Three possible mechanisms of borehole failure associated with different scales are considered: (1) slip on a fault in a vicinity of the opening (large scale); (2) multiple sliding distributed over oriented systems of parallel joints, cracks, or rock foliation (intermediate scale); and (3) propagation of pre-existing randomly oriented micro-cracks that result in spalling or borehole breakout (small scale). For each of the failure types 2-D models of the mechanisms of borehole instability have been developed. In the case of a near-by fault, the extent of its slip essentially depends on the fault orientation. In laminated rock, where strength anisotropy exist, inelastic zones in the anisotropic material are qualitatively different from those appearing in isotropic rock where there is clear asymmetry. Finally when the previous two mechanisms cannot be invoked, extensive unstable growth of pre-existing micro-cracks in compression that parallel the borehole surface takes place followed by buckling of the thin rock plates between the growing cracks as the mechanism of borehole breakouts.
When drilling circular openings (tunnels or boreholes) in rock, different types of instabilities can occur near the borehole wall due to in situ stress concentration (Figure 1). Elongation of the instabilities are, in most cases, aligned perpendicular to the minimum principal stress. The petroleum industry experiences financial losses because of borehole breakouts caused by a particular type of instability. Furthermore, similar phenomena in mining and tunneling lead not only to financial problems but also to human casualties. Although the orientation of breakouts can, in some cases, indicate the principal directions of the in situ stress tensor (e.g., Mastin, 1988), any attempt to correlate area or volume of broken rock to stress magnitude will require an understanding of the failure mechanism (s) responsible for the final borehole geometry (Detournay andRoegiers, 1986). Conventionally, borehole instabilities have been related to shear failure dictated either by the stress conditions (e.g., Zoback et al., 1985; Cheatham, 1993; Haimson and Song, 1993; Ong and Roegiers, 1993) or by stresses acting on pre-existing plains of low or no cohesion. The problem, however, is to reconcile the assumption of shear failure with the shapes of spalled rock pieces observed in the opening walls (Figure 1). That is, spalling suggests tensile failure, a brittle mechanism (e.g., Santarelli and Brown, 1989; Ewy and Cook, 1990; Brady and Zoback, 1993; Lee and Haimson, 1993). Additionally, the failure mechanisms depend on the initial rock structure which, accordingly, should be taken into account. This paper attempts considers and compares different mechanisms of borehole instabilities associated with natural fracture systems of different scales similar to the multiscale approach used by others (e.g., Dyskin et al., 1992). The first, obvious type of instability results in slipping on a previously stable fault situated in a vicinity of the opening (large scale). This sliding can occur due to the opening- induced disturbance in the original stress field while a fault can be generally recognized as the weakest discontinuity having a better potential for failure in comparison to smaller defects (cracks).
