Analysis of rock stress and rock stress measurements with application to Äspö HRL

Sammanfattning: The process of choosing a site for a nuclear waste repository means that many aspects have to be taken into consideration. One of these is that the repository has to be mechanically stable for a long time. The mechanical stability of the rock is very difficult to determine. One of several factors, which determine the mechanical stability, is the virgin state of stress. The thesis project consists of two parts. In the first part the state of stress at Äspö Hard Rock Laboratory had to be defined. This was done based on earlier rock stress measurements conducted during the years 1988 to 1997. Two different measurement techniques have been used, hydraulic fracturing and overcoring. During the overcoring two types of cells have been used, CSIRO HI-cell and a cell developed by the Swedish State Power Board (SSPB). In the second part of the project, investigation of the correlation between the stress and geological structures are made using numerical modelling tools such as FLAC, UDEC and 3DEC. The rock stress measurements using the hydraulic fracturing gave orientations of the horizontal stress that coincide with earlier hydraulic fracturing measurements conducted in Scandinavia. The magnitudes of rock stresses are slightly lower than the earlier reported stress magnitudes for the Scandinavian part of the earth crust. The rock stresses obtained from the overcoring resulted in higher stresses than what was predicted by the hydraulic fracturing measurements. However, the orientation of the maximum horizontal stresses coincides well between the two techniques. The orientation is also more or less constant with respect to increasing depth. The state of stress at Äspö is defined by using the results from the hydraulic fracturing and the measurements conducted by SSPB-cell. The measurements from the SSPB-cell are used since these have a Poisson's ratio that corresponds well with the uniaxial tests of rock samples and since the measurements have been done at a distance from the opening where no influence from the openings can be expected. Since the magnitudes of the rock stresses differ between overcoring and hydraulic fracturing, some efforts have been made to find possible causes for this. The rock stresses when conducting overcoring gave higher values overall, which could be explained by high Poisson's ratios and a minor influence from the opening as the stress measurements might have been done in the disturbed zone. The high Poisson's ratio may depend on the stress-induced microcracks, which might be initiated during the overcoring of the cell, during the drilling of the pilot borehole, in which the cell is installed, and during biaxial testing. Statistical analysis showed that there is significant differences between the mean values of Poisson's ratio obtained from biaxial tests of cores containing the CSIRO HI-cell and the SSPB-cell. Poisson's ratio is about 0.34 for CSIRO HI-cell while the SSPB-cell gave a Poisson's ratio of 0.23. The analysis also showed that Young's modulus does not differ between the techniques. The modelling in FLAC was made to simulate the overcoring and biaxial testing. The result show that it is possible to obtained extensional strain in the core during overcoring if the major principal stress is perpendicular to the borehole axis. This may lead to microcracking occurring in the core causing high Poisson's ratio, which results in higher stresses. It can also be seen from the simulation of the biaxial testing that extensional strain is achieved even if the hollow core is not damaged during overcoring. The analyses using UDEC was made to study the effect of different properties of a discontinuity, such as the dip angle, Young's modulus, Poisson's ratio, density and the normal and shear stiffness. The analyses showed that an inclined discontinuity affects the stresses especially if sliding occurs. So, the dip angle does not solely, determine the amount of disturbance of the state of stress around a discontinuity. If slip will occur or not depends, thus, on a combination of dip angle, friction angle and the far field state of stress. A dip angle of 30 degrees affected the major principal stress most, while the minor principal stress is most affected by a dip angle of 45 degrees, for a friction angle of 10 degrees. The results from the simulation of a thick zone showed that the elastic properties of the zone material mainly affect the stresses within the zone. However, higher values of Young's modulus and Poisson's ratio in the zone than in the side rock resulted in higher stresses within the zone than outside. The orientation of the major principal stress becomes more perpendicular to the zone. The 3-dimensional analyses using 3DEC was made in order to investigate if the stresses at Äspö could be correlated with the major geological structures. The results show that the increase in the horizontal stress seen both in KAS02 and KAS03 is obtained in the model when using a bilinear stress state that is based on the measurements performed at Äspö. However, a satisfying coincidence is not obtained with the measured stresses in KAS02, KAS03, KAS05 and KZXSD8HL, which were the boreholes used as reference boreholes. One of the reasons for the disagreement may mainly be that the discontinuities used in the 3DEC model are more or less vertical. The least dip angle used is 60 degrees. Another reason may be that the measured stresses are influenced by far more parameters than are used in the 3DEC- model, such as different rock types, smaller discontinuities and mineral grains.