SPE/ISRM 78163

Mud Chemistry Effects on Time-Delayed Borehole Stability Problems in Shales

Erling Fjr, SPE/SINTEF Petroleum Research, Rune M. Holt, SPE/NTNU and SINTEF Petroleum Research,
Olav-Magnar Nes, SPE/SINTEF Petroleum Research, and Eyvind F. Snsteb/SINTEF Petroleum Research

Copyright 2002, Society of Petroleum Engineers Inc.

This eliminates the effective support from the mud to the
This paper was prepared for presentation at the SPE/ISRM Rock Mechanics Conference held borehole wall, and thus reduces stability over time.
in Irving, Texas, 20-23 October 2002.
There is also the possibility of chemical interaction
This paper was selected for presentation by an SPE/ISRM Program Committee following
review of information contained in an abstract submitted by the author(s). Contents of the
between the water phase and the shale. Various salts such as
paper, as presented, have not been reviewed by the Society of Petroleum Engineers or KCl and CaCl2 are frequently added to the mud in order to
International Society of Rock Mechanics and are subject to correction by the author(s). The
material, as presented, does not necessarily reflect any position of the Society of Petroleum improve borehole stability. Although positive effects of these
Engineers, International Society of Rock Mechanics, its officers, or members. Papers
presented at SPE/ISRM meetings are subject to publication review by Editorial Committees of
additives are reported, their presence in the mud does not
the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part always cure the problem1.
of this paper for commercial purposes without the written consent of the Society of Petroleum
Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more As indicated above, and confirmed by operators in the
than 300 words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
field, borehole stability is time dependent. That is, an initially
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435. stable hole may become unstable after some time. Wellbore
instability is however a problem only for a limited period
Abstract (until casing has been set). Hence it is of great importance
A combination of small sample testing and numerical when deciding drilling fluid composition to be able to quantify
modelling offers a powerful tool for identification of mud the impact of shale - mud interaction on rock mechanical
chemistry effects on wellbore stability in shales. The tool behaviour, and in particular on the time scale involved.
allows for modelling of how the stability limits for the mud- The objective of this study has been to identify and
weight varies with time, based on measured time scales and quantify relevant parameters for shale-mud interaction, and to
magnitudes for the various processes involved. It may also enable quantitative estimation of the consequences of KCl and
assist in estimation of the optimal amounts of chemical CaCl2 additives on wellbore stability. To achieve this, several
additives in the mud. shale materials have been studied in the laboratory, by
specially designed tests that reveal how the materials respond
Introduction mechanically to exposure of water with various salt
Shale instability remains a costly drilling problem. The concentrations. The observed response is compared to
industry usually estimates that 5 10 % of the drilling time is mathematical models for the expected behaviour, and relevant
lost due to borehole stability problems. Although significant values are obtained for the model parameters characterizing
scientific progress has been made over the last decade, the the shale properties. By transforming the mathematical models
practical borehole stability problems have not been reduced to borehole geometry, expected behaviour of the shales around
accordingly. This is of course a result of the complexity of the a borehole can be estimated. The borehole model can then be
problem, difficulties in establishing a proper diagnosis once used to predict and analyse how the mud chemistry may affect
the problem occurs, and lack of implementation of research borehole stability problems in shales.
results into drilling operations.
The mudweight is the main parameter to control in order to Effects of fluid exposure on shales
counteract borehole instability. However, field experience Borehole stability problems are related to mechanical
shows that the drilling fluid as such may also have an impact. instabilities of the rock around the hole, and the problems are
For instance, it is found that water-based mud usually creates therefore directly related to the mechanical properties of the
more problems than oil-based mud. A possible explanantion material, and to the effective stresses. The problems may
for this is that capillarity prevents the oil phase from entering however have non-mechancial causes.
low permeability shale; hence the borehole wall is effectively A shale material consists of clay minerals and water with
impermeable for an oil-based mud. This allows the well dissolved ions in a state of mechanical, chemical and
pressure to be kept higher than the pore pressure so that the electrostatic equilibrium, specific for the actual condition of
mud may effectively support the borehole wall. Water may on the material. If the material is cut or broken, such that a free
the other hand enter the shale, thus the pore pressure in the surface is exposed, this state of equilibirium is disturbed.
vicinity of the well will gradually approach the well pressure. Several processes may then occur, which affect the materials
mechanical properties and the effective stresses.
2 E. FJR, R.M. HOLT, O.-M. NES, E.F. SNSTEB SPE/ISRM 78163

Pore pressure. If the fluid pressure outside the shale is initial effect of the chemical activity is controlled by pore
different from the initial pore pressure, there will be a pressure diffusion, while the time scale of the relaxation
disturbance in pore pressure that propagates by diffusion from process is controlled by ion diffusion, which is usually a
the surface into the rock. This may give rise to a delayed slower process. Thus, even if the membrane efficiency is zero,
instability in a drilling situation: If the well pressure is higher there will be a temporary effect of the chemical activity.
than the pore pressure, the drilling fluid provides a support for Traditionally3,4, the effects of mud chemistry have only
the borehole wall, and may prevent failure. As the pore been assigned to the chemical activity. It has been shown
pressure inside the rock gradually increases towards the well experimentally however5, that some ions typically cations
pressure (provided there is no isolating membrane or like K+ and Ca2+ may be exchanged with Na+ in the clay
mudcake at the borehole wall), the support of the well mineral lattice structure, resulting in a change in the size of the
pressure will fade away, and the borehole wall may become structure. Thus the rock may expand or shrink as these alien
mechanically unstable. ions move in. The consequences for a borehole are similar to
Temperature. If the temperature of the fluid outside the the effect of a difference in temperature, giving a shift in the
rock is different from the initial temperature of the rock, the axial stress and the hoop stress at the borehole wall:
step in temperature will propagate by diffusion from the
surface into the rock. Rocks, like most other materials, expand E
or shrink when the temperature is changing. In a borehole z = = ch (3)
geometry, thermal expansion is restricted (except for the radial 1
direction). A difference between the mud temperature (Tm) and
the formation temperature (Tf) will therefore result in changes ch is the chemically induced strain (positive in contraction).
in the axial stress (z) and the hoop stress () at the borehole It is assumed to be a function of the cation exchange capacity
wall, given as2 and the relative amount of swelling clay minerals (e.g.
smectite) in the rock. The degree to which the ion exchange
process is expected to take place is assumed to depend on the
E
z = = Th (Tm T f ) (1) level of concentration of alien ions in the pore fluid of the
1 shale, up to a certain limit where the lattice is saturated. The
ion exchange process is considered to be much faster than the
E is the Youngs modulus, the Poissons ratio, and Th the diffusion process that brings the ions into the shale; hence the
coefficient of thermal expansion. It is apparent from Eq. (1) latter process will effectively control the time scale for the
that a cooling of the drilling fluid will tend to reduce the axial total process.
stress and the hoop stress, and thus enable the hole to Diffusion. Each of the processes described above implies
withstand lower well pressures. transport of some quantity from the surface into the rock, and
Chemical effects. The pore fluid in the shale has a the transport mechanism is diffusion. Thus, the distribution in
chemical activity (ash) given by the type and amount of time and space of the processes are controlled by the diffusion
dissolved ions. If the surface of the rock is exposed to a fluid equation
with a different chemical activity (adf), water molecules will
move into or out of the shale to compensate for this Q
disturbance, thus creating a disturbance (p) in pore pressure = C D 2Q (4)
at the surface3:
t
where Q is the actual quantity (ion concentrations, pore
adf pressure or temperature) and CD is the diffusion constant.
p = coT f ln (2)
ash The consequences of these processes will therefore be
delayed relative to the time of exposure of the free surface,
and the delay is given by the diffusion constants and the size
This disturbance (commonly called osmotic pressure) will and geometry of the rock involved.
propagate into the rock like any other pore pressure
disturbance, with the same consequences. p may be positive Laboratory tests on small samples
or negative, depending on whether the activity of the fluid is Diffusion in shales is a rather slow process, due to the low
larger or smaller than the activity of the shale. If adf < ash, we permeability. Typically, the diffusion constant is in the order
have that p < 0, which is beneficial for borehole stability of 10-8 10-9 m2/s for pore pressure diffusion, while for ion
since that increases the effective support of the drilling fluid diffusion the value is in the order of 10-10 10-11 m2/s. The
on the borehole wall. However, if the rock surface does not act typical time needed to propagate a distance L by diffusion is
as a perfect membrane (i.e. a membrane that allows only water given as
molecules to pass) ions will also move through the surface and
gradually change the activity of the shale towards that of the L2
= (5)
drilling fluid. After a while, the pore pressure shift p is CD
reduced to p, where is a number (called the membrane
efficiency) between zero and one. The time scale for the
SPE/ISRM 78163 MUD CHEMISTRY EFFECTS ON TIME-DELAYED BOREHOLE STABILITY PROBLEMS IN SHALES 3

Thus, the time needed for ion diffusion over a distance of concentrations of KCl make the shale shrink faster, however
19 mm (= radius of a standard 1.5 x 3 core plug) is typically the final amount of shrinkage is not correspondingly higher.
10 100 hrs for pore pressure diffusion, and 42 420 days for Thus the effect seems to reach a saturation level even at
ion diffusion. For a sample of radius 5 mm (which is the relatively low concentrations of the salt. The effect of
typical sample size used in the tests reported here) the reducing the salt concentration at a later stage is seen to be
corresponding diffusion time is reduced by a factor of 14. relatively low.
Notice that the ions will follow the water molecules if there is Quantitative analysis of these tests is not straightforward,
a pore pressure gradient causing the pore water to flow. This because the measured strain represents the integrated response
effect (called advection) will slow down or speed up the over the entire sample, while the various parts of the sample
overall movement of the ions, depending on the direction of respond differently according to the distance to the exposed
fluid flow versus the direction of ion diffusion. surface. As the deformation of the sample is actually resticted
Clearly, the aspect of time consumption favours the use of by the pistons, the sample must deform uniformly. This
smaller samples for laboratory studies of diffusion processes implies that parts of the sample will experience restricted
in shales. Furthermore, limited availability of core material deformation, such that the stress will vary over the sample,
and difficulties in sample preparation also encourage the use however in such a way that the total stress remains constant.
of smaller samples. The following test procedure has therefore Denoting as loc(r,t) the strain that would have been induced at
been established for fluid exposure test on shales6: the distance r from the centre of the sample at time t, if the
The shale sample is shaped like a small cylinder, typically deformation were not restricted, we find that the measured
10 mm in diameter and 2 mm high. The sample is positioned strain of the sample at time t, (t), is given as
between two pistons and subjected to a predefined, constant
uniaxial load, while the free surface of the sample is exposed R
2
to various fluids. The induced axial deformation of the sample ( t ) = 2 r loc ( r , t ) dr (6)
is measured as displacement of the pistons. The test geometry R 0
is shown in Fig. 1. Dehydration of the samples is carefully
avoided during the mounting procedure, to prevent capillarity where R is the sample radius. This implies that the measured
effects. response is dominated by the events close to the surface of the
In a typical test, the sample is first exposed to a 3.5 w% sample, since the part of the sample outside the radius
NaCl solution or sea water (which is assumed to be fairly
close to the salinity of the pore water) and left to stabilise for 1 RM = R / 2 will have the same impact as the part inside RM.
- 2 days. Then the sample is exposed to the solution of Furthermore, when the solution of the exposing fluid is
interest, and the response of the shale in terms of induced changed, several processes are triggered simultaneously. The
deformation is monitored. observed strain is thus also a sum of several effects. In order to
For the ion exchange process, the deformation of the facilitate quantitative analysis of the tests, a mathematical
sample is directly related to the chemically induced strain. For model, based on Eq. (4) for the actual geometry, has therefore
the effect induced by a contrast in chemical activity, the been established. The model estimates loc(r,t) due to pore
deformation of the sample is an elastic response to the change pressure diffusion (induced by chemical activity) and ion
in effective stress caused by the shift in pore pressure. Thus, exchange processes. By fitting the model to the measured
both effects have an impact on the sample that is detectable by strain data, the key parameters for the various processes can be
the measurement of sample deformation. extracted.
Results from tests on a field core material are shown in The modelled strain is shown in Fig. 2 as solid lines. The
Fig. 2. The material is a Tertiary shale from the North Sea, figure shows that the match between model and experiment is
with about 28% porosity7. The total clay content is about 44%, reasonably good. The key parameters extracted from these
and about half of this is smectite. Prior to the tests the samples KCl exposure tests were:
were stabilised in a 3.5 w% NaCl solution. In the first test
shown in Fig. 2, the sample was exposed to 5 w% KCl. It is CD (ion diffusion) = 4.210-11 m2/s
seen that the sample shrinks (the strain is positive) due to this Total strain amplitude ch = 0.0102
exposure. As the chemical activity of 5 w% KCl is almost Saturation level for ion exchange = 6.6 w%
exactly the same as for 3.5 w% NaCl, we can deduce that
chemical activity is not responsible for the observed effect, The theoretical value8 for the pore pressure diffusion
and we assume that it is caused by an ion exchange process5,7. constant is about 710-9 m2/s for a material with the actual
In the second test the sample was first exposed to 10 w% KCl. permeability and stiffness. For diffusion over a distance of 1.5
After about 24 hours the salinity was reduced to 5 w% KCl. In mm ( R - RM) the characteristic time is then about 5 minutes.
the third test the sample was first exposed to 20 w% KCl, Thus, on the time scale seen here, the chemical activity effect
followed by 5 w% KCl after about 24 hours. For the 10 w% occurs almost immediately. For the modeling, the theoretical
and 20 w% cases the activity of the water is lower than for the value was therefore used for the pore pressure diffusion
3.5 w% case (0.956 and 0.892 versus 0.982, respectively). constant.
Thus for these cases we expect to see effects of both chemical The membrane efficiency is expected to vary with the
activity and ion exchange. It is seen that the higher salinity of the exposing fluid. This has been accounted for, in
4 E. FJR, R.M. HOLT, O.-M. NES, E.F. SNSTEB SPE/ISRM 78163

accordance with the model presented by Lomba et al.9. The by Eq. (5)). The corresponding time for ion diffusion is about
values used here were = 0.17, 0.12 and 0.10 for KCl 60 hours.
concentrations of 5 w%, 10 w% and 20 w%, respectively. Fig. 6 shows the effect of using a 5 w% KCl solution for
The effect of fluid exposure clearly depends on the type of the drilling. This solution has the same chemical activity as the
ions involved. Fig. 3 shows the response to a CaCl2 solution in solution shown in Fig. 4; hence the differences between Fig. 4
comparison to a KCl solution for the same shale (but different and Fig. 6 are due to the ion exchange process only. It is seen
batches) as shown in Fig. 2. The samples shown in Fig. 3 had that the shrinkage induced by the ion exchange process relaxes
been exposed to sea water during the stablisation period prior the axial and tangential stresses in the vicinity of the borehole.
to the tests. Notice that the sample exposed to CaCl2 expands, This significantly improves the stability of the borehole with
while the sample exposed to KCl contracts. The total strain respect to possible shear failure, as it reduces the shear stress
amplitude estmated for the CaCl2 exposure was ch = -0.0012. (Fig. 7). The effect appears gradually, at about the same time
Other tests indicate that the expansion induced by CaCl2 as the positive effect of the chemical activity is declining (Fig.
exposure depends on the concentration of the solution. An 5). For practical applications, one should try to obtain the
apparent difference in saturation level for K+ ion exchange combined benefits of both effects, such that the chemical
between Fig. 2 and Fig. 3 may possibly be due to the activity ensures short term stability, while the ion exchange
difference in chemical content of the stabilizing fluid (3.5 w% takes over and ensures stability when the chemical activity
NaCl versus sea water). effect is declining.
The laboratory tests on small samples represent a fast and The borehole model has the ability to simulate the stress
inexpensive way to obtain information about how shales state in the vicinity of a borehole in a fairly general way, as it
respond to various mud compositions. In particular, they can also includes the general expressions (based on linear
tell us the magnitude of the response, and the characteristic elasticity) for stresses in the vicinity of a wellbore10.
time scale for the response process. This is essential Anisotropic horizontal stresses and deviating boreholes are
information for modelling the effect of mud composition on thus accounted for. Tests of the effective stresses versus
borehole stability. criteria for shear and tensile failure are also included, and the
model thus represents a powerful tool for analysis and
Numerical modeling of borehole stability prediction of borehole stability.
The effects of fluid exposure observed in the laboratory tests The model uses analytical solutions extensively, in order to
are likely to occur also in a borehole. To illustrate the reduce the computing time. Some of these solutions are exact
consequences for wellbore stability, a model based on the only at the borehole wall, and have reduced precision deeper
same equations and principles as the one used for the into the formation. This is not considered to be a significant
laboratory tests has been established for borehole geometry. problem for the model as a tool for borehole stability analysis,
Figs. 4 - 6 show examples of calculations using this model. as the critical events with respect to stability occurs in the
The data for the shale are representative for the material vicinity of the borehole wall.
described in the previous section. A simple geometry and
stress state has been chosen here in order to illustrate the Discussion
effects of mud chemistry: The well is vertical, the horizontal The behaviour of shales in the vicinity of a borehole is fairly
stresses are equal, and the well pressure is equal to the in situ complex, and several aspects have not been included in the
pore pressure. The shale properties are the same as derived modeling, or have only been treated approximately. Some of
from the laboratory tests described in the previous section. these are:
Data for the simulation are given in Table 1. - Anisotropy of chemical strain. Shales are highly anisotropic
Fig. 4 shows the situation when the drilling fluid is in on a microscopic scale, and to some extent the anisotropy
perfect balance with the pore fluid. There is an induced pore prevails also on a macroscopic scale. This may a.o. imply that
pressure effect, and no ion exchange effect. The stresses and the strain observed in the laboratory tests (Fig. 2) is
the fluid pressure remain constant in time. representative only for that specific orientation of the material.
Fig. 5 shows the effect of using a 10 w% NaCl solution for To compensate for this possibility, the chemically induced
the drilling fluid. This composition gives a chemical activity shrinkage has been scaled by a factor 1/3 in the borehole
(osmotic) effect, but no ion exchange effect. It is seen that the model. This would be the correct value if there is no strain in
pore pressure in the vicinity of the borehole is affected, and the lateral directions in the laboratory tests (not measured),
hence the effective stresses are also affected. The effect is and the shale downhole consist of a randomly oriented
positive for borehole stability, since it implies that the assembly of particles, each of which behaving like the sample
borehole wall gets some support from the fluid pressure in the observed in the laboratory.
well. This shifts the stress state away from the failure - Alteration of strength. It is possible (in fact evident in the
envelope, as illustrated in a Mohr-circle plot (Fig. 7). The data shown by Horsrud et al.5) that the shift in ion
positive effect declines gradually with time (as ions from the composition of the pore fluid, and the exchange of ions in the
drilling fluid move into the rock and reduces the contrast in shale microstructure, also have an impact on the strength of
chemical activity). With the actual data set, the characteristic the shale. Such an effect would of course have a significant
time for fluid pressure diffusion from the borehole wall to a impact on the stability of the borehole, as it would shift the
point 3 mm into the formation is about 20 minutes (as given position of the failure envelope (Fig. 7).
SPE/ISRM 78163 MUD CHEMISTRY EFFECTS ON TIME-DELAYED BOREHOLE STABILITY PROBLEMS IN SHALES 5

- Temperature. Chemical processes are always sensitive to p = Pore pressure, MPa

temperature. This adds some uncertainty to the use of data R = Sample radius, m
obtained at room temperature for simulation of downhole RM = Main point radius for the small sample tests, m
conditions. Measurements6 at elevated temperature (80C) Tm = Mud temperature, K
have so far, however, not revealed any large temperature Tf = Formation temperature, K
effects compared to room temperature data. Th = Coefficient of thermal expansion, 10-3/K
- Coupling. The effects discussed here are all coupled in z = Axial stress increment, MPa
various ways. For instance, the chemically induced strain is = Hoop stress increment, MPa
likely to have an impact on the permeability5, which in its turn = membrane efficiency
has a direct impact on the diffusion constants. Coupling = Strain, 10-3
effects are likely to be increasingly important at higher ch = strain induced by ion exchange, at saturation, 10-3
concentrations of salt in the drilling fluid. loc = Local (virtual) strain, 10-3
- Acceptable instability. We have here discussed borehole
= Poissons ratio
stability based on considerations about what happens at the
= Characteristic time for diffusion over a length L, s
borehole wall. For drilling purposes, failure of a thin sector at
the borehole wall needs not be a problem at all, unless the
References
condition propagates into the formation. To provide a 1. Stjern, G., Horsrud, P. and Agle, A.: Improved drilling
complete borehole stability model applicable for drilling performance in troublesome clay formations in the Heidrun
operations, the differentiation between acceptable and field. IADC/SPE 59219 (2000).
unacceptable instabilities also needs to be clarified. 2. Guenot, A. and Santarelli, F.J.: Influence of mud temperature on
While the tests involving KCl appears to be quite deep borehole behaviour, in Rock at Great Depth, V. Maury
consistent and predictable, the tests involving CaCl2 seem to and D. Fourmaintraux (eds.) Balkema, Rotterdam (1989).
indicate a difference in behaviour for higher concentrations 3. Mody, F.K. and Hale, A.H.: A borehole stability model to
than for low ones. More work is thus needed to fully couple the mechanics and chemistry of drilling fluid shale
understand the effect of CaCl2 in the mud. interaction. SPE/IADC 25728 (1993).
4. Tare, U.A., Mody, F.K. and Tan, C.P.: Mitigating wellbore
stability problems while drilling with water based muds in
Conclusions deepwater environments. OTC 14267 (2002).
We claim that the effects of mud chemistry on borehole 5. Horsrud, P., Bostrm, B., Snsteb, E.F., Holt, R.M.,:
stability problems in shales may have at least two different Interaction between shale and water-based drilling fluids:
causes: 1) a contrast in chemical activity, which induces a shift Laboratory exposure tests give new insight into mechanisms and
in pore pressure and thereby changes the effective stresses field consequences of KCl contents. SPE48986; in Proc. ATCE,
near the borehole wall, and 2) a strain induced by exchange of New Orleans, USA; September 1998.
ions in the shale microstructure, which changes the stress state 6. Nes, O.-M., Snsteb, E:F, Horsrud, P., and Holt, R.M.:
near the borehole wall. By suitable choice of salt types and Dynamic and static measurements on mm-size shale samples,
SPE 47200, expanded abstract, Eurock98, (1998).
concentrations both effects can be utilized to stablilise the
7. Snsteb, E.F. and Holt, R.M.: Brine Exposure Effects on a
borehole. Both effects are controlled by diffusion processes, Tertiary North Sea Shale, in Proc. DCRocks, Washington DC,
which imply that their impact on borehole stability changes 7-10 July; A.A.Balkema (2001).
gradually over time. 8. See for instance: Fjr, E. Holt, R.M., Horsrud, P., Raaen, A.M.
The combination of small sample testing and numerical and Risnes, R.: Petroleum Related Rock Mechanics. Elsevier
modelling offers a powerful tool for identification of mud (1992).
chemistry effects on borehole stability, and it may also assist 9. Lomba, R. F. T., Chenevert, M. E., Sharma, M. M.: The ion-
in estimation of the optimal amounts of such additives. selective membrane behavior of native shales, Journal of
Petroleum Science and Engineering 25 9-23 (2000)
10. Bradley, W.B.: Failure of inclined boreholes, J. of Energy
Acknowledgements Resources Tech., Trans., ASME, p 232-239 (1979).
The authors would like to acknowledge the technical and
financial support from the companies supporting the IDE
(Improved Drilling Efficiency by better handling of shale Table 1 Data for the numerical simulations shown in Figs. 4 6.
problems) project at SINTEF Petroleum Research. These Vertical stress 33 MPa
Horisontal stress 28 MPa
companies are: BP, Conoco, ENI-Agip, Norsk Hydro, PDVSA Pore pressure 20 MPa
and Statoil. Youngs modulus 0.7 GPa
Poissons ratio 0.1
Nomenclature Porosity 28.4 %
Permeablity 1.23 nD
adf = chemical activity of drilling fluid Diffusion constant, pore pressure -9 2 -1
ms
6.410
ash = chemical activity of pore fluid in shale Diffusion constant, ion diffusion 4.310
-11 2 -1
ms
co = Relative pressure amplitude = 0.046 MPa/K Relative amount of smectite 0.21
CD = Diffusion constant, m2/s Cation exchange capacity 42.7 meq/100g
Chemical activity of shale 0.98
E = Youngs modulus, GPa
L = diffusion length, m
6 E. FJR, R.M. HOLT, O.-M. NES, E.F. SNSTEB SPE/ISRM 78163

Constant load 6
Exposed 5 w% CaCl2
4
surface

Strain (mStr)
2
Piston
0
Measured
Sample -2 0 10 20 30 40 50
deformation
-4
Piston
-6
Time (h)

12
10

Strain (mStr)
5 w% KCl
8
Fig. 1 Fluid exposure test for small samples.
6
12 4
10 2
Strain (mStr)

8 5 w% KCl 0
6 0 10 20 30 40 50
4 Time (h)
2 Fig. 3 Effect of fluid exposure on a shale material, for two
0 different types of solutions. Solid line: mathematical model
0 20 40 60 80 including chemical activity effects and ion exchange.

Time (h)
3.5 w% NaCl
12
40
10 10 w% KCl
Strain (mStr)

8
6 35
Axial
Principal stresses [MPa]

4
Tangential
2 5 w% KCl 30
0
0 20 40 60 80
25
Time (h) Radial

12 20 w% KCl 20 Fluid
10 pressure
Strain (mStr)

8
15
6
0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7
4
Distance from borehole centre (relative to hole radius)
2 5 w% KCl
0 Fig. 4 Principal stresses and fluid pressure in the vicinity of the
borehole wall, as predicted by the borehole model, for a drilling
0 20 40 60 80 fluid containing 3.5 w% NaCl. Formation properties and in situ
Time (h) stresses are given in Tab. 1. Solid lines: 0.1 days after drilling;
Dashed lines: 1 day after drilling; Dotted lines: 10 days after
drilling. (The lines fall on top of each other.)
Fig. 2 Effect of fluid exposure on a shale material, for various
concentrations of KCl. Solid line: mathematical model including
chemical activity effects and ion exchange.
SPE/ISRM 78163 MUD CHEMISTRY EFFECTS ON TIME-DELAYED BOREHOLE STABILITY PROBLEMS IN SHALES 7

15 w% NaCl

40 Failure envelope

35

Shear stress
Axial
Principal stresses [MPa]

Initial state
Tangential
30

25
Radial
Pore pressure
reduction
20 Fluid
Shrinkage
pressure
Effective normal stress
15
0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 Fig. 7 Mohr-circle plot illustrating schematically the initial stress
state at the borehole wall, and the effects of pore pressure
Distance from borehole centre (relative to hole radius) reduction (induced by a contrast in chemical activity) and
shrinkage (induced by ion exchange).
Fig. 5 Principal stresses and fluid pressure in the vicinity of the
borehole wall, as predicted by the borehole model, for a drilling
fluid containing 15 w% NaCl. Formation properties and in situ
stresses are given in Tab. 1. Solid lines: 0.1 days after drilling;
Dashed lines: 1 day after drilling; Dotted lines: 10 days after
drilling.

5 w% KCl

40

35
Axial
Principal stresses [MPa]

Tangential
30

25
Radial

20 Fluid
pressure

15
0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7
Distance from borehole centre (relative to hole radius)

Fig. 6 Principal stresses and fluid pressure in the vicinity of the

borehole wall, as predicted by the borehole model, for a drilling
fluid containing 5 w% KCl. Formation properties and in situ
stresses are given in Tab. 1. Solid lines: 0.1 days after drilling;
Dashed lines: 1 day after drilling; Dotted lines: 10 days after
drilling.