Selection of fracturing fluid for stimulating tight gas reservoirs

Virtually all wells completed in tight gas sands and shales require fracturing. A flowchart was developed to help pick the right fluid for the job.

Raj Malpani* and Stephen A. Holditch, Texas A&M University 

Essentially all producing wells drilled in tight gas sands and shales are stimulated using hydraulic fracture treatments. The development of optimal fracturing procedures, therefore, has a large impact on the long-term economic viability of the wells. The industry has been working on stimulation technology for more than 50 years, yet practices in use may not always be optimal.

Using information from the petroleum engineering literature, numerical and analytical simulators, surveys from fracturing experts and statistical analysis of production data, a flowchart was developed to guide selection of the appropriate facture fluid in most gas shale and tight gas reservoirs. Various parameters are considered, such as the type of formation, the presence of natural fractures and reservoir properties.

HYDRAULIC FRACTURE STIMULATION

Hydraulic fracturing plays a key role in producing unconventional gas resources. In the first stage, a small quantity of fluid, known as “pre-pad,” is pumped down the well to fill it, break down the formation, and make sure the mechanical condition of the well is satisfactory. Then, a second fluid known as “pad” is pumped. The hydraulic pressure generated by pumping the pad causes the fracture to propagate into the reservoir. The pad fluid also cools down the wellbore and the rock near the fracture walls. Subsequently, a slurry consisting of fluid and proppant is pumped into the fracture. High fluid viscosity is required to carry proppants deep into the fracture and prevent proppant settling. After these steps, it is critical to break the fluid and reduce the viscosity using additives so the fluid can flow back and the well can clean up. The fracture must close on the proppant to prevent settling and to create a long conductive fracture.

Industry experience suggests that successful stimulation of tight gas sands requires creation of a long (several hundred feet or more) and conductive fracture. To achieve this, large volumes of proppants must be pumped at high concentrations using large volumes of fluids to transport proppant deep into the fracture.

An ideal fracture fluid would have enough viscosity to transport proppant deep into the fracture at a reasonable cost relative to other fracture fluids. The fluid should have low to moderate friction properties and should be stable at the reservoir temperature during pumping. After pumping ends, the fluid’s viscosity should decrease to enhance the cleanup, leaving little to no residue in the fracture that would reduce conductivity.

Failure of a fracturing treatment can be caused by fracturing out of zone, poor choice of proppant or fracture fluid, poor reservoir characterization, proppant settling, inefficient fracture cleanup and/or damage to the fracture. For one or more of these reasons, a poor fracture treatment design does not result in a fracture long enough or conductive enough to optimize gas recovery.

HISTORY

Virtually all wells completed in tight gas sands and shales require fracturing to achieve economic gas flowrates and recovery. The first hydraulic fracture treatment was pumped in the Hugoton gas field in July 1947.¹ Four gas-productive limestone layers were fractured using gasoline gelled with napalm. By the mid-1960s, the use of large volumes of low-cost water as the fracture fluid was normal to stimulate many low-permeability gas wells. In the early 1970s, viscous fluids emerged as improved fracture fluids that were capable of carrying higher concentrations of proppants (4-5 ppg).

Since the 1980s, fracture treatments have used water gelled with polymers that could be cross-linked so that large volumes of propping agents at high concentrations (8-10 ppg) could be pumped. These treatments often work very well, especially at high temperature (> 300°F) where stabilizers must be used in the gelled fluid. At lower temperatures (200°F or less) and for low reservoir pressures, the industry typically uses foam fluids. Foam fluids will break and clean up when the BottomHole Pressure (BHP) is reduced during flowback.

At formation temperatures of 200-250°F, cross-linked gel fluids can still be used, but the fluid must be carefully designed so that sufficient breaker is used to break the fluid after the treatment is completed. If the appropriate type and amount of breakers are not used at BottomHole Temperatures (BHTs) less than 250°F, then the unbroken fracture fluid may cause damage to the fracture. Under certain conditions, minimal effective stimulation may result, sometimes leading to sub-economic wells,² especially if fracture fluid cleanup problems occur.

In some medium-temperature reservoirs, like those in the Cotton Valley Formation in East Texas, it was observed that some cross-linked fracture treatments were not very successful in creating long fractures. As an experiment, some operators began pumping water fracture treatments to see if less expensive treatments could provide adequate stimulation.

Water fracture treatments were initially designed to generate fractures by injecting low-viscosity fracturing fluid composed of water, clay stabilizers, surfactants and friction reducer. Most of the proppant is pumped at concentrations of 0.5-1.0 ppg. Near the end of the treatment, proppant concentrations are ramped up to a maximum of 2 ppg to achieve higher near-wellbore conductivity.³ Water fracture treatments’ main advantage is that they cost less than a comparable gel fracture treatment, because less polymer, fewer chemicals and less proppant are pumped. However, since lower proppant volumes and concentrations are used, issues concerning effective fracture length and conductivity must be analyzed.

Nitrogen foam is 65% foam quality or higher. At lower temperatures (200°F or less) and low reservoir pressures, the industry typically uses foam fluids. Foam fluids will break and clean up when the BHP is reduced during flowback.

Miceller fluid is created by adding an electrolyte, such as quaternary ammonium salt, to water along with a special surfactant that creates long, wormlike micelles. The micelles create viscosity in a similar manner to that of long-chain polymers in gel fluids. Miceller fluid has great appeal, as it develops reasonable viscosity and has reasonable proppant transport without requiring polymers. In this case, the produced hydrocarbons are the breaker system; oil and gas entering the fracture break the micelles and assist fluid cleanup. Miceller fluids have been used for several years. They have not dominated the market, because of temperature limitations and the cost of surfactant. However, if surfactants can be developed with higher temperature stability and the costs can be reduced, then miceller fluids could be the ideal fluid for many tight gas reservoirs.

Recently, a new, hybrid fracture treatment has been used by some operators with reasonable results. The hybrid treatment combines the benefits of a cross-linked gel and a water treatment. Slick water is pumped as the pad fluid to create the fracture geometry with theoretically little hydraulic width development and minimal out-of-zone height growth. Subsequently, a more viscous cross-linked gel is pumped to create fracture width and carry proppant into the fracture. In one field, hybrid fracturing seems to generate longer effective fracture half-lengths and conductivities than either gel or water treatments.⁴

The current understanding of when and where to apply various types of stimulation treatments such as gel fracture treatments, water fracture treatments, hybrid fracture treatments, miceller fracture treatments or foam fracture treatments is limited. In “medium”-temperature reservoirs, it appears that hybrid fracture treatments or miceller fracture treatments may provide the best stimulation alternative.

EVOLUTION OF WATER TREATMENTS

In 1986, slick water treatments were reintroduced to fracture stimulate horizontal wells completed in the Austin Chalk Formation, resulting in greatly increased production.⁵ The treatment uses slick water as the pad to create the initial fracture geometry, followed by 20-30 lb/1,000 gal of linear gel for the proppant-laden stages. Explanations for the success of water fracturing in the Austin Chalk include imbibition, gravity effects, opening of multiple fractures, skin removal, cleanup of old fracture fluid residue, dissolution of salt, reservoir re-pressurization and rock mechanics effects.⁶ We believe that much of the benefit came from the removal of old gel that never broke when the well was originally fracture treated and had been plugging the natural fractures around the wellbore for years.

In the mid-1990s, a few operators started pumping water fracture treatments in the Cotton Valley sands, partly because of success in the Austin Chalk. It was believed that gel fracture treatments in the Cotton Valley were not cleaning up effectively, resulting in short effective fracture lengths. The lower cost of slick water treatments compared with gel was also a major reason for the switch. The early water fracture treatments pumped in the East Texas Basin primarily used slick water as the fracturing fluid without any linear or cross-linked gel, with very little proppant. Higher injection rates were used to help transport the proppant and minimize leak-off.⁷

By 1997, water fracture treatments were being pumped in the naturally fractured Barnett Shale, reportedly resulting in better stimulation than gel treatments.^3,7 This success is attributed to permeability and porosity, gross thickness and the existence of a natural fracture network.⁸ A common feature of the Barnett Shale and the Austin Chalk is the existence of a natural fracture network.

Water fracturing must be evaluated on both a technical and an economical basis. Clearly, the treatments work well in naturally fractured reservoirs like the Austin Chalk and Barnett Shale. However, it is not clear from the literature whether water fracture treatments provide optimal stimulation in medium-temperature, tight gas sands that are not naturally fractured, such as the Cotton Valley Formation.

Generally, the success of water fracturing in many cases depends upon the existence of natural fracture systems and their favorable response to the injection of fracture fluid and proppant. Other possible reasons that water fracturing works well include imbibition, the creation of a wide fracture network due to opening of multiple natural fractures, shear dilation and asperities, and the absence of cleanup problems in the fracture because very little gel is used.

Imbibition. Imbibition is a process by which the wetting phase displaces the non-wetting phase. For water-wet, naturally fractured rocks, water will displace oil and gas from the pores in the matrix, expelling the oil and gas into the natural fractures where it can flow to the hydraulic fracture and eventually to the wellbore. Numerous studies have shown that significant imbibition occurs in the matrix of Austin Chalk cores. Analysis of water injection in the low-permeability, naturally fractured Spraberry Formation has indicated the importance of imbibition in reservoir performance.⁶

Creation of a fracture network. Slick water pumped at very high injection rates is able to open existing fractures and, perhaps, create new ones. The new fracture geometry may be very complex.^5,6 Often a network of fractures will be created rather than a single, planar fracture, such as expected when treatments are performed in homogenous rock. The process may induce fracture offset and branching, thus enhancing the reservoir permeability. Microseismic mapping of water fracture treatments often indicate the creation of extremely complex fracture networks, resulting in an increased fracture surface area, Fig. 1.⁹

Fig. 1. Fracture geometry of hydraulic fractures ranging from a single, planar fracture to a wide fracture network.

Shear dilation and asperities. When a fracture (natural or man-made) opens, rock-face mismatches and asperities may be created, which may prevent the fracture from completely closing when the pressure within it falls below the in situ stress. Propagating fracture fluid can open existing faults and planes of weakness by shear slippage. During pumping, the pressure inside natural fractures is elevated, and thus the stress distribution around the fracture changes. Beyond a threshold pressure, rock material around the fracture fails by sliding, instead of opening as considered in conventional hydraulic fracturing. At the end of pumping, asperities of the rough fracture surfaces may not come back to the original position, and thus the fracture may remain open.¹⁰

Absence of unbroken fluids and proppants. Water fracture treatments primarily use slick water with little or no polymer, without any cross-linker so that the fluid does not have to be broken to flow back and clean up. The fracture created remains clean due to a lack of unbroken and/or degraded polymer. In addition, the fracture face remains un-damaged and open to gas flow unless fluid loss additives were used in the fluid. The interaction of the proppant with the natural fractures appears to hinder fracture growth, and allow for the re-direction of fluids in the reservoir.¹¹

FRACTURE FLUID SELECTION

Based on relevant literature, numerical and analytical simulators, surveys from fracturing experts and statistical analysis of production data, the researchers developed a flowchart for selection of the appropriate fracture fluid for particular sets of conditions, Fig. 2. The flowchart includes eight key parameters to guide engineers to the appropriate fluid: bottomhole temperature and pressure, presence of natural fractures, type of lower and upper barrier, modulus of the formation, height of the pay, and desired fracture half-length.

Fig. 2. Flowchart for fracture fluid selection.

The fluids among which the engineer can choose are:

1. Cross-linked gel

2. Low-concentration (20-25 lb/ 1,000 gal) cross-linked gel

3. Gelled water

4. Hybrid fracture treatment fluid

5. Miceller fluid

6. Foam fluids.

Figure 2 shows the flowchart for selection of the appropriate fracture fluid to stimulate the reservoir in various situations. In wells with low bottomhole temperature (less than 200°F) and low reservoir pressure gradient (< 0.2 psi/ft), nitrogen foam fracture treatments work well.

In deep and hot (BHT > 270°F) wells such as in the Vicksburg sands and the Wilcox-Lobo sands of South Texas, where polymers break down rapidly and gel must be stabilized, cross-linked gel fracture treatments are an effective selection. This option is also appropriate for wells with high temperature and high reservoir pressure gradient (> 0.2 psi/ft). For situations of high temperature and low reservoir pressure gradient (< 0.2 psi/ft), either carbon dioxide-assisted or nitrogen-assisted cross-linked gel treatments should be used.

For medium temperature (200°F-270°F) and low reservoir pressure gradient (< 0.2 psi/ft), either carbon dioxide-assisted or nitrogen-assisted hybrid fracture treatment should be pumped.

When to pump water treatments. As shown in the flowchart, there are various situations in which pumping water fracture treatments may be the best option.

When there are many pre-existing natural fractures in the formation, such as in the Austin Chalk naturally fractured oil reservoir, water treatments help to clean out fractures, are imbibed into the rock and expel oil. The Barnett Shale is another naturally fractured reservoir where water fracture treatments work well by creating a wide fracture network. For developments like these, with temperature less than 270°F, high reservoir pressure gradient (> 0.2 psi/ft) and a naturally fractured reservoir, water fracture treatment is appropriate.

Another situation in which water fracture treatments are appropriate is one with:

• Few or no natural fractures
• Temperature less than 270°F
• High reservoir pressure gradient (> 0.2 psi/ft)
• A strong lower barrier
• A weak or moderate upper barrier
• A thin pay zone (< 75 ft)
• A desired fracture length of less than 400 ft.

Finally, water treatments should be used in situations with all of the above parameters, except that the lower barrier is of moderate strength with a high Young’s modulus.

When not to pump water treatments. When BHT is greater than 270°F, then cross-linked gel fracture treatments should be used to provide better proppant transport than water treatments offer.

In reservoirs with a low pressure gradient (< 0.2 psi/ft) at any temperature, a foam- or gas-assisted treatment system should be used.

In reservoirs with few or no natural fractures, BHT less than 270°F, high reservoir pressure gradient (> 0.2 psi/ft) and a weak lower barrier or moderate-strength lower barrier with a low Young’s modulus, water treatment should be avoided. In these situations, use of water treatment would allow all the proppant to settle down in the zone below the pay zone, because gelled water is not viscous enough to retain and transport proppant deep into the fracture. In this situation, hybrid fracture treatments, low-concentration cross-linked gel fracture treatments, or miceller fracture treatments should be used to provide effective stimulation.

Even for reservoirs like those above, but with a strong lower barrier or a moderate-strength lower barrier with a high Young’s modulus, water treatment should be avoided if the pay zone is thick (i.e., greater than 75 ft). In this case, water treatment would allow all the proppant to settle in the lower part of the pay zone, leaving the upper portion of the pay without any proppant. In this situation, hybrid fracture treatments, low-concentration cross-linked gel fracture treatments, or miceller fracture treatments should be used.

These treatments should also be used instead of water treatments in reservoirs with:

• Few or no natural fractures
• BHT less than 270°F
• High pressure gradient (> 0.2 psi/ft)
• A strong lower barrier or moderate lower barrier with a high Young’s modulus
• A thin pay zone (< 75 ft)
• A strong upper barrier.

In this situation, though the zone is too thin for the proppant to settle at the bottom, the strong upper barrier locks the proppant into too small a vertical space, which will cause proppant to settle near the wellbore. This creates a risk of proppant bridging and screen-out.

Finally, water fracture treatments (and also low-concentration cross-linked gel fracture treatments) will not be useful to create fractures of 400 ft or more in length in reservoirs with:

• Few or no natural fractures
• BHT less than 270°F
• High pressure gradient (> 0.2 psi/ft)
• A strong lower barrier or moderate-strength lower barrier with a high Young’s modulus
• A weak or moderate upper barrier
• A thin pay zone (< 75 ft).

If fracture lengths greater than 400 ft are desired, hybrid fracture treatments or miceller fracture treatments should be used. 

LITERATURE CITED

¹ Gidley, J. L. et al., Recent Advances in Hydraulic Fracturing, Monograph Series, SPE, Richardson, Texas, 1989, pp. 1-2, 83-89, 95-106.
² Rushing, J. A. and R. B. Sullivan, “Evaluation of a hybrid water-frac stimulation technology in the Bossier tight gas sand play,” SPE 84394 presented at the SPE Annual Technical Conference and Exhibition, Denver, Colo., Oct. 5-8, 2003.
³ Mayerhofer, M. J. and N. D. Meehan, “Waterfracs: Results from 50 Cotton Valley wells,” SPE 49104 presented at the SPE ATCE, New Orleans, Sept. 27-30, 1998.
⁴ Sullivan, R. B. et al., “Optimizing fracturing fluid cleanup in the Bossier Sand using chemical frac tracers and aggressive gel breaker deployment,” SPE 90030 presented at the SPE ATCE, Houston, Sept. 26-29, 2004.
⁵ Tschirhart, N. R., “The evaluation of waterfrac technology in low-permeability gas sands in the East Texas Basin,” MS thesis, Texas A&M University, College Station, Texas, 2005.
⁶ Meehan, D. N., “Stimulation results in the Giddings (Austin Chalk) Field,” SPE 24783 presented at the SPE ATCE, Washington, D.C., Oct. 4-7, 1992.
⁷ Mayerhofer, M. J. et al., “Proppants? We don’t need no proppants,” SPE 38611 presented at the SPE ATCE, San Antonio, Texas, Oct. 5-8, 1997.
⁸ Coulter, G. R., Benton, E. G. and C. L. Thomson, “Water fracs and sand quality: A Barnett Shale example,” SPE 90891 presented at the SPE ATCE, Houston, Sept. 26-29, 2004.
⁹ Mayerhofer, M. J. et al., “East Texas hydraulic fracture imaging project: Measuring hydraulic fracture growth of conventional sandfracs and waterfracs,” SPE 63034 presented at the SPE ATCE, Dallas, Oct. 1-4, 2000.
¹⁰ Hossain, M. M., Rahman, M. K. and S. S. Rahman, “A shear dilation stimulation model for production enhancement from naturally fractured reservoirs,” SPE Journal, 7, No. 2, June 2002, pp. 183-195.
¹¹ Urbancic, T. I. and S. C. Maxwell, “Microseismic imaging of fracture behavior in naturally fractured reservoirs,” SPE 78229 presented at the SPE/ISRM Rock Mechanics Conference, Irving, Texas, Oct. 20-23, 2002.

THE AUTHORS
	Raj Malpani is a completion and production solutions engineer for Schlumberger Data and Consulting Services. He is involved in designing and evaluating hydraulic fracturing treatments in tight gas sand and gas shale formations. He is also engaged in planning and completing horizontal wells. Mr. Malpani earned an MS degree in petroleum engineering from Texas A&M University in 2006. He can be contacted at rajmalpani@slb.com.
	Stephen A. Holditch is head of the Department of Petroleum Engineering at Texas A&M University. He previously worked for Schlumberger, and he served as president of S. A. Holditch & Associates, a full-service petroleum engineering consulting firm, from 1977 to 2000. His firm provided petroleum engineering technology involving the analysis of low-permeability gas reservoirs and the design of hydraulic fracture treatments. Dr. Holditch also has been a production engineer at Shell Oil Company in charge of workover design and well completions. He joined the Texas A&M faculty in 1976.