor tests conducted under the simultaneous displacement mode, an increase of the fill height led to a decrease in the minimum pressure and its corresponding trapdoor displacement. Furthermore, the ultimate normalized pressure and recovery rate exhibited a decreasing trend with an increase of the fill height. As compared to the single and double trapdoor tests, the multiple trapdoor tests had similar initial soil arching, minimum normalized pressure, and its corresponding trapdoor displacement but a relatively lower stress recovery rate. In the trapdoor tests conducted with the successive displacement mode, the pressure on the first settled trapdoor decreased while it increased on the later settled trapdoor during the settlement of the first settled trapdoor. Conversely, during the settlement of the later settled trapdoor, the pressure on the first settled trapdoor increased, while it decreased on the later settled trapdoor. Analyzing the GRCs on the later settled trapdoor under the successive displacement mode revealed that the initial slope and ultimate pressure of the GRC decreased as the fill height increased.