Synopsis of the genesis of micro-cracks in brittle rock

Category Other
Group GSI.IR
Location International Geological Congress,oslo 2008
Author Bنckstrِm, Ann; Lanaro, Flavio
Holding Date 08 October 2008

In this contribution we argue that during investigations at large depth, the identification of pre-existing micro-crack patterns must be taken into account when evaluating strength of brittle rock samples.
For specimens of the brittle Ävrö granite from the Äspö HRL, Sweden, it was found that a set of pre-existing micro-cracks interfered with the cracks induced during uniaxial compressive testing. 15 of the specimens were tested in uniaxial compressive conditions, whereas one specimen was kept unloaded.
A set of pre-existing cracks is observed in all specimens and displays similar orientation independent from the ductile fabric of the rock. Clearly, the fabric did not influence the orientation of these micro-cracks. In addition, the pre-existing cracks did not seem to have been reactivated during the uniaxial compressive test.
The real orientation in space of the pre-existing micro-cracks is sub-vertical. This orientation is about 10° from the orientation of the major principal in-situ stress and is perpendicular to the minor principal stress at the site. The major principal stress has a magnitude of 30 MPa. The difference between the minor and major principal stress is about 20 MPa. The tensile strength of the Ävrö granite is about 13±1.5 MPa.
Uniaxial compression test and Brazilian test results on Toki Granite, Japan, show a very strong negative correlation with the difference between the major and minor principal in-situ stress: in other words, the strength of the samples diminishes as the in-situ stresses become higher. Differences between the major and minor principal stresses of about 20 MPa already produce lower strength values. Such reduction is imputed to micro-cracks either pre-existing or drilling induced.
Numerical simulations of core drilling using FRACOD2D indicate that the pre-existing micro-cracks can be induced when the difference in magnitude between the major and minor principal stresses is relatively small. Moreover, numerical simulations of Brazilian tests also show that the presence of pre-existing cracks in the samples greatly affects the strength.
For brittle rocks like granite, the formation of micro-cracks may depend on: i) intrinsic flaws and cleavage planes in minerals (e.g. feldspars); ii) in-situ stresses; iii) effect of the drilling induced stresses on the core. In our investigations, the effect of the mineral matrix does not seem to govern the orientation of the micro-cracks, whereas simulations show that micro-cracks depend on the differences between the principal in-situ stresses. These pre-existing micro-cracks will influence the crack pattern during mechanical testing affecting the laboratory strength of the specimens. Thus, further investigations of the origin of the observed pre-existing micro-cracks in brittle rocks should be conducted where the stress field is known to confirm that the mechanical strength measured in laboratory might depend on the level of pre-existing micro-cracking of the specimens.