Journal of Geotechnical and Geoenvironmental Engineering

Field monitoring data on lateral wall deflection and ground settlement from 14 deep excavation case histories in Hong Kong involving cast in situ concrete wall were collected and analyzed. Mixed ground conditions consisted of successive layers of fill, marine deposit and/or alluvium, and decomposed geomaterials. The measured data were differentiated into two groups based on the standard penetration test $N$ values of the ground $-N\le 30$ (Group A) and $N>30$ (Group B) at half of the excavation depth. The mean values of the maximum wall deflection $\left({\delta }_{\mathrm{hm}}\right)$ and ground settlement $\left({\delta }_{\mathrm{vm}}\right)$ in the excavations in Group A are 0.23 and 0.12% of the excavation depth $\left(H\right)$ , respectively. For the excavations in Group B, the mean value of ${\delta }_{\mathrm{hm}}$ is $0.13\%H$ which is similar to other excavations in predominantly other stiff soils $({\delta }_{\mathrm{hm}}∕H=0.10–0.16\%)$ , but ${\delta }_{\mathrm{vm}}$ is only $0.02\%H$ which is small when compared to other similar database $({\delta }_{\mathrm{hm}}∕H=0.1–0.2\%)$ . The settlement influence zone of the excavations in both groups reaches a distance of $2.5H$ , which lies between that observed in other soft clays and sands $(d=2H)$ and in other stiff clays $(d=3H)$ . The measured wall deflection and ground settlement are relatively insensitive to the system stiffness of the support system using cast-in-situ concrete wall.

Tensile strength characteristics of unsaturated sands are examined through a combined theoretical and experimental study. The characteristics of tensile strength in all three water retention regimes of pendular, funicular, and capillary are examined. A simple direct tensile strength apparatus is employed to determine tensile strength for sands with a broad range of particle sizes from silty sand to fine sand and medium sand over a full range of degree of saturation. Tensile strength characteristic curves (TSCC) are established experimentally for these sands and are used to validate the existing theories for tensile strength in the pendular regime. The TSCC for sand characteristically exhibits two zeros at 0 and near 100% saturation, and two peak values occurring in the pendular and capillary regimes, respectively. A minimum tensile strength is observed in the dense fine sand, indicating that either water bridges or pore pressure contributes exclusively to the tensile strength in the funicular regime of this sand. The maximum tensile strength for the silty sand is $\mathrm{1,448}\mathrm{Pa}$ , the fine sand is $\mathrm{1,416}\mathrm{Pa}$ , and the medium sand is $890\mathrm{Pa}$ . Comparison between the soil–water characteristic curves obtained for these sands indicates that the peak tensile strength in the capillary regime is highly correlated to the air-entry pressure. Photographs of the failure surfaces clearly delineate distinct geometric characteristics for different water retention regimes. Analysis of the patterns of failure surfaces in different water retention regimes indicates that the effective stress principle is valid for tensile stress failure in unsaturated sands.

For a complex layered site, development of a shear wave velocity profile from inversion of seismic surface wave data can yield nonunique results. One class of solutions provides depth-averaged profiles that are useful but do not detail the local velocity anomalies. Simulated annealing inversion can be used to establish whether nonuniqueness is a concern for a particular case and, if so, focus results to allow a solution that is expected based on independent knowledge of site conditions. A two-step process is recommended whereby the model parameters are first optimized using simulated annealing and then fine tuned by unconstrained linearized inversion. The process also yields useful confidence measures. One example of a complex layered site where the proposed procedure is helpful is a desert profile with high-velocity cemented layers interspersed among uncemented sediments.

The key purpose of this study was to test the use of a permeable blanket made up of a geocomposite drainage layer (GDL) for leachate recirculation in municipal solid waste (MSW) landfills and to predict the observed leachate travel in the blanket using a numerical model. A $34\mathrm{m}$ long by $12\mathrm{m}$ wide permeable blanket made up of GDL was constructed at an active MSW landfill located in Michigan. Leachate was injected in the GDL using a perforated pipe placed centrally above the GDL along its length. Moisture content sensors, pressure transducers, thermistors, thermocouple sensors, and a vertical load sensor were embedded immediately below the GDL blanket to monitor the flow of injected leachate. After the blanket was covered with waste, leachate was injected into the blanket at rates ranging from 0.9 to $2.6{\mathrm{m}}^3∕\mathrm{h}$ per meter length of the blanket. Data collected from the embedded sensors indicated that the injected leachate traveled at rates ranging from 5 to $18\mathrm{m}∕\mathrm{h}$ through the blanket depending upon the leachate injection rate. Only pressure transducers and thermistors were consistently able to detect migration of injected leachate once the blanket got saturated. Moisture content sensors could not register any change in readings once the blanket became saturated. Leachate injection pressure monitored over a period of about 12 months indicated no signs of clogging of the blanket. The leachate pressures measured immediately below the blanket were less than the net leachate injection pressure indicting that there was a head loss in the GDL blanket. Numerical modeling of liquid flow in the blanket indicated that predicted leachate travel in the blanket was consistent with the field data for assumed values of the waste hydraulic conductivity. In the absence of measured representative hydraulic properties of the waste, absolute verification of the field data was not possible.

This paper presents experimental results from a number of column tests on cement-bentonite (CB) mixtures permeated with chloride $\left({\mathrm{Cl}}^{-}\right)$ . Factors influencing the hydraulic conductivity $\left(k\right)$ , the hydrodynamic dispersion coefficient $\left(D\right)$ and the retardation factor $\left(R_d\right)$ are investigated. The significant factors considered are the curing period, seepage velocity $\left(v\right)$ , reference concentration $\left(C_R\right)$ , and mean particle size $\left(d_{50}\right)$ . Other factors such as porosity $\left(n\right)$ and mix design are also discussed. Results show that a longer curing period lowers all three transport parameters. Furthermore, as the seepage velocity increases, $R_d$ decreases and $D$ increases. The yielding velocity at which mechanical dispersion $\left(D_{\mathrm{md}}\right)$ starts prevailing is observed at about $0.0001\mathrm{cm}∕\mathrm{s}$ . Both $R_d$ and $D$ are seen to reduce with increasing $C_R$ . Thus, the concept of $C_R$ could be used to explain the data scattering in a CB hydrodynamic curve.

This paper reports results of a full-scale field test to assess the performance of dilute colloidal silica stabilizer in reducing the settlement of liquefiable soil. Slow injection methods were used to treat a $2\text{-}\mathrm{m}$ -thick layer of liquefiable sand. Eight injection wells were installed around the perimeter of the $9\text{-}\mathrm{m}$ -diameter test area and 8% by weight colloidal silica grout was slowly injected into the upper $2\mathrm{m}$ of a $10\text{-}\mathrm{m}$ -thick layer of liquefiable sand. A central extraction well was used during grout injection to direct the flow of the colloidal silica towards the center of the test area. Details of the field injection are described. Subsequently, the injection wells were used to install explosive charges and liquefaction was induced by blasting. After blasting, approximately $0.3\mathrm{m}$ of settlement occurred versus $0.5\mathrm{m}$ of settlement in a nearby untreated area. The mechanism of improvement is thought to be bonding between the colloidal silica and the individual sand particles; the colloidal silica gel encapsulates the soil structure and maintains it during dynamic loading.

Often, the use of traditional techniques in geotechnical engineering faces obstacles of economical and environmental nature. The addition of cement becomes an attractive technique when the project requires improvement of the local soil. The treatment of soils with cement finds application, for instance, in the construction of pavement base layers, in slope protection of earth dams, and as a support layer for shallow foundations. However, there are no dosage methodologies based on rational criteria as exist in the case of the concrete technology, where the water/cement ratio plays a fundamental role in the assessment of the target strength. This study therefore aims to quantify the influence of the amount of cement, the porosity and the moisture content on the strength of a sandy soil artificially cemented, as well as to evaluate the use of a water/cement ratio and a voids/cement ratio to assess its unconfined compression strength. A number of unconfined compression tests, triaxial compression tests, and measurements of matric suction were carried out. The results show that the unconfined compression strength increased linearly with the increase in the cement content and exponentially with the reduction in porosity of the compacted mixture. The change in moisture content also has a marked effect on the unconfined compression strength of mixtures compacted at the same dry density. It was shown that, for the soil-cement mixture in an unsaturated state (which is usual for compacted fills), the water/cement ratio is not a good parameter for the assessment of unconfined compression strength. In contrast, the voids/cement ratio, defined as the ratio between the porosity of the compacted mixture and the volumetric cement content, is demonstrated to be the most appropriate parameter to assess the unconfined compression strength of the soil-cement mixture studied.

A viscoplastic cap model of the Perzyna type was developed for simulating high strain rate behaviors of soils. An associative viscous flow rule was used to represent time-dependent soil behaviors. The viscoplastic cap model was validated against experimental data from static and dynamic soil tests. The model was also compared with soil behaviors under creep and stress relaxation with good agreement. However, the model was unable to represent tertiary creep where strain softening became significant. The model was subsequently integrated into LS-DYNA for finite-element simulations of high strain rate behaviors of sandy and clayey soils in explosive tests. The significance of strain rate effect on the soil responses is presented herein. It is concluded that the viscoplastic cap model is adequate for simulations of soil behaviors under high strain rate loading, creep, and stress relaxation, covering a wide range of time-dependent problems.

Decomposed granite (DG) soils are very erosive and, when disturbed, are difficult to revegetate. Sediment eroding from DG roadcuts can severely impact nearby drainage basins. Two ways of reducing surface erosion are to increase the surface saturated hydraulic conductivity $\left(K_{\mathrm{sat}}\right)$ of the soil and to stabilize the surface with vegetation. This study examined the ability of unscreened yard waste compost to increase the $K_{\mathrm{sat}}$ of these soils and facilitate revegetation. A compost application rate of 24% by volume increased postconstruction $K_{\mathrm{sat}}$ of a DG roadcut to levels comparable to a revegetated reference site. The establishment of vegetation was important to maintain the initial increase in $K_{\mathrm{sat}}$ associated with tillage and compost amendment. By the second year, above-ground biomass significantly increased with increasing compost amendment rate. There was also a significant increase in soil $K_{\mathrm{sat}}$ in the 12 and 24% compost treatments in the second year compared to the first year, indicating the strong influence of plant growth on soil surface hydrology.

Three-dimensional response of buried pipes under circular surface loading is investigated using the finite-element method. Previous work by Poulos in 1974 is reexamined, considering the longitudinal behavior of pipes under surface loading. Analyses are performed for pipes of varying stiffnesses and embedment depths. When stiff pipes are located close to the ground surface, the burial depth has little impact on the peak deflection. However flexible pipe deflections decrease significantly as embedment depth increases. Not surprisingly, peak moments increase with pipe stiffness and decrease as the pipes become more remote from the ground surface. The comparison of the new results with those of Poulos indicates that his Mindlin solution calculations are somewhat conservative relative to the finite-element solutions for deeply buried pipes, but unconservative at shallow burial.

Clear water was observed percolating from the tops of fresh auger-cast piles and from the surrounding ground. This transient event, typically lasting less than an hour, occurred in approximately 40% of nearly 700 piles. The percolating water was apparently due to liquefaction of saturated loose to very loose sand, induced by the auger during installation of the piles. The liquefaction water was observed during this project because of a locally high water table; however, temporary flow of excess pore water through fresh grout must be a regular occurrence in loose sands, but is usually unnoticed because the groundwater table is normally well below the ground surface.

In column testing of reactive solute transport, the retardation factor and column Péclet number can be simultaneously calculated by the nonlinear regression of the observed breakthrough curve. Several alternative methods exist to determine the retardation factor, but alternate techniques for determining the Péclet number have not been proposed. This paper presents a graphical method to obtain the Péclet number with examples of the proposed method. The graphical method provides results that are quite consistent with the nonlinear regression technique; and is a reasonable alternative, depending on the scatter in the effluent data.

Analytical and numerical models of capillary barrier performance commonly use hydraulic characteristics measured using pure water. However, the potential exists for an infiltrating solution to have a surface tension lower than that of pure water due to the presence of surface-active contaminants (surfactants). A lower surface tension solution may impact capillary barrier performance due to the dependence of capillarity on surface tension. An existing analytical solution for capillary diversion length $\left(L\right)$ was modified to include the effect of surface tension reduction on steady-state capillary barrier performance during uniform and constant infiltration. The $L$ for a surfactant-contaminated system was found to be less than for a pure water system and equal to $L$ for a pure water system multiplied by the relative surface tension. Numerical modeling using HYDRUS-2D also showed that diversion was less in the surfactant-contaminated system and that the difference in the performance of the two systems was due to the fact that the fine layer in the capillary barrier retains less liquid when wetted with surfactant solution.