Low-polymer borate frac fluid cleans up thoroughly

Mike Wiskofske
Marathon Oil Co.
Midland, Tex.

Matt Wiggins
Halliburton Energy Services Inc.
Midland, Tex.

Joe Yaritz
Halliburton Energy Services Inc.
Duncan, Okla.

Reduction in fracture-fluid gel concentration with a simplified low-guar, borate-crosslinked (LGB) has resulted in a two-to-five-fold production increase in Marathon Oil Co.'s Permian basin wells stimulated with this fluid system.

Thorough post-frac cleanup is believed to be the most important reason for the production increase.

In these stimulation treatments, the base fluid is a water-based guar mixed at 15-25 lb/1,000 gal. The low gel concentration contributes to quick and complete cleanup. Much less polymer is left in the formation, and the low gel concentrations do not hinder fluid rheological properties.

LGB

Marathon, as have many Permian basin operators, sought a cost-effective gel that would improve the production of marginal wells and make treating these wells more feasible. In its first LGB application, the Marathon and Halliburton Energy Services alliance used the frac fluid to stimulate a well in the Blinebry formation, Lea County, N.M.

After sand fracturing with 145,000 gal of 25 lb/1,000 gal LGB fluid and 338,000 lb of 16/30-mesh sand, the well flowed 70 bo/d and 2,500 Mcfd. Production rates have remained relatively steady for the past 16 months.

Before using the LGB in the area, Marathon had experimented with several frac fluids, including 30 lb/1,000 gal borate-crosslinked fluids, 65-70 quality CO2 foams, binary foams, and crosslinked binary foams. Typical pre-LGB frac stimulation production, in this area, ranged from 15 bo/d and 300 Mcfd to 30 bo/d and 1,400 Mcfd.

Formulation of the LGB system is simple because the buffer and crosslinker are combined into a single component that buffers the fluid to the required pH regardless of the initial source water pH.

The system is easy to mix because extra additives are not needed. This simplifies field operations so that less crew time on location is required.

Because the gel system has 30-40% less polymer than conventional fracturing fluids containing guar, it leaves far less gel and insoluble residue in the formation. Cleanup is thorough because LGB is compatible with both enzyme and oxidizing breakers.

In early 1996, LGB was introduced specifically for stimulating the low-temperature formations common in the Permian basin.

The LGB fluid system that resulted from research efforts to address conditions in the Permian basin has now found applications in Canada and South America.

To date, more than 1,500 wells in North America have been fractured with the LGB fluid system. Formations treated include sandstones, carbonates, dolomites, and other formations at low-temperature (70-140° F.).

Borate fluids

Borate-crosslinked fracturing fluid (BCF) systems have found widespread use for the following reasons:¹

Borate systems have high viscosity.
Borate gels reheal under shear because the borate crosslinks break and reform continually, with very little degradation.
Because borate salts are low-cost and naturally occurring, borate fluid systems decrease overall fracturing cost. The manner in which a borate provides crosslinking to a guar fluid is inherently different from other crosslinkers. The rapid exchange equilibria of the borate ion on the guar polymer and the low bond energy of the borate crosslink to the guar backbone contribute to the rehealing character of borate-crosslinked fluid systems.²

Borate-crosslinked fluids are less sensitive to shear under normal conditions, and the crosslinks are consequently broken and reformed readily. Borate crosslinks are also chemically reversible and will uncrosslink when the pH is lowered.

These properties along with the reduced guar concentration required for LGB lessen formation damage and allow for a more complete proppant pack cleanup than conventional systems. A more thorough formation cleanup and reduced proppant pack impairment provides for better fracture conductivity and increased production.^{3 4}

Conventional borate-crosslinked fluids generally do not break reliably between 70 and 120° F. Persulfate breakers used with borate systems are slow to react at temperatures below 120° F.; therefore, they require the addition of a catalyst to increase reaction rate.

Because enzyme breakers are hindered by the alkalinity requirements of conventional borate-crosslinked fracturing fluids, additives must be used to lower pH levels to be compatible with enzymes.

LGB has been designed to overcome these problems and to be compatible with enzyme breakers, especially at the temperatures that oxidizing breakers are inconsistent and require a catalyst.

Traditionally, non-Newtonian fluids with higher viscosity have shown better proppant transport than those of lower viscosity, although the exact correlation is an ongoing topic of discussion.^{5 6}

Fluid pH values of borate-crosslinked fluids have been generally equated to more stable viscosity and thus better proppant-transport capabilities. That is, if the pH remains at high levels, the fluid viscosity will be more stable.

Fluid stability provides better proppant transport and lower proppant-settling rates. Previous studies have indicated that at least a 9.5 pH and a 35 lb/1,000 gal polymer concentration are minimum specifications to provide consistently good proppant transport.^{7 8} The LGB system has shown to be an exception to this norm.

Effective fracturing fluids have good proppant-transport properties and cause minimal proppant-pack damage. Reduced fracture conductivity may be caused by one or more of the following conditions; remedies are listed in parentheses:

Proppant embedment into the formation. (Design higher proppant-bed concentrations.)
Formation debris or fluid residue plugging the formation. (Use special additives to stabilize the formation.)
Fines from crushed proppant. (Select proper size and type of proppant.)
Gel filter cake. (Reduce polymer loading or add breaker that will remain in filter cake.)
Unbroken gel plugging the proppant pack. (Design fluid and breaker system more effectively for well conditions.)

LGB rheology

LGB is a mixture of freshwater and 15-25 lb/1,000 gal guar, crosslinked by a single-buffer/crosslinker additive. Fig. 1 [35,966 bytes] shows the viscosity of 20 and 25 lb/1,000 gal LGB fluids compared to viscosities of 30, 35, and 40 lb/1,000 gal conventional borate-crosslinked, guar-based fluids at 120° and 140° F.

LGB has been developed to have viscosity equivalent to conventional borate-crosslinked fluid systems (BCF) having 10-15 lb/1,000 gal higher gel concentrations.

Crosslinking is accomplished by adding a single component buffer/crosslinker additive. This adds a crosslinker to the system and buffers fluid pH to the desired level, regardless of the source-water pH, or the influence of resin-coated proppants, resin-coated activators, or other additives.

The buffer consistently maintains the optimum pH level without further adjustments by frac operators.

Viscosity of crosslinked fluids can vary widely with variations in field parameters such as polymer concentration, crosslinking agent, pH, temperature, field water used, and shear regime.^{2 9}

The simplified LGB has demonstrated enhanced viscoelastic properties in Model 50-type Nordman viscometer and Bohlin rheometer testing.

Throughout the recommended temperature range, LGB has shown comparable or better rheological properties than conventional borate systems that use 30-40% more polymer.

LGB was designed to be compatible with both oxidizing breakers and enzyme breakers at low temperatures, without the need for additional components to lower the fluid pH level.

Fig. 2 [20,258 bytes] shows typical laboratory test results of breaker systems used with LGB. It is necessary to add an activator for the oxidizing breaker at 120° F. to obtain predictable break profiles.

Fracture conductivity

Past studies have indicated that fracture conductivity impairment and proppant-pack damage depends on the fracturing fluid used.^{3 4}

Reduced polymer loading inherently reduces the amount of gel and insoluble residue pumped into formation. This has been shown to improve retained conductivity.¹⁰

A study sponsored by the Gas Research Institute indicated that unbroken fracturing fluids can plug permeability to the point that it reduces recoverable gas reserves by 30% and initial gas rates by 80%.¹¹ Fracturing fluid cleanup can be delayed by weeks or months.

Conventional systems have shown 20-25% regained permeability, while LGB-type systems return greater conductivity, up to 50-75% of the original permeability.

Improved conductivity and breaker performance result in faster well cleanup and improved long-term production.

Fluid loss

Conventional BCF systems are extremely efficient at controlling fluid loss. LGB is no exception to this rule. As shown in Table 1 [42,939 bytes], the fluid efficiency of LGB is similar to the efficiency of conventional BCFs tested in low-permeability core.

On the Ohio sandstone core, a 35 lb/1,000 gal conventional borate fluid has a fluid-loss coefficient (Cw) of 0.0044 ft/min1/2, while a 25 lb/1,000 gal LGB fluid has a comparable Cw of 0.0039 ft/min1/2. But for high-permeability cores, LGB has improved fluid-loss control over conventional BCF systems.

On the 200-md Berea core, the 25 lb/1,000 gal LGB fluid at 140° F. has a spurt loss of 0.23 gal/sq ft and a Cw of 0.0030 ft/min1/2 in comparison to a conventional 35 lb/1,000 gal BCF with a spurt loss of 0.63 gal/sq ft and a Cw of 0.0035 ft/min1/2.

The better fluid-loss control on high-permeability cores would suggest that the LGB fluid system would be appropriate for frac-pack treatments.

Proppant transport

The dynamic proppant transport capabilities of a 25-lb/1,000 gal LGB fluid system were compared to those in a 35-lb/1,000 gal conventional BCF system.

A transparent slot was designed to simulate the fracturing-fluid performance as it is pumped through a perforation into a fracture.

The 2-ft high by 10-ft long transparent-slot model had a 3/8-in. gap width. The perforation was at two thirds of the face height. Fluid exited through an opening at the other end.

The 25 lb/1,000 gal LGB fluid was compared to a conventional 35 lb/1,000 gal BCF. These proppant-laden fluids had 20/40-mesh proppant at 5-ppg concentrations and were pumped through the perforation at rates of 1/2-1/6 bbl/min, resulting in shear rates in the fracture from 120 sec-1 to 40 sec-1.

To determine if there were any detrimental effects to the initial proppant-carrying capacity of these fluids, researchers also tested the fluids with breakers.

Tests showed both the conventional BCF and LGB systems seem to exhibit good dynamic proppant transport capabilities, but the LGB fluid has demonstrated more uniform proppant distribution throughout the fracture height.

The conventional BCF showed signs of stratification and had a more pronounced proppant concentration gradient, with the highest concentration of proppant near the bottom of the fracture simulator.

The LGB fluid also showed a near-perfect laminar flow profile whereas the conventional borate fluid established distinct layers of proppant-laden fluid traveling at different velocities and possessing different apparent viscosities.

Both the laminar flow profile and uniform proppant distribution in the LGB fluid result in better proppant placement throughout the length and height of the fracture.

Proppant-laden LGB fluids with typical field breaker concentrations were also tested. Breaker added to the fluid caused no immediate effect on LGB capability to provide proppant transport through the fracture simulator.

Field results

LGB was developed in response to Permian basin needs; however, use of the low-polymer principle is now widespread. Some LGB applications in North America by operators other than Marathon are as follows.

Permian basin-One 1,600-ft, 80° F. well was drilled next to offset wells that typically produce at 2-3 bo/d on pump, after conventional fracture stimulation. After stimulation with 15,000 gal LGB and 20,000 lb 20/40 sand, the well flowed at 16 bo/d. Economic value from the oil production increase was $117,000/year.
Indiana-Production in a well had declined to 4 bo/d. Attempts to fracture the well (which is in a sandstone formation) with 30 lb/1,000 gal linear gel failed because of sandouts at 1-ppg proppant concentration. An LGB treatment of 6,000-gal fluid carrying 6,000 lb 20/40 sand mixed at 2 ppg was pumped at 8 bbl/min and 3,000-psi treating pressure. The well cleaned up immediately with no proppant production. Initial production was 35 bo/d and after 2 months production was steady at 20 bo/d. The treatment paid out in 15 days and the incremental net present value (NPV) of 3 months was $43,240.
West Virginia-A 6,000-ft gas-storage well required fracture stimulation to increase capacity and ratio of gas stored to gas delivered back for sale.

Other storage wells in the area had been successfully treated by foam-frac; this technique requires high wellhead pressure, however, which causes operators to cut sand early in some treatments.

High incidence of screenout also prevented placement of planned amounts of sand from reaching the formation.

An LGB treatment of 25 lb/1,000 gal gel was run, with gel mixed on-the-fly from liquid gel concentrate. No extra gel was left over to require disposal. The treatment consisted of 500 bbl of fluid and 250 sacks, pumped at 3 ppg.

The well showed a 2.5-fold increase in deliverability, and the treatment cost less than previous ones.

Pennsylvania-An operator working the Medina formation had tried using gels up to 40 lb/1,000 gal to efficiently fracture-stimulate wells. The highest sand concentration possible was 4 ppg Ottawa 20/40 sand.

Using LGB (25 lb/1,000 gal), the operator placed 6 ppg Ottawa 20/40 sand in fluid gelled on-the-fly from liquid gel concentrate.

Gas production increased above the field average from an area of the field that traditionally produced far below the field average. On-the-fly mixing saved 4 hr of premixing effort, reduced service company manpower expense, and reduced equipment required on location.

Alberta-A 10-well stimulation program in the Doe Creek oil formation was split evenly between service companies. One company used a surfactant-based, clear fracturing fluid. The other used LGB as the fracturing fluid.

In 44 days of production testing, the five wells treated with LGB produced $155,000 worth of gas more than the five wells fractured with the surfactant-based fluid. This amounted to over $1.2 million/year in economic value for the operator.

References

Harms, W.M., "Application of Chemistry in Oil and Gas Well Fracturing," Oilfield Chemistry ACS Symposium Series 396, American Chemical Society, Washington DC, 1989, pp. 76-78.
Harris, P.C., "Chemistry and Rheology of Borate-Crosslinked Fluids at Temperatures to 300° F.," JPT, March 1993, p. 264.
Roodhart, L., Kuiper, T.O., and Davies, D.R., "Proppant Rock Impairment During Hydraulic Fracturing," Paper No. SPE 15629, 61st Annual SPE Technical Conference and Exhibition, New Orleans, Oct. 5-8, 1986.
Norman, L.R., Hollenbeak, K.H., and Harris, P.C., "Fracture Conductivity Impairment Removal," Paper No. SPE 19732, 64th Annual SPE Technical Conference and Exhibition, San Antonio, Oct. 8-11, 1989.
Report on the Investigation of the Rheology and Proppant-Carrying Capacity of Common Fracturing Fluids, Stim-Lab Completions Technology Consortia, 1997.
Report of proceedings of the Technical Advisory Group Meeting on the Fracturing Fluid Characterization Facility, Norman, Okla., Apr. 10, 1997.
De Kruijf, A., Roodhart, L.P., and Davies, D.R., "The Relation Between Chemistry and Flow Mechanics of Borate Crosslinked Fracturing Fluids," Paper No. SPE 25206, International Symposium on Oilfield Chemistry, New Orleans, Mar. 2-5, 1993.
McMechan, D.E., and Shah, S.N., "Static Proppant Settling Characteristics of Non-Newtonian Fracturing Fluids in a Large-Scale Test Model," Paper No. SPE 19735, 64th Annual SPE Technical Conference and Exhibition, San Antonio, Oct. 8-11, 1989.
Shah, S.N., Harris, P.C., and Tan, H.C., "Rheological Characterization of Borate Crosslinked Fracturing Fluids Employing a Simulated Field Procedure," SPE Production Technology Symposium, Hobbs, N.M., Nov. 7-8, 1988.
Nimerick, K.H., and Temple, H.L., "New pH-buffered Low-Polymer Borate-Crosslinked Fluids," Paper No. SPE 35638, Gas Technology Conference, Calgary, Apr. 28-May 1, 1996.
Voneiff, G.W., Robinson, B.M., and Holditch, S.A., "The Effects of Unbroken Fracture Fluid on Gas Well Performance," Paper No. SPE 26664, 68th Annual SPE Technical Conference and Exhibition, Houston, Oct. 3-6, 1993.