No abstract

Introduction Sands are known to crush under stresses at high confinement. Particle crushing is more predominant under shear stresses than under isotropic compression. Stresses in soil under deep foundations often reach the range where crushing becomes significant. Calcareous sands crush more readily than terrigenous quartzitic sands. Calcareous sands occur as skeletal remains of marine organisms and/or as nonskeletal oolites of calcareous material. Skeletal calcareous sand grains are characterized by the presence of intraparticle voids that increase the tendency of these sands to crush under stress. Presence of thin plate-like shell fragments further increases the crushability of calcareous sands.Deposits of calcareous sands are encountered extensively on the continental shelves lying between latitudes 30 degrees N and 30 degrees S. In many such marine areas, jacket-type offshore structures for oil production are being founded on calcareous sand stratae.g., in the Arabian Sea offshore India, in the Bass Strait offshore Australia, and in the Persian Gulf offshore Saudi Arabia. The foundations of these structures consist of deep-penetration piles. These piles derive their axial load capacity from the skin friction resistance and the end bearing resistance offered by the soil strata they penetrate. Static pile load tests in calcareous sand strata have exhibited unusually low values of unit skin friction and unit end bearing capacity. It is believed that particle crushing that occurs during pile driving and static loading is one of the causes for the low pile capacity. For design of deep-penetration piles in calcareous sand strata, low values of limiting unit skin friction and limiting unit end bearing capacity currently are adopted in comparison to those being used for terrigenous quartzitic sands. To arrive at a rational design practice for piles in calcareous sands, one must begin by (1) identifying a method for quantitatively expressing the sands' propensity to crush, (2) determining the extent of crushing of different types of calcareous sands in comparison to terrigenous sands, and (3) isolating the influence of crushing on shear behavior, which partly controls the end bearing and skin friction offered by such sands.This paper describes the results of an experimental investigation designed to achieve these three objectives and, in part, offers an explanation for the low skin friction and end bearing values in field tests. Quantitative Expression for Degree of Crushing Different methods of quantitatively expressing the degree of crushing have been proposed by various investigators. Lee and Farhoomand studied crushing of material under anisotropic compression and have defined a parameter designated as relative crushing, which is equal to D15i/D15a, where D15i is the diameter through which 15% of particles of the original material pass and D15a is the diameter through which 15% of the particles pass after being subjected to anisotropic compression. Ramamurthy and Lal have expressed quantitatively the degree of crushing as the area between the grading curves of a sand obtained before and after it has been subjected to shear. Lowrison suggested that the amount of crushing can be estimated by evaluating the increase in the surface area of the particles after crushing. SPEJ P. 77＾

It is difficult to obtain enough undisturbed submarine samples from a particular location to carry out triaxial compression tests in the conventional manner using three samples at three different cell pressures. Multistage triaxial tests offer an attractive means for determining shear strength parameters of submarine soils since they use only a single sample. This paper presents a study through which a rational procedure for multistage undrained triaxial testing of submarine soils has been evolved. For this purpose tests were conducted on laboratory sedimented kaolinite. This rational procedure was then used to determine shear behavior of a submarine soil from off the west coast of India and then verified by comparing it with conventional tests. It is shown that to ensure sufficiently good correspondence between results obtained from conventional and multistage testing, the sample in multistage testing must not be sheared to failure in each stage. The sample should be sheared to about 2 to 4% axial strain in each stage, and then failure condition predicted by using Kondner's hypothesis and extrapolating to a finite strain. The influence of finite strain as well as the cell pressure to be used in successive stages of the multistage test has been studied and suitable recommendations have been made.

No abstract

It is now well recognized that the presence of carbonate material in submarine soils often affects their engineering behavior significantly because of (1) crushing of carbonate particles under stress and (2) cementation by carbonate material. This paper shows that depending on the nature of carbonate material, the magnitude of crushing and degree of cementation both vary over a wide range and have a markedly different influence on engineering behavior of carbonate soils. Existing classification systems do not take adequate note of this fact and hence are not totally satisfactory from an engineering viewpoint. A system of description is proposed for carbonate soils which, it is hoped, will in due course provide information to evolve a useful engineering classification system.