The viscous properties of a variety of poorly graded unbound granular materials were investigated by direct shear tests on 12 cm-cubic specimens. A number of natural sands having different particle shapes and sizes as well as uniform glass beads having different particle sizes were used. The viscous properties were evaluated by changing the shear displacement rate many times during otherwise monotonic loading (ML) at constant shear displacement rate and normal pressure. Creep loadings were performed in two tests. Different types of viscous properties, which are affected by the particle shape but essentially independent of the particle size, are reported. The viscosity type varies as the shear displacement increases from the pre-peak regime towards the residual state. A new viscosity type, called “Positive & Negative”, was found with relatively round granular materials in the pre-peak regime and with relatively angular granular materials in the post-peak softening regime and at the residual state. Peculiar “rate-independent unstable behaviour” is observed with round natural sands and glass beads in the post-peak regime, which is more significant and frequent with glass beads. Controlled by the particle size, this behaviour is caused by the so-called stick/slip phenomenon. The viscous properties observed in the DS tests are quantified by the rate-sensitivity coefficient defined in terms of the shear and normal stresses, which are then converted to those defined in terms of the major and minor principal stresses, β₁₃. These β₁₃ values are consistent with those directly obtained by the triaxial and plane strain compression tests. The effects of particle size on the β₁₃ value are negligible and the β₁₃ value tends to decrease as the particle shape becomes more round.

View full abstract

The seismic behavior of asphaltic concrete used in embankment dams subjected earthquake loads has been studied. In order to evaluate the dynamic behavior, an extensive series of monotonic and cyclic tests were carried out on triaxial specimens of asphalt concrete used in hydraulic structures. The MTS-dynamic equipment at the Norwegian Geotechnical Institute (NGI) was used for this purpose. Temperature and frequency effects on specimen behavior and on specimen degradation have been studied under the cyclic loads in both isotropic and anisotropic initial stress conditions. For investigation of the fatigue behavior, thousands of cyclic loads were imposed on some of the specimens. Moreover, to study any sign of material degradation due to the cyclic loading, the post-cyclic monotonic stress-strain curve was compared with the corresponding curve for specimens that were not first subjected to cyclic loading. Geotechnical parameters to be used in dynamic numerical analysis models are also presented.

View full abstract

A series of plane strain compression (PSC) tests were performed on large sand specimens unreinforced or reinforced with prototype geosynthetic reinforcements, either of two geogrid types and one geocomposite type. Local tensile strains in the reinforcement were measured by using two types of strain gauges. Sustained loading (SL) under fixed boundary stress conditions and cyclic loading (CL) tests were performed during otherwise monotonic loading at a constant strain rate to evaluate the development of creep deformation by SL and residual deformation by CL of geosynthetic-reinforced sand and also residual strains in the reinforcement by these loading histories. It is shown that the creep deformation of geosynthetic-reinforced sand develops due to the viscous properties of both sand and geosynthetic reinforcement, while the residual deformation of geosynthetic-reinforced sand during CL (defined at the peak stress state during CL) consists of two components: i) the one by the viscous properties of sand and reinforcement; and ii) the other by rate-independent cyclic loading effects with sand. The development of residual deformation of geosynthetic-reinforced sand by SL and CL histories had no negative effects on the subsequent stress-strain behaviour and the compressive strength was maintained as the original value or even became larger by such SL and CL histories. The local tensile strains in the geosynthetic reinforcement arranged in the sand specimen subjected to SL decreased noticeably with time, due mainly to lateral compressive creep strains in sand during SL of geosynthetic-reinforced sand. This result indicates that, with geosynthetic-reinforced soil structures designed to have a sufficiently high safety factor under static loading conditions because of seismic design, it is overly conservative to assume that the tensile load in the geosynthetic reinforcement is maintained constant for long life time. Moreover, during CL of geosynthetic-reinforced sand, the residual tensile strains in the geosynthetic reinforcement did not increase like global strains in the geosynthetic-reinforced sand that increased significantly during CL. These different trends of behaviour were also due to the creep compressive strains in the lateral direction of sand that developed during CL of geosynthetic-reinforced sand.

View full abstract

An extensive program of laboratory model tests was undertaken to study the effect of slope on p-y curves due to surcharge load in dry sand. The paper concerns the method developed in a series of laboratory model tests to experimentally determine p-y curves. Bending moment curves are differentiated by using curve fitting method of cubic polynomial function. The study includes effect of slope angle and relative density on bending moment, lateral soil resistance, lateral deflection and non-dimensional p-y curves. The non-dimensional p-y curves for piles on sloping ground under surcharge load are developed modifying API RP 2A (2000) method by including a Reduction Factor (R) using the experimental results.

View full abstract

For the purpose of establishing the method for estimating in-situ mechanical behavior of artificial buffer materials, stress-deformation behavior of bentonite-sand mixtures were investigated through oedometer test, consolidated undrained triaxial compression test and expansive stress-strain measuring test by changing the clay content as 30, 50, 70 and 100%, and by changing the range of initial dry density of mixture from 1.4 to 1.8 g/cm³. Oedometer test results suggest that the magnitude of consolidation yield stress almost coincides with the maximum expansive stress (p′_s)_max irrespective of bentonite-sand mix proportion, initial density of mixture and the magnitude of molding stress at the specimen making. Strong correlation between consolidation stress and initial tangent modulus during undrained triaxial compression test is observed, and it is found that the reduction rate of rigidity is hardly dependent on the specimen making method, molding stress and the consolidation stress. From the two series of expansive stress-strain measuring tests, it is recommended to perform the measurement of expansive stress by feed back system with the load cell installed at the base of the specimen. A unique relationship is found between the maximum expansive stress (p′_s)_max versus bentonite specific volume v_b, which is defined as the specific volume calculated by excluding the volume of sand particles. The line showing the unique log v_b versus log (p′_s)_max relationship can be recognized as the state boundary line prescribing one-dimensional expansive stress-strain behavior of the bentonite-sand mixtures.

View full abstract

This paper describes the results of large-scale shake-table experiments involving a 3×3 pile-group. The pile-group was embedded in dry sand and subjected to sinusoidal waves and an earthquake motion recorded from the 1995 Hyogo-ken Nanbu (Kobe) earthquake. The load transfer between soil and pile was derived and the group effect was captured. Numerical simulations were also performed using a Beam-on-Nonlinear-Winkler-Foundation approach with a new hysteretic p-y curve. A comparison of the experimental and numerical results revealed that the numerical simulation is capable of accounting for the soil-pile interaction observed in the experiment.

View full abstract

Risk assessment and design of remediation methods at soil sites polluted with gaseous phase contaminant require an accurate description of soil-gas diffusion coefficient (D_p) which is typically governed by the variations in soil air-filled porosity (v_a). For undisturbed volcanic ash soils, recent studies have shown that a linear D_p(v_a) model, taking into account inactive air-filled pore space (threshold soil-air content, v_{a, th}), captured the D_p data across the total soil moisture range from wet to completely dry conditions. In this study, we developed a simple, easy to apply, and still accurate linear D_p(v_a) model for undisturbed volcanic ash soils. The model slope C and intercept (interpreted as v_{a, th}) were derived using the classical Buckingham (1904) D_p(v_a) power-law model, v_a^X, at two soil-water matric potentials of pF 2 (near field capacity condition) and pF 4.1 (near wilting point condition), and assuming the same value for the Buckingham exponent (X=2.3) in agreement with measured data. This linear D_p(v_a) prediction model performed better than the traditionally-used non-linear D_p(v_a) models, especially at dry soil conditions, when tested against several independent data sets from literature. Model parameter sensitivity analysis on soil compaction effects showed that a decrease in slope C and v_{a, th} due to uniaxial reduction of air-filled pore space in between aggregates markedly affects the magnitude of soil-gas diffusivity. We recommend the new D_p(v_a) model using only the soil-air contents at two soil-water matric potential conditions (field capacity and wilting point) for a rapid assessment of the entire D_p-v_a function.

View full abstract

When geosynthetic clay liners (GCLs) are applied as bottom liners at waste containment facilities, they are naturally prehydrated by absorbing moisture in the underlying base layers. In order to evaluate the effects of cations contained in waste leachates, this study investigated the effects of the water content distribution of the GCLs prehydrated with actual soils on their hydraulic conductivities against CaCl₂ solutions. The “prehydration tests”, which were conducted prior to the hydraulic conductivity tests, showed that the water content distribution of the prehydrated GCLs depends on the properties of the GCLs and the base layers. In particular, drastic differences between GCLs with powdered bentonite and GCLs with granular bentonite were observed in the prehydration water content and its distribution. Prehydrated GCLs with powdered bentonite had a higher water content and a more homogenous distribution than those with granular bentonite. The hydraulic conductivity tests showed that most of the prehydrated GCLs exhibit a low hydraulic conductivity of k≈1.0×10^-8 cm/s against CaCl₂ solutions with 0.1-0.5 M. However, GCLs with granular bentonite may be difficult to homogeneously prehydrate and exhibit an unstable hydraulic conductivity, which varies from k=2.9×10^-9 cm/s to k=1.5×10^-6 cm/s. The homogeneity of the water content distribution has been considered an important factor to obtain a required barrier performance under prehydration conditions, which are naturally generated in actual sites.

View full abstract

Soil vapor extraction and bio-venting have been utilized for purification of contaminated soil or groundwater. It is necessary to predict the movement of gas phase components in soil for the design of soil vapor extraction and bio-venting systems. Though chemical substances migrate with advection and diffusion in gas phase of soil, we investigated multi-component diffusion systems in gas phase of soil. Numerical modeling for multi-component diffusion is useful to the prediction of the movement of components. A dusty gas model for multi-component diffusion systems has not so far been formulated by the Finite Element Method; furthermore it has not been applied for assessing the movement of components in the gas phase of soil. Accordingly, a dusty gas model for three gas phase components was formulated by the Finite Element Method in this study, and the concentrations of components in binary and multi-component gas systems were calculated by numerical methods developed in this study. As a result, it was found that the dusty gas model must be applied for study of diffusion in a multi-component gas system; and the study showed that the difference between molecular weights of gas phase components influenced the movement of components in the gas system.

View full abstract

For a wide excavation in soft soil, the excavation can be stabilized by an embedded improved soil berm to increase wall stability and control soil movement. An embedded stiff berm essentially behaves like a horizontal pile subjected to a load applied by the retaining wall and derives its resistance to horizontal movement from both end bearing and interfacial shear resistance on the top and bottom of the berm. This resistance helps to restrain the wall from moving inwards to the excavated side. However, to date, there is no known reported literature on the determination of the undrained ultimate bearing capacity of such a berm, especially for the unit end bearing, q_b. In this paper, the undrained end bearing of an improved berm under a plane strain condition was determined. The undrained end bearing capacity was first derived using a solution from a proposed upper bound analysis based on observations from centrifuge tests and then modified, taking on the basis of an equivalent finite element analyses. The proposed end bearing capacity factor N_c lies between the upper bound and lower bound solutions. The solution showed that the undrained end bearing capacity is not a constant but decreases during the excavation process. Furthermore, it was shown that the existence of an improved soil berm will provide an additional pressure relative to the passive pressure to control the wall displacement.

View full abstract

A number of cyclic triaxial tests were performed on Monterey No. 0/30 and Sacramento River Sand to investigate the effect that loading frequency has on the response of sands. The tests were performed on dense, air pluviated sand with loading frequencies of 0.1 and 1.5 Hz at varying confining pressures, cyclic shear stresses, and peak shear stresses. Under certain loading conditions, the frequency of loading did have a noticeable effect on the response of the sand; larger axial strains were measured in the samples that were subjected to the lower frequency of loading. This difference in response measured at the two loading frequencies occurred mainly in the first few cycles of loading, when the difference in the strain rates was the greatest. Conditions that resulted in larger axial strains, such higher stress levels and larger cyclic shear stresses, also resulted in a greater difference between the axial strains measured at the two loading frequencies.

View full abstract