Page 215 - Reservoir Geomechanics

196 Reservoir geomechanics

More on wellbore breakouts

We discuss here several aspects of breakout formation that will be important when we
use breakout observations to estimate stress magnitudes (Chapters 7 and 8) and examine
excessive breakout formation associated with wellbore instabilities (Chapter 10).
As discussed above, breakouts form in the area around a wellbore where the stress
concentration exceeds the rock strength. As first pointed out by Zoback, Moos et al.
(1985), once a breakout forms, the stress concentration around the wellbore is such that
breakouts will tend to deepen. This was illustrated theoretically as shown on the left side
of Figure 6.15a (Zoback, Moos et al. 1985). Subsequent work on the manner in which
breakout growth would eventually stabilize confirmed this result (Zheng, Kemeny et al.
1989), as did laboratory studies of breakout formation by Haimson and Herrick (1989)
who presented photographs of breakouts formed in laboratory experiments (Figure
6.15a,b) and also found an excellent correlation between measured breakout widths
and the theoretically predicted ones (right side of Figure 6.15b) using a relatively
simple failure theory presented by Zoback, Moos et al.(1985). While this will not be
true for extremely weak formations such as uncemented sands, it appears to be the
case for cemented rocks of at least moderate strength. As discussed in Chapter 10, the
fact that after initial formation, breakouts deepen (until reaching a stable shape) but
do not widen allows us to establish a relatively simple criterion for assessing wellbore
stability. As long as drilling conditions result in breakouts that do not have excessive
width, wells can be drilled successfully.
In general, it is quite difficult to predict the evolution of the failure zone around a
well once a breakout has formed. Zheng, Kemeny et al.(1989) attempted to model this
analytically, but the rather pointed breakout shapes they predicted are not generally
seen when viewing actual breakouts in cross-section. This appears to be because as
the rock begins to fail, strain energy is absorbed through inelastic deformation, thus
allowing the breakout shape to stabilize with a relatively flat-bottomed shape.
In the same manner that the stresses induced by cool drilling mud in the wellbore
affect the formation of tensile fractures, thermally induced stresses also affect the
formation of breakouts. The decrease in circumferential stress at the wellbore wall will
decrease the tendency for breakouts to occur although the effect is relatively small. Note
that for the case in which the hoop stress decreases by several MPa (as in the Visund
example cited above), the temperature change would only decrease the maximum hoop
stress shown in Figures 6.2 and 6.3 by about 2%. Such a decrease in hoop stress would
notbeaseffectiveinincreasingwellborestabilitybecausetherewouldbenocomparable
increase in σ rr ,as mentioned above. Thus, the initial area of wellbore failure when
10 Cof cooling occurs (as shown in Figure 6.14c) is only slightly smaller than that
◦
when there is no cooling (Figure 6.3). However, with time, cooling changes both σ θθ
and σ rr in such a way as to lessen the tendency for rock failure away from the wellbore