This report critically examines the state of the art and understanding of the mechanisms controlling the failure of boreholes and other circular openings from a multidisciplinary point of view. Although such excavations have been built and used since ancient times, it became quite apparent during this last decade that the classical approaches based on elastic considerations could not always explain the strength of some circular holes. Such abnormal stabilities, combined with attempted excavations in more difficult environments, as well as the search for optimization led to increased research activities. This explains why such a classical problem as a circular hole excavated in a rock formation needed to be revisited.
A critical review of the literature across many disciplines quickly revealed the difficulty in arriving at a common definition of failure. This concept is indeed dependent on the level of stability which is appropriate for each individual application. Hence, in this report failure was taken to mean, in its broadest sense, encompassing local instabilities as well as rock bursts or total collapses.
When considering the life of a circular opening, from excavation to abandonment, time-dependent effects start to play an important role. While excavation or drilling proceeds, the influence of temporary supports needs to be taken into account to minimize the overall cost. A recent breakthrough has been the realization that fractures do not necessarily initiate at the borehole wall, and that the maximum stress concentration can occur within the rock mass itself, in the vicinity of the wall. Such situations can be created by involving nonlinear rock properties or strong coupling effects. Such mechanisms can also explain delayed instabilities without the loss of rock strength nor a decrease in support capacity. However, there are some remaining mysteries requiring additional attention and research: (1) breakouts that align themselves in an unexpected direction, (2) circular openings
that show no distress although the formation strength has apparently been exceeded, and (3) apparently more instabilities generated by the use of the Polycystalline Diamond Compact (PDC) bits, a gentler cutting tool.
When reviewing the successes/failures of the geophysical techniques used to assess the character and condition of a geological formation either ahead, during, or after completion of the excavation, the data interpretation is strongly dependent on the specified model(s). In addition, the selection of the technique poses a problem of having to choose between the depth of investigation and the scale of resolution as it is essentially a question of wavelength compared to the size of feature one wants to determine. Because none of the geophysical methods provides a direct measurement of the rock property, one needs to assume the existence of a fundamental relationship; this leads to problems of uniqueness. Therefore, most investigations have relied on a combination of techniques to reach confirmation via overlapping data. Recent developments in vertical seismic profiling show promise in refining the detection of the fracturing extent. The problem of interpreting the seismic data to determine the static rock properties remains unsolved and still relies on synthetic logs or empirical correction relationships. Radar sounding and tomography as well as measurement-while-drilling (MWD) are still in development stages but have already shown great promise. One common problem requiring attention is the development of a technique to detect abnormal pore-pressure conditions ahead of any excavation.
Documenting the failures of circular openings of varying scales has highlighted similarities over a wide range of scales. It has also pointed out the complex relationships between the stress field and the material characteristics. For example, axial splitting parallel to the direction of maximum principal stress seems to occur in stronger rocks only where curved shear fractures prevail in a weak formation. However, these phenomenological observations are still obscure. Periodicity and localized failure are also documented more now that such patterns are known to exist to the extent that symmetrical failures have become an exception rather than the rule.
Exercising sophisticated models has helped to explain some of the above-mentioned mechanisms and has even revealed new failure possibilities. The stability analysis of circular openings is a complex problem as the stress state redistributes itself not only during the excavation process, where it switches from a fully 3D situation to a 2D problem, but also during fracture propagation. By combining continuum and fracture mechanics concepts, it is possible to predict the path a crack will take to simulate, for example, the progressive formation
of the characteristic dog-ears. Bifurcation theories, although still in their infancy, have already contributed to quantifying scale effects and failure modes. At present, considerable research efforts are spent in investigating surface instabilities as well as the influence of preexisting discontinuities on the overall stability of underground openings.
Finally, substantial progress has been made in assessing the distress of the wall of the opening. Downhole cameras and fracture identification logs have allowed detailed screening of fractures at great depth. However, most of these developments are still limited to detecting the fracture traces, and the depth of investigation within the rock mass is rather limited.