This thesis is concerned with the general and fundamental engineering characterisation of a geological formation within Wianamatta group, known as Bringelly shale. Bringelly shale is the neighbouring member of Ashfield shale; both are soft rocks of Triassic age within a geological structure known as the Sydney basin in the state of New South Wales, Australia.
Bringelly shale rock and its residual material cover an approximate area of 700km2. It is found inland, to the west of the city of Sydney, where most of the new residential, commercial and industrial development is taking place. There is a limited amount of existing experimental data in part due to the technological difficulties in obtaining specimens and this has contributed to the uncertainties surrounding the engineering behaviour of the rock.
In this research, efforts have been made to identify index property tests useful for determining the engineering characteristics of the material. Further aims were to explore the reasons for the difficulty of obtaining core specimens using standard water flush drilling techniques and, to ascertain and explain why Bringelly and Ashfield shales behave differently in many aspects of their engineering performance, even though they are members of the same geological group.
Qualitative and semi-quantitative analysis by the X-ray diffraction technique was used to evaluate the clay mineralogy of the Bringelly shale materials at different degrees of weathering. Thin sections were examined by optical microscopy to study the nature of cementation and bonding. Polished sections of natural and reconstituted specimens were examined by electron microscopy to investigate the internal structure of each material and its mineral composition. It has been found that the presence of a significant amount of swelling clay and microcracks in the plane of laminations are responsible for increasing the swelling potential of the Bringelly shale. There is little evidence of induration and only apparently weak bonding due to re-crystallisation of mica at particle contacts. Changes in particle alignment following failure were also observed.
Because of the difficulty of obtaining specimens suitable for UCS testing, correlations were established between the point load strength index and the measured values of uniaxial compressive strength in the direction perpendicular to laminations. The strength anisotropy from the point load index was also determined. In this research, it was found that due to the limited number of specimens tested for UCS, the determined correlation factor could over-predict the strength of the shale.
Durability and swelling of the shale were also investigated. The durability of Bringelly shale was found to vary from medium for fresh intact material to very low for extremely weathered material. To further investigate the mechanisms responsible for the durability of the shale, unconfined and confined swelling tests were performed. Volumetric strains of 6-8% were measured for cube specimens with a volume of 27000 mm3, however, the material has shown an inverse relationship between its volumetric swelling and specimen dimensions. The chemical composition of the fluid into which the specimen was immersed was found to have a major influence on volume changes in the intact material. The results of the investigation confirmed that potassium chloride solution can be used to reduce swelling potential, and further, to improve core recovery during drilling.
An extensive experimental program to investigate the engineering performance of the shale has involved the use of conventional and specialised high pressure triaxial equipment. The program investigated the volumetric compression and shearing behaviour of three different forms of specimen. These were natural core specimens, and reconstituted specimens created from crushed shale by either pressing dry powder in a mould or by compression of a slurry.
Isotropic consolidation tests over a wide range of stresses were performed. The program has also involved a series of drained and undrained triaxial strength tests on the three different forms. The series has covered a wide range of confining effective stresses from 20 to 60,000kPa, degree of saturation from 65-100%, and porosities from 10% to 60%. These tests have provided an extensive set of data to investigate the influence of stress, saturation, suction and internal structure on the compression behaviour of the reconstituted and natural rock. Analysis of these data has been conducted in terms of cementation, swelling, saturation, confining stresses, and frictional resistance.
A series of standard direct and ring shear tests has been carried out on the reconstituted Bringelly shale at normal stresses in the range from 50 to 200 kPa, and a residual friction angle was determined. It was found that this value has not been affected by the reorientation of clay particle despite the high clay fraction content of the material.
The results of this research indicate that the general pattern of behaviour for reconstituted material that has experienced a maximum effective stress of less than 6 MPa is consistent with the assumptions of critical state soil mechanics and similar to many other reconstituted materials. This pattern of behaviour shows a significant deviation from the framework of critical state when the same material (slurry or core form) is subjected to a maximum effective stress of 60MPa. The significance of bonding and structure of the intact shale could be detected from investigating the same material at reconstituted state. However, further development of the critical state framework is required to take into account the reduction in strength caused by the high degree of alignment of clay platelets. The OCR seems to have minor effect on the strength of the material.