Ayling, Mark Raymond; (1991) An experimental study of physical property changes in crustal rocks undergoing triaxial deformation. Doctoral thesis , UCL (University College London).

Preview

Text Ayling_thesis.pdf Download (15MB) | Preview

Abstract

A laboratory investigation has been undertaken to examine changes in a number of physical parameters of deforming brittle rocks. The experiments were carried out in a pressure-balanced. gas-medium triaxial cell. which is capable of simulating lower crustal conditions. During sample deformation. contemporaneous measurements were made of differential stress, axial strain, compressional wave velocity (Vp), shear wave velocity (Vs) and received elastic waveforms, which were stored for later analysis. In an alternative operational mode, simultaneous measurements were made of differential stress, axial strain, acoustic emission (AE) statistics and porosity changes through direct pore volumometry. Four different sedimentary rocks have been systematically examined: Darley Dale sandstone, Gosford sandstone, Solenhofen limestone and Tennessee sandstone, at confining pressures up to 200MPa and at ambient temperature. A number of major improvements have been made to the triaxial deformation system and to the electronic data acquisition and control equipment during this study. Specifically. the commissioning and performance evaluation of a newly-built servohydraulic actuator is described. The integration of a more powerful control and data-logging computer with new elastic wave velocity measurement and display equipment is also described. An important aspect of this developmental stage of the study was the enhancement of the acoustic signal transmission/reception system which allowed simultaneous measurements of both Vp and Vs to be made; therefore, transducer theory is considered in detail. Also during the course of this study, a new servo-controlled pore-fluid pressure intensifier and volumometer was commissioned. This device was interfaced with the triaxial cell so that direct measurements of changes in sample porosity could be made during deformation. Simultaneous compressional and shear wave velocity measurements were carried out on Darley Dale sandstone samples at confining pressures ranging from 30MPa up to 200MPa. Under initial deviatoric loading Vp and Vs increase slightly (1% to 5%) with Vp increasing relatively more than Vs. As dilatancy ensued. Vp and Vs both decrease (generally from 8% to 20%) with Vs decreasing more sharply and proportionately more than Vp. In general, as the confining pressures were increased, the velocity increases became less and the velocity decreases became more gradual. Similar trends were observed for the Gosford sandstone tested under comparable conditions. These velocity changes are interpreted in terms of a model of microcrack populations closing and opening as a function of stress, strain and time. The recording of both velocities simultaneously has enabled changes in the effective moduli (including the dynamic Young's modulus, and dynamic Poisson's ratio) along with the crack density parameter (£) to be estimated. The changes in these parameters are discussed for Darley Dale sandstone samples showing both catastrophic brittle deformation and pseudo-ductile deformation. As dilatant cracking occurs the effective Poisson's ratio increases, and this is interpreted as resulting from inelastic radial strains. At the same time the effective Young's modulus decreases, and the vertical crack density parameter (which reflects the cracking aligned in the axis of maximum stress) also increases. Furthermore, for the first time, these measurements have allowed seismic Q values to be determined for a deforming rock sample using the spectral ratio method. The seismic Q values for a Darley Dale sandstone sample at SOMPa confining pressure \W~ound to increase sharply under initial deviatoric loading to approximately twice their initial values, and then to decrease steeply as dilatant crack growth occurred. Velocity measurements on Solenhofen limestone samples at confining pressures up to 200MPa are also reported. However, in contrast to the results for the sandstones, Vp was found to decrease proportionately more than V s during ductile deformation. This result is interpreted as showing plastic processes may be occurring in this rock type, at least at elevated confining pressures. Experiments were also carried out at SOMPa confining pressure on samples of Tennessee sandstone in order to examine the effect of its strong bedding on the mechanical and acoustic properties of the rock when the bedding was orientated parallel and perpendicular to the maximum stress direction. The rock was found to be markedly stronger and failure to be less brittle when the bedding plane is perpendicular to the axis of maximum load. Results from drained and undrained water saturated tests conducted at l00MPa confining pressure on samples of Darley Dale sandstone and Solenhofen limestone are also described. These results show that the strength of sandstone is reduced markedly by the presence of interstitial water. Sandstones are weakest in the undrained tests. Experiments were also carried out in which the pore-fluid pressure was maintained at a constant level by means of the servo-controlled pore-fluid pressure intensifier. Three tests are described in detail for Darley Dale sandstone samples exhibiting catastrophic localised failure, distributed pseudo-ductile deformation, and a failure mode transitional between these two. The strength of the sample showing localised brittle failure is comparable to that seen for dry samples of this rock at equivalent effective confining pressures. However, the stress drop after peak stress is less in the saturated sample. The strengths of the samples showing transitional and pseudo-ductile behaviour are less than would be expected from application of the simple effective stress law. This discrepancy is considered to be the result of incomplete sample drainage causing local pore-fluid gradients which reduce the effective confming pressure. AE waveforms were recorded and stored for the test on the sample exhibiting localised brittle deformation. The amplitude spectra of these waveforms show that after peak stress the amplitude of the waveforms increase by an order of magnitude, and the dominant frequency in the waveforms decreased substantially. These observations are interpreted as reflecting an increase in the crack source dimensions following peak stress. Direct porosity change measurements which were also made during these tests, are brought together with crack density values determined for dry Darley Dale sandstone samples under comparable effective confining pressures to determine, for the fust time, the crack aspect ratio of dilatant cracking accumulating during brittle and pseudo-ductile deformation. The experiments performed during this study have led to further insights into the damage processes involved in brittle deformation and failure of rocks. These insights have enabled a new delineation of the stages of brittle failure to be made for Darley Dale sandstone. It is considered that this delineation can be held as general for many brittle upper-crustal rocks. The linking of far-field acoustic monitoring techniques and the in situ stress field, the evolving state of damage, and the fluid transport properties of rock, is of great importance in many areas of geological interest, including earthquake source processes, oil exploration and exploitation, and nuclear waste disposal.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item