One of the major unsolved problems in geotechinical engineering is how to decide the shear strength parameters of the soil in a shear zone to analyze landslide behavior. It is often reported that the shear strength parameters obtained in laboratory tests on undisturbed soils sampled from landslides disagree with the back calculated ones. Therefore, the authors investigated the practical application of direct shear test to decide the relevant shear strength parameters of the soil in shear zones, and the following conclusions were obtained. 1) The stress state on the horizontal plane, i. e. the apparent shear plane, would ultimately reach the critical condition as the result of the restriction of the shear plane, satisfying the Mohr-coulomb failure criterion in direct shear tests. Therefore, the inclination angle of rupture line (φ''_d) ovserved in direct shear tests is considerably larger than the one of Mohr-Coulomb rupture line in τ_yx, σ'_y Space. 2) The friction angle at Mohr-Coulomb rupture (φ') can be theoretically predicted from the result of direct shear tests by the modified Cam-Clay theory. 3) The value of φ' predicted above almost agreed with the back calculated one for normally consolidated soils. 4) This prediction method of φ' could be also applied to lightly overconsolidated soils.

View full abstract

One method currently being considered for the disposal of high level radio-active waste is burial in the seabed. When a hot object is buried in soil, the temperature of the soil adjacent to the object is raised and there is transient heat flow away from the object. The rise in the temperature of the soil causes excess pore pressures to be generated and transient pore water diffusion also takes place. A finite element formulation of the coupled heat flow and consolidation problem is presented. The computer implementation of this formulation is found to compare favourably with analytical solutions based on the same physical principles.

View full abstract

It is important to thoroughly understand the particle breakage phenomenon and to be able to comprehend quantitatively its effect with regard to engineering properties. In this paper, the particle breakage phenomenon observed during soil testing is investigated, and an attempt is made to clarify the particle breakage characteristics of granular soils by analyzing the soil test data presented by several researchers using uniformly graded soil samples. As a result, several facts have been revealed, one of which is that a certain regularity is recognized among grain size accumulation curves obtained from tests conducted under various testing conditions. Based on these findings, consideraton is given to the meaning of the "Comminution theory" that is currently used in other fields.

View full abstract

Partial drain behavior of soft saturated clay foundation is discussed under embankment loading. The partial drain effect both on a consolidation settlement and on a bearing capacity is totally examined in terms of both soil permeability and load intensity, the latter of which is shown to be normalized by the undrained bearing capacity that should be evaluated at the stage just before embanking execution. Through a series of numerical computation and the case record study, the following findings are obtained : (1) The range of soil permeability which yields the variation of partial drain behavior has been found 10^-7cm/sec<k<10^-4cm/sec when a clay deposit of about 25 meters thick is considered. This is also corresponding to the usual loading rate of embanking in engineering practice of soft clay. Out of this particular range of parmeability the clay foundation will behave almost in a perfectly undrained manner or in a fully drained manner. (2) When soil permeability lies on this particular range the bigger the load intensity, the more the partial drain effect should be expected. (3) The soil permeability representative of a clay foundation, sometimes referred to as mass permeability, can be evaluated inversely based on the observation of consolidation settlement when a load intensity is normalized with the undrained bearing capacity of the foundation. The procedure is shown to be still applicable, through a case record study, even for a clay foundation that is improved by a number of vertical sand drain piles. In this context, the inference is also made in the present study that the efficiency of sand drains is to be observed in a limited range of soil permeability and under a particular magnitude of load intensity.

View full abstract

A shear strength reduction technique for finite element slope stabilityan alysis has been developed by the authors. An important orginal point in the proposed method is that the slope failure is defined according to the shear strain failure criterion. The aim of this paper is to verify the shear strength reduction technique for the finite element slope stability analysis. Are presented the detailed background behind the shear strength reduction technique, the elucidation of the physical meaning of the critical shear strength reduction ratio in regard to the total shear strain and shear strain increment for both embankment and excavation slopes and its practical application to a field test on a reinforced slope cutting. As the results, the critical shear strength reduction ratio agrees with the safety factor by the Bishop's method if total shear strain is used for analyses of embankment slopes. In the case of the natural excavation slopes, in which total shear strain is difficult to be assessed, the safety factor can be related to the average of the local safety factors along the failure slip surface obtained by the shear strength reduction technique. The predicted behavior of the reinforced slope cutting agrees with the field test data and site observation. Agreement between the shear strength reduction technique and a modified Fellenius' method is satisfactory. Consequently, applicability of the proposed method to practical design works is demonstrated.

View full abstract

In order to evaluate soil parameters free from regional variations both in soil behaviour and technical quality, a new procedure to determine the shear strength for design use, τ_f(mob) as well as consolidation yielding stress, σ'_y is proposed. Various methods to determine τ_f(mob) in current practice are first reviewed and it is then revealed that regional difference in technical quality is significant. It is indicated that cone penetrometer, sheathed field vane and direct shear tests are recommended as the tests to minimize the effect of quality difference together with that the direct shear strength offsets the strength anisotropy and be uniquely related to τ_f(mob). Results of investigations with the proposed method consisting of three tests described above are presented for three marine clays. Point resistance, field vane, direct shear and unconfined compression strengths were investigated and the correction factor to determine τ_f(mob) from these strengths are shown together with those proposed in various methods. It is demonstrated that σ_y'can be obtained from the shear strengths both in the field and in the normally consolidated state, and τ_f(mob) can uniquely be related to σ'_y irrespective of clay properties. Further, it is also shown that τ_f(mob) and σ'_y can be predicted without undisturbed samples when vane shear and direct shear strengths in the normally consolidated state are obtained.

View full abstract

Effects of fabric anisotropy on the mechanical behavior of granular materials have become well known. Without the introduction of variables describing fabric anisotropy into the constitutive formulation, some important deformation features of granular materials can not be formulated in a reasonable way. Although the principle of material objectivity and the thermodynamical restrictions in the continuum mechanics give the framework to a class of reasonable constitutive equations associated with anisotropic features, the framework is still too wide to obtain a practical constitutive equation. This fact indicates the necessity of conventional methods within the fundamental restrictions. Modified stress tensors being a consecquence of the linear transformation depending on fabric anisotropy from the stress tensor are poposed and discussed, with a particular emphasis on the elasto-plastic behavior of anisotropic granular materials. It is shown that : (a) classical hardening models (isotropic, kinematic, and combined) are obtained as special cases of the present method ; (b) the shape change of yield surfaces during plastic deformation is a natural consequence if relevant terms are included in the transformation rule. Anisotropic deformation and strength behaviors of granular materials are formulated based on the modified Drucker-Prager yield condition and the non-associated flow model for two dimensional problems in order to demonstrate the effectiveness of the proposed method.

View full abstract

The effects of undrained cyclic loading on clay behaviour were investigated using triaxial tests. The current investigation is divided into behaviours during and after cyclic loading. The authors present concepts and methods for predicting both behaviours. The following conclusions were made using those methods : (1) The stress reversal observed during undrained cyclic loading accelerates excess pore pressures and residual shear strains, leading to cyclic failure. (2) Undrained strength decreases with the amplitude of cyclic-induced excess pore pressures, while post-cyclic undrained strength followed by drainage increases in proportion to the amplitude of cyclic-induced pore pressure. (3) Post-cyclic recompression volumetric strains are governed by the magnitude of cyclic-induced pore pressures. Those values can be approximately estimated using the recompression index obtained from isotropic consolidation tests.

View full abstract

A series of triaxial tests was performed on undisturbed samples of soft Bangkok clay, which were consolidated under different constant stress ratios, to study the pore pressure development under monotonic loading conditions. The test results are interpreted through a pore pressure : stress ratio relationship. The behaviour of isotropically consolidated samples in the normally consolidated state shows a linear relationship between the pore pressure and the stress ratio under monotonic loading conditions and the same patterns were also established for other clays such as kaolin, Drammen clay and Boston Blue clay. Overconsolidated samples, and anisotropically consolidated samples showed bi-linear variation on the same plot. The linear/bilinear variation in the pore pressure-stress ratio relationship was used to develop simple expressions for the undrained stress paths for different initial conditions of the samples, using simple soil constants.

View full abstract

Undrained strength, S_u normalized with the effective overburden stress, σ'_vo or the consolidation stress in the laboratory tests is an important index in evaluating engineering properties of a clay. Extensive studies have been made on this subject and the results mostly suggest that there is a correlation between the normalized undrained shear strength and plasticity index, I_p. In this paper, undrained strength in the normally consolidated state, S_un normalized with the vertical consolidation stress, σ'_vc obtained from K₀-consolidated triaxial compression and extension, and direct shear tests is first presented for various clays with I_p=5 to 100 together with discussions on previous studies on this subject. Undrained strength in the field, S_uf normalized with σ'_vo and the consolidation yielding stress, σ'_y investigated by field vane, direct shear and oedometer tests is then presented for 7 clays in Arabian Gulf, Indonesia and Japan. Further, a strain rate effect on S_u is also investigated. Conclusive points obtained from this study are : -1) Normalized undrained shear strengths, S_un/σ'_vc and S_uf/σ'_vo as well as strength anisotropy are not correlated to I_p, 2) S_uf of clays investigated is expressed by S_un+k₁ to a certain σ'_vo and S_un× k₂ beyond this σ'_vo where k₁ and k₂ are constants, 3) S_uf/σ'_y shows a constant value being close to S_un/σ'_vc irrespective of overconsolidation ratio and I_p, and 4) the effect of strain rate on S_u is independent on I_p.

View full abstract

To study the mechanism of permanent ground displacement due to liquefaction, shaking table tests on 24 soil models were conducted. In these models, the slope of the ground surface, the thickness of a liquefied layer, the density of the liquefied layer and some other factors were changed. Permanent ground displacements did not occur at the boundary between the liquefied layer and the lower non-liquefied layer, but occurred in the liquefied layer with a constant shear strain. Relationships between the displacement and the thickness of the liquefied layer or the slope of the ground surface were also obtained. Then, vane shear tests and cyclic torsional shear tests were carried out to measure the rate of reduction of shear strength or shear modulus. By the cyclic torsional shear tests, it was clarified that the shear modulus decreases to a very small value due to liquefaction. Based on these test results, a simplified procedure for the analysis of permanent ground displacement was proposed. To confirm the accuracy of this procedure, it was applied to the models used for the shaking table tests and to typical soil cross sections of soil at Niigata City. Finally, some measures to prevent permanent ground displacement were discussed based on the shaking table tests and analyses.

View full abstract

This paper represents a series of tests performed to evaluate the static undrained shear degradation of saturated clays after cyclic loading. Both remolded and undisturbed clays were used for this purpose, and the samples are selected to cover both normally and over-consolidated states. First, the stress-strain-pore pressure behavior of these types of clays was presented and compared with that without a cyclic loading effect. Secondly, particular attention has been paid to the equivalent over-consolidation phenomenon observed for the normally consolidated specimens, in their subsequent response to undrained shearing, and the equivalent over-consolidation hypothesis was established. Thirdly, based on the experimental results and the equivalent over-consolidation hypothesis, the shear characteristics degradation relationships were constructed for both remolded and undisturbed normally consolidated clays. Finally, a method for predicting the stress-strain behavior after cyclic loading of random amplitudes was proposed and reliable results were observed.

View full abstract

By examining a bulk of laboratory test data on sands obtained by using a simple shear apparatus, a family of curves was established in which the volumetric strain resulting from dissipation of pore water pressures is correlated with the density of sand and conventionally used factor of safety against liquefaction. Thus, given the factor of safety and the density in each layer of a sand deposit at a given site, the volumetric strain can be calculated and by integrating the volume changes throughout the depth, it becomes possible to estimate the amount of settlements on the ground surface produced by shaking during earthquakes. The outcome of the data arrangements as above was put in a framework of methodology to estimate the liquefaction-induced settlements of the ground. The proposed methodology was used to estimate the settlements at several sites devastated by liquefaction during the 1964 Niigata earthquake. The estimated values of the settlements were examined in the light of known performances of the sand deposits at respective site. It was shown that the proposed methodology may be used for predicting liquefaction-induced settlements with a level of accuracy suitable for many engineering purposes.

View full abstract

A new limit equilibrium analysis for assessing the stability of a reinforced embankment on soft clay against rotational undrained failure is presented. The analysis considers strain compatibility along the slip surface to deduce the tension mobilized in extensible reinforcement at collapse condition, and models strain softening of soils along the slip surface. The load extension responses of the reinforcement can be dependent on confinement by soils. Non-uniform increase in undrained cohesion during construction can be accounted for. The analysis yields a 'Safety Factor Curve' instead of a unique value for safety factor. A parametric study for illustrating the relevance of the proposed analysis is reported.

View full abstract

The multiple regression model with parameters of N-value and depth implies a form of shear wave velocity with a specific physical meaning, but these two parameters are not independent of each other. The research introduced statistical analysis techniques of the stepwise selection, the "rule of thumb" test, and the "RSM" test to overcome the multicollinearity problem. By using 491 data collected from the Taipei basin, the results show that only SM soil failed to pass the criteria while the CL and ML soils successfully did. Thus, the multiple regression model in a form of V_s=aN^bD^c may be used in some soils cases.

View full abstract

This paper describes not only the application of Forchheimer's formula, which is a discharge equation for flat, open-bottom impermeable casing wells, to the estimation for the discharge of dewatered excavation as a large circular well but also the practical utility of a proposed method for estimating the radius of influence which was derived from Forchheimer's formula. The investigation was conducted by comparing the observed results of an actual dewatered excavation with the calculated results. Also, the study was made of the application of the existing methods for estimating the radius of influence in the dewatered excavation as a large circular well with open-bottom impermeable casing. The results of the investigation are as follows : (1) The calculated results of discharge by Forchheimer's formula are in good agreement with the observed results ; the error between these two results is under 14 per cent. (2) The calculated results of the radius of influence by the proposed equation with s_R = 0.2m agree approximately with the observed results ; the error between these two results is under 20 per cent. (3) The calculated results of the radius of influence by the existing methods are quite different from the observed results.

View full abstract