In recent years, LNG (Liquefied Natural Gas) has been used in Japan. Because LNG is stored at a very low temperature (-162°C), the use of LNG underground tanks presents new geotechnical problems, such as thermal analysis of the soil around the tank, the mechanical properties of frozen soils and an evaluation of freezing earth pressure.This paper describes two kinds of tests undertaken to evaluate freezing earth pressure in stiff cohesive soil, for the purpose of checking an assumed design pressure adopted for the first LNG underground tank in Japan. This checking study comprises a preliminary laboratory test carried out to obtain an approximate value of freezing earth pressure and subsequently, a field test carried out on a model tank 2 m in diameter and 1 m high. An estimated value of maximum freezing earth pressure was obtained from the result of the field test, and it compared well with the results of analysis by the two-dimensional theory of elasticity. The relevance of the use of this experimentally obtained maximum value of freezing earth pressure seems to be proved, to some extent by the fact that the prototype LNG tank designed fortuitously on this value, has performed well for 10 years.

View full abstract

Many research works on the method of the reliability-based design for the geotechnical engineering field have been published for this decade and the applications to the practical design are tried now. In this phase, two problems which bother the engineers occur. One is difficulty and trouble to calculate the probability of failure and the other is the engineers' feeling to resist to accept the new design method. The present study mainly aims to overcome the former difficulty. The optimum design problems of an embankment constructed on a saturated clay layer are taken up and the charts to easily obtain the probability of failure are given after the general formulations are carried out. In order to soften the second difficulty mentioned above, the procedure to decide the probability of failure through the safety factor is introduced. It is possible for the engineers through this procedure to compare the results of the reliability-based design with those by their traditional methods. In addition, the loss functions to estimate the design alternatives are also formulated and for the purpose to make it easy to understand how to use the charts, several numerical examples optimizing the design of an embankment are shown and the results are discussed in comparison with those by the traditional safety factor method.

View full abstract

The effect of reconsolidation on the undrained strength was experimentally studied for cohesive soils subjected to disturbance due to stress release and sample preparation. Three types of cohesive soils were used; Kawasaki clay with plasticity index of about 50 and two artificial mixtures of Kawasaki clay and Toyoura sand with plasticity index of about 20 and 10. Reconsolidation pressures employed were approximately 0.5, 1, 2 and 4 times the preconsolidation pressure of 98 kN/m². Pore pressures were monitored by an embedded transducer during the whole stages of preconsolidation, stress release, sample preparation, reconsolidation and shear tests. It was found that soils with lower plasticity indices were more susceptible to disturbance, showing the loss in undrained strength of 50 to 60% of original strength. With respect to the relationship between the void ratio and consolidation pressure, the reconsolidation pressure four times the preconsolidation pressure did not completely bring the trimmed specimens back to the normal consolidation state. The undrained strength increased and the strain at the maximum deviator stress decreased with the increase in reconsolidation pressures. Even at the reconsolidation pressure four times the preconsolidation pressure the effect of disturbance did not vanish for the ratio of undrained strength of reconsolidated specimen to that of ideal samples, while for the effective stress paths the effect was almost negligible. The ratio of increase in undrained strength with consolidation pressure of trimmed specimens became nearly equal to that of ideal samples at the reconsolidation pressure twice the preconsolidation pressure

View full abstract

Soft clay deposits have generally been aged for hundreds to thousands of years since deposition and consequently they posses an additional strength developed by aging effect. Because of this additional strength, soft clay deposits are usually in an overconsolidated state even though they have not been subjected to release of the overburden. This type of clay is so called normally consolidated aged clay.Undrained strength characteristics of this type of clay are discussed in detail in this paper. Fundamental concept required for evaluating the properties of clay subjected to aging effect is first presented and a method to evaluate the effect of aging is indicated. Undrained strength of normally consolidated aged clay in terms of both total and effective stresses are discussed and various characteristics obtained are presented. Discussions are also given to the stress-strain behaviour and the K₀ value of this clay. It is demonstrated that these characteristics of clay subjected to aging effect are quite different from those of overconsolidated clay formed by release of the overburden.

View full abstract

Undrained shear strength of clay is generally increased due to secondary compression or delayed compression. Though this increased strength is regarded to be derived from densification of clay particles, metamorphosis of clay minerals and dissolution of cemented matters into free water, mechanism of increase in strength during long term consolidation has not completely been clarified.In the present paper, described is a method for estimating the amount of increase in undrained shear strength of clay due to secondary compression. The most important postulate in derivation of the basic equation is that the quasi-overconsolidation of clay induced by secondary compression should be equivalent to overconsolidation due to release of over-burden pressure. In this way, the simplified relation for estimating the secondary compression-induced increase in strength is expressed by means of comparison with the strength of normally-consolidated clay. The relation suggests the increase in strength to be a power function of the elapsed time during secondary compression, including three parameters which can be obtained from oedometer tests and triaxial compression tests. The proposed method is proved to be valid for estimating the increase in shear strength of a reconsolidated clay and an undisturbed clay in direct shear tests. Additionally, the strength increment ratio defined by means of the equivalent consolidation pressure of Hvorslev (1960) is shown to be independent of the elapsed time of secondary compression.

View full abstract

A strain measurement system is developed to determine exact strain distributions in soil in model tests using an X-ray technique. This method can determine displacements and strains in soil by tracing images of small lead shot recorded on X-ray films. The lead shot are embedded in advance in soil in a lattice pattern. The reliability of the method therefore depends upon the accuracy of determining the coordinates of lead shot images. An equipment called "X-ray Film Reader" which satisfies this requirement is developed. The method has also in advantage of detecting rupture lines on X-ray films as dark curves, which represent the failure mechanism of soils, since the void ratio exceedingly increases there. As an example the method is applied to model tests to study the behaviour of sand around a buried pipe, where the adjacent sand settles. In this case the failure mechanism of sand is usually based upon the projection condition of the Marston-Spangler theory. Several interesting observations are made from the model tests that the configuration of rupture lines largely depends on the initial void ratio of backfilled sand and differs from the sliding planes assumed in the Marston-Spangler theory, and that the maximum shear strain is prominent along rupture lines recorded on X-ray films.

View full abstract

This paper presents the theoretical and experimental studies on the behavior of soil in front of laterally loaded piles. Based on the three-dimensional elastic theory considering the displacement of soil in the plastic zone, the theory to predict horizontal displacements of soil being in the elastic and plastic conditions in front of piles is proposed. In applying the proposed theory to the test data, the methods to obtain the elastic modulus of the soil and the ultimate resistance of the soil in the field are also discussed.The results of seven field tests carried out by authors and other investigators are examined, and the horizontal displacements of soil in these results are compared with the computed ones based on the proposed theory and the elastic theory in which the plastic state of soil is not considered. It is found that the proposed theory can predict the test results fairly well, but the values obtained by the elastic theory depart greately from the measured values.

View full abstract

There is a limit in application of the conventional analytical methods by the elastic and plastic theories to complicated soil conditions. The finite element method is capable of considering the nonlinear characteristics of soil and of dealing with a complicated ground.Contrary to such an advantage of the finite element method, problems lie in its analysis techniques and errors related thereto. In this study, analyses of the pullout resistance of a buried anchor were performed by an finite element analysis program based on the Lade's constitutive equation to clarify ground stresses, maximum pullout resistance, etc.The results were compared with one of the pullout test of an anchor by a centrifugal testing machine and the effects of various factors used in the analyses on their results were investigated. The following items have been clarified : (1) the analysis of a shallow anchor requires finer meshes to be taken diagonally from the anchor edge to the ground surface; (2) the Jaky's equation for the coefficient of earth pressure at rest to obtain the initial stress gives a greater pullout resistance for φ<35°; and (3) a region extending to the side more than three times the anchor width must be included for analyzing a shallow anchor.

View full abstract

A questionnaire survey was conducted to ascertain the current practice of standard penetration test (SPT) in Japan. All 53 respondents perform the test according to the Japan Industrial Standard using drilling mud to stabilize boreholes in sand below groundwater table. The cathead and rope method is more popular than the trip monkey method, although about 75 percent of the respondents have used either method. The cathead and rope method is typically performed by turning a 15 mm manila rope twice aroung a 120 mm cathead.A level site in Niigata where the soil profile consisted of horizontal sand strata enabled a direct comparison of the N-values determined by the cathead and rope method and those by the trip monkey method. The test results show that the cathead and rope method yielded significantly higher N-values than the trip monkey method where N exceeded about 24, whereas either method produced about the same N-values where they were below 12.

View full abstract

The present paper is the part of a series of study to find out a simple engineering method of performing the reliability-based design of an embankment constructed on a poor clay layer. For the purpose, it is necessary to make the accurate charts and simple equations to obtain the central safety factor and the probability of failure. The discussion from such a viewpoint is carried out associated with a method of improvement of a clay layer by sand compaction piles and some numerical examples of its optimization are shown in order to verify the usefulness of the simple engineering method. The charts to determine the location of a slip circle and the probability of failure are not involved in this paper since the same charts previously published in another paper (Matsuo and suzuki, 1983) are also used in the present study.

View full abstract

The model test results for the ultimate uplift capacity of piles of Das, Seeley and Pfeifle (1977) have been extended. Based on the model test results, it has been found that the unit skin friction at the soil-pile interface increases linearly with depth up to a critical embedment ratio. The critical embedment ratio is dependent on the relative density of the soil. Based on the laboratory experimental results, a tentative procedure for estimation of the ultimate uplift capacity of piles has been presented.

View full abstract