NEW FOAMS INTRODUCE NEW VARIABLES TO FRACTURING

July 16, 1990
David J. Mack Western Co. Oklahoma City Larry J. Harrington Western Co. The Woodlands, Tex . Two recent developments in foam fracturing technology are delayed crosslinked carbon dioxide (CO2) and the combination of nitrogen (N2) with CO2. The characteristics of these products are different than conventional CO2 or N2 foams. To obtain good fracs with these newer foams, special considerations must be made in the design and setup of the job.
David J. Mack
Western Co.
Oklahoma City
Larry J. Harrington
Western Co.
The Woodlands, Tex
.

Two recent developments in foam fracturing technology are delayed crosslinked carbon dioxide (CO2) and the combination of nitrogen (N2) with CO2.

The characteristics of these products are different than conventional CO2 or N2 foams. To obtain good fracs with these newer foams, special considerations must be made in the design and setup of the job.

Rheological properties, flowback procedures, pumping schedules, and friction effects are some of the variables that are different from conventional CO2 and N2 foams.

THE NEED FOR FOAMS

Foams, as hydraulic fracturing systems, have been used for over 20 years. The first reported usage was in 1968.1 Their usage continued to increase throughout the 1970s2 3 4 and was highlighted by the issuing of a patent5 to Blauer and Durborow.

The initial reasons for using a foamed fracturing fluid were two fold. First, the N2 or CO2 would provide necessary energy to clean up treating fluids in underpressured (less than hydrostatic gradient) reservoirs.67 This was done by opening the wellhead to reduce the back pressure.

Because the gas was now at a lower pressure, it would expand.

The path of least flow resistance was into the well bore and out the wellhead. Liquid would be carried along with the expanding gas, thus assisting the natural reservoir pressure in cleaning up the treating fluid.

The second major benefit was in formation compatibility. Formations which are high in clay content generally do not respond well to water based treating fluids.

By removing a substantial portion of the water from the system and replacing it with a compatible gas, damage to natural permeability is reduced.

A third major benefit of a foamed system has appeared recently. Disposal of treating water has become very expensive and environmentally sensitive. With a majority of the treating fluid volume being gaseous, water disposal costs are reduced and the negative impact on the environment is reduced.

FOAM PROPERTIES

By definition, a foam is a gas in liquid dispersion stabilized by the inclusion of a surfactant in the base liquid. The liquid can be water, acid, hydrocarbon (e.g., diesel), or a mixture of water/hydrocarbon (e.g., methanol).

Acid and acid/hydrocarbon foams are typically used in treatments of carbonates and generally do not include proppants. Well bore cleanup is accomplished using a water based foam, and hydraulic fracturing with proppant laden fluid will use water, hydrocarbon, or the water/hydrocarbon mixture.

Generally, water is used as the base liquid for hydraulic fracturing.

The gas, or internal phase, can be any gas available. Although air is probably the most readily available gas, it is not used due to its volatility. The same is true for natural gas. N2 and CO2 are the two common internal phases used. Their availability, compatibility, and cost make them the most logical choices.

Regardless of the internal phase of the system, there are several properties which are always consistent. The first is the compressibility of the overall system. This is because 53-96% of the system is a gas (or a liquid in the case of CO2 foams below 87.8 F.) which is highly compressible.2 8 Typically, gas percentages, commonly called "quality," range from 60 to 85.9

When discussing foams, there are two types of qualities discussed. The first is "Mitchell quality." It is defined (see nomenclature) as the ratio of the volume of gas to the volume of fluid (foam).

VG

Fm = ________ (1)

VG + VL

The second type of quality is "slurry quality" or sometimes called total internal phase.

This is defined as the ratio of the volume of gas plus the volume of proppant to the total foam slurry.

VG + VP

z Fs = ____________ (2)

VG + VP + VL

The effects of this compressibility are many and are controlled by changes in temperature and pressure. System volume and density are directly affected, and flow rate and friction pressure are consequently affected.

Whether the gas is N2, CO2, or any other, its behavior is controlled by the real gas law:

 PV = znRT (3)

The technique used to determine volumetric changes in the gas phase as temperature and pressure change is outlined in Reference 10.

Although the technique converts reservoir volume to surface volume, the relationships are valid for any set of conditions. This can be accomplished by changing the subscripts from R and SC to 1 and 2, respectively. The volumetric relationship now becomes:

P1 V1 Z2 T2

V2 = ___________ (4)

Z1 T1 P2

The method used to determine volumetric relationship of nitrogen to water is:

PR F

B = 199.3 _____ ___ (5)

TR ZR 1 - F

CO2 volumetric relationships are derived using pp. 46 of Reference 10 and p. 2 of Reference 12.

The z-factor used in volumetric relationships can be obtained from several sources. A presentation for both N2 and CO2 can be found in Amyx, Bass, and Whiting13 and for CO2 only as Fig. 4 of Reference 10.

A mathematical determination for nitrogen z-factor is shown on pp. 17-18 of Reference 11.

The most current and widely accepted mathematical determination for both N2 and CO2 are presented by Hendricks, et al.14

Also common to foam fracs is proppant addition rate.

Since the gas portion of the system must be injected via a closed system, proppant can be added to the liquid portion only. This requires the conversion from a desired downhole proppant concentration (based on a gallon of downhole foam) to a surface blender concentration (based on a gallon of liquid at the blender). Fig. 1 is based on the equation:11

CDH

CB = ______ (6)

(1-FM)

Liquid injection rate of a foam is affected by changes in the proppant concentration in the same manner as a conventional field. However, only a portion of the carrier fluid is liquid. Taking into account foam quality:

QL =

QT

________________________ (7)

1 + DCDH(1 + FM) (FM)

____ + ____

1-FM 1-FM

The liquid/proppant slurry rate out of the blender will be:

Qs =

Fm QL

QL + DCDH (QL + _____)(8)

1-FM

The required proppant addition rate in lb/min is:

42 CDHQT

QP = ________ (9)

1 + DCDH

Values for absolute volumes, D, for various proppants are obtained from Table 1.

Also common to all foams is the requirement for a more sophisticated postfrac cleanup technique. Due to the added energy of the system, it is more difficult to flowback a well without bringing proppant along with the load fluid. Also with foams, it is important to flowback at a sufficiently high rate to maintain a good gas/liquid dispersion to maximize liquid removal.15 16

Table 2 is a guide for choke size selection.15 This table is meant as a guide only and should be modified based on field experience in the area.

N2 VS. CO2

Although many of the job design and application techniques of N2 and CO2 foams are common, there are many physical and chemical properties which are unique to each.

Both N2 and CO2 are brought to location as cold liquids; however, N2 is heated to the range of 80-110 F. and injected into the treating line as a gas. CO2, on the other hand, is not heated, and it is injected as a liquid.

Thus, each has unique pumping equipment requirements.

N2 requires a specially designed cryogenic liquid pump and a heat exchanger. CO2 allows for the utilization of conventional triplex pumps with cryogenic components.

CO2 does, however, require a prepump or "booster" pump to condense any free gas and to supercharge the triplex pump.

Heating Of CO2 prior to the foam entering the formation is accomplished from four areas, as discussed on pp. 710 of Reference 10.

Cleanup or load recovery tendencies of N2 foams tend to be more rapid, but not as complete as with a CO2 foam. This is due to the low solubility of N2 in water.

The N2 gas tends to dissipate from the system, causing a quick "blow," thus reducing recovery efficiency as the nitrogen leaves the water behind.

CO2 is appreciably more soluble than N2. Its cleanup tendencies are generally slower, but more complete as it creates a solution-gas drive effect. Fig. 210 shows the solubility of CO2 and nitrogen in water.

This solubility, along with CO2 chemical reactivity, creates another area of variance. Fig. 310 shows the effect of pH change as a function of CO2 solubility. Carbonic acid is created. Although N2 is slightly soluble in water, no pH change takes place as it is chemically inert in its N2 foam.

N2'S inertness and heated pumping temperature provide some advantages over CO2 foams. The inert nature allows N2 to be pumped with any fluid and into any formation. CO2 foam should not be pumped into zones which contain asphaltenitic crudes.

Increased system temperature of N2 foams reduces the thermal shock or tubulars and reduces paraffin deposition probability in low pour point crudes.

Table 3 presents a comparison of CO2 and N2 with regards to physical and chemical properties.

RHEOLOGY

Initial investigations into the Theological properties of foams were performed using N2 foams.2 8 In these studies, foam was described as having Bingham plastic properties, yield point, and plastic viscosity.

The effective viscosity of a foam is a function of both foam quality and shear rate. For slot flow, flow through a fracture, the effective viscosity is:

1182 ty W2 H

me = mp + ____________ (10)

Q

Values for mp and ty are found on p. 30 of Reference 11. Western's FOAMFRAC design program is based on this relationship.

Further studies have revealed that a more accurate description of a foam would be to call it a yield pseudo plastic.17 18 The studies showed that the power-law properties of n' (flow index), and K' (consistency index), could be calculated and were a function of foam quality, temperature, and base fluid (water, gel, diesel, etc.) properties.

The equations used to calculate n' and K' are derived in Reference 18. They are as follows:

n't =

n'75e(0.0028-0.0019FM)(TF-75) (11)

C1 = 4(n'75)1.8 (12)

C2 = e-(3.1 + 3n'75) (13)

K't =

K'75 e(c2FM - 0.018)(TF - 75) (14)

K'f = K't e(c1FM + 0.75 FM2) (15)

Because of equipment limitations, the testing was performed using N2 only as the internal phase. The authors18 do state that they did observe a variation in the yield point and c1 values. They also suggest that only a small refinement of the equation would be necessary to "unify the rheology equations for both N2 and CO2 foams."18

In 1985 and 1986, Western R&D constructed a foam flow loop and performed a series of tests. The first of these tests was with N2 foam.

The purpose was to observe data correlation with the previous studies.17 18 Within experimental error, the Harris, et al.18 equations were verified. However, when CO2 foams were tested on the flow loop, calculated values showed poor correlation to flow loop values. As a result, two documents19 20 of CO2 rheologies were published.

Comparisons of calculated CO2 foam rheologies to Newsletter20 rheologies are presented in Figs. 4-6. The base gel in each case is a 50 lb/1,000 gal HPG (hydroxy propyl guar). Foam quality is held at 70. It is obvious that there is considerable difference between the two techniques.

The higher equation viscosities would result in wider calculated fracture widths and a possible false sense of security regarding potential treatment screenouts. Table 4 shows two WIDTH computer program outputs that demonstrate the width variation. Rheology at 125 F. was used.

FLUID LOSS

Foam fracturing fluids were first generated using only water (no gel) as the base liquid. Based on these fluids, leakoff control was considered to be influenced by the fluid loss coefficients Cv and Cr.3 The leakoff viscosity used to calculate Cv was considered to be the same viscosity as that of the foam in the fracture.

At conditions of low shear rate, wide and tall fractures at low injection rates, leakoff would be viscosity controlled. Cv values smaller than 0.0001 were common. This would result in very high apparent fluid efficiencies, yet treatment screenouts were common.21 22 This technique is used in the FOAMFRAC program.

Subsequent studies23-25 have shown that leakoff is not Cv but Cw controlled. In general, it was found that Cw was a function of gel loading, temperature, formation permeability and, to a small degree, foam quality.

Values for Cw have been published;23-25 however, there are some variations. "Rule-of-thumb" values have proven to be adequate. In general:

  • For low permeability situations (k < 0.1 md) use Cw = 0.0015.

  • Moderate permeabilities (0.5 k 0.1) have a Cw = 0.002

  • In high permeability situations (2.0 k 0.5), Cw = 0.003

  • For a very high permeability (k 2.0), a Cw = 0.005.

Spurt loss is generally considered as zero; however, in a very high permeability situation some small value, 2 or 3 cc, should be assigned.

TREATMENT DESIGN

There have been many papers written with regards to hydraulic fracturing treatments using foams. A paper by King26 and a series of articles by Holcomb27 provide references. These referenced papers provide additional information regarding proppant concentrations, crosslinked based fluid, and additional fluid loss additives.

In earlier sections of this article, references were made to the FOAMFRAC design program. It was based on the then "state-of-the-art" foam description-Bingham plastic with Cv and Cr leakoff control. As has been pointed out earlier, the industry's thinking has changed.

To show the differences in treatment design results, computer runs were made using FOAMFRAC and our power-law model PROP. Job parameters, similar to treatments performed in the Appalachian basin, were used. The fluid used is a 70 quality nitrogen foam.

Table 5 is the FOAMFRAC run for this treatment. The admittance criteria for sand is determined using:

W

Cmax = ____ (16)

0.18

For this run, Cmax = 1.672 psf. Based on this, our treatment design is safe and acceptable.

Now looking at Table 6, the PROP program, a different conclusion is reached. Here Cmax = 1.561 psf. Both the 2 lb and the 3 lb stages show screenout. This is precisely the situation we saw in Ohio.

Foam fracs designed with a Bingham plastic, Cv and Cr model, were screening out. When jobs were designed using a power-law, Cw model, the success ratio dramatically increased.

Based on field results of treatments designed with both fracturing models, hydraulic fracturing treatments with foam fluids tend to be more successful when designed with Cw controlled and using current foam theory and power-law.

FRICTION PRESSURE, DENSITY

As pointed out by King,26 friction pressure of foamed fluids varies with foam quality. Even with tight constraints on gas/liquid ratios, foam quality will vary due to pressure and temperature changes as it travels down the well tubulars.

To obtain a more accurate prediction of friction pressure, foam quality must be recalculated at numerous locations throughout the treating string. Graphs such as those on pp. 32-41 of Reference 11, can then be used to determine the friction pressure at subsequent pipe locations. Such a technique is used in our FOAMP program. In general, N2 foam has an appreciably lower friction pressure gradient than CO2 foam.

It has also been pointed out by Mack, et al.,28 and Baumgartner, et al.,29 that there are substantial variations in foam density throughout the tubulars. These changes must be taken into account when estimating surface treating pressure (ISDP) or calculating bottom hole treating pressure for real-time net pressure analysis.

Mack and Baumgartner28 29 further stated that proppant addition affects both fluid density and friction pressure. The effects on density are the same for any internal phase foam. However, friction pressure effects are different for N2 foams and CO2 foams.

The friction pressure effects on N2 foams were found to follow the relationships for conventional fluids as described by Hannah, et al.;30 CO2 foams, on the other hand, do not. Here, proppant tends to become part of the internal phase of the foam. Changes in friction more closely match the changes in foam quality.

Western's TMV (treatment monitoring vehicle) programs are designed to provide accurate descriptions of bottom hole treating pressure using these relationships.

NEW TECHNOLOGY

Up to this point, the discussion has centered around N2 and CO2 foams. These two systems were the only ones available until late in 1988 when Western developed two new high internal phase fluids. These fluids, Binary Westfoam and CO2 blend, provide the industry with additional alternatives.

BINARY FOAM

In binary foam, two gases, N2 and CO2, are used. This combines the advantages of CO2 (low pH, solution gas drive, and increased hydrostatic head) with the advantages of N2 (rapid early cleanup and reduced friction pressure) into one system.

The binary system does not require any new or additional pumping equipment or chemicals. Standard CO2 pumps and N2 pumps are tied into the treating line separately and in the same manner as if each type were the only one on the line.

Foaming agents and surfactants are the same and used in the same quantities as though it were a single phase CO2 foam.

Customers,31 have said that the binary foam appears to clean up more rapidly and more completely than N2 or CO2 foams alone. Also observed were that initial production rates have been at equal to or higher levels. Several operators32 have reported that due to the reduced volume of CO2 in the system, they are able to return the gas to the sales line several days sooner, as compared to a straight CO2 foam.

Surface treating pressures on tubing fracs have been generally lower with the binary system as compared to CO2 foam. An analysis of binary treatments has shown that the general range of friction pressure for binary foams, down tubing, is 45-65% of the friction of fresh water.

A value of 55% is used for job design as compared to 68% for CO2 foam.

Fig. 7 shows the surface treating pressure relationship for binary and CO2 foams, pumped down 10,000 ft of 2 7/8-in. tubing. Casing fracs using the binary foam have exhibited friction of approximately 150% that of freshwater. However, here there are only a few jobs and the data scatter is very large, 77-284%.

Proppant friction effects appear to be similar to straight CO2 foam, but a definitive relationship has not yet been developed.

From an operator's and a logistics viewpoint, a binary job is more difficult than a straight N2 or CO2 foam job. Compared to a single gas phase foam, the binary system requires the monitoring and recording of an additional fluid.

To simplify job monitoring, new software has been developed for Western's existing frac monitors. This software will track all four flow rates (slurry, N2, CO2, and proppant), calculate Mitchell and slurry qualities at surface and downhole, as well as provide all the standard totaling and charting capabilities.

Treating design can be accomplished using rheology developed specifically for binary foams.34 This shows the binary system to behave as a power-law fluid. The PROP design program should be used.

Development of a pumping schedule is just as complex as the actual pumping of a treatment. N2, CO2, proppant, and water all have different, and sometimes variable, densities, compressibilities, and heat capacities. N2 and CO2 are also soluble in water.

To create the required pumping schedule, a new design program, BINARY, was written. BINARY will build a series of rate and volume tables, based on surface conditions, to provide the treater and engineer sufficient information to set up and execute a binary foam frac.

CO2 BLEND

The second new system is called CO2 blend. It is a high quality, greater than 53% system, utilizing a delayed crosslink to stabilize the CO2 in water dispersion.

Stabilization is accomplished by the same mechanism that gives most crosslinked gels the perfect proppant transport characteristic. After the CO2 has been dispersed into tiny droplets by the shear conditions, the delayed crosslink will form a continuous "net" around each droplet, keeping the droplets in place with a continuous external crosslinked water phase.

To date, no rheology has been developed for CO2 blends.

Values for CO2 foams from References 19 and 20 are currently used. Some isolated tests have been performed showing the CO2 blend to have a much longer half-life than a similar gel loading CO2 foam.

The friction pressures of CO2 blend have shown the same characteristics as binary foam (55% of freshwater for tubing and 150% for casing). As was the case for binary foams, friction effects of proppant on CO2 blend treatments appear to be similar to straight CO2 foam.

ACKNOWLEDGMENT

We would like to thank: Bill Strickland and Jim Wilke for their assistance in the development of the BINARY Program; Jim Wilke for preparing the computer graphs; and Western Co. for permission to publish this article.

REFERENCES

  1. Grundman, S.R., and Lord, D.L., "Foam Stimulation," SPE 9754, Production Operations symposium, Oklahoma City, Mar. 1-3, 1981.

  2. Blauer, R.E., Mitchell, B.J., and Kohlhass, C.A., "Determination of Laminar, Turbulent and Transitional Foam Flow Losses in Pipes," SPE 4885, California Regional Meeting, San Francisco, Apr. 4-5, 1974.

  3. Blauer, R.E., and Kohlhass, C.A., "Formation Fracturing with Foam," SPE 5003, 49th Annual Technical Conference, Houston, Oct. 6-9, 1974.

  4. Blauer, R.E., and Holcomb, D.L., "Foam Fracturing - Application and History," Southwest Petroleum Short Course, Lubbock, April 1975.

  5. Blauer, R.E., and Durborow, C.J., "Formation Fracturing with Stable Foam," U.S. Patent No. 3,937,283, Feb. 10, 1976.

  6. Bullen, R.S., "Combination Foam/Fluid Fracturing," J. Con. Pet. Tech., July-September 1980, pp. 51-56.

  7. Ainley, B.R., and Charles, J.G., "Fracturing Using Stabilized Foam Pad," SPE/DOE 10825, Symposium on Unconventional Gas Recovery, Pittsburgh, May 16-18, 1982.

  8. Mitchell, B.J., "Viscosity of Foam," PhD dissertation, University of Oklahoma, 1969.

  9. Strang, D.L., and Morton, J.L., "Foamed Sand Provides Improved Stimulation Results from Devonian Shale," SPE 2312, Eastern Regional Meeting, Champion , Pa., Nov. 9 11, 1983.

  10. Mack, D.J., and Harrington, L.J., "CO2 West Foam Design," Completion Engineering Newsletter, Western Petroleum Services, Fort Worth, October 1982, pp. 4-5.

  11. Mack, David, West-Foam Design Manual, Western Petroleum Services, Fort Worth, January 1982, p. 12.

  12. Mack, David, "CO2 West-Foam," Completion Engineering Newsletter, Western Petroleum Services, Fort Worth, May 1984, p. 2.

  13. Amyx, J.W., Bass, D.M., Jr., and Whiting, R.L., Petroleum Reservoir Engineering, McGraw-Hill Book Co., New York, 1960.

  14. Hendricks, R.C., Baron, A.K., and Peller, I.C., "GASP: A Computer Code for Calculating the Thermodynamic and Transport Properties for Ten Fluids: Parahydrogen Helium, Neon Methane, Nitrogen, Carbon Monoxide, Oxygen, Fluorine, Argon and Carbon Dioxide," National Aeronautics and Space Administration, Lewis Research Center, Cleveland, February 1975.

  15. Mach, David, "West-Foam/Energized System Flowback and Other Operational Considerations," Completion Engineering Newsletter, Western Petroleum Services, Ft. Worth, March 1983.

  16. Mack, D., Mooney, M., and Terry, C., "Correct Procedures Improve Fracturing Treatments," Petroleum Engineer International, November 1983, pp. 64-74.

  17. Reidenback, V.G., Harris, P.C., Lee, Y.N., and Lord, D.L., "Rheological Study of Foam Fracturing Fluids Using Nitrogen and Carbon Dioxide," SPE 12026, 58th Annual Technical Conference, San Francisco, Oct. 5-8, 1983.

  18. Harris, P.C., and Reidenback, V.G., "High Temperature Rheological Study of Foam Fracturing Fluids," SPE 13177, 59th Annual Technical Conference, Houston, Sept. 16-19, 1984.

  19. Couchman, D.D., and Phillips, A.M., "High Temperature Rheology of CO2 Foam Fracturing Fluids," Laboratory Services Newsletter, Western Petroleum Services, Ft. Worth, March 1986.

  20. Couchman, D.D., and Phillips, A.M., "High Temperature Rheology of CO2 Foam Fracturing Fluids-Second Edition," Technical Services Newsletter, Western Petroleum Services, Ft. Worth, December, 1986.

  21. Wendorf, C.L., and Ainsley, B.R., "Massive Hydraulic Fracturing of High Temperature Wells with Stable Foam Fracs," SPE 10257, 56th Annual Technical Conference, San Antonio, Oct. 5-7, 1981.

  22. Goelitz, R., and Evertz, G.E., "Foam Fracturing in the Uinta Basin-A Field Study of the Dakota and Wasatch Formations," SPE 10883, Rocky Mountain Regional Meeting, Billings, Mont., May 1921, 1982.

  23. King, G.E., "Factors Effecting Dynamic Fluid Leakoff of Foam Fracturing Fluids," SPE 6817, 52nd Annual Technical Conference, Denver, Oct. 9-12, 1977.

  24. Harris, P.C., "Dynamic Fluid Loss Characteristics of Foam Fracturing Fluids," SPE 11065, 57th Annual Technical Conference, New Orleans, Sept. 26-29, 1982.

  25. Harris, P.C., "Dynamic Fluid Loss Characteristics of CO2 Foam Fracturing Fluids," SPE 13180, 59th Annual Technical Conference, Houston, Sept. 16-19, 1984.

  26. King, G.E., "Foam and Nitrified Fluid Treatments Stimulation Techniques and More," SPE 14477.

  27. Holcomb, D.L., "Foam for Fracturing and Acidizing Stimulation Parts 1, 2, and 3," Drilling, January, February, and May 1982.

  28. Mach, D.J., and Baumgartner, S.A., "Friction Pressure of Foamed Stimulation Fluids Evaluated With an On-Site Computer," SPE 15631, 61st Annual Technical Conference, New Orleans, Oct. 5-8, 1986.

  29. Baumgartner, S.A., and Mack, D.J., "On-Site Monitoring of Foamed Stimulation Fluids," SPE 17531, SPE Rocky Mountain Regional Meeting, Casper, Wyo., May 11-13, 1988.

  30. Hannah, R.R., Harrington, L.J., and Lance, L.C., "The Real Time Calculation of Accurate Bottomhole Fracturing Pressure from Surface Measurements Using Measured Pressures as a Base," SPE 12062, 58th Annual Technical Conference, San Francisco, Oct. 5-8, 1983.

  31. Personal conversations with Ted Jacobson of Kaiser Francis, and Brian Dennis of Amoco Production.

  32. Personal conversations with Fred Toney of Western Co.'s Woodward district.

  33. For additional information regarding monitor conversion to Binary, contact Ronnie Lindley at network 423-7019.

  34. "High Temperature Rheology of CO2/N2 Foam Fracturing Fluids," Technical Newsletter-Western Co., Ft. Worth, June 1, 1989.

Copyright 1990 Oil & Gas Journal. All Rights Reserved.