Chapter 3. Rock Strength and Failure Introduction Rock strength refers to the stress state that a rock may sustain before it fails. The [NN study of rock strength is very important for evaluating wellbore stability, sand production, Hydraulic fracturing and Faulting. In this chapter different types of rock failures are discussed. The microstructure of the rock and the confinement conditions control the failure state of the rock. Laboratory test methods are used to determine stress-strain relationships and rock strengths, As well as Empirical equations ond correlations are presented for rock strengths in relation to petrophysical and geophysical properties across different lithologies. 1. Why Rock Strength is important? Rock strength is one of the most important parameters for: (encerendsand (2) Cemented sido + evaluating wellbore stability and + sand production, + Hydraulic fracturing sta uu wii euin 2. Overview of types of rock failure * 3 = Rock yield (plastic deformation) and failure can happen 3 = due to tensile stresses, shear stresses, compressive F3 S stresses, and a combination of the three. The following Hu itl “att sections explore these types of rock damage separately. ea oe rman, ceninginducedtens acs eefaubreatod Ge ean compaction 3. Laboratory tests Laboratory tests usually consist of simple experiments appropriate to the nature of the rock in which important quantities, often stress and strain, are determined. Different types of laboratory tests can be used to obtain rock strength, Young's modulus, and other mechanical properties, ‘+ Uniaxial compression test (A): the cylindrical rock specimen is only compressed in the axial direction by ‘the axial load or stress (61): “ ) © + Uniaxial strain test (B): the cylindrical rock specimen is only loaded in ‘the axial direction, and the lateral direction is constrained (no lateral displacement); + Tensile test (C): the cylindrical rock specimen is applied tensional force (stress) in the axial direction: + Hydrostatic test (D): the cylindrical rock specimen is compressed with the same stress in all directions: + Triaxial compression tests (E): the cylindrical rock specimen is compressed in both axial and lateral directions. The axial stress (1) is greater than the lateral stress (03, confining pressure): ‘+ Polyaxial (true triaxial) compression test (F): three independent and “i * mutually perpendicular stresses (o1, 02, and os) load to the faces of a rectangular prismatic specimen,4. Tensile failure 4.1 Direct tension: Point Load Test The application of a tensile stress (with negative sign) in the axial direction results in tensile strains (negative +00). The maximum tensile stress that the rock may resist is called tensile strength. The 1D (point load) tensile strength test for metals is difficult to implement in rocks due to the difficulty of grabbing the rock on its sides (or glue it since the rock may break at those “grabbing” points). One alternative is to "machine" the rock to a convenient shape, so that, we can pull it without using glues or grabbing jaws. However, rocks are not easy to "machine" in general (as steel, aluminum, wood), and thus this ‘test becomes impractical in many situations. ‘Metals are usually “ductile” ond deform after reaching a peak stress. When unstressed, the bar does not recover its original length but remains with "plastic deformation’. Rock failure in simple tension usually displays “brittle” failure, no plastic strains follow after reaching tensile strength, Tt just breaks quickly. So, the unloading elastic part can't be done. Strength is the maximum stress that a material can support under a given type of failure. Tf pica tense test 5 rock od reeow_ detorm ae ‘eal atain GUL 4 a Pal strain BU Tense Tensile mtn en dessin Strength + 4.2 The Brazilian test ‘As we saw previously, the simple direct tension test has the following shortcomings: - Tt is very difficult to grab the rock on its sides, ~The rock breaks quickly following a "brittle" failure, no plastic strains follow after reaching tensile strength. To overcome these problems, we usually use the Brazilian test. Tt uses short cylindrical samples (L must be < D) and takes advantage of the shape of the rock specimen to create tensile stresses with application of a compressive force along the sample diameter, which produces a bending of the principal stresses field around the rock trying to take advantage of all the surface in order to transfer that load, This stress bendi produces a tensile stress in the perpendicular direction. That's why this test is called indirect (because it is the compression that causes the tensile stress). The fracture will, initially, open at the middle then it propagates to the sides. Compression + cylindrical shape > tension ‘The compressive stress in this spherical shape can be composed into two main parts (1) vertical compressive stress that cancel out and (2) horizontal tensile stresses that cause tensile stress. +Q. is it possible to perform Brazilian test using a cubic sample? No, in that case, the compressive stress will not cause a tensile component fame se 7 > this is why the shape is very important. A solution of the state of stress within the rock (assuming a linear elastic homogeneous material) using the Navier equations yields the tensile strength value equal to: ‘cn —_ Where Psis the peak compressive force, L is the specimen length, and D is the specimen diameter. Applying a compression stress (P:) along the diameter will cause a tension stress in the perpendicular direction, which will cause the rupture of the rock (the plan of rupture will be perpendicular to Ts) if the tensile strength is reached. Typical values of tensile strength for cemented sedimentary rocks range from 0.5 MPa to 10 MPa. Uncemented sediments -very common in sedimentary basins- have zero tensile strength. radius R length L 2500 Cylindrical sample: Streamlines of 3° ppincpal sreses g Peonpresson 2000 2.0 rg Zo F, E 1000 3 Fo 500 Displacement rate: SSK SS SESS oS ose mms oe Fe Fee PF A ; 5 AMS i 0 10 2 30 40 50 Time [s) Inhydraulic fracturing, tensile strength is crucial because high-pressure fluids induce fractures by creating tensile stresses, unlike the natural compressive in-situ stresses in deep reservoirs. When fluid pressure exceeds the rock's tensile strength at the fracture tip, fractures initiate and propagate, staying open to allow oil or gas flow. Although in-situ stresses are compressive, the injected fluid locally overcomes them, causing tensile failure. After injection stops, some fractures may close due to compressive stresses, but ‘those that remain open enhance permeability and hydrocarbon extraction. 5. Compression failure: Pore collapse Uniaxial compression test is used to determine different properties of rocks: uniaxial compressive strength (UCS), Poisson's ratio, and Young's modulus. This test mimics the unconfined condition (Fig. A). In this test, a circular cylinder of rock sample is compressed parallel to its longitudinal axis. Consequently, ‘the axial load and the axial and lateral strains are measured and the volumetric strain is calculated. + (© 64:): during the initial stages of loading, initially existing cracks orientated at an angle to the applied load close at stress of cc. The stress-strain response is nonlinear, exhibiting an increase in axial stiffness (E). © (Giz > oc): once the majority of existing cracks have closed, linear elastic deformation takes place. + {at aucs): once the applied load exceeds the peak strength (UCS), rock starts failure, and it loses to its partial or all capacity to bear loading dependent on the plasticity of the rock. For a brittle rock, itmay lose the capacity; however, for ductile (plastic) rock, it may still have capacity to partially bear the applied load. UCS can be calculated using the following equation: A Where Prox is the maximum load exerted on the rock sample; A is the area of the cross section of the sample. ucs = Inreservoir conditions, if compression stresses are high enough, grains can crush filling the pore space. This pore collapse can happen due to rock burial or due to reservoir depletion. In both cases effective stresses increase in all directions. When effective stress goes over the yield stress, significant plastic irrecoverable deformations occur and may decrease permedbility. In long and thin reservoirs, depletion does not cause strains in all directions but predominantly in the. vertical direction. This type of deformation is called “uniaxial-strain" condition (Fig. B). In some severe cases, the rock may compact so much that the surface (or sea floor) starts to go down > phenomenon of subsidence. Tf rocks fail in shear or tension, they will have a higher permeability than if they fail in compression. Reservoir depletion is a result of pore pressure reduction. A decrease of pore pressure (Pi-Po) results in + increased effective stress in the depressurized region (Sv remains constant above the reservoir but the Pp decreases > a, increases > oy increases (a), = -*a,)). + reservoir compaction and porosity reduction (7 = ye») as Pp decreases, «, increases, then porosity decreases. + affects neighboring formations and faults. ‘+ subsidence (changes in ground surface elevation) ‘+ Differential displacements can result in casing damage and shearing and casing buckling, ‘+ Reduction in total horizontal stress within the reservoir t0 Etc Ver Stes Pa) circ Stes Pa 6. Shear Failure/Strength Shear strength is the capacity of rocks to resist shearing stresses just before shearing occurs. It comes from: ‘+ Friction between particles which creates resistance to sliding, ‘+ Cohesion and adhesion, ‘+ Interlocking of particles: particles make stuck together like a puzzle structure. d6.1 Frictional Strength The shear strength of rocks depends on the cohesive strength Sond the internal frictional strength of the rock characterized by the friction coefficient (x) The force Fr needed to displace an object resting on a surface depends on the friction coefficient (1) and the applied normal force Fu, such that Fy = 1 Fy. Fr increases linearly with Fu NB. if Fx=0, then Fr=0, Similarly, uncemented sediments can resist shear stresses (Similar to Fx) if an effective “confining” compressive stress (Similar to Fx) is applied The maximum shear stress (r) in uncemented sands is proportional to the normal effective stress (0,) through an “internal” friction coefficient (1) (red line: the failure line). Luss Real life examples 1, For dry uncemented sand, there are no confining stresses (no water to play this role), so the sand is loose and it seems as if it has no strength. 2. Sand castles can be built in wet moist (but not saturated sand) because there is confining stress and sand will develop some strength. 3. Tfavacuum-sealed package of coffee is emptied of air (vacuum P=0), then the air pushes from the outside by an effective stress Pora which gives the coffee grains a given strength that holds its shape. 4. Over-balanced drilling implies (Pw > Pp) which favors the formation of a "mud-cake” on the wall of the wellbore which permits adding stress support on the wellbore wall. The resulting effect is similar to an impermeable and elastic membrane applying a stress on the wellbore wall. This is Particularly useful for loose sand layers with zero internal cohesiveness. ill = 6.2 Mohr Circle As a result of applying principal stresses , and o,, normal and shear stresses (a,,r) are developed at the granular level between grains ona plane. To understand when it may fail, we need to determine the stress state at each point in ‘the subsurface. At each point, there are countless planes passing through it. Each with different stress magnitudes acting on them, ie, a different combination of stresses —_ (ou0). a Tf the shear stress becomes higher than the shear strength of a given plane, the soil (or rock) fails according to this plane. This plane is called the failure plane. The normal and shear stresses that apply on that plane are given as: O 2401) | Oa~ 9 oy = AED 4 Ca 0) coscopy (a-0,) 5 sin(zp) & Where: 4,9, are axial and radial (confining) applied principal stresses A Gt are normal and shear stresses Bis the failure angleThe German civil engineer “Christian Otto Mohr" realized that these stresses at a point can be represented graphically using a circle: cos*(28) -4,)] pes | |e ee [2] seen _ a+ 0) 2 This equation represents the equation of a circle with center ). This is called Mohr circle. pee (5) o-01) = Each point in the Mohr circle represents the normal and shear stresses (0,,) acting ona particular plane passing through that point in the subsurface (I/ANES@ FIGURES} ‘+ for plane (1): there is only normal stress (0,) with no shear stress > point (1) in Mohr circle. + for plane (2): there is only normal stress (a,) with no shear stress > point (2) in Mohr circle. + For plane (3): the state of stress will be a combination of (a;, = f(a;0,)) and (x) which will be characterized by an angle (y). ‘+ As the plane moves from (1) to (2), the stress moves along the Mohr circle from point (1) to (2). So, the points on the Mohr circle are representations of (0,,1) of all the planes. + Aswe rotate from plane (1) to (2) # is between 0° and 90°, whereas, in Mohr circle, » is between 0° and 180°. 6.3 Failure criterion The point on the top of the Mohr circle represents the maximum shear stress acting on a given plane within +the subsurface. You may think that failure occurs along that plane of maximum shear stress. BUT this is not always the case, To estimate the plane and angle of failure and the magnitude of stresses that may cause that 1 failure, several criterions were developed in the literature. ‘As we saw previously, shear stress required to cause shear failure depends on normal stress. :—>| consequently, Mohr concluded that failure occurs at a critical combination of normal and shear stresses acting on the failure plane. T Later, Coulomb developed Mohr-Coulomb failure criterion which states that: “shear failure : ‘occurs when shear line (r= j4;0,) intersects the state of stress represented by the Mohr circle”. at toi: Q. rocks fail at the intersection of the failure line and the Mohr circle. WHY? Rocks unlike metals do not fail due to the max shear stress because in rocks there are the internal friction coefficient. Because if it is so, the plane of failure will be always with an angle ( = 45°) because it is the ‘apex of the Mohr circle. Alternatively, rocks fail at the point where ) is maximum. The higher the ratioof (2), the more likely we get a rock failure. Along the Mohr circle, we get the max value of this ratio at ‘the intersection point of the line and the circle. The shear (failure) line represents the maximum axial stress we can apply on the rock without causing shear failure, ie, it represents the shear strength of the rock. The Mohr circle is a graphical representation of all possible stress states depending on the plane at which + and 6, is measured. From all these possible states of stresses, there is just one state of stress that intersects the line r = 04. At that plane, the shear failure will occur. ! ! qos eens Failure condition Stable and safe condition The friction coefficient y; is related to the friction angle y as u; = tan* @ . Typical values of 1; vary from 0.4010. The failure angle f is related to the friction angle » by _@," B=$45 ' Q. the angle is different in the two figures Why? | In the 1" figure, the plan between a, and a, is x, whereas in the 2°! is 5. So, it is as if the plan in figure 1 is 2*figure 2. Wha} deere TARVER AAD tttttt To obtain normal and shear stresses at failure as well as the failure angle, we proceed as follows: 6.3.1 For uncemented confined rocks 1. For a fixed confining stress (0,), apply an increasing axial stress till failure then take the corresponding «; (for ex, green circle). 2. Draw the first Mohr circle by calculating normal and shear stress (failure angle is varied from 0 to ». 3. Repeat steps (1) and (2) for different values of confining stress (0:) (for ex, blue circle) > the higher ‘he confining radial stress, the higher the ollowable axial stress). 4, The resulting circles will be limited by the same failure line which is an intrinsic property of the rock. Draw this failure line possing by * the apex of Mohr circles. 5. The slope of this line is the friction coefficient u, x = 1 04 6. Calculate the friction angle p = tan“? 7. Calculate the failure angle p = £+26.3.2 For uncemented unconfined rocks For uncemented unconfined sand, radial stress = 0 (c, = 0) > oxial stress required to cause shear failure = 0 (0, = 0) > it can't resist shear stress > there will be no ‘Mohr circle > material strength = 0. ‘That is why at the beach; the unconfined sand is loose while the confined sand is strong (capillary forces due to the presence of water holding the grains together lke in sand castles). 6.3.3 For unconfined cemented rocks Cemented rocks can bear shear stresses with zero lateral confining stress. Contrary to uncemented rocks, even for unconfined conditions (a, = 0), cemented rocks can resist the load till a given maximum stress colled Unconfined Compression Strength (UCS) is achieved, (0) Unonind Londng Ue 6.3.4 For confined cemented rocks Tf an ef fective compressive “confining” stress is applied o, + 0, the measured peak stress is higher than the peak stress without confining stress, The increment in peak stress will be a function of the internal frictional strength of the rock. Hence, the maximum shear stress r will bea function of both the reck cohesive strength ‘So and the applied normal effective. compressive stress o,, through the Coulomb failure criterion expressed in the following equation: T= 55+ Hi Sa is the cementation coefficient of the rock fii read Tt presets plysialy the sheer aresa ot on TULL ae : W0,+5, effective normal stress equal to 0. . : NB. for uncemented rocks, there will be no . green Mohr circle (Ft will be a point since for $o:0 and oy = 0, 7= 0), the failure line will Hill = a pass through the origin, To characterize the failure, we consider an unconfined condition at the beginning in order to determine the cementation coefficient So. Then, we proceed «3s in the previous algorithm. Tt is easier to compute shear failure parameters based on principal stresses rather than on normal and shear stress. Coulomb's failure criterion can be rewritten as: 0, = UCS +q0r Where a, is the effective maximum principal stress at failure, a; is the effective minimum principal stress and q is the friction parameter function of the friction angle. Tt can be shown that) + sing T=singFor atypical = 30°, 93. The friction coefficient can be calculated from the friction parameter q: w= 2 = UCS can also be expressed in terms of cohesive strength as: ves =25 (14808) asin asap) et wD) 4a 6 ee ye | rn a, | AL) _\, 7. To % How to draw a Mohr circle? 1. Preform a compression test without the confinement and get the corresponding rock strength UCS, ‘then draw the green Mohr circle. 2. Repeat the test, for a given confinement stress o,, (cst), keep increasing the deviatoric stress till foilure occurs and get the corresponding o;:, then draw the blue Mohr circle. 3+ Repeat the same test many times with increasing confinement stress «,, and get the corresponding rock strength o,, then draw the Mohr circles. 4- Plot (a, = USC +4 4;) +o get USC and q. Be get u, andy. 6- Draw the failure line. 6.4 Triaxial test Although shear strength of rocks depends on effective stresses, not on total stresses, in field and laboratory, we usually measure total stresses (Si, Sz, Ss) instead of ef fective stresses (01, 02, 03). The shear strength of rocks is measured in a cylindrical triaxial cell. A confining pressure P: (or stress Ss), a deviatoric stress ov (0i-02), and pore pressure Pp are applied on the rock sample.tT . \ae The confining stress is applied using a deformable sleeve, asus which is pressurized by a confining fluid (typicelly oil or water) Vi supplied by @ fluid pump. The sleeve prevents the confining | ) fluid from entering the rock's pores, thereby avoiding contamination or mixing with the pore fluid inside the rock. Ina typical deviatoric triaxial test, the confining pressure Pe is maintained constant throughout the test, while the axial load is varied using a piston that compresses the rock sample along its axial direction to simulate deviatoric (differential) stress o>=S;Ss, where Si represents the total maximum stress applied by the frame, and $3 is the minimum total stress, which is equal to the confining pressure Pc (i.e. $3=P.). The principal stresses satisfy S2=Se, so the deviatoric stress can also be expressed as Sp = 5, — Sy = 0; ~ 03. The pore pressure P; must always be less than the confining pressure Pc to prevent the sleeve from inflating inside the pressure vessel. That is why another fluid pump is connected Yo the pore space to ensure that the pore pressure Py is ‘maintained constant throughout the experiment while the confining pressure and axial load are varied. A load cell is placed at the lower part to measure the deviatoric stress (So) 7. Yield/failure locus Q: in a given condition, which type of failure will occur to the rock? Let's combine all rock failure types in a single (01,1) plot ‘+ The rock fails in tension if the Mohr circle goes further to the left of T.. ‘+ Tt fails in shear if it touches the red line r = Sy + i. ‘+ Tt develops significant compressive plastic strains if it crosses the yield cap (blue line). The rock fails in tension, K will increase. Tf it fails in compression, it will decrease. Whereas, if it fails in shear, K may decrease or increase. So, using geomechanics, by ensuring the right type of failure, we may increase the permeability. That is why, both tension and shear failures are very important in hydraulic fracturing. At the beginning, the reservoir was in initial in-situ stress. This stress state may be altered by drilling wellbore, by adding a hydraulic fracture (which may cause a tension failure), by depleting the reservoir (which may cause compression failure), or by injecting fluids (which may cause a shear failure caused by fault reactivation).