1

WELLBORE STABILITY, STABILIZATION AND STRENGTHENING

F.B. Growcock, N. Kaageson-Loe, J. Friedheim, M.W. Sanders, J. Bruton, M-I SWACO, Houston, USA

This paper was presented at the Offshore Mediterranean Conference and Exhibition in Ravenna, Italy, March 25-27, 2009. It was se-lected for presentation by the OMC 2009 Programme Committee following review of information contained in the abstract submitted by the authors. The Paper as presented at OMC 2009 has not been reviewed by the Programme Committee. ABSTRACT Technically challenging drilling environments are now encountered in many of the world’s hydro-carbon plays. These are often manifested by narrow drilling windows that arise either from the geo-logical setting or from depletion of the hydrocarbons within the field over an extended period. Nar-row drilling windows require strict control of the equivalent circulating density (ECD) of the drilling fluid or the utilization of technologies that manipulate the near-wellbore environment so that drilling is allowed to progress using mud weights outside the prescribed mud weight window. There is a diverse range of such wellbore stabilization technologies, though most can be distilled into the con-cepts of wellbore isolation and wellbore strengthening. These are distinctively different from ma-naged pressure drilling or underbalanced drilling technologies. The drilling window defines the range of hydraulic pressures required to maintain wellbore integrity while avoiding fracturing or collapse of the hole. Essentially the hydraulic pressure provides the balance between the physical rock properties and the earth’s stress field. By manipulating one or several of these components, it is possible to change the state of equilibrium, thus preserving well-bore integrity while using sub-optimal mud weight. The authors review various wellbore stabilization technologies available in the industry, including physical, chemical, thermal, and mechanical methods.

INTRODUCTION According to a James K. Dodson Co. study (1993-2002 NPT analysis of shelf wells in the Gulf of Mexico), instability-related problems account for 44% of non-productive time (NPT) during drilling of oil and gas wells.1 These problems include lost circulation, stuck pipe, flows, kicks, sloughing shales and wellbore collapse. All of these are facets of wellbore instability, wherein the wellbore loses integrity and results in loss of time and/or materials to enable drilling through to the target. Lost circulation in particular has been and still is one of the biggest contributors of NPT. With the advent of extended-reach drilling and with the increased emphasis of deepwater drilling during these last few years, lost circulation now accounts for an even larger share of NPT than deter-mined in the 1993-2002 analysis. The risks associated with drilling into depleted zones are also increasing in importance as fields mature and reservoir pressures decline. Many producing reser-voirs have an overpressured shale cap-rock and, in these and other cases, may also have relative-ly impermeable interbedded shale layers. Mud densities sufficiently high to stabilize the shales can generate very high overbalances in the underlying or surrounding depleted sands: overbalances in the North Sea may be on the order of a few thousand psi, but some in the Gulf of Mexico as high as 13,000 psi have been encountered. Such high overbalances increase the likelihood and severi-ty of lost circulation.

2

In addition to the costs associated with NPT, loss of drilling fluid to the formation and the addition of products to stem those losses contribute large costs to the operation. This is particularly true for operations using NAF (non-aqueous fluids), i.e. oil-based and synthetic-based muds (OBM/SBM), though operations using water-based muds (WBM) are not immune. This paper provides a review of current industry practices for treating and preventing lost cir-culation during drilling operations.

MECHANICS OF LOST CIRCULATION Fundamentals of Wellbore Breakdown Lost circulation is classified/categorized to facilitate design of appropriate prevention and mitigation solutions. The most common classification schemes use (a) the rate or magnitude of fluid loss and (b) the loss mechanism. The rate of fluid loss may be classified as seepage (1-10 bbl/hr), par-tial loss (10-100 bbl/hr) and severe loss (>100 bbl/hr). The loss mechanism, which is the preferred classification scheme because it focuses on the source of the problem, may be classified as losses in the matrix, in existing fractures and in induced fractures. Losses in existing or induced fractures occur because the hydrostatic pressure in the wellbore ex-ceeds the stresses in the borehole wall, causing a fracture to open. The stresses in the borehole wall are themselves functions of the borehole’s presence in a previously undisturbed rock mass and are a consequence of the rock attempting to close around the hole. It is this closure that gives rise to the elevated stress in the vicinity of the hole (“hoop” stress). In essence the near-wellbore stress can be defined as being composed of two parts: the larger part is generally the horizontal far field stress (which itself is a function of the overburden), and the smaller part is the additional stress that arises from the disequilibrium caused by the hole. The near-wellbore stress is depen-dent on the depth and trajectory of the borehole; for passive sedimentary basins that characterize the majority of the world’s oil provinces, the near-wellbore stress typically has a minimum value that is 75 to 85% of the overburden. In planning a well, it is necessary to know the maximum mud weight that the well will support at any given depth to avoid fracturing the wellbore. This maximum well pressure is either calculated or measured. Measurements are made with Formation Integrity Tests (FIT) or Leak-Off Tests (LOT); these are usually run just below the last casing shoe and determine the maximum mud weight for that interval. An extended LOT, such as that shown in Fig. 1, increases the wellbore pressure beyond formation breakdown in order to observe fracture propagation and closure.2 The fracture gradient (FG) or leak-off pressure (LOP) is determined as the point where the pres-sure-vs-volume (or pressure-vs-time at constant pump rate) line deviates from linearity, and is often described as that point where non-propagating microfractures begin to form or open. The fracture breakdown pressure (FBP) is that point where a fracture opens to such an extent that it propagates unchecked. At this point the pump cannot keep up with the rate of loss of fluid. Some workers equate FG to the FBP, but here and in most drilling applications, FG is taken to be the lower LOP. Fracture propagation is generally manifested as a nearly constant pressure over time or pumped fluid volume after a fracture has achieved FBP; this pressure plateau is defined as the fracture propagation pressure (FPP). If pumping is stopped, pressure in the fracture bleeds off, and the fracture closes. The pressure at which the fracture closes—fracture closure pressure (FCP)—is normally associated with the far-field minimum horizontal stress (Shmin). If pumping resumes (e.g., in multi-cycle extended leak-off tests), the fracture can re-open and continue to propagate; usually the re-opening pressure is similar in magnitude to FCP. Extensive field experience suggests that Shmin generally correlates with LOP, hence this point is used to define the upper safe mud weight limit (fracture gradient) during well planning. It should be noted, however, that for deviated wells in which there is considerable anisotropy in the stress field around the wellbore, Shmin may be signifi-cantly lower than LOP.

3

The difference between LOP and FBP depends on many factors, including lithology, tensile strength of the rock, drilling fluid rheology and drilled solids content of the drilling fluid. Nonethe-less, in most cases LOP is about 95% of FBP; for a typical, normally pressured wellbore, this equates to a difference of 0.3 to 0.5 lb/gal between LOP and FBP.

Time or Volume Pumped (constant pump rate)

Pressure Fracture ClosurePressure (FCP) 

(= Minimum Horizontal Stress, Shmin)

Fracture PropagationPressure (FPP)

Fracture Breakdown Pressure (FBP)Leak‐OffPressure(LOP)

FractureGradient (FG)

Fig. 1: Generalized Results from Extended Leak-Off Test

Formation integrity tests (FIT) attempt to measure the wellbore pressure that a formation can with-stand up to the LOP or to some predetermined maximum well pressure. These tests generally do not fracture the formation. For many drilling operations, the lower limit of the mud weight is the pore pressure (PP) gradient and the upper limit is the FG. Drilling with a hydrostatic pressure less than PP will lead to fluid in-fluxes from the formation into the wellbore fluid; if not controlled, this influx may cause significant endangerment to the rig and personnel. A further hazard is from hole collapse, as the hydrostatic pressure may be too low to support the rock. For poorly consolidated or fragile formations, the low-er prescribed mud weight limit is commonly determined by something called the ‘hole collapse’ pressure, which is greater than the pore pressure. The range of pressures (mud weights) that are bounded by the pore pressure/hole collapse gradient on one side and the fracture pressure on the other is referred to as the “stable mud weight window,” or the drilling margin. The drilling margin can vary with hole deviation, rock properties and pore pressure. For deviated wellbores in poorly consolidated formations and for overpressured shales, the elevated hole collapse gradient narrows the drilling margin. In depleted reservoirs, the fracture gradient is also reduced, since it is coupled to the pore pressure, thus again narrowing the drilling margin. It is not an uncommon drilling di-lemma for the minimum mud weight required to stabilize an overpressured cap rock to exceed the fracture gradient in a depleted reservoir immediately below it. Location of the Loss Location of the loss zone is as important as the type of loss. If losses first occur while drilling ahead, or are accompanied by a change in torque or a drilling break (including the bit dropping), then the losses are likely to be on bottom. If losses occur while tripping or increasing the mud weight, then the losses may be off bottom. In these cases, losses are invariably located close to the previous casing shoe, as it is generally here that the fracture gradient is lowest for that hole section. If necessary, a temperature, pressure or spinner survey can be run to locate the loss zone. If an MWD resistivity tool is part of the BHA, then time lapse resistivity logs can be used to locate the loss zone.

4

PREVENTION AND CONTROL OF LOST CIRCULATION Much research and development is being devoted to avoidance of induced fractures, as well as minimizing losses in existing fractures and permeable zones. Wellbore Stability Analysis Obtaining an accurate geomechanical picture of the planned wellbore is of paramount importance. This will help to determine casing points and mud weights and quantify risk of hole collapse and lost circulation (hazard mapping). This means using data and wellbore stability models that gener-ate locally accurate PP, Shmin and FG, rather than average gradients. Much of the necessary in-formation is obtained from offset wells and documented drilling experience. Risk and wellbore sta-bility assessment can also be updated while drilling using real-time logging and data processing techniques. This requires real-time monitoring of downhole pressure, condition of the well and drilled cuttings volume and morphology. Drilling Techniques The technique used to drill a well plays a major role in determining and controlling the hydrostatic pressure throughout the well during the drilling operation. Thus, it is important to give a lot of thought to the method of drilling: i) Conventional drilling

While tripping in, break circulation at the shoe and every ~300 m in openhole; Circulate for at least 5 min; Bring the pumps up slowly after connections; Rotate the pipe before turning on the pumps; While tripping out, pump out for the first few stands/singles off bottom; Keep tripping speeds low across areas of potential lost circulation; Consider use of lubricants to reduce drag; Consider use of sweeps to clear cuttings from the wellbore prior to POH to run casing. This

will minimize cuttings bridges when RIH casing and cementing; Use annular fluid velocity that is just sufficient to clean the hole; Control ROP to avoid loading the annulus; Reduce the length of the exposed loss zone and reduce influx size.

ii) Managed Pressure Drilling (MPD): MPD techniques3 should be investigated to determine if

they are economically viable. Unlike underbalanced drilling operations and power drilling, the primary objective with MPD is to obtain a stable wellbore within a narrow operating PP/FG win-dow, and influx of formation fluids is avoided. MPD effectively manipulates the pressure win-dow so that the fluid “walks the line” between wellbore collapse and wellbore failure (fracturing, ballooning) with greater certainty. An important goal of MPD technology is to stretch or elimi-nate casing points. In a typical MPD application, the fluid system is closed utilizing (a) a Rotat-ing Control Device (RCD) and a drilling choke to restrict and control the exposed wellbore pres-sure profile, and (b) a casing pump to provide back-pressure when required. However, other configurations are also used, and the range of possibilities is expanding rapidly.

iii) Casing While Drilling (CWD):4 In CWD, a well is drilled and cased simultaneously using stan-dard oilfield casing. The BHA is latched into the bottom joint of casing and is run and retrieved through the casing via wireline. For directional or horizontal wells, the BHA can be fitted with conventional directional equipment, such as mud motors and measurement-while-drillinng (MWD) tools. Since these tools are run and retrieved inside casing, they are protected from the harsh downhole environment while in transit. This eliminates problems that typically occur dur-

5

ing tripping operations, such as kicks, unintentional sidetracks, casing wear, and wellbore in-stability due to surge/swab pressures and formation sloughing/swelling.

iv) Others: Underbalanced Drilling (UBD), Coiled Tubing Drilling (CTD), Expandable Liners Hardware Minimizing hardware restrictions and optimizing hardware performance is equally important. For surface equipment, (a) use solids control equipment that is able to maintain a designed concentra-tion of low-gravity solids in the mud; (b) remove pump strainers; (c) line up surface piping so that at least one mud pump can be rapidly switched to water or seawater; (d) have all surface equipment pressure-tested for leaks before downhole losses occur, and offshore have the ROV/SSTV check the riser for leaks on a daily basis; and (e) ensure that no mud transfers, additions, or dilutions are carried out while drilling towards or in a loss zone. For downhole equipment, (a) remove bit nozzles if large losses expected; (b) minimize the BHA, including the number of drill collars and heavy-weight drill pipe, and use no stabilizers; (c) restrict angle build by maintaining high RPM and low weight; (d) avoid running tools with limited flow paths or restrictions where possible, including core barrels, MWD, mud motors, floats and survey rings; and (e) avoid running drill pipe casing protectors, as these can swell and act like packers. Drilling Fluid & Wellbore Hydraulics To minimize wellbore instability, the drilling fluid should be designed and maintained so as to mi-nimize changes in equivalent static and circulating densities (ESD and ECD) that arise from changes in the drilling environment (especially flow rate, temperature, pressure and contamina-tion). Thus, it is important to:

Accurately calculate hydraulics profile of the well and monitor it at the rig; Use good hole-cleaning practices; Optimize solids control equipment configuration and performance; Use minimum mud weight while drilling, and change mud weight slowly; Use maximum low-shear-rate viscosity and flat gels;5 Maintain low fluid loss and thin filter cake; Follow prescribed tripping schedules.

Drilling fluids that can help to prevent or mitigate lost circulation create less stress at the wellbore or reduce the rate of loss of fluid into permeable or fractured formations. These are fluids that in-trinsically minimize ECD or are so shear-thinning that their rate of invasion in loss zones is slowed at the low shear rates encountered in the pores/fractures of the formation. Several types of drilling fluids meet at least one or preferably both of these criteria:6

Conventional Reservoir Drilling Fluid – non-damaging for reservoirs, of course Underbalanced Drilling Fluid – depleted zones Aphron Drilling Fluid – depleted zones Brine-Weighted or Micronized Fluid – deep wells, reservoirs Flat Rheology (temp-insensitive) Fluid – narrow PP/FG window, such as deepwater and

ERD (extended-reach drilling) Another manner in which drilling fluids can reduce the risk of loss is to isolate the wellbore, i.e., create a mechanical barrier between the fluid and the wellbore. Wellbore isolation can have the ef-fect of not only limiting contact and transfer of fluid but also reducing pressure transmission. Ideal-ly such a fluid would create “casing” in the act of drilling, thereby protecting the integrity of the just-drilled formation. This can be approximated in practical situations through implementation of cas-ing-while-drilling or liner-while-drilling. Fluid strategies can approach this ideal, too, through use of materials that plaster the wellbore; indeed, casing-while-drilling operations are claimed to stabilize

6

wellbores by “smearing” drilled solids onto the face of the formation. In conventional drilling opera-tions, several drilling fluids are able to achieve some degree of wellbore isolation. These include.

Wellbore-Isolating Drilling Fluids – systems based on silicates, aluminates, asphaltic and asphaltenic materials, ultra-low solids

Fluid/pressure isolation can also be achieved through formation of relatively impermeable plugs us-ing swellable or cross-linkable polymers. Even tight, well-consolidated filter cake can achieve some degree of fluid/pressure isolation (see next section). Irrespective of the type of drilling fluid used, maintaining good drilling fluid properties is also critical. Key properties for minimizing lost circulation are as follows:

• Keep the mud weight as low as possible; • Maintain gel strengths, yield point, and viscosity at the lowest levels which will effectively

clean the hole; • Maintain low MBT levels; • Keep fluid loss low to prevent excessive filter cake buildup. There is some evidence that

using filter-cake building materials with high compressive strength may aid in forming strong bridges that help to resist fracture initiation or opening.7

Additives for Preventing and/or Curing Losses Treatments with additives to prevent or mitigate lost circulation can be classified as low-fluid-loss or high-fluid-loss.8 Low-Fluid-Loss Treatments These are effective where the openings in the formation can be sealed relatively rapidly; they in-clude materials such as cement, resins, cross-linkable materials and particulates that pack tightly at the wellbore or within the openings of the loss zone. Sealing the wall of the wellbore can be effec-tive if the fraction of larger particles can form a stable external barrier (or a plug just inside the mouths of the openings) which can then be sealed with smaller materials. If a bridge can be created internally, the seal is more permanent, inasmuch as fluid and mechanical motion will not dislodge it as easily. Low-fluid-loss treatments generally make use of lost circulation materials (LCM). To cure losses, LCM are usually administered as high-concentration pills. To prevent losses, the whole drilling flu-id may be treated with LCM to provide a “background” concentration of the material, or it can be administered as 20- to 100-bbl pills that are added regularly, e.g., every 30 to 100 ft, depending on the drilling operation and type of loss zone expected to be encountered. For permeable and natu-rally fractured zones, general prescriptions typically are based on blends of sized CaCO3 and syn-thetic graphite, perhaps supplemented with a fiber; however, a large number of types particulates may satisfy the requirements. For whole mud treatment, a total concentration of LCM on the order of 20 to 30 lb/bbl is usually sufficient; for pills to be squeezed or used in sweeps, the concentration may be multiplied by 3 to 5. For severe losses, gunk squeezes, cement, swellable materials and cross-linkable polymers may provide some relief. The LCM product blend should include very coarse particles to plug or bridge the largest openings in the formation, be they fractures or pores. Typically plugging is thought to occur when the D90 of the LCM is greater than the aperture of the formation openings; bridging is defined as the structure that is built when the D90 of the LCM is less than half the aperture.8 Thus, plugging tends to occur at or near the mouths of the openings, whereas bridging occurs internally. Whether the formation openings are plugged or bridged, finer particles are also necessary to fill the voids between the coarse particles, and even finer particles are necessary to produce a tight filter cake, thus produc-ing a seal and fluid loss control. However, with normally weighted fluids, the weighting material is of

7

a size and shape that it takes on the role played by fine LCM; consequently, in weighted fluids, the concentration of fine LCM can be reduced or even eliminated. High-Fluid-Loss Treatments These treatments form seals only within the openings of a loss zone and are especially effective for sealing fractures. Thus, the seals are relatively stable and difficult to dislodge by normal drilling practices, though the seals are even more difficult to dislodge if the treatment material can adhere to the walls of the openings. High-fluid-loss treatments are generally particle-based. To promote fluid loss, the particle size dis-tribution is relatively narrow (uniform) or the particles have uneven shapes or open structures. In relative terms, the particle size of the LCM should be smaller than the fracture opening. This is ne-cessary to ensure that the material enters into the fracture and is then deposited by a process of de-fluidization as the carrier fluid leaks-off. Since the success of the treatment requires high fluid loss, contamination by drilling mud or other fines-laden fluid can significantly impair its effec-tiveness. Therefore, this type of treatment is better suited to the spotting and squeezing of pill-based LCMs. Nevertheless, several jobs have been run successfully with WBM in which the whole drilling fluid was treated to provide high fluid loss.9 High-fluid-loss treatments may not be effective for sealing very wide fractures (> 2 mm). Excessive flow rates in such fractures may prevent the deposited material from completely plugging the frac-ture opening. In addition, very large volumes of material may be required. Under these circums-tances, the high-fluid-loss treatment may be used to slow the rate of loss sufficiently and followed with settable materials like cement or gunk to plug the zone. Generally, high-fluid-loss treatments are effective only in high-permeability formations or fractured formations that exhibit high fluid loss. Even when using LCM in the whole mud, it is prudent to have an LCM pill on hand. A minimum of 100 bbl pumpable volume in a slug pit should be available. This should be mixed at the highest concentration of LCM that the agitators can handle. Additional LCM (as much as 80 lb/bbl) can be administered by dumping straight into the top of the pits or via big bags. As contingency, it is also critical to have a large volume of reserve mud prepared. Concentrated slurries containing as much as 250 lb/bbl LCM have been used successfully on Ekofisk to ameliorate the logistics problems in-volved in treatment of large volumes of whole mud with high concentrations of LCM. Wellbore Stabilization In addition to selecting a fluid which intrinsically limits fluid invasion and fracturing and choosing preventive or curative LCM treatments, formations may be stabilized by: • Changing chemical composition of the formation • Altering thermal gradient to increase the formation temperature, e.g., via mud heaters • Mechanically altering downhole stresses

Drilling fluids may change the chemical composition of the formations when they come in contact with them. Although this often has a destabilizing effect on the formation, some fluid/formation in-teractions may provide stabilization. For example, dehydration of water-sensitive formations by NAF has the well-known effect of reducing pore pressure, thus causing consolidation of the shale. This in turn causes a reduction in hoop stress and reducing the risk of wellbore collapse. However, excessive dehydration can lead to an increase in fracturing, spalling and disintegration of the shale. Also destabilizing is the reaction of water and some components of WBM with clay-laden forma-tions, generally reducing structural integrity. However, exchangeable cations in WBM (e.g., potas-sium, K+) can exchange with cations in clays, causing the clay lamellae to compress and consoli-date. As for the NAF discussed above, this consolidation reduces the hoop stress and “stabilizes” the wellbore. Again if the concentration of such additives is too high, it can result in spalling and disintegration of the shale. Compensating for these chemical effects is the nature of filter cakes produced by WBM. Generally WBM produce thicker filter cakes than do NAF, due to the WBM’s

8

intrinsically higher fluid loss. Relative to a fracture, the thicker WBM filter cake will isolate more of the fracture interior, thus reducing the risk of fracture propagation. Because of this, it is generally considered permissible to run WBM at somewhat higher densities (0.3 to 0.5 lb/gal higher) than in-vert fluids. Changes in wellbore temperature can affect both the fracture gradient (fracture pressure) and the collapse pressure. Heating the drilling fluid, and consequently the wellbore temperature, increases the hoop stress and, to a lesser extent, the pore pressure. This increases the fracture gradient but also the possibility of shear failure and wellbore collapse. In an onshore well in South Texas, the fracture gradient determined in LOT tests appeared to increase by 1.5 lb/gal when the WBM was heated from 92°F to 153°F (33°C to 67°C).10 Two different, but in many respects complementary, concepts are being used to mechanically alter downhole stresses, particularly to control losses while drilling depleted zones. The so-called “Stress Cage” concept is used primarily as a preventative measure that involves treatment of the whole mud system (also using regular, repetitive pills). The “Fracture Closure Stress” concept has been commonly used as a remedial treatment involving squeezing pills into existing fractures, though recently it has also been implemented as a whole mud treatment for WBM. These are de-picted in Fig. 2.

Stress Cage Fracture Closure Stress

Fig. 2: Wellbore Strengthening Concepts

Stress Cage Concept Building a Stress Cage theoretically involves changing the stress state of the target formation near the wellbore, rather than altering the physical strength of the rock itself. LCM is added continuous-ly at relatively low concentrations while drilling. The drilling fluid is overbalanced with respect to the FG of a target formation, thereby inducing shallow fractures in the near-wellbore region. Sized LCM particles are driven into the opening of the incipient fracture, prop it open and form an hydrau-lic seal near the fracture mouth. Once the fluid is isolated, fluid within the fracture leaks off through the fracture walls, thereby limiting further fracture propagation. Furthermore, as pressure in the fracture subsides, the fracture closes but is prevented from fully closing by the LCM that is wedged in it. This places an additional compression on the surrounding rock, thus causing a local increase in the hoop stress. The result is that a higher wellbore pressure is required to fracture the forma-tion. Hence, in theory, one can drill with mud weights that exceed the original fracture gradient.11-14 A novel approach calls for cooling the mud to reduce the hoop stress at the borehole wall, then set-ting the stress cage and allowing the mud temperature to increase.14 This has the effect of creating a more permanent stress cage and even higher ECDs. The stress cage concept appears to be effective and well proven for controlling fracture propaga-tion in permeable zones. There is less conclusive evidence for the effectiveness of the technique in impermeable formations such as shale. However, recent successful experience (as yet unpub-

As the slurry loses liquid, which filters into the formation, it consolidates.

15

The residual solid plug supports the fracture and isolates the tip.

9

lished) with drilling a weakened shale overburden above a depleting North Sea field indicates that this method can result, when engineered correctly, in wellbore pressures exceeding FG by 600 psi. In another successful trial, a settable adhesive chemical gel pill was used in conjunction with spe-cially sized and selected particulates, so-called stress cage solids, to reinforce induced fractures and raise the apparent fracture reopening pressure by 550 psi and the FBP by 150 psi.16 This set-table gel technology has been enhanced further with the development of a new type of system de-signed to promote adhesion of LCM to the shale formation, thus reinforcing the fracture seal and the stress cage effect.17 On the other hand, other studies suggest that sealing microfractures in shales can increase the integrity and apparent strength of the formations without increasing the hoop stress.18,19 Given that the LCM needs to form a propped seal in proximity to the fracture mouth, the type and size distribution of particles is critical. Various proprietary models describe the optimum matching of LCM to the fracture width, and this is an area of active research within the in-dustry. Commonly used models share the same particle plugging and packing theory as that used for selecting LCM to generate effective filter cakes for reservoir drilling fluids. Typically these mod-els match the D90 of the LCM particle size distribution to the maximum size of the openings.20 The role of the fine particles is to minimize fluid loss between the larger particles. For stress caging, the LCM also must have sufficient compressive strength that they will resist the fracture closure stresses involved in the operation. Thus, suitable LCM are generally large, granu-lar and tough. This sub-group of LCM is referred to as loss prevention materials or LPM. Some materials that qualify as LPM are shown in Figure 3.

Stress cage treatments usually require treatment of the whole mud with at least 15 lb/bbl LPM. Typical treatments use blends of sized synthetic graphite and crushed sized marble (CaCO3). Siz-ing of the LPM is determined first by calculating the maximum fracture width that would be gener-ated by the desired wellbore pressure. Propping of these fractures so that they maintain that frac-ture width and sealing them using an optimized LPM blend would allow the drilling operation to go ahead with the elevated wellbore pressure. One technique for calculating the sizing of the LPM is based on linear elastic fracture mechanics theory. This approach allows the fracture width to be calculated for a given fluid pressure and fixed fracture length. In many applications the fracture length is assumed to be 6 in; this means that the fracture lies within the wellbore stress field. Input to such models includes the elastic properties of the rock (Young’s Modulus and Poisson’s Ratio), the far-field principal stresses (overburden, minimum horizontal and maximum horizontal stresses), hole size, and deviation/orientation of the wellbore. The LPM concentration is determined from semi-empirical particle packing models that describe how the particles distribute within the fracture. Fig. 4 illustrates the interface for one stress cage design software package.

Fig. 3: Suitable Loss Prevention Materials

10

Fig. 4: Software used to Design LPM for Stress Cage Application When the Stress Cage technique is applied as a continuous treatment of the whole mud, mainten-ance of the required PSD of the mud is essential. The logistics of managing large amounts of large particles in the mud while maintaining acceptable mud properties is quite challenging. Regular monitoring of the drilling fluid PSD is most convenient if it can be handled on site, which requires a granulometer to monitor at least the trend in the PSD. Electrical or optical methods like laser light scattering can be used; an even simpler technique is wet sieve analysis, which has the advantage that drilled fines and weighting material can be removed so that the measurements reflect the PSD of the LPM. Such a device is shown in Fig. 5. To manage the required concentration and distribution of LPM in the mud optimally would be done by removing the large cuttings and fines from the flowline and returning the middle fraction back to the active mud system. However, if the interval to be stress caged is relatively short – a few hun-dred meters – it may be possible to by-pass the solids control equipment entirely and simply con-trol the concentration of drilled fines via dilution. If the interval is longer than a few hundred meters, it is typically more economical to use shale shakers with very coarse screens – using only the top level – to remove the cuttings, again employing dilution to control the concentration of drilled fines. This method is likely to be used mostly in smaller hole sizes, such as 12¼-, 8½- and 6-in. sections. Perhaps the best way is to separate the drilled fines and cuttings simultaneously while recovering the majority of the LPM for recycling through the active pit. A couple of solids-control equipment configurations that function in this manner as LPM recovery units are shown in Fig. 6.

11

Fig. 5: Wet Sieve Particle Size Analyzer

Fig. 6: LPM Recovery Devices Fracture Closure Stress (FCS) Concept The FCS technique theoretically involves creating, enlarging and subsequently plugging short frac-tures in a weak formation using high-fluid-loss pills after losses have begun.9,21,22 In common with the stress cage method, filling and sealing the fracture with LCM prevents it from fully closing, thus placing the surrounding rock in additional compression and increasing the hoop stress in that part of the wellbore. Unlike the stress cage approach, the treatment is generally applied as a pill con-taining high concentrations of LCM. In the stress cage method described previously, fracture propagation is minimized by quickly seal-ing the fracture mouth with LPM containing a broad PSD. In contrast, the FCS method entails squeezing LCM of fairly uniform size (analogous to stimulation fracture packing) or of disparate shape. Although these particles must have high compressive strengths, they are neither neces-sarily large nor granular and, consequently, they do not meet the requirements of LPM. The par-ticle size distribution of the LCM initially ensures a high fluid loss so that, as the LCM is squeezed into the fracture, it loses liquid to the formation and becomes an immobile mass within the fracture. This process isolates the interior of the fracture, thus halting any further propagation. Squeeze pressures are used that equal or exceed the mud weight or calculated ECD (wellbore pressure) re-quired for onward drilling. From linear elastic fracture mechanics, the squeeze pressure corres-

12

ponds to some increased fracture width; the LCM fills, seals and prevents the fracture from closing once this width is attained, thus maintaining the elevated hoop stress. The FCS method is typically applied using hesitation squeezes, often with stepped increases in application pressure. As a general rule, since the technique relies on elevated fluid loss, multiple hesitation squeezes are required in low-permeability rock to maximize the effectiveness of the treatment; fewer hesitation squeezes are required in high-permeability formations. Relatively high concentrations of LPM, typically > 100 lb/bbl (~300 kg/m3), are used to implement the FCS con-cept. The FCS method has also been applied as a whole mud treatment for WBM in depleted for-mations, resulting in apparent increases in FG of 1 to 3 lb/gal.9 Nevertheless, the most common approach is application through squeeze treatments of pills. As smaller and more discrete volumes of material are used overall, the FCS approach is consi-dered by some to be more adaptable to larger hole sections than the Stress Cage method, which typically relies on continuous addition of LPM. When using LPM, the FCS approach is limited to application in permeable formations, as is the case for the Stress Cage approach. However, with cross-linked polymer plugs, the FCS method may also be used in impermeable formations. When Losses Still Occur If, in spite of preventative efforts, lost circulation still occurs, a suitable LCM remediative treatment is the most likely prescription. If the treatment calls for use of a single pill, it is advisable to pull the drillstring to the shoe before attempting the pill. Common practice calls for having enough openhole volume below the bit to accommodate the full treatment. Squeeze treatments are typically more successful than sweeps. This is particularly true for individual or occasional pills. However, if regu-lar, repetitive pills are used, e.g., every 20 to 50 m, squeezing may not be necessary, in which case the pills can be administered as sweeps without interruption of the drilling process. If there is time and capacity, a pilot test of the treatment can help to evaluate the probability of suc-cess of the treatment. Existing test methods include the API Permeability Plugging Test, produc-tion screen tests and sand sealing tests, as might be appropriate for the loss zone. In addition, so-phisticated techniques are now available to study and develop techniques for prevention and con-trol of lost circulation. One such facility is shown in Fig. 7, wherein fracture reopening and closure stresses as well as fracture sealing in impermeable and permeable media can be quantified.23

Fig. 7: Fracture Test Device

13

Various risks accompany most incidents of lost circulation. Cuttings often settle around the BHA during loss of fluid, and the pipe may become mechanically stuck. Settling cuttings will act as a packer and exacerbate the problem if the losses are occurring below the pack-off point, so it is prudent always to keep the pipe moving. As loss zones may be at low pressure, differential stick-ing is also a risk. In addition, reactive clays overlying the loss formation may become unstable if exposed to uninhibited fluids, so it is important to ensure that the clays are chemically stabilized at all times. If losses occur in a highly permeable gas-bearing formation, even with the annulus closed, the likelihood of gas invasion into the drilling fluid is high. This will cause gas to migrate up the wellbore, displacing the mud in the well. If bull-heading is used, its rate must never be less than 600 gal/min and should be monitored carefully to track the hydrostatic pressure. SUMMARY Various techniques are now available that can complement and even obviate conventional lost cir-culation remediation practices. These techniques rely upon a comprehensive strategy for stabiliz-ing the wellbore, which includes implementation of or improvements in

Drilling Practices - locally applicable, more reliable wellbore stability modeling and ECD management practices, new techniques like MPD, CWD, UBD, CTD

Drilling Fluid Selection and Optimization of Mud Properties - choosing drilling fluids that pro-vide better control of ECD and drilling fluid invasion into the formation;

Surface and Downhole Hardware – Minimize obstructions and ECD surges; Wellbore Stabilization Techniques - Hoop stress enhancement methods including stress

cage and fracture closure stress. GLOSSARY Dxx – The particle size below which xx % of the particles exist, e.g., for D90 = 200 μm, 90% of the particles are of a size less than 200 μm equivalent diameter. Equivalent Circulating Density (ECD) – The surface mud weight, adjusted for temperature and viscosity under downhole circulating conditions. Fracture Gradient (FG) – Equivalent to the Leak-Off Pressure and, under isotropic conditions, in a homogeneous reservoir, Shmin. Fracture Closure Pressure (FCP) – The well pressure at which a propagating fracture closes. It is considered equivalent to the fracture re-opening stress and Fracture Gradient, and under iso-tropic conditions in a homogeneous reservoir, Shmin. Hoop Stress – Elevated tangential stress in the vicinity of the wellbore created by the presence of the wellbore in the rock mass. Hoop Stress Riser – Linear elastic stress response in the near-wellbore region caused by an in-crease in the circumference of the wellbore, e.g., via creation of a filled fracture. This is often now referred to as a “stress cage.” PSD – Particle Size Distribution, generally determined either using laser light diffration or reflection or by using, wet sieve analysis. Both methods are suitable for drilling applications. Shmin – Minimum horizontal stress. This is determined during a flow-back test after fracture break-down as the point at which the pressure decrease rate increases, i.e., when the fractures closes and is equated with the fracture closure pressure. In a homogeneous formation in an isotropic stress field, Shmin is often equated to the fracture gradient. Stress Cage – The increase in near-wellbore hoop stress brought about by creating, filling and sealing small hydraulic fractures . Filling and sealing the fracture prevents it from fully closing, thus placing the surrounding rock in additional compression leading to a local increase in hoop stress. Wellbore Strengthening – A procedure designed to widen the available mud weight window (drill-ing margin). This can be achieved by locally reducing the pore pressure or hole collapse gradient (e.g., with a low-water-activity OBM/SBM) or locally increasing the fracture gradient.

14

REFERENCES 1. James K. Dodson Co., Drilling Trouble Zones Forum 2004-2005, Galveston, TX. 2. Nolte, K.G., “Fracture Design Considerations Based on Pressure Analysis,” SPEPE (Feb 1988), p 22-30. 3. McCaskill, J., Kinder, J. and Goodwin, B., “Managing Wellbore Pressure While Drilling,” Drilling Contrac-

tor, p. 40, March/April 2006. 4. World Oil, Casing While Drilling Handbook, 2006. 5. Adachi, J., Bailey, L, Houwen, O.H., Meeten, G.H., Way, P.W., Growcock, F.B. and Schlemmer, R.P.,

“Depleted Zone Drilling: Reducing Mud Losses Into Fractures,” IADC/SPE 87224, 2004 IADC/SPE Drill-ing Conf., Dallas, TX, March 2-4, 2004.

6. Growcock, F., Paiuk, B. and Lopez, S., “Soluciones para la Perdida de Circulacion en Formaciones Per-meables y Fracturadas,” VII SEFLUCEMPO, Isla de Margarita, Venezuela, May 19-23, 2008.

7. Aadnoy, B.S., Belayneh, M., Arriado, M. and Flateboe, R., “Design of Well Barriers to Combat Circulation Losses,” SPE/IADC 105449, 2007 SPE/IADC Drilling Conf., Amsterdam, The Netherlands, Feb 20-22, 2007.

8. Kaageson-Loe, N., Sanders, M.W., Growcock, F., Taugbøl, K., Horsrud, P., Singelstad, A.V. and Oml-and, T.H., “Particulate-Based Loss-Prevention Material - The Secrets of Fracture Sealing Revealed!,” IADC/SPE 112595, presented at the 2008 IADC/SPE Drilling Conf., Orlando, FL, March 4-6, 2008.

9. Dupriest, F.E., Smith, M.V., Zeilinger, C.S. and Shoykhet, I.N., “Method to Eliminate Lost Returns and Build Integrity Continuously with High-Filtration-Rate Fluid,” IADC/SPE 112656, 2008 IADC/SPE Drilling Conf., Orlando, FL, March 4-6, 2008.

10. Gonzales, M.E., Bloys, J.B., Lofton, J.E., Pepin, G.P., Schmidt, J.H., Naquin, C.J., Ellis, S.T. and Laur-sen, P.E., “Increasing Effective Fracture Gradients by Managing Wellbore Temperatures, “ IADC/SPE 87217, IADC/SPE Drilling Conf., Dallas, TX, March 2-4, 2004.

11. Alberty, M.W., Aston, M.S. and Mclean, M.R., “Drilling Method,” US Patent 7,431,106, Oct 7, 2008. 12. Aston, M.S., Alberty, M.W., McLean, M.R., de Jong, H.J. and Armagost, K., “Drilling Fluids for Wellbore

Strengthening”, IADC/SPE 87130, IADC/SPE Drilling Conf., Dallas, TX, March 2-4, 2004. 13. Alberty, M.W. and McLean, M.R. “A Physical Model for Stress Cages.” SPE 90493, SPE Ann. Tech.

Conf. & Exhib., Houston, TX, Sept 26-29, 2004. 14. Song, J.H and Rojas, J.C. “Preventing Mud Losses by Wellbore Strengthening.” SPE 101593, SPE Rus-

sian Oil and Gas Tech. Conf., Moscow, Oct 3-6, 2006. 15. Gil, I., Roegiers, J. –C and Moos, D., “Wellbore Cooling as a Means to Permanently Increase Fracture

Gradient,” SPE 103256, 2006 SPE Ann. Tech. Conf. & Exhib., San Antonio, TX Sept 24-27, 2006. 16. Aston, M.S., Alberty, M.W., Duncum, S., Bruton, J.R., Friedheim, J.E. and Sanders, M.W., “A New

Treatment for Wellbore Strengthening in Shale,” SPE 110713, 2007 SPE Ann. Tech. Conf. & Exhib., Anaheim, CA, Nov 11-14, 2007.

17. Scorsone, J.T., Sanders, M.W. and Patel, A.D., “Development of a Novel Oil-based Chemical Gel Sys-tem for Improved Wellbore Stabilization and Strengthening,” Paper n. 30 Drill 05/04, Offshore Mediterra-nean Conf. & Exhib., Ravenna, Italy, March 25-27, 2009.

18. Wang, H., Towler, B.F. and Soliman, M.Y., “Fractured Wellbore Stress Analysis – Can Sealing Micro-Cracks Really Strengthen a Wellbore?,” SPE/IADC 104947, 2007 SPE/IADC Drilling Conf., Amsterdam, The Netherlands, Feb 20-22, 2007.

19. Wang, H., Towler, B.F., Soliman, M.Y.and Shan, Z., “Wellbore Strengthening without Propping Fractures: Analysis for Strengthening a Fracture by Sealing Fractures Alone,” IPTC 12280, International Pet. Tech. Conf., Kuala Lumpur, Malaysia, Dec 3-5, 2008.

20. Dick, M.A., Heinz, T.J., Svoboda, C.F. and Aston, M., “Optimizing the Selection of Bridging Particles for Reservoir Drilling Fluids,” SPE 58793, 2000 SPE International Symposium on Formation Damage, La-fayette, LA, Feb 23-24, 2000.

21. Fuh, G-F., Beardmore, D. and Morita, N. “Further Development, Field Testing, and Application of the Wellbore Strengthening Technique for Drilling Operations,” SPE 105809, SPE/IADC Drilling Conf., Ams-terdam, Feb 20-22, 2007.

22. Dupriest, F.E., “Fracture Closure Stress (FCS) and Lost Returns Practices,” SPE/IADC 92192, 2005 SPE/IADC Drilling Conf., Amsterdam, The Netherlands, Feb 23-25, 2005.

23. Hettema, M., Horsrud, P., Taugbol, K., Friedheim, J., Huynh, H., Sanders, M.W. and Young, S., “Devel-opment Of An Innovative High-Pressure Testing Device For The Evaluation Of Drilling Fluid Systems And Drilling Fluid Additives Within Fractured Permeable Zones,” Paper N. 41, Offshore Mediterranean Conf. & Exhib., Ravenna, Italy, March 28-30, 2007.

ACKNOWLEDGMENTS Thanks to M-I SWACO for permission to publish and present this work.