Chemical hooks keep proppant in place

Philip D. Nguyen, Jimmie D. Weaver
Halliburton Energy Services Inc.
Duncan, Okla.

Operators stimulating wells by hydraulic fracturing with water-based fluids can chemically modify the proppant surface to give each grain a "tacky" coating that enhances proppant-pack porosity and fracture conductivity.

This enables the proppant pack to resist being dislodged and becoming mobile during production.

These surface modification agents are compatible with proppant-flowback prevention methods such as resin-coated proppants and resin coatings applied on-the-fly.

Proppant surfaces modified to give a tacky feel permit more aggressive well cleanup procedures after stimulation. Under some conditions, aggressive flowback can improve stimulation performance.

When curable resin-coating is used as flowback prevention, however, production must be delayed to allow the resin to set, and the effect of aggressive cleanup is lost. Totally unrestrained proppant production defeats the purpose of placing proppant in the fracture, reduces production, and damages tubulars and surface equipment.

Numerous applications to date of surface modification techniques show that wells have been flowed back immediately after pumping or after an overnight shut-in period without showing evidence of proppant flowback.

Fig. 1 [21,726 bytes] illustrates a molecular-model where the proppant surfaces are modified to add the extreme tacky or adhesive qualities to proppant grains. The surface-modification agent (SMA) molecule model shows the hook-like structural features that mesh together with other SMA molecules to reduce slippage between SMA-treated proppant grains.

Flowback

During hydraulic fracturing treatments, proppant grains suspended in a viscous fluid are pumped into created fractures. When the pumping pressure is released, the proppant remains in the fractures, holding the fractures open and forming a conduit for fluid flow into the well bore.

If the proppant flows back into the well bore, the fracture channel width will decrease, reducing fracture treatment effectiveness. Also, produced proppant erodes production equipment, requiring costly repairs and frequent downtime. Sand fill can restrict production and make cleanout necessary, delaying time-to-first-sale.1

Variations in formation in situ stress and mechanical properties can lead to nonuniform fracture closure and open channels above the proppant pack.2 Fluid tends to flow preferentially along such channels, and high velocities may increase the tendency of proppant detachment and flowback in this region. The following factors influence proppant flowback:

• Flowback rate and fluid rheology
• Formation closure stress and rate
• Proppant size, distribution, and angularity
• Formation hardness
• Fracture height, width, and tortuosity
• Perforation size, density, and orientation
• Proppant binding forces.

Screens or slotted liners installed in the well bore can stop the proppant or formation material flowback; however, this equipment limits the production rate, particularly when the screens become plugged with scale, formation debris, fines, and corrosion byproducts.

Other flowback prevention techniques include forced closure, resin-coated proppant, on-the-fly resin-coating, and fibers. All of these methods are applied as tail-end treatments, and are done as cost compromises. However, the last portion of a frac treatment may go toward the fracture tip and not remain near the perforations. With tail-end treatments, every perforation must be covered with the control treatment.

Forced closure

In a "forced closure" technique, treating pressure is released as soon as pumping is complete. Under ideal conditions, the fracture closes while most of the proppant is still suspended across the pay zone, preventing the proppant from settling at the bottom of the fracture.

Improved stimulation often results from more-effective fracture-fluid removal. A disadvantage of using the forced-closure technique is the increased possibility of proppant back-production into the well bore.

True forced closure requires measurement of frac fluid efficiency. The frac fluid is "dosed" with sufficient breaker to lower viscosity so that the fluid/proppant slurry is fluidized and viscosity no longer enhances proppant recovery. Special surface equipment is required for forced closure, adding to treatment expense.

Operators may also observe large proppant volumes being produced with poor production results; therefore, many operators are reluctant to use the forced-closure method.

Resin-coated proppant

Precoated, curable proppants generally are treated with 2-4% phenolic resin, which hardens after exposure to temperatures above 140° F. Although these materials are easy to apply, they add significantly to material and chemical costs.

Further, some precoated proppants tend to react quickly at elevated temperatures so that they cure before the proppant grains become compacted. This results in a low-strength consolidated pack.

Some precoated proppants have demonstrated incompatibility with fracturing fluids.

On-the-fly resin coating

With on-the-fly resin coating, epoxy resin is applied directly to proppant during fracturing treatments. Because the resin is liquid, it flows to the contact points to form stronger bonds at much lower resin concentrations than curable, precoated proppant.

Cure time of liquid resin is adjustable so that curing occurs after fracture closure.

While on-the-fly resin coating systems are economical and perform well, they add complexity to fracturing treatments.

Fibers

Fibers can be added to fracturing fluid as part of the proppant slurry to bridge across constrictions and orifices in the proppant pack and help prevent proppant flowback. Mixing fibers with proppant, however, tends to reduce the conductivity of the proppant pack within the fracture.3

Generally, fibers break during pump shearing and broken fibers form a weak framework in the fracture. Fragments and fibers produced out with the fracturing fluid can cause significant production loss and present long-term problems with fluid filtrates.

SMA application

Fig. 2 [7,830 bytes] illustrates SMA properties. In the figure, both glass jars contain equal volumes of 20/40-mesh Ottawa frac sand and water. Before the photograph was taken, the jars were shaken, causing the water shown in the jar on the left to cloud as the frac-sand fines became suspended. A ruler held behind the cloudy water is barely visible through the water.

In the jar on the right, the ruler can be seen clearly because the SMA added to this jar caused fines to adhere to sand grains.

During fracture treatments, large volumes of fines are generated from crushing of proppant and formation cracking or sloughing. Tacking these fines in place enhances fracture conductivity.

When producing from a formation containing fines, the tacking helps prevent fines from migrating through the proppant pack and causing plugging of the pack porosity.

In the jar on the right, the SMA volume added was proportional to amounts used in field applications.

Although the two jars contained identical sand volumes at the beginning, sand in the right-hand jar shows a marked increase in volume. This increase results from a pore-volume increase that occurred because the tacky sand grains resist settling and cannot slide into a dense pack.

As shown in Fig. 3, the sand with SMA also resists erosion from fluid flow. The graph in Fig. 4 [20,754 bytes] illustrates that SMA-treated proppant settles more slowly, and to a less compact proppant bed than untreated proppant.

Fig. 3a [5,424 bytes] shows untreated sand in the upright portion of a 1/2-in. ID glass T-tube. Water is pumped through the horizontal portion of the tube, creating a shear stress across the loose sand which erodes away some sand.

In this test, the water was pumped at a rate proportionate to formation production of 10 b/d across the top of a proppant pack.

In Fig. 3b [5,555 bytes], the sand has been modified by applying SMA in a ratio simulating field use. The flow equivalent is 54 b/d passing through the horizontal section of the tube. Apparent streaks coming off the sand are turbulence.

The simulated proppant pack was not disturbed by the water flow.

Treatment procedure

Proppant is treated on-the-fly by adding the SMA, through standard liquid-additive metering pumps, directly to the frac blender tub along with the proppant. Normal concentration is 0.1-0.2 gal/sack of proppant. A recommended rate is 0.12 gal/sack.

No special handling is required and no changes to usual fracturing procedures are necessary. The SMA does not require shut-in time, and variations in metering rate can be tolerated.

Laboratory testing

Laboratory testing included a large-scale slot model, flow-cells, and determination of such parameters as proppant stability, proppant settling rate, conductivity, and relative permeability.

Large-scale slot

A slot model, 2 ft x 10 ft x 3/8 in., was the laboratory fixture used to test treated and untreated proppant to compare proppant stability in the fracture.

Untreated proppant became fluidized, and the top layers of the proppant pack moved toward the well bore area. Treated proppant beds were stable. High-velocity flow eroded only enough to create a channel to support fluid velocity, then erosion ceased.

Flow cells

A laboratory flow cell was constructed to allow water to flow above a slot filled with proppant, thereby testing the capability of the proppant to resist flow.

Resuspension flow tests of treated proppant showed that flow rate could be much greater than for untreated proppant without resulting in proppant flowback.

Testing showed that higher SMA concentrations allowed treated proppant to withstand high rates of production flow. However, a range of concentrations is recommended to effectively coat the proppant without causing mixing and proppant transport difficulties.

Proppant-settling rate

SMA-treated proppants settled more slowly and formed a less compact (10% less density) proppant pack. The hindrance to settling resulted in higher porosity and permeability in test columns. Implications of settling tests are that treated proppant will yield superior vertical proppant distribution (Fig. 4) within a fracture.

This property should promote more conductivity in the pack, requiring less post-frac cleanup and allow higher production rates and total recovery.

Conductivity flow testing

Standard API conductivity flow tests on proppant treated with SMA (Fig. 5 [21,151 bytes]) typically showed a 10-20% increase in conductivity below 3,000 psi closure stress, whereas fibers (and thermoplastic strips) tested reduced proppant pack conductivity.

Another unexpected test result was that fines and particulates resulting from proppant crushing were found to bind to uncrushed proppant grains. Under equal stress loads, fines from untreated proppant moved freely within the pack, causing conductivity loss.

The API conductivity study also demonstrates that SMA material provides improved conductivity to the proppant pack by enhancing frac-gel cleanup.

SMA rapidly coats onto the proppant, preventing the guar-based polymer from adsorbing onto the proppant surface. This surface coating significantly increases the cleanup efficiency of commonly used gel breakers (Fig. 6 [24,439 bytes]).

Relative permeability

After two identical fracturing-fluid slurries were prepared, one was treated with SMA. Two identical, parallel flow cells were filled with equal volumes of slurry, then the gel slurries were allowed to break, and the proppant settled to maximum pack density.

Flow rates were determined for both water and oil, using identical differential pressure in both cells. Testing showed (Fig. 7 [34,780 bytes]) that the treated pack was more permeable to both oil and water than the untreated pack.

During identical flow time used for cleanup, the treated pack allowed a much greater volume of cleanup fluid to pass. This result indicates that faster and higher production can be obtained from the well.

Field examples

Field results indicate that SMA used as a flowback control agent permits more aggressive flowback procedures following conventional fracturing treatments. Of the wells treated with SMA to date, all have demonstrated that proppant flowback has been reduced and conductivity has been increased. Wells properly treated with SMA have not produced back any significant volume of proppant.

A Michigan well was fracture-stimulated with a 70-quality nitrogen foam treatment. SMA, 0.12 gal/sack, was included with the 200 sacks of 20/40-mesh sand in the first stage, and the 400 sacks of 12/20-mesh sand in the second stage. Bottom hole temperature was 65° F.

The well was flowed back hard 1 hr after the second stage was finished. No sand was produced and the treatment was successful. Normally, the operator would have waited 1-2 days to allowed resin-coated proppant tail-in sand to set up.

A New Mexico well was stimulated with 25 lb/1,000 gal low-polymer gel at 40 bbl/min. SMA was included at 0.12 gal/sack; proppant volume was 2,850 sacks of 16/30 mesh sand. Well temperature was 100° F. No sand has been produced. The operator eliminated waiting time for resin-coated proppant tail-in.

References

1. Nguyen, P.D., Weaver, J.D., Parker, M.A., and King, D.G., "Thermoplastic film prevents proppant flowback," OGJ, Feb. 5, 1996, pp. 60-62.
2. Milton-Taylor, D., Stephenson, C., and Asgian, M.I., "Factor Affecting the Stability of Proppant in Propped Fractures, Results of a Laboratory Study," SPE paper No. 24821, SPE Technical Conference and Exhibition, Washington, D.C., Oct 4-7, 1992.
3. Howard, P.R., King, M.T., Morris, M., Feraud, J.P., Slusher, G., and Lipari, S.A., "Fiber/Proppant Mixtures Control Proppant Flowback in South Texas," Paper No. SPE 30495, Annual Technical Conference and Exhibition, Dallas, Oct. 22-25, 1995.

The Authors