Page 101 - Reservoir Geomechanics

85 Rock failure in compression, tension and shear

Not many years ago, the concept of a critically stressed crust with stress magnitudes
controlled by the frictional strength of the crust (as discussed in this chapter) may have
seemed unrelated to hydrocarbon development. It is clear today, however, that fault
slip can be induced during both fluid injection and production, slip on faults may be
responsible for damage to cased production wells in some fields (requiring many wells
to be redrilled) and it is becoming common practice in some areas to try to intention-
ally induce micro-seismicity to enhance permeability in low-permeability formations.
Hence, knowing that relatively small perturbations of stress may induce fault slip is
important in field development in many regions. Second, slip on faults can be a source
of wellbore instability (Chapters 10 and 12) and third, active faults provide efficient
conduits for fluid flow in fractured reservoirs and influence the seal capacity of reservoir
bounding faults (Chapter 11).
Before discussing rock strength, it is helpful to start with a clear definition of terms
and common test procedures as illustrated in Figure 4.1.
Hydrostatic compression tests (S 1 = S 2 = S 3 = S 0 ) are those in which a sample is

subjected to a uniform confining pressure, S 0 . Such tests yield information about
rock compressibility and the pressure at which pore collapse (irreversible porosity
loss) occurs. Use of an impermeable membrane around the sample isolates it from
the liquid confining fluid and allows one to independently control the pore pressure
in the sample as long as P p < S 0 .
Uniaxial compressive tests (S 1 > 0, S 2 = S 3 = 0) are those in which one simply

compresses a sample axially (with no radial stress) until it fails at a value defined
as the unconfined compressive strength (usually termed either the UCS or C 0 ). As
sample splitting (and failure on pre-existing fractures and faults) can occur in such
tests, an alternative method for determining UCS is described below.
Uniaxial tension tests (S 1 < 0, S 2 = S 3 = 0) are not shown in Figure 4.1 as they

are fairly uncommon. Tensile strength (T =−S 1 at failure) is generally quite low
in sedimentary rocks (see below) and this type of test procedure tends to promote
failure along pre-existing fractures or flaws that might be present in a sample.
Triaxial compression tests (S 1 > S 2 = S 3 = S 0 ) are the most common way of mea-

suring rock strength under conditions that presumably simulate those at depth. Such
tests are unfortunately named triaxial as there are only two different stresses of inter-
est, the confining pressure S 0 and the differential stress S 1 − S 0 . The strength of
the sample at a given confining pressure is the differential stress at which it fails.
The confining pressure is held constant as the sample is loaded. As in the case of
hydrostatic compression tests, it is relatively straightforward to include the effects of
pore pressure in such tests.
Triaxial extension tests (S 1 = S 2 > S 3 where S 3 acts in an axial direction) can also

be used to measure the compressive rock strength but are typically carried out only
as part of specialized rock mechanical testing programs. One advantage of such tests
is that they are useful for studying strength at low effective stress. A combination