A study on brittleness of granitic rocks in Hong Kong

Abstract

Brittleness is one of the crucial factors that controls rock breakage and deformation behaviour. The objective of this study is to extract relevant test results from routine laboratory testing records on granitic rocks in Hong Kong for brittleness assessment and analysis. It focuses on both the pre-peak and post-peak strength parameters, as well as the elastic and plastic properties under uniaxial compression. It also discusses the calculated brittleness indices, its practical use, application in the fields of TBM tunnelling and aggregate mining in Hong Kong and its limitations. The term ‘brittleness’ has been viewed differently among researchers in the academia without any consensus reached on the definition and formulation. More than 60 definitions have been proposed based on the analysis of rock strain, strength, mineralogy or energy accumulation and consumption under uniaxial compression tests or triaxial compression tests. After considering the influencing factors on rock behaviour, 3 hypotheses are put forward and to be testified under this study. The first hypothesis is that the rock brittleness increases with grain size. Longer grain boundaries provide more weakness planes, leading to reduced energy for crack propagation. Minerals with cleavage planes, such as feldspars, manifest better in larger-sized grains and promote the ease of crack propagation. The second hypothesis is that the rock brittleness increases with quartz content as being brittle is a physical property of quartz. The third hypothesis is that the rock brittleness increases with the presence of shearing features. For rocks experienced shearing, small cracks are expected to be present already. This eliminates the energy required to form new cracks and allows pre-existing cracks to propagate more easily. In order to conduct the analysis and testify the hypotheses, relevant information on uniaxial compressive strength test results was extracted from the existing ground investigation logs and HOKLAS accredited laboratory testing records from Ginfo, Dr. Louis N. Y. Wong and the 1:20000 HKGS Series HGM 20 Geological Maps. The computer software “ImageJ” was used to calculate the strain and energy from the stress-strain curve. A preliminary selection of 17 brittleness indices based on strength, strain and energy characterizations were assessed and eventually 4 of them were used for analysis due to limited data availability. The results could not affirm the hypotheses and the 4 calculated brittleness indices were distributed over a small variation range in all the textural parameters, which is possibly attributed to weak assumptions, imprecise factor definitions and difficulties in obtaining post-peak parameters. Although no significant correlations could be observed between all the textural parameters and brittleness indices, this study has gone some way towards enhancing our understanding on the laboratory testings practice and the application of brittleness indices, which are TBM tunnelling and aggregate mining, in Hong Kong. The use of brittleness index has to be seen in light of some limitations. One is related to the deficiencies in the brittleness equations and the other is the applicability to the practical world. It should be further studied and validated by a larger sample size on the postpeak stress-strain curves and more accurately defined parameters.