Preventing differential sticking and mud losses when drilling

Deep Shelf Drilling

Field tests and lab work indicate mud losses in highly depleted sands can be reduced with a new formation sealing product.

Saddok Benaissa, Baker Hughes Drilling Fluids; Alain Bachelot, Jean Ricaud and Gerard Arquey, TOTAL E&P USA; and Seehong Ong, Baker Atlas

A new preventative approach with water-based mud has been applied in several fields, while drilling through a series of highly depleted sands, and has proven to be effective in preventing differential sticking and mud losses.

Although operationally successful, geomechanics and fluid design resulting in these successes are not well understood. A geomechanical analysis indicates that two mechanisms might contribute to the success:

1. The near wellbore region is turned into a non-porous rock because the particles in the new mud tend to block the pore spaces. The theory of poroelasticity indicates that fracturing pressure is increased by reducing the difference between mud pressure and the pore pressure immediately behind the borehole, which for non-porous rock is zero.
2. Because of this blockage, it is possible that the near-wellbore rock strength is increased. This strengthening effect increases the fracturing pressure.

The geomechanics model can be used to define operational limits of various mud weights with proper drilling fluids design. This model would enable a consistent and focused approach on drilling fluid design to effectively mitigate massive fluid losses associated with drilling through severely depleted sands or in narrow pore pressure/ fracture gradient environments.

This article highlights the following:

• Introduction of a newly developed Deformable Sealant (DS) that can be added to a water-based fluid at 2 to 4% volume. The DS sealing additive is a component of a newly developed High Performance Water Based Mud (HPWBM).
• Field case studies show the evidence of using this new mud system to prevent, rather than cure, lost circulation when drilling through a series of highly depleted sands. Several wells were drilled successfully in different South Texas fields using this novel WBM. These applications have reduced mud losses and have proven to be successful in preventing differential sticking. Leak-Off Tests (LOT) conducted with and without the deformable sealant reveal a gain in fracture pressure of at least 1.4 ppg (800 psi).
• Data from logs, hydraulic fracturing, LOT, along with pore pressure and fracture gradients were evaluated and inserted into a geomechanical model to further predict the upper limit of the recommended mud weight used to drill wells in the area.

INTRODUCTION

Generally, the types of formations that are prone to lost returns are cavernous and vugular, naturally occurring or induced fractures, unconsolidated sands, highly permeable and highly depleted tight sands. Well known lost circulation control techniques such as bridging, gelling and cementing are typically used, with varying degrees of success. These remedies can sometimes complicate problems associated with lost returns. Attempts to cure lost circulation can be difficult and costly, especially when considering the associated non-productive time.

Lost circulation problems related to drilling through depleted sands are compounded by the low fracture gradient in the sands and the high mud weight required to minimize compressive failure in adjacent shales. For depleted sands, the best way to manage lost circulation is to prevent, rather than cure, the problem. This can be achieved using a combination of a geomechanics and a fluids approach. A literature survey indicates that significant work had been done in this area. ^1-13

Loss prevention materials (LPM) were developed to increase fracture initiation or fracture propagation pressure. Recently, a theory of using stress cages to increase fracturing resistance has been developed and demonstrated successfully in the field. ² Sand bridging or "smearing effect" that is generated by casing while drilling techniques has also been applied. ⁴

FIELD PROJECT DRIVERS AND GEOLOGICAL SETTING

Difficulties encountered when drilling through depleted formations to reach deeper objectives required optimization of drilling practices and drilling fluid design. The project drivers for the South Texas fields are numerous, and include

• Need to drill trouble free, and reduce non-productive time (NPT) associated with mud losses and stuck pipe.
• Highly depleted formations, which vary from well to well due to complex subsurface geology.
• High differential pressure that creates risk of differential sticking.
• Complex casing designs are often required to met the production objective.
• Historical difficulties were experienced in logging and running liners.
• Hole stability maintenance occurs in reactive and often pressured shale sections.
• Use of WBM for environmental reasons.

Several mature gas fields, located in Hidalgo and Starr counties in South Texas, are operated by TOTAL E&P USA. Table 1 outlines general well profiles in each field. The main reservoirs are in the deeper well sections. Several highly depleted, tight and low-permeability sands are present throughout the entire well. These 10 to 150-mD permeability sands can be exposed to as much as 8,000 psi differential pressure. These series of highly depleted sands present drilling challenges when trying to reach deeper reservoir targets. Hydrostatic pressure to control over-pressured shale varies from field to field and can be as high as 14,000 psi, requiring an 18.5-lb/gal mud density.

This area has a complex subsurface geology, with extensive faulting in the deeper intervals. A sequence of stacked sands separated by shale and lateral fault barriers is typical of the stratigraphy of the area.

This geological setting makes it difficult to predict the depletion rate in the sands to be drilled, especially in deeper intervals. The severity of depletion of these sands increases risk of mud losses and differential sticking. This often has a direct impact on casing-point selection and can jeopardize well architecture. Typical pore pressure and fracture gradient profiles in these fields show a sand depletion as low as 350 psi. Fig. 1 indicates a typical pressure regime in these fields in South Texas.

Fig. 1. Typical formation and pore pressure regime.

Numerous faults in the deeper formations often act as permeability barriers to the pressured-sand reservoirs. All of these variables, along with depletion due to offset drainage, affect pressure distribution throughout the well profile and make it difficult to predict pressure in sandstone formations from well to well.

Excessive drilling cost has been experienced in the past in these fields. This has been a result of the high differential between hydrostatic and depleted sand formation pressure, bringing about lost circulation and associated differential sticking. When the thief zone is severely depleted, problem-solving approaches such as cement plugs, squeezes, and expandable liner and casing while drilling can be costly solutions. Use of the DS additive was investigated and successfully applied to the drilling operation.

MECHANICAL STABILITY AND WELL PLANNING

Drilling successfully through this series of depleted zones often requires a delicate balance between mechanical shale stability and controlling hydraulic fracturing in the sands. Planning a directional well through a series of highly depleted sands separated by over-pressured shale presents significant technical challenges.

These depleted sands have low permeabilities and small pore sizes, and require a special sealant to provide adequate bridging of the pore throat openings. These South Texas formations have been extensively studied by many operators. The Wilcox sand series ranges in permeabilities from 0.1 to 150 mD.

Numerous types of LCM have been used with varying degrees of success. However, when it comes to lost returns prevention in depleted sand, they can be ineffective. Most conventional lost circulation materials are too large and, therefore, would be ineffective in bridging the low-permeability sands. It is possible that these large-size LCM materials could also be harmful to the fluid system because they tend to contribute to the external filter cake thickness and create a potential for pipe sticking.

DRILLING FLUID DESIGN

An inhibitive WBM with a low fluid loss was selected to maximize shale inhibition and minimize filtrate invasion into the depleted sands. DS is added to the drilling fluid prior to drilling depleted sand formations. Sealing agent concentration varies from 2 to 4% by volume. The degree of depletion, permeability and pore throat sizes of the depleted sand generally dictate the concentration of DS to use. Refer to Table 2 for typical field mud properties and composition.

The deformable sealing agent is a modified liquid-insoluble polymer that has excellent sealing characteristics. It is designed to reduce pore pressure transmission by internally bridging pore throats of the low-permeability sands and shale micro-fractures. It is water insoluble and highly dispersible in fresh-to- saturated saltwater-base mud. Because of its unique characteristics, DS can bridge very small pores.

Because DS comprises deformable colloidal particles, it will bridge at the borehole interface of low-permeability sandstone formations. This bridging not only creates an internal filter cake, but also contributes to external filter quality. Having a small, precise diameter and being deformable allows it to enter into pore throats of a low-permeability sand and rapidly form an internal cake seal. Data in Table 3 show evolution of fluid loss invasion into a 50-mm disk. Note the reduction in filtration rate with respect to time, indicating a reduction in permeability at the inner disk face.

These bridging and sealing characteristics appear to enhance "rock strength," hence, increasing formation fracturing resistance. By increasing fracturing resistance, the initial pore pressure-fracture pressure window is widened. This "effective" rock strength enhancement will allow the depleted sand to be drilled with appropriate mud weight required to control inter-bedded, pressured shale, while potentially reducing mud losses in depleted sand formations.

The internal filter cake resulting from this bridging material results in a rapid reduction in differential pressure between depleted formation and fluid hydrostatic head. It is believed a reduction in permeability develops in the internal cake interface. Figs. 2,3 are schematics of the effect of DS on high-permeability external and low-permeability internal filter cakes with relation to differential pressure sticking.

Fig. 2. Deformable Sealant (DS): External and internal mud cake and differential sticking.

Fig. 3. DS: Reduction in differential pressure.

LABORATORY EVALUATION

Lab tests were initiated to better understand and assess the mechanism of filtrate invasion and internal cake bridging on low-permeability tight sandstone cores. Fluids with and without DS were compared. Cores were saturated with a standard brine. Three runs were made, one with brine to establish a base line, the other two with base mud, with and without DS.

Preliminary PPT results show that mud containing 4% DS has less pore pressure transmission, indicating the DS formed an internal cake and reduced rate of filtrate invasion. Additional testing is presently in progress to confirm pressure transmission reduction by the DS, as well as better understand other mechanisms that could contribute to enhancement of fracturing resistance.

GEOMECHANICAL STUDY

While operationally successful in the wells discussed, it was not clear how significant mud losses were avoided from the viewpoint of geomechanics. It is speculated that because of the mud system's characteristics, pore spaces and throats are blocked in the depleted sands. This reduces fluid communication between borehole and depleted sand. This argument has been supported by preliminary lab pore pressure transmission (PPT) tests. It has been well documented that reducing pore pressure transmission can increase the pressure that triggers lost circulation. ⁹ An extreme case is that the DS fills all pore spaces and throats in the vicinity of a borehole. A close to non-porous internal filter cake will then contribute to the substantial increase of the breakdown pressure.

To illustrate this idea, a geomechanical study was conducted on three wells of similar geological and pressure profiles, namely Well 55, drilled without DS, and Wells 44 and 47 drilled with DS. For Well 55, fracturing and mini-fracturing data were available for the Lower Hensley sand (TVD 14,200 ft), and were used to calibrate the in-situ stresses models. It was estimated that there are minor differences between horizontal stresses, which are 1.01 psi/ft and 1.02 psi/ft using a Poisson's ratio of 0.25, Biot's constant of 0.96, and overburden gradient of 1.0 psi/ft.

Using the lower Hensley as a calibration basis and considering the effect of reservoir depletion on maximum/ minimum horizontal stress ratio, stresses in other sands were estimated, Fig. 4. Using the constrained rock mechanics parameters and near-borehole stress models, such as the one presented by McLenan and Addis, ¹³ the breakdown pressure was determined for permeable wall/ impermeable wall/ internal cake cases.

Fig. 4. Mud weight, minimum horizontal stress and breakdown pressure for Well 55.

For Well 55, no DS was used in drilling. Geomechanical analysis indicates that mud weight used in several depleted sands was above breakdown pressure (for permeable/ impermeable cake cases). This indicates that fracturing occurred. It can also be seen that mud weight was higher than minimum horizontal stress, thus causing the induced fracture to propagate. ⁶ As a result, significant mud losses occurred.

For Well 44, Fig. 5, using the same geomechanical parameters used in Well 55, it was found that mud weight was higher than breakdown pressure (for permeable/ impermeable cake cases) and minimum horizontal stresses. Several depleted sands would have been fractured and significant mud loss would have occurred if no DS had been used. However, because of the sealing capability of the DS, an internal mud cake was formed, which increased the breakdown pressure.

Fig. 5. Mud weight, minimum horizontal stress and breakdown pressure for Well 44.

For the Hansen sand, located at TVD 9,490 ft, with a pore pressure of 5.27 lb/gal, DS increased mud weight 1.7 lb/gal, compared with an impermeable-wall case. For the Hensley sand, located at TVD 14,137 ft with a pore pressure of 11 lb/gal (original pore pressure 18.2 lb/gal), DS can increase mud weight 2.2 lb/gal. Similarly, for Well 47, it was found that for the Hansen sand, with a current pore pressure of 8.28 lb/gal, DS can increase mud weight by 2.65 lb/gal.

LOT/ FIT AND FRACTURING RESISTANCE ENHANCEMENT

Well 29 of Pharr field had to deal with the following sequences:

• From 8,000 to 10,500 ft, presence of a pressure transition zone required mud weight to be raised to 16.3 lb/gal from 12 lb/gal.
• From 10,500 to 11,000 ft, a depleted zone (tight, low K sand) with pore pressure less than 3 lb/gal.
• From 11,000 to 14,000 ft, the pressured pay sand requires a mud weight between 17.0 and 18.5 lb/gal.

Well architecture addressing the variations in pressure profiles was designed in accordance with the following casing/ liner program:

• 9-7/8-in. casing was set at top of the first zone.
• First zone was drilled using a 9-1/2-in. (8-1/2-in. bi-centered) bit with progressively increasing mud weights (from 12.5 up to 16.3 lb/gal. The section was cased with an expendable liner.
• The second zone was drilled using an 8-1/2-in. (7-1/2-in. bi-centered) bit with a lighter mud of 11 lb/gal and cased off with a 7-in. liner.
• The third zone was drilled using a 6-in. bit with a heavy mud of greater than 17 lb/gal.

The 9-7/8-in. casing was set as planned and a formation integrity test (FIT) was performed 15-ft below the shoe to ensure that scheduled mud-weight increases were possible. A 16.5 lb/gal EMW was required. The FIT result was disappointing, indicating fluid leak-off started at 15.3 lb/gal EMW and pumping was stopped at 16.1 lb/gal EMW to avoid risk of fracturing the formation. Line pressure decreased progressively and was bled off at 15.1 lb/gal EMW with the pressure yet to reach perfect stabilization.

Data indicated maximum possible safe mud weight was now only 14.6 lb/gal (taking into consideration trip margin, etc.) resulting in the inability to drill the transition zone in one section as initially planned. As the selected hole diameters were already small, there was no room for contingency casing, and the entire well architecture was seriously jeopardized. It was then decided to: add a 3% DS to the mud, drill 1,500 ft, perform another FIT, and then assess the possibility to drill further without running the liner.

At 9,640 ft and six days later, the FIT was performed. The pressure was increased straight up to 16.5 lb/gal without injection. The test was satisfactory and drilling resumed. The planned TD (10,514 ft) was reached within nine days, with a 16.1 lb/gal mud and the liner was run to bottom and cemented. There was no seepage loss during these operations, further supporting evidence that the DS did enhance fracturing resistance. Fig. 6 shows LOT data prior to and after treating with DS.

Fig. 6. Leak off test (LOT): Formation strength enhancement.

CASE HISTORIES

Case history 1. When drilling the 12-1/4, 9-1/2 and 6-1/2-in. intervals, a series of depleted sand sections were to be drilled. Some of these sandstone formations were expected to have an extremely low depletion pressure. Differential pressure as high as 8,000 psi is often observed in these fields.

It was planned to use the DS in conventional water-based mud to eliminate differential sticking, mud losses and logging problems that had been suffered by the operator on offset wells. The DS was to be added in a concentration of 3% by volume. This concentration was to be maintained throughout the three intervals.

On the first two wells, all sections were successfully drilled without any differential sticking or mud losses in the depleted sands. All logging runs were made without incident. A reduction of several days in rig time was achieved, compared to offset wells. This reduction in drilling days and cost was achieved with improvements to the basic water-based mud used previously in this field. This was considered a major achievement because problems had occurred on most wells drilled by the operator in this field.

Case history 2. Severe losses and logging problems have been consistently encountered on offset wells. Several depleted sand formations were planned to be drilled in the 14-1/4-in. intervals of these wells. The adjacent shale required a 13 lb/gal mud weight for stability control. The reason for adding DS to the water-based mud was to reduce non-productive time (NPT). The DS was used to eliminate mud losses while drilling the depleted sand sections by adding a concentration of 3% by volume prior to drilling the first depleted sand.

The 9-1/2 and 6-1/2-in. sections were successfully drilled without any mud losses or differential sticking in the depleted sands. Most important, logging problems were not observed in these wells. The flexibility in allowing the DS to be added to the drilling fluid while drilling aided in reducing operating costs. This well reached TD several days less than the predicted drilling curve.

CONCLUSION

Field case histories supported by preliminary lab work and geomechanical studies indicate that mud losses associated with severely depleted tight sands can be reduced with use of a newly developed Deformable Sealant (DS) product. Good drilling and mud engineering practices contributed to success of the operation. Several conclusive benefits were observed with addition of this bridging/ sealing agent. With a better understanding of the sands' pore bridging properties and use of an internal mud cake approach, highly depleted tight sands can be drilled successfully and economically. Some of the accomplishments are:

• Improved drilling curve
• Reduced cost
• Stable and gauge hole
• Reduction in mud losses
• Reduction in NPT
• Establishing a new learning curve.

Fig. 7 highlights the achievement that resulted in cost reduction and improvement in the drilling curve. There still are some downtime incidents such as pipe sticking due to gas kicks and problems setting casing at desired depth. It is believed that they can be resolved with improvements in drilling, mud engineering management and wellbore integrity management.  

Fig. 7. Evolution of field performance.

ACKNOWLEDGMENT

This article was prepared from paper SPE/ IADC 92266, presented at the 2005 SPE/ IADC Drilling Conference, Amsterdam, the Netherlands, February 23-25, 2005. The authors wish to thank TOTAL E&P USA, Baker Hughes Drilling Fluids and Baker Atlas for permission to publish this article. The work, test protocol, data and valuable discussions provided by staff from both companies (TOTAL: Patrick Zaugg, Brian Pregger, Jon Caron, Bernard Sanseau, Jason Peterson and BHDF: David Clark) are very much appreciated. Special thanks go to Mike Knight with Dominion Exploration for providing sands data, and to Xianjie Yi for his valuable contribution.

LITERATURE CITED

¹ Ali, A., C. Kallo and U.B. Singh, "Preventing lost circulation in severely depleted unconsolidated sandstone reservoirs," Paper SPE/ IADC 21917, SPE/ IADC Drilling Conference, Amsterdam, March 1991.

² Aston, M.S. et al., "Drilling fluids for wellbore strengthening, " Paper SPE 87130, IADC/ SPE Drilling Conference, Dallas, Texas, March 2-4, 2004.

³ Benaissa, S. et al., "Down-hole simulation cell for measurement of lubricity and differential pressure sticking," Paper SPE/ IADC 52816, SPE/ IADC Drilling Conference, Amsterdam, February 1999.

⁴ Fontenot, K., R.D. Strickler and P. Molina, "Improved wellbore stability achieved with casing drilling operations through drilling smear effect," World Oil Casing While Drilling Technical Conference, Houston, March 2004.

⁵ Fuh, G.F., et al., "A new approach to preventing lost circulation while drilling," Paper SPE 24599, presented at the 67th Annual Technical Conference and Exhibition of the SPE, Washington DC, October 4-7, 1992.

⁶ Ito, T., M.D. Zoback and P. Peska, "Utilization of mud weights in excess of the least principal stress to stabilize wellbores: Theory and practical examples,"SPE Drilling & Completion, December 2001, pp. 221-229.

⁷ Jones, J.F., "Successfully managing drilling-fluids losses in multiple, highly depleted sands," Paper OTC 13107, Offshore Technology Conference, Houston, April 2001.

⁸ Monroy, R., et al., "Pemex controls lost circulation in South Mexico,"Oil & Gas Journal, August 1999. Also SPE/ IADC 67735, Amsterdam, February 2001.

⁹ Morita, N. and A.D. Black, "Theory of lost circulation pressure," Paper SPE 20409, Presented at the 65th Annual Technical Conference and Exhibition of the SPE, New Orleans, Louisiana, September 23-26, 1990.

¹⁰ Rojas, J.C., et al., "Minimizing down hole losses," Paper SPE/ IADC 39398, SPE/ IADC Drilling Conference, Dallas , Texas, March 1998.

¹¹ Singh, B. and N. Emery, "Fracture-gradient predictions in depleted sands in the Gulf Coast sedimentary basin,"AAPG Memoir 76, pp. 125-129, 2002.

¹² Tare, U.A., D. Whitfill and F.K. Mody, "Drilling fluid losses and gains: Case histories and practical solutions," Paper SPE/ ATC 71368, September 2001.

¹³ McLean, M.R. and M.A. Addis, "Wellbore stability: The effect of strength criteria on mud weight recommendations," Paper SPE 20405, presented at the 65th Annual Technical Conference and Exhibition of the SPE, New Orleans, Louisiana, September 23-26, 1990.