Foam Applications Manual

Foam Applications Manual

Halliburton® Energy Services

Notices All information contained in this publication is confidential and proprietary property of Halliburton Energy Services, a division of Halliburton Company. Any reproduction or use of these instructions, drawings, or photographs without the express written permission of an officer of Halliburton Energy Services is forbidden. ©Copyright 1992, Halliburton Company All Rights Reserved. Printed in the United States of America. Printing History: First Release 1992 Second Release Reprinted

TOC Foam Applications Manual Table of Contents

Section 1: Introduction Foamed Acid ...........................................................................................1-3 Hydraulic Fracturing Stimulation.............................................................1-3 Foam Cement ......................................................................................... 1-3 Other Applications .................................................................................. 1-3 Section 2: Nitrogen Properties Introduction .............................................................................................2-3 Physical Properties .................................................................................2-4 Nitrogen Expansion ................................................................................2-4 Nitrogen in Foam ....................................................................................2-4 Section 3: Nitrogen Safety Physical Properties .................................................................................3-3 Nitrogen in the Air ...................................................................................3-3 Cryogenic Thermometer .........................................................................3-3 Safety Precautions for Handling Liquid Nitrogen .................................... 3-4 First Aid Procedures for Cold Liquid Frostbite (Freeze Burns)...............3-4 Liquid Air Hazard ....................................................................................3-5 Oxygen Deficiency Hazard ..................................................................... 3-5 Liquid Nitrogen Equipment Safety ..........................................................3-5 Section 4: Foam Applications in Acidizing Acidizing with Foam ................................................................................4-3 Advantages of Foamed Acid ...................................................................4-3 Foamed Acid Penetration .......................................................................4-4 Foam Stability ......................................................................................... 4-6 Table of Contents

TOC-1

Foam Diversion .......................................................................................4-6 Fracture Acidizing ................................................................................. 4-10 References ........................................................................................... 4-15 Other References ................................................................................. 4-15 Section 5: Foam Applications in Hydraulic Fracturing Introduction ............................................................................................. 5-3 Types of Foams Used in Hydraulic Fracturing ....................................... 5-4 Foam Rheology.......................................................................................5-4 Crosslinked Foams .................................................................................5-5 Foam Fluid Loss .....................................................................................5-7 Fracture Conductivity ............................................................................ 5-12 Treating Pressure Response ................................................................ 5-14 Fluid Recovery ...................................................................................... 5-22 Treatment Designs for Hydraulic Fracturing ......................................... 5-22 Minifractures ......................................................................................... 5-31 Conclusions .......................................................................................... 5-34 References ........................................................................................... 5-34 Additional References .......................................................................... 5-35 Section 6: Foam Cementing Introduction ............................................................................................. 6-3 Foam Generation ....................................................................................6-4 Downhole Behavior .................................................................................6-6 Cement and Additives .............................................................................6-9 Job Considerations ............................................................................... 6-10 Design Considerations ......................................................................... 6-11 Evaluating Foam Cementing Results.................................................... 6-13 Section 7: Other Nitrogen Applications Sand Washing ........................................................................................7-3 Unloading Wells .....................................................................................7-7 Gas Displacement ................................................................................ 7-10 Pressurizing Medium ............................................................................ 7-11 Commingled Gas .................................................................................. 7-12 Sand Consolidation............................................................................... 7-13 Leak Detection Service ......................................................................... 7-17 References ........................................................................................... 7-19

TOC-2

Table of Contents

Section 1 Introduction Contents Foamed Acid ......................................................................................... Hydraulic Fracturing Stimulation .......................................................... Foam Cement ....................................................................................... Other Applications ................................................................................

Introduction

1-3 1-3 1-3 1-3

1-1

1-2

Introduction

Introduction Nitrogen has been used in the well service industry for more than 30 years. Nitrogen is an inert gas that allows pressure to be applied downhole without causing damage to sensitive formation surfaces. Nitrogen can be used in well testing and wellbore cleanout or in creating foam fluids to stimulate oil and gas production. This manual discusses the physical properties of liquid and gaseous nitrogen, important safety considerations for personnel and equipment, and some of the more popular applications for nitrogen foaming.

Foamed Acid Foamed acid is a finely dispersed mixture of nitrogen gas bubbles within hydrochloric acid liquid. Foaming the acid increases the volume of the active acid and improves penetration. Foam also helps divert fluid from high permeability zones into lower permeability zones. Expansion of the gas after treating helps remove created fines and lessens damage to the conductive fracture.

Hydraulic Fracturing Stimulation Nitrogen is widely used in hydraulic fracturing stimulation. High quality foams produce high viscosity for proppant transport. The foam also has a low liquid content to protect formations that are sensitive to fluids. In addition, Foam helps to control

Introduction

fluid loss, maximize fracture conductivity, and provide gas expansion to assist flowback. Foam fracturing fluids are especially beneficial for under-pressured or depleted reservoirs, but have been used in high-pressure and high-temperature reservoirs as well. Foam can also be used in minifrac analysis, aiding in fracture design.

Foam Cement Nitrogen in foam cement provides a means of producing very lightweight cement. Foamed cementing slurries in the range of 4 to 15 lb/gal can develop relatively high compressive strength in a minimum period of time. Due to the inert character of nitrogen, Halliburton's conventional additives can be used in foam cement.

Other Applications Nitrogen is frequently used in other applications such as sand washing, well unloading, drillstem testing (as a pressurizing medium), sand consolidation, and leak detection. Halliburton has a variety of fluids, additives, and engineering computer programs to properly design nitrogen-assisted service for your well.

1-3

1-4

Introduction

Section 2 Nitrogen Properties Contents Introduction .............................................................................................2-3 Physical Properties .................................................................................2-4 Nitrogen Expansion ................................................................................2-4 Nitrogen in Foam ....................................................................................2-4

Nitrogen Properties

2-1

2-2

Nitrogen Properties

Nitrogen Properties Introduction Nitrogen (N2) was first introduced to well servicing in 1956 when it was used as a gas cushion to control well flowing pressure during drillstem tests. Although quantities and pressures were limited, this service did allow operators to control well liquids and pressures by using an inert gas. In 1959, cryogenic transports and pumps were introduced for use with liquid N2. This allowed great volumes of liquid N2 to be converted to gas and placed in the well system under any combination of pressure and rate that the job might require. Liquid N2 is readily available at several industrial complexes. As a manufactured byproduct of industrial gases, it is usually

100°C

212°F

20°C

70°F

0°C

32°F

created during the air separation process used to obtain liquid oxygen. Because N2 is an inert gas, it cannot react with hydrocarbons to form a combustible mixture. In addition, N2 is only slightly soluble in water and other aqueous liquids, which allows it to remain in bubble form when commingled with these fluids. Nitrogen is a nontoxic, colorless, and odorless gas naturally found in the atmosphere (78% of air is N2). Nitrogen is brought to the work site in liquid form in cryogenic bottles at temperatures below -320°F (Fig. 21). The nitrogen is then pumped through a triple-stage cryogenic pump at a desired rate and forced into an expansion chamber. The expansion chamber allows the N2 to absorb sufficient heat from the environment to vaporize into dry gas. The N 2 gas is then

Cryogenic range -78.4°C

-109.3°F CO2 sublimes (dry ice)

-183.0°C -195.8°C

-297.3°F Liquid oxygen -320.4°F Liquid nitrogen

-273.16°C

-459.7°F Absolute zero

Fig. 2-1: Cryogenic thermometer showing relative coldness of liquid nitrogen.

Nitrogen Properties

2-3

displaced by positive displacement pumps out of the expansion chamber and down the service piping to perform the prescribed job.

expands 696 times its volume in going from a liquid at -320°F to a gas at 70°F, as shown in Fig. 2-2.

Physical Properties

Nitrogen in Foam

Table 2-1 lists some of the physical properties of N2 at atmospheric pressure (14.7 lb/ in.2).

Nitrogen is most often used as the gas phase of foams. Because foam has low fluid loss, low density, low liquid content, and high viscosity, it can be used when stimulating, drilling, and cleaning low-pressure and water-sensitive formations effectively. Foam quality is the ratio of gas volume to foam volume at a given pressure and temperature. Usually, the pressure and temperature are the same as bottomhole treating or circulating conditions. To determine foam quality (Qf), use the equation below:

Nitrogen Expansion Nitrogen expands greatly as it absorbs heat from the environment. Nitrogen Table 2-1: Physical Properties of Nitrogen Boiling point

-320.36°F

Liquid weight density

6.745 lb/gal

Gas weight density

0.0724 lb/scf

Heat required to convert liquid to 70°F gas

184 btu/lb

Expansion ratio of liquid to gas

1 to 696*

Solubility in water

2.35 parts nitrogen in 100 parts water at 32°F 1.55 parts nitrogen in 100 parts water at 68°F

* One gallon of liquid nitrogen at -320°F expands to 93.11 scf gas at 70°F

Heat

Liquid Nitrogen

Gaseous Nitrogen

Qf =

N 2 volume , ... (2-1) liquid volume + N 2 volume

In the 0 to 52-quality range, gas bubbles in the foam are spherical and do not contact each other. Foam in this quality range has rheology similar to the liquid phase. In the 52 to 96-quality range, gas bubbles in the foam interfere with one another and deform during flow. This causes foam viscosity and yield point to increase as quality increases. In this particular range, foam behaves like a Bingham plastic fluid, where yield stress must be overcome to initiate fluid movement. Above 96 quality, foams may degenerate into a mist. The thin liquid layer cannot contain the larger volume of gas, causing the foam bubbles to rupture. The liquid phase of foam can be either water-, methanol-, or hydrocarbon-based. Usually less than 1% foaming agent by volume is added to generate the foam.

Fig. 2-2: Liquid N2 expands to 696 times its liquid volume when heated to 70°F.

2-4

Nitrogen Properties

Section 3 Nitrogen Safety Contents Physical Properties .................................................................................3-3 Nitrogen in the Air ...................................................................................3-3 Cryogenic Thermometer .........................................................................3-3 Safety Precautions for Handling Liquid Nitrogen .................................... 3-4 Wear protective clothing .............................................................................. 3-4 Avoid skin contact ........................................................................................ 3-4

First Aid Procedures for Cold Liquid Frostbite (Freeze Burns)...............3-4 Symptoms .................................................................................................... 3-4 What to Do ................................................................................................... 3-4 What Not to Do ............................................................................................ 3-4

Liquid Air Hazard ....................................................................................3-5 Oxygen Deficiency Hazard ..................................................................... 3-5 Liquid Nitrogen Equipment Safety ..........................................................3-5 Cryogenic Materials and Components ........................................................ 3-5 Cryogenic Materials ............................................................................... 3-5 Cryogenic Components ......................................................................... 3-5 Noncryogenic Material and components ..................................................... 3-5 Noncryogenic Material ...........................................................................3-6 Noncryogenic Components.................................................................... 3-6 Equipment Precautions ................................................................................ 3-6 Pressure Buildup.......................................................................................... 3-6

Nitrogen Safety

3-1

3-2

Nitrogen Safety

Nitrogen Safety Physical Properties

Nitrogen in the Air

Below are some of the important properties of nitrogen (N2) at atmospheric pressure (14.7 lb/in.²). The importance of each property is explained in the following sections.

Air contains 78% nitrogen, confirming that nitrogen gas is colorless and odorless and is not toxic or irritating. Nitrogen gas neither burns nor supports combustion, does not support life functions, and is a poor conductor of heat, preventing cold liquid N2 from instantly collapsing hot pressure building gas. Oxygen is the component of air that supports combustion and life functions.

Boiling point Liquid density Heat required to convert liquid to 70°F gas Expansion ratio of liquid to gas

-320°F 6.745 lb/gal 184 btu/lb 1 to 696*

*One gallon of liquid nitrogen at -320°F expands to 93.11 scf gas at 70°F

Cryogenic Thermometer The cryogenic thermometer below shows the extraordinarily cold nature of liquid nitrogen. Note: Water boils at 212°F, and liquid nitrogen boils at -320°F.

Components of Air Nitrogen- 78 %

100°C

212°F

20°C

70°F

0°C

32°F

Cryogenic range -78.4°C -183.0°C -195.8°C -273.16°C

-109.3°F CO2 sublimes (dry ice) -297.3°F Liquid oxygen -320.4°F Liquid nitrogen -459.7°F Absolute zero

Other- 1 % Oxygen- 21 %

Fig. 3-1: Chart showing the amount of nitrogen in the air.

Nitrogen Safety

Fig. 3-2: Cryogenic thermometer showing relative coldness of liquid oxygen and liquid nitrogen.

3-3

Safety Precautions for Handling Liquid Nitrogen

First Aid Procedures for Cold Liquid Frostbite (Freeze Burns)

Wear protective clothing

Symptoms

Safety goggles or face shield Insulated gloves Long-sleeved shirts Cuffless trousers

Avoid skin contact

Liquid leaking from equipment Cold equipment surfaces

Liquid nitrogen is hazardous! Contact of human tissue with severe cold will destroy tissue in a manner similar to high-temperature burns. Freeze burns will result from contact with the actual liquid or contact with the cold surfaces of piping and equipment containing the liquid. An increased dimension of hazard is added when the liquid N2 is under pressure. These facts emphasize the need for protective clothing and safety attitudes by the nitrogen equipment operator. Safety goggles or a face shield should be worn if liquid ejection or splashing may occur or cold gas may issue forcefully from equipment. Clean, insulated gloves that can be easily removed and long sleeves are recommended for hand and arm protection. Cuffless trousers should be worn outside boots or overshoes to shed spilled liquid. Liquid N2 causes immediate eye damage that is usually beyond repair! The severe nature of eye injuries emphasizes the extreme importance of wearing eye protection. One drop of liquid N2 to the eyeball could damage the eyeball instantaneously. For one second of unsafe practices, someone could be blind for life.

3-4

Skin pink just before frostbite develops Skin changes to white or greyishyellow as frostbite develops Initial pain that quickly subsides Victim feels cold and numb; he or she is often not aware of frostbite

What to Do

Cover the frostbitten part with a warm hand or woolen material. If fingers or hand is frostbitten, have victim hold hand in his or her armpit, next to body. Bring victim inside as soon as possible. Place frostbitten part in lukewarm water or warm by air at room temperature. Gently wrap the part in blankets if lukewarm water is not available or is impractical to use. Let circulation reestablish itself naturally. When the part is warmed, encourage the victim to exercise fingers and toes. Give victim a warm, nonalcoholic drink.

What Not to Do

Do not rub with snow or ice. Rubbing frostbitten tissue increases the risk of gangrene. Do not use hot water, hot water bottles, or heat lamps over the frostbitten area.

Nitrogen Safety

Liquid Air Hazard Because oxygen condenses and liquifies at a higher temperature than nitrogen, air that has supercooled, from condensing on cold liquid nitrogen equipment surfaces, will rapidly become oxygen-enriched. This condensed air can contain up to 52% oxygen, causing normally noncombustible material to become flammable and normally flammable material to burn at an increased rate.

he or she may not know to move to a well ventilated area. One full breath of pure nitrogen will strip blood of necessary oxygen, resulting in a loss of consciousness. Maintain proper ventilation to prevent asphyxiation.

Liquid Nitrogen Equipment Safety

Cryogenic Materials and Components

Oxygen Deficiency Hazard Cold N2 gas will displace warmer air containing vital oxygen for breathing. As seen below, oxygen is necessary for people to function correctly. A slight oxygen deficiency results in deeper respiration, faster pulse, and poor coordination. As the oxygen deficiency increases, ones judgment becomes so poor,

Most construction materials are adversely affected by extreme low temperatures. It is imperative that the components engineered for use in cryogenic service be chosen from suitable approved materials.

Cryogenic Materials

Table 3-1: Symptoms of Oxygen Deficiency Amount of Oxygen in the Air

Symptoms

21%

Normal

14%

Deeper breathing Faster pulse Poor coordination

12%

Giddiness Poor judgment Blue lips

10%

Nausea Vomiting Ashen complexion Approaching loss of consciousness

8%

Death within 8 minutes At 6 minutes, 50% will die At 4 minutes, all will recover with treatment

4%

Coma in 40 seconds Convulsions Death

Nitrogen Safety

Copper and brass Stainless steels300 series Aluminum (open-ended only and low psi)

Cryogenic Components

Inner tank of nitrogen tank Nitrogen low-pressure piping Nitrogen fluid ends Nitrogen high-pressure piping

Noncryogenic Material and components Most of the components of nitrogen pumping units are constructed of materials that cannot withstand cryogenic temperatures. Do not expose these components to extreme cold.

3-5

Carbon steel cannot withstand rapid contraction. Nitrogen can shrink the inside of treating iron so fast that it separates from the outside, causing severe breaks.

Noncryogenic Material

Carbon steels Low-alloy steels Most rubbers Most plastics

Pressure Buildup

Noncryogenic Components

Treating iron Cryogenic tank casing Trailer frame Power train Structural components Hydraulic lines Tires

Equipment Precautions Treating iron will not withstand cryogenic temperatures! Allowing liquid N 2 in the carbon steel treating iron is one of the most dangerous mistakes an operator can make. Carbon steel becomes brittle at approximately -40°F. When this occurs, any shock could cause treating iron to break like glass.

Nitrogen to be used as a gas is often stored and transported as a liquid for economy and convenience. It is easier to pump as a liquid than as a gas. However, there is a continuous, unavoidable, and inexhaustible heat leak into liquid N2. This heat increases the temperature of the liquid or boils the liquid at a constant temperature. Nitrogen expands 696 times its volume in going from a liquid at -320°F to a gas at 70°F, as shown in Fig. 3-3. One cubic foot of liquid nitrogen (50.46 lb) at -320°F exerts 0 psi. When this same volume warms to 70°F, it will exert 42,500 psi while in the same space. As an example of the extremely high pressure, 12 ft of 3-in. treating iron full of N2 at 10,000 psi has the same energy as 90 lb of nitroglycerin! This possible high pressure is why Halliburton pumping systems are designed using a primary safety relief valve and a secondary bursting disk assembly at any place N2 could be trapped.

Heat

Liquid Nitrogen

Gaseous Nitrogen

Fig. 3-3: Liquid N2 expands to 696 times its liquid volume when heated to 70°F.

3-6

Nitrogen Safety

Section 4 Foam Applications in Acidizing Stimulation Contents Acidizing with Foam .............................................................................. 4-3 Advantages of Foamed Acid ................................................................ 4-3 Foamed Acid Penetration ..................................................................... 4-4 Foam Quality ...............................................................................................4-4 Fracture Temperatures ...............................................................................4-5 Fracture Width .............................................................................................4-5 Pump Rate ...................................................................................................4-6

Foam Stability ....................................................................................... 4-6 Foam Diversion ..................................................................................... 4-6 Diverting Agents ..........................................................................................4-7 Types of Diversion Systems .......................................................................4-8 Mechanical Systems .............................................................................4-8 Chemical Systems ................................................................................4-8 Foamed Systems ..................................................................................4-8 Using Foam Diverters—Pointers and Recommendations ........................4-9 Commingled Nitrogen and Acid .................................................................4-9

Fracture Acidizing ............................................................................... 4-10 Results of Fluid-Loss Tests ..................................................................... 4-10 Results of Fracture Flow Capacity Tests................................................ 4-13

References .......................................................................................... 4-15 Other References ............................................................................... 4-15

Foam Applications in Acidizing Stimulation

4-1

4-2


Foam Applications in Acidizing Stimulation Acidizing with Foam As oil and gas wells age, many of them show similar characteristics. One of the most obvious is, of course, reduced bottomhole pressure that can contribute to the formation of paraffins, asphaltenes, and scales. Many old wells have had repeated acid treatments. Following conventional acid treatments, large amounts of insoluble fines such as quartz, gypsum, and feldspars may reduce fracture conductivity. All of these factors related to old wells can be controlled through foamed acid stimulation. Treating wells with characteristics as outlined above with a conventional nonfoamed acid treatment will probably be beneficial. However, the high liquid content of conventional fluids may increase clay problems. Also, low viscosity of the spent acid may leave a large amount of insoluble fines in the well. In addition, low bottomhole pressure may require swabbing to clean up the well. Nitrogen (N2) is the most widely used material in foam treatments. Volumetric gas content (foam quality) is generally between 65 and 85% (comprising 65 to 85% gas and only 15 to 35% liquid), although qualities as high as 95% have been used. The liquid phase of the foam may contain 0.5 to 1.0% surfactant and 0.4 to 1.0% inhibitor.

Advantages of Foamed Acid Foamed acid has widespread applications in both oil and gas wells and offers the following characteristics to virtually


eliminate the problems mentioned in the previous section: Low liquid content- Foamed acids used in fracture acidizing generally range from 60 to 80 quality. The low liquid content is extremely important when treating a liquid-sensitive formation where large amounts of liquid may cause swelling in the formation and reduce the permeability of the formation to the produced fluids. Reduced fluid loss- The high apparent viscosity of the foamed acid results in reduced fluid loss, allowing deeper acid penetration than a comparable nonfoamed or conventional acid system. In low permeability reservoirs, the bubbles of the foam may be sufficient to prevent leak-off to the matrix. This can reduce the affect of wormholing (channeling). Also, since no fluid loss additive is necessary in low permeability reservoirs, there is a reduced chance of impairment of formation conductivity due to the solids in some additives. High apparent viscosity- Viscosity is difficult to obtain in a nonfoamed acid system since the acid used frequently is not compatible with the gelling agent. A viscous acid provides the advantage of better pumpability, wider fracture, and improved fluid loss when used in fracture acidizing. Increasing the viscosity of the acid before it is foamed will give these benefits plus help to increase foam stability. Better cleanup- The built-in gas assist derived from using a foamed acid treatment now makes recovery of treating fluids from low-pressure reservoirs more effective than nonfoamed treatments. The built-in gas assist plus the high apparent viscosity of the foamed acid enable the acid insoluble formation fines to be returned to the surface on flow back rather than stay in

4-3

the formation where they could hamper production. This means a faster cleanup that reduces liquid damage to water-sensitive formations. Also, it may eliminate the need to swab the well after the treatment. Improved solids transport- Another advantage of foamed acid is its capability to suspend fines. Often in conventional acid treatments, large amounts of insoluble fines such as quartz, gypsum, and feldspars will be left behind because of the low viscosity of the spent acid. This may reduce fracture conductivity, but with the additional viscosity provided by foaming, more of these fines are suspended and removed from the well during cleanup. Less formation damage- Foamed acid has a low liquid content. Normally, foamed acid is 60 to 80 quality. Less liquid contacts the formation, thus reducing the opportunity for damage to occur. Minimum well shut-in time- Foamed acid treatments should have minimum well shut-in time after pumping. The foamed acid should be flowed back as soon as possible following the treatment to reduce the chance of liquid and nitrogen separation. The longer the foamed acid is allowed to remain in a static, nonflowing condition, the easier it is for liquid to drain from the foam bubbles and for suspended fines to settle out of the foamed acid. Better control- Foamed acid also provides better control. Flow can be better controlled by adjusting the amount of nitrogen, thereby changing the acids density. Because acid is normally heavier than the formation water, acid treatments tend to sink. Foamed acid can be made to stay higher in the fracture by being less dense than the formation water. Foaming the acid also helps control the reaction rate by reducing its diffusion. Foam increases the viscosity of an acid system, so the acid can be prevented from entering more permeable or low-pressure zones. This

4-4

allows for more uniform coverage without the use of other diverters. Foamed acid can also carry any of the conventional diverting systems such as Perf Pac ball sealers or granular diverter. Foamed acid offers other advantages. It has less thermal demand, causing less thermal contraction in the tubing. In cold treatment conditions, this can save having to reset the tubing due to tubing shrinkage. Nitrogen-foamed acid systems reduce asphaltene sludge by diluting the concentration of carbon dioxide (CO2) formed from acid reactions. In addition, foamed acid treatments can be displaced with straight nitrogen, leaving the hole with no hydraulic column to impede load recovery.

Foamed Acid Penetration Tests have been conducted to calculate the effect of various parameters on acid penetration distance. 1 Foamed acid reaction rate tests were performed on a laboratoryprepared fracture. These tests show that the spending of HCl in a fracture is governed primarily by the mass transfer of the acid to the fracture wall. This is referred to as a "mass transfer or diffusion" controlled spending. These tests also show that in a dolomite formation at low temperaures, the foamed acid spending is primarily controlled by the surface reaction kinetics. The effects of various factors on the spending of foamed acid are discussed in the following sections.

Foam Quality The calculated effect of foam quality on acid penetration distance (defined as the distance the live acid would travel before its concentration is spent to 0.1%) at various temperatures is shown in Figs. 4-1 and


ture. However, the surface reaction rate does change with temperature, but this reaction already is fast compared to the mass transfer to the fracture face in the HCL-limestone reaction. The dolomite acid penetration distance does decrease with an increase in temperture. This is because the surface reaction rate is the controlling factor. The surface reaction rate changes as the temperature changes. This effect of surface reaction rate can be determined experimentally by rotating disc tests at various temperatures.

Fracture Width The wider the fracture, the longer it will take for hydrogen ions to reach the carbon ate rock surface. Thus, the acid will travel farther down the fracture before spending. Fig. 4-1: Penetration distance vs. fracture temperature for limestone.

4-2 for limestone and dolomite, respectively. Four curves are shown representing 60-, 70-, 80-, and 90-quality foamed 28% HCl in each figure. An increase in foam quality results in a decrease in acid penetration distance. The higher the quality of the foam the lower the acid content of the foam. The less acid present in the foam the lower the foam's rock dissolving power. This is true for both the limestone and dolomite cases.

Fracture Temperatures The effect of temperature on acid penetration distance in limestone is negligible in the test calculations. The experimental mass transfer coefficients were measured at 70°F [21.1°C] and assumed to be independent of temperature. This may be approximately correct as long as the foam texture does not change substantially with tempera-


Fig. 4-2: Penetration distance vs. fracture temperature for dolomite.

4-5

Halliburton's Pen-5, HC-2, and SPERSEALL surfactants have been found to be effective foaming agents providing stable foams in both active and spent acid systems (see Table 4-1).

Foam Diversion In most cases, formations will be comprised of zones possessing different permeabilities or zones that may have sustained differing degrees of damage during drilling, completion, or workover operations. When acidizing treatments are performed on such formations, the treating fluids naturally enter the zones that present the least resistance to flow. This can result in placing the acid in zones that require the least stimulation. Diversion can be used to alter the fluid injection profile of a treatment. Because Fig. 4-3: Penetration distance vs. fracture width.

This is true for foamed acids as well as for nonfoamed acids. Fig. 4-3 shows this width effect.

Pump Rate If the pump rate is increased and the fracture height remains constant, the distance that the foamed acid will travel down a fracture before spending will increase. This is true in foamed acidizing of both limestone and dolomite formations. The effect of pump rate is shown in Fig. 4-4.

Foam Stability Stability of the foam is an important consideration. If a foam is stable in spent acid, foam can be returned to the surface when the well is opened, bringing the fines with it. This also helps improve formation conductivity.

4-6

Fig. 4-4: Penetration distance vs. pump rate.


fluids across the entire perforated interval, diverting agents such as Temperature Agent Charge Concentration insoluble sand, benzoic acid flakes, Limit solid organic acids, deformable Pen-5 Nonionic 250°F 0.5 to 1% solids, mixtures of waxes and oilsoluble polymers, acid-swellable HC-2 Amphoteric 275°F 0.5 to 1% polymers, and mixtures of inert 275°F 0.5 to 1% solids (silica flour, calcium carbonSPERSE-ALL Nonionic 275 to 300°F 2% ate, rock salt, oil-soluble resins, etc.) are frequently used to form tempofluids will choose the path of least resisrary filter cakes on the higher permeability tance, diversion is primarily a resistance or least damaged zones. This then forces problem; the goal is to alter injection rate the treatment into the rest of the interval. per unit of area so that all zones accept the One concern when using such materials is proper proportion of the treatment. Reserthat the filter cakes are sometimes slow to voir properties that can vary the injection dissolve in the produced fluids, thus rerate per unit of area are permeability, quiring remedial treatments for diverting differential pressure, and length; if these agent removal. properties are not in the correct proportion, In the mid 1980's, foam was introduced diversion should be considered. This as a diverting agent in place of particulatedisproportion can result from the followtype diverting agents for acidizing through ing: gravel packs. Such foams achieve diversion zones having differing due to their high apparent viscosity and the permeabilities plugging effect of the gas bubbles in the zones having differing formation foam as they enter the pore network of the pressures formation. Diversions have been accom zones containing fluids with differplished with 60 to 80 quality foam. The ent compressibility better the quality of the foam, the better its zones containing fluids with differdiverting ability. ent viscosity Foams possess several distinct advan zones having natural fractures tages over particulate diverting agents. One A goal of acid treatment is to cause main advantage is that since no solid zones of similar permeability to produce at particles are used, and because foams higher rates by increasing the permeability degrade fairly rapidly, the concern about in the critical near-wellbore area. Diverdiverting agent cleanup is eliminated. A sion helps reach this goal by forcing acid second advantage becomes evident when into damaged areas to allow the entire zone acid treatments are performed on gravel(assuming near equal permeability distripacked wells. If particulate-type diverters bution) to be productive. are used on such wells, the particles have to be sized such that they will be able to Diverting Agents pass through the gravel-pack sand and still be able to form a filter cake on the formaDiverting agents have been used in tion. This dramatically limits the types of stimulation treatments for years to help material that can be used. Foam, however, ensure treatment distribution over the easily passes through the gravel-pack sand entire perforated interval. In order to while still providing effective diversion on provide uniform placement of the reacting Table 4-1: Halliburton Acid Foaming Agents


4-7

the formation without concern about sizing or cleanup considerations.

Types of Diversion Systems Three types of diversion systems are presented herein: mechanical, chemical, and foam. For the purposes of this discussion, foams are treated separately from other chemical systems because they possess several different characteristics.

Mechanical Systems Mechanical systems may be used to create diversion. Examples of such systems are (1) straddle packers, (2) bridge plugs and packers, and (3) perforation ball sealers. More information on mechanical systems may be obtained from the SPE monograph, Hydraulic Fracturing (Sections 7.5 and 8.8).

Chemical Systems Some commonly used diverting techniques involve chemical systems; however, these are also more difficult systems to use properly. Chemical diverters can be used on perforations, in the perforation tunnel, in fractures, and on the formation face. The choice of chemical diverter to use for a particular application is determined by type of production, bottomhole static temperature (BHST), placement fluids, type of well completion, and type of treatment. In addition, the chemical diverter chosen usually has these characteristics: soluble in production fluids insoluble or marginally soluble in placement fluids relatively inert to other additives used in the treatment a melting point above the BHST The carrier fluid for a chemical diverter can be either a brine, an acid, a gel, a hy-

4-8

drocarbon, an emulsion, or a foam. If the diverter is soluble in the carrier fluid, it is important to saturate the carrier fluid with the diverter. Enough excess of diverting solid should be used to satisfy carrier fluid solubility of the diverter at bottomhole temperature conditions. Advantages of chemical diverters include low cost and a wide range of application (perforated, openhole, gravelpacked, and fractured formations). Disadvantages include uncertain diversion and secondary formation damage potential. These are examined in the following sections. Chemical diverters can cause secondary formation damage. This occurs when a diverter has completely shut off part of a zone, and removal of the diverter is dependent on producing formation fluids. Dissolution of the diverter may not occur in a reasonable length of time. Uncertain diversion is one of the major limitations of continuous chemical diverting in chemical treatments. A diversion in a fracturing treatment can be indicated by a pressure surge at the surface. In matrix treatments, however, solids introduced into the formation can reduce permeability. A high-permeability zone can act as a lost circulation zone, diverting the fluid away from the damaged or low-permeability zones and into the higher permeability zones. This is what diversion is designed to prevent. Another diversion technique involves pumping an immiscible mixture of two fluids (emulsions or foams). Nonfoamed immiscible mixtures, emulsions, are difficult to work with because their surface characteristics can be dramatically altered by (1) the shear encountered during injection down the tubing string and by (2) forcing the emulsion to flow through formation capillaries. This is further complicated by their high friction pressures in tubing.


Foamed Systems Foam passes easily through a gravel pack but has difficulty flowing into a formation. Because of this, foam was introduced as a diverting agent and has been used successfully in sandstone acidizing for almost any type of completion or production. Foamed water-based fluid diverters have been applied either continuously or as staged slugs. Foams have several characteristics that make them effective diverting agents. The physical nature of foams (bubbles consisting of discrete cells) helps control leakoff and limits the reaction rate at any given site, thus allowing deeper penetration. Foams can flow as liquids and remain motionless like a solid. Major advantages of using foam diverters include the following: suitability over a range of pressures, temperatures, and permeabilities enhanced treatment flowback improved gravel transport into perforations transportation of released fines Foam quality increases as it flows away from a wellbore, which is advantageous when treating a multizone interval with varying pressures. In the lower pressure zone, the foam will have higher quality and potentially more diversion effect. The relationship among pressure, quality, and viscosity is such that as pressure is lost, quality and viscosity increase (until 90 quality is reached). Foams also exhibit different flow properties as a function of permeability. The specific foam diversion technique used in a treatment design depends on individual well characteristics and the stimulation objective; therefore, it should be expected that foams may exhibit diverting properties as a result of differences in permeability or reservoir pressure.


A significant benefit of foamed diverters is their capability to transport released fines and insoluble particles out of the near-wellbore area during flowback. This property is especially important in underpressured reservoirs. Foam slugs (partially foamed treatments) offer the same advantages as using foamed fluids, but at considerably lower cost and less risk of system upsets during treatment flowback.

Using Foam Diverters—Pointers and Recommendations

Foams having 60 quality and higher provide a greater reduction in flow than lower quality foams. More importantly, the duration of the diversion lasts much longer when using foams with 60 to 90 quality. In certain types of rock, brine foams give more resistance and longer diversion than acid foams. This is most prevalent in either high porosity and/or high permeability limestones. Alternating stages of foamed diverter and either nonfoamed or commingled acid are more effective than a single stage of foam diverter. Foam effectively diverts acid from a nondamaged core to a damaged core. For successful diversion, the differences in zone permeabilities should not be greater than a factor of 10. Wormholes play an important role in acidizing. When there is no fluidloss control, the distance that an acid will penetrate is controlled by the development of wormholes. Foamed diverters discourage wormhole formation because the discrete cells help control leakoff and limit the reaction rate that can occur at

4-9

any given site. By not producing wormholes, the acid stays active longer in the fracture, develops a deeper penetration, and produces more flow capacity.

Commingled Nitrogen and Acid When enough N2 is introduced into an acid to impart energy for load recovery and hydrostatic column reduction, but the amount of gas is not sufficient to cause bubble bumping, it is not considered a foam. This condition greatly decreases the load recovery time by providing a gas assist. This type of gas addition can also aid in reducing the total weight of the treatment column by helping place it effectively in low bottomhole pressure wells.

Fracture Acidizing

ing agents on fluid loss control. Conventional 15% HCl channeled through a six-in. core in less than 1 minute and exhibited little or no fluid-loss control. Fig. 4-5 shows the face of this core and several large wormholes indicating where acid breakthrough occurred. Fig. 4-6 shows the face of the core across which the 90-quality foamed acid, 15% HCl plus 1% foamer, was flowed for 36 minutes. No fluid loss for 36 minutes and the large number of small holes on the face of the core indicates 90quality foamed acid gave good fluid-loss control. These same results were noted for 80-quality foamed acid. When the quality of this foamed acid was lowered from 80 to 70, breakthrough occurred after 18 minutes. At breakthrough, foam, rather than separate gas and liquid phases, came through the core. Bubble size in this foam was much larger than when the foam was initially generated. A 60-quality foamed acid broke through

Use of foam in fracturing treatments has gained widespread acceptance. Low liquid Table 4-2: Effect of Foam Quality and Foaming Agents upon Fluid content, good fluid-loss Loss of Foamed Acid (pressure diff.= 100 psi) control, and quick Foam Rock Permeability Breakthrough 36-min N2 cleanup are just a few Test Solution Quality to N2 at 110°F (md) Time (min) Fluid Loss (L) reasons why foams are 15% HCl 0 0.85 <1 being used. Halliburton 90 0.83 >36 has investigated the effects of foam quality, 80 0.72 >36 15% HCl + foam stability, and 1% Foamer A 70 0.84 18 chemical compatibility 60 0.66 7 on fluid loss and fracture 2 flow capacity. The 90 1.21 0 results are summarized 90 0.26 0 in the following sections.

Results of Fluid-Loss Tests Table 4-2 shows the effect of foam quality and two different foam-

4-10

15% HCl + 1% Foamer B

80

0.63

0

80

0.61

0

70

0.88

0.07

70

1.14

0.69

60

0.69

0.22

60

1.83

0.46


Fig. 4-5: Fluid-loss test core showing effect of conventional 15% HCl. Note the "wormhole."

Fig. 4-6: Fluid-loss test core showing effect of 90quality foamed 15% HCl.

the core in 7 minutes. Tests were repeated substituting a different foamer, and results indicated no acid or foam fluid loss occurred for 36 minutes Table 4-3: Effect of Foam Quality and Acid Concentration when any of these four quality upon Fluid Loss of Foamed Acid (pressure diff.= 100 psi) foamed acids were tested. Foam Rock Permeability 36 min N2 Fluid Test Solution However, nitrogen loss did Quality to N2 at 110°F (md) Loss (l) occur when the 70- and 6090 1.21 0 quality foamed acids were 90 0.26 0 tested. Results show the effect of 80 0.63 0 chemical compatibility in a 80 0.61 0 15% HCl + 1% foamed acid system. The first Foamer 70 0.88 0.07 foamer made a stable foamed 70 1.14 0.69 acid with 15% HCl, but when this foamed acid came in con60 0.69 0.22 tact with a large amount of 60 1.83 0.46 spent acid, such as when a 6090 0.48 0 or 70-quality foamed acid was run, the foam apparently col90 0.41 0 lapsed and subsequently broke 80 0.55 0 through the core. The second 80 0.69 0 28% HCl + 1% foamer appeared to be more Foamer 70 0.47 0.06 compatible with spent acid than the first, so no foam break70 0.70 0.25 through occurred. It is impor60 0.71 0.20 tant that all chemicals used in a 60 0.78 0.97 foamed acid system be checked


4-11

tained fluid-loss control for 25 minutes before foamed acid breakthrough. The more Foam Rock Permeability Breakthrough Test Solution permeable 0.41 and 0.53 md Quality to N2 at 110°F (md) Time (min) cores experienced foamed acid 90 0.43 3 breakthrough in 2 minutes. 90 0.39 3 Increase in differential pres28% HCl + 1% sure from 100 psi to 500 psi 80 0.53 2 Foamer changed the fluid-loss control 80 0.15 25 of the foamed acid consider80 0.41 2 ably. Comparison of the 90 0.41 4 foamed 28% HCl results from Table 4-3 with the results 28% HCl + 90 0.37 4 HAC + 1% given in Table 4-4 clearly 80 0.32 3 Foamer illustrates the difference. 80 0.31 2 A similar trend was noted for conventional acids containfor compatibility in spent acid as well as in ing solid fluid-loss material. With increasthe live acid. ing pressure differential, it is more difficult Effects of foam quality and acid concento maintain fluid-loss control. One way to tration on foamed acid fluid loss are shown help minimize the effects of the increased in Table 4-3. No acid fluid loss occurred for pressure differential is to stabilize the any of the four qualities of foamed 15% foamed acid. This can be accomplished by HCl. Nitrogen loss did occur with the 60- and 70-quality Table 4-5: Acid-Etched Fracture Flow Capacity with Conventional and Foamed 28% HCl foamed 15% HCl. This same trend was shown when acid Temperature- 110°F, Pressure- 1,500 psi, Closure Pressure- 1,000 psi concentrations were increased from 15 to 28% HCl. Fracture Flow Etching Core Removed Effects of acid type, formaTest Capacity Time (min) (in.) (md-ft) tion permeability, and pressure differential are illustrated in 9 9,691 0.044 Table 4-4. The two types of acid 18 12,960 0.056 No. 1- 200 ml/min studied, 28% HCl and a mixture 28% HCl 27 26,691 0.068 of mineral and organic acid, 36 40,255 0.088 foamed equally well and gave virtually the same fluid-loss 9 4,833 0.058 control. When foam break18 6,990 0.074 No. 2- 20 ml/min through occurred, bubble sizes 28% HCl 27 7,535 0.091 of the foams were about equal to the bubble sizes just after 36 28,409 0.109 generation. 9 17,533 0.066 No. 3- 180 ml/min Upon examination of the 80N2 + 20 ml/min 18 12,329 0.084 quality foamed 28% HCl sys28% HCl + 1% 27 70,000+ 0.130 tem, it was noticed that the 0.15 Foamer 90-Quality Foam md permeability core main36 70,000+ 0.153 Table 4-4: Effect of Acid Type upon Fluid Loss of Foamed Acid (pressure diff.= 500 psi)

4-12


Table 4-6: Effect of Foam Quality on Acid Etched Fracture Flow Capacity Temperature- 110°F, Pressure- 1,500 psi, Closure Pressure- 1,000 psi Test

No. 1- 180 ml/min N2 + 20 ml/min 28% HCl + 1% Foamer, 90-Quality Foam No. 2- 80 ml/min N2 + 20 ml/min 28% HCl + 1% Foamer, 80-Quality Foam No.3- 47 ml/min N2 + 20 ml/min 28% HCl + 1% Foamer, 70-Quality Foam No. 4- 30 ml/min N2 + 20 ml/min 28% HCl + 1% Foamer, 60-Quality Foam

Etching Time (min)

Fracture Flow Core Removed Capacity (md-ft) (in.)

9

17,533

18

12,392

27

70,000+

36

70,000+

9

8,613

18

21,537

27

70,000+

36

70,000+

9

12,392

18

41,464

27

36,026

36

27,259

9

14,678

18

28,977

27

38,443

36

37,234

increasing the viscosity of the acid before it is foamed. Both 80- and 90-quality foamed acids showed only N2 fluid loss but no acid fluid loss for 36 minutes where previously they broke through the core in 2 to 3 minutes. Both 60- and 70-quality foamed acids maintained fluid-loss control for 10 to 11 minutes. Increasing the acid viscosity to help stabilize a foamed acid and improve fluid-loss control without the use of wallbuilding additives is keeping with the idea of a true foamed acid. Extremely large pressure differentials and large formation permeabilities may, however, require adding conventional fluid-loss additives to the foamed acid system. Fluid loss in high


permeability formations can be reduced by using a pad fluid ahead of the foamed acid.

Results of Fracture Flow Capacity Tests

Tests have shown that foamed acid can give good 0.084 fluid-loss control. However, a 0.130 successful fracture acidizing treatment does not depend only 0.153 on good fluid-loss control. 0.037 Adequate fracture flow capac0.070 ity must be established by the acid system used. Quantity of 0.139 rock removed and the pattern 0.175 in which it is removed from the 0.036 fracture faces are important. 0.074 Fracture flow capacity depends on the nature of the 0.096 rock and the volume, type, and 0.120 concentration of acid used. In 0.030 order to eliminate some of the variables, Bedford Indiana 0.075 limestone was selected as a 0.097 homogeneous rock and was 0.120 tested with one concentration of acid (28% HCl). Table 4-5 shows the results of equal velocities of treating solution as well as equal amounts of acid. Tests No. 1 and 3 were both conducted at a total flow rate of 200 ml/min. The foamed acid in Test No. 3 was only one-tenth the amount of 28% HCl as compared to the conventional acid in Test No. 1 and created more fracture flow capacity. Comparison of Tests No. 2 and 3, which used equal amounts of 28% HCl, indicated that foamed acid created more fracture flow capacity. Also, the foamed acid system removed more core than either of the two conventional acid systems tested. It was noted in Test No. 3 that some fracture flow capacity was lost between the first and second time intervals. This effect, 0.066

4-13

Table 4-6: Effect of Foam Quality and Foam Stability on Acid Etched Fracture Flow Capacity Temperature- 110°F, Pressure- 1,500 psi, Closure Pressure- 1,000 psi Test

No. 1- 47 ml/min N2 + 20 ml/min 28% HCl + 1% Foamer, 70-Quality Foam No. 2- 47 ml/min N2 + 20 ml/min 28% HCl + 1% Foamer + 4% Foam Stabilizer, 70-Quality Foam No.3- 30 ml/min N2 + 20 ml/min 28% HCl + 1% Foamer, 60-Quality Foam No. 4- 30 ml/min N2 + 20 ml/min 28% HCl + 1% Foamer + 4% Foam Stabilizer, 60-Quality Foam

Etching Time (min)

Fracture Flow Core Removed Capacity (md-ft) (in.)

9

12,392

0.036

18

41,464

0.074

27

36,026

0.096

36

27,259

0.120

9

11,314

0.048

18

30,126

0.067

27

70,000+

0.076

36

70,000+

0.087

9

14,678

0.030

18

28,977

0.075

27

38,443

0.097

36

37,234

0.120

9

30,695

0.037

18

70,000+

0.054

27

70,000+

0.063

36

70,000+

0.070

fluid-loss control. Acid viscosity was increased, and 60- and 70-quality foamed acids generated. Table 4-7 compares these results. Both 60- and 70-quality foamed acids achieved maximum fracture flow capacity and showed no signs of overetching. Smaller amounts of rock were removed from the core faces, but pattern of removal was more effective. These tests show that foamed acid achieves better fracture flow capacity when compared to conventional acid at equal velocities of treating solution as well as equal amounts of acid.

called overetching, is quite common in homogeneous cores where rock is often removed evenly. Effect of foamed acid quality on fracture flow capacity is shown in Table 4-6. Excellent fracture flow capacities were obtained when any of the four qualities of foamed acid were used. Large amounts of core were also removed in each of the four cases. Niether 60- nor 70-quality foamed acids obtained the maximum fracture flow capacity that the 80- and 90-quality foamed acids obtained. Overetching effects were also more pronounced in the 60- and 70quality foamed acids. Foam stability affects acid-etched fracture flow capacity the same as it affects

4-14


References 1. Ford, W., and Roberts, L.: "The Effect of Foam on Surface Kinetics in Fracture Acidizing," paper SPE 11120 presented at the 1982 Annual Fall Technical Conference and Exhibition of the SPE, New Orleans, LA (September 26-29). 2. Ford, W.: "The Use of Foamed Acid in Fracture Acidizing," paper SPE 9652 presented at the 1981 SPE Middle East Oil Technical Conference, Manama, Bahrain, March 9-12.

Thompson, K. and Gdanski, R.D.: Laboratory Study Provides Guidelines for Diverting Acid with Foam, paper SPE 23436 presented at the 1991 SPE Eastern Regional Meeting, Lexington, Kentucky (October 22-25). Williams, B.B., Gidley, J.J., and Schechter, R.S.: Acidizing Fundamentals, SPE/ AIME monograph, Vol. 6., Dallas, 1979.

Other References Burman, J.W. and Hall, B.E.: Foam as a Diverting Technique for Matrix Sandstone Stimulation, paper SPE 15575 presented at the 1986 SPE Annual Technical Conference and Exhibition, New Orleans, (October 5-8). Ford, W., Burkleca, L., and Squire, K.: "Foamed Acid Stimulation: Success in the Illinois and Michigan Basins," paper presented at the 1980 Annual Fall Technical Conference and Exhibition of the SPE, Dallas, Texas (September 2124). Kennedy, D.K., Kitziger, F.W., and Hall, B.E.: Case Study on the Effectiveness of Nitrogen Foams and Water Zone Diverting Agents in Multistage Matrix Acid Treatments, paper SPE 20621 presented at the 1990 SPE Annual Technical Conference and Exhibition, New Orleans (September 23-26). King, G.E.: "Foam and Nitrified Fluid TreatmentsStimulation Techniques and More," paper SPE 14477presented as a Distinguished Lecture during the 1986-86 Distinguished Lecturer Program.


4-15

4-16


Section 5 Foam Applications in Hydraulic Fracturing Stimulation Contents Introduction .............................................................................................5-3 Types of Foams Used in Hydraulic Fracturing ....................................... 5-4 Foam Rheology.......................................................................................5-4 Crosslinked Foams .................................................................................5-5 Foam Fluid Loss .....................................................................................5-7 Fluid-loss Coeffecients ................................................................................ 5-8 Test Results of Factors Affecting Foam Performance ................................ 5-8 Test Conclusions ....................................................................................... 5-12

Fracture Conductivity ............................................................................ 5-12 Proppant Pack Permeability ...................................................................... 5-12 Proppant Grain Size............................................................................. 5-12 Effective Closure Stress ...................................................................... 5-12 Multiphase Flow Effects ....................................................................... 5-13 Fracturing-Fluid Residue Damage .......................................................5-13 Filtercake Buildup ...................................................................................... 5-13

Treating Pressure Response ................................................................ 5-14 Constant Internal Phase ............................................................................5-15 Increased Proppant Concentration ............................................................5-17 Field Treatment Results of Constant Internal Phase ................................5-17 Conclusions................................................................................................5-22

Fluid Recovery ...................................................................................... 5-22 Treatment Designs for Hydraulic Fracturing .........................................5-22 PROP Hydraulic Fracture Design Program ...............................................5-22 FracPac II ................................................................................................... 5-24 Candidate Selection ............................................................................. 5-24 Wireline Logging ..................................................................................5-25 Formation Strength .............................................................................. 5-26 Fracpressure Log .................................................................................5-26 Perforating............................................................................................ 5-26 Fracture Design.................................................................................... 5-26 3-D Fracture Design Simulator ............................................................5-29 Prefracture Testing .............................................................................. 5-29 Downhole Tools ...................................................................................5-30 Example Procedure ............................................................................. 5-30

Foam Applications in Hydraulic Fracturing Stimulation

5-1

Contents (cont.) Minifractures ......................................................................................... 5-31 Minifracture Analysis Technique ................................................................ 5-31 Analysis Theory.................................................................................... 5-31 Minifracture Test Results ...........................................................................5-33 Well Data.............................................................................................. 5-33 Minifracture Fluids ................................................................................ 5-33 Treatment Fluids ..................................................................................5-33

Conclusions .......................................................................................... 5-34 References ........................................................................................... 5-34 Additional References .......................................................................... 5-35

5-2


Foam Applications in Hydraulic Fracturing Stimulation Introduction Foams are being used in a number of petroleum industry applications that exploit the foams' high viscosity and low liquid content. Some of the earliest applications for foam dealt with its use as a displacing agent in porous media and as a drilling fluid. Following these early applications, foam was introduced as a wellbore circulating fluid for cleanout and workover applications. In the mid-1970s, nitrogen- (N2) based foams became popular for both hydraulic fracturing and fracture acidizing stimulation treatments. In the late 1970s and early 1980s, foamed cementing became a viable service, as did foamed gravel packing. Most recently, carbon dioxide (CO2) foams have shown usefulness in hydraulic fracturing stimulation. The early widespread use of foams as fracturing fluids was to help low-pressure gas reservoirs in returning the liquid phase of the foam. The internal phase of the foam typically consisted of 65 to 80% by volume (quality) of N2 gas, with an external phase of water and a foaming agent (surfactant). These simple N2 foam fluids, coupled with the pumping technology of the 1970s, were able to transport sand concentrations of 1 to 2 lb/ gal [120 to 240 kg/m3] into fractures. Such low proppant concentrations gave beneficial results in low-pressure sandstone, carbonate, and shale reservoirs. Much of the success of the early treatments was due to the capability of N2 gas to expand and remove substantial quantities of the liquid phase from the reservoir. Gelling agents were not originally used, so no gel filtercakes were formed to damage proppant beds.

With the success of simple N2 foams in low-pressure gas reservoirs and the development of high-pressure N2 pumping equipment, the application of foam fluids was extended to higher pressure gas reservoirs and oil wells. These zones placed additional requirements on foam fracturing fluid, such as higher viscosity, better leakoff control, higher temperature stability, and greater proppant carrying capacity. The need for higher viscosity was met by using water soluble polymers, such as guar and hydroxypropyl guar (HPG) gelling agents, to increase viscosity of the liquid phase and the foam. Adding gelling agents served to improve fluid leakoff control by building a thin gel filtercake on the face of the fracture. Higher temperature stability was improved by the development of surfactants that were capable of stabilizing foams to greater than 392°F [200°C].1 Mechanical improvements in highpressure slurry pumping equipment allowed significantly higher concentrations of proppant in slurries to be pumped. By 1980, technology had developed to the point that massive hydraulic foam fracturing treatments were conducted that placed over 1 million lb [454,000 kg] of sand at concentrations up to 4 lb/gal [480 kg/m3] in a formation with a temperature of 270°F [130°C].2 Deeper reservoirs were made accessible with the introduction of CO2 foams. CO2, pumped into the wellbore as a liquid, has a greater density than N2 gas, allowing surface pumping pressures to be lower than with N2 for a corresponding depth. CO2 foams are formed when reservoir temperatures warm the fluid to above the critical temperature of liquid CO2. The resulting mixture of gaseous


5-3

CO2 with water is considered to be a foam, but the density of the CO2 remains high as long as pressure is maintained. The structure of CO2 foams is similar to N2 foams, but the proppant transport capability of CO2 foams is inherently greater because of its higher bouyancy. As the number of treatments using CO2 increased, it became apparent that the friction pressure of CO2 foams was higher than for N2 foams, especially in the high concentration proppant stages. In an effort to overcome high friction problems, a new method of producing sand-laden foams was developed. The technique of constant internal phase held the liquid phase volume constant while balancing the sum of gas plus sand to equal the desired internal phase volume.3 Delayed crosslinked gelled fluids were introduced as fracturing fluids in the early 1980s, and these crosslinking agents were soon applied to producing crosslinked N2 foams. The higher viscosity produced by crosslinking the gelling agent in the foam produced higher viscosity foam fluids that were able to place higher proppant concentrations than noncrosslinked foams. Crosslinking of CO2 foams was introduced at a later date and extended the advantages of CO2 foams to deeper, hotter reservoirs.4 Proppant concentrations as high as 12 lb/gal [1,440 kg/m3] have been successfully placed with crosslinked CO2 foams using the constant-internal-phase technique. Throughout the development of foam fracturing fluids over two decades, foam fracturing fluids have been used in liquid sensitive formations because of their capability to minimize liquid contact with the reservoir and their capability to rapidly recover the majority of the treatment fluid. Even though the cost of foam treatments is typically 10 to 20% greater than nonfoamed crosslinked stimulation treatments, quick fluid recovery and minimal damage to the reservoir have given foamed fluids a place

5-4

among the leaders in number of treatments performed. This section presents some of the technical benefits of foam as a minimal damage fluid for fracturing.

Types of Foams Used in Hydraulic Fracturing A wide variety of liquid phases are available for N2 foam fracturing. The base liquids include water, water-alcohol mixtures, and hydrocarbons. Water is the most economical liquid phase available. When watersensitive clays are likely to be encountered, salts, such as potassium chloride or Cla-StaTM additives may be used to help protect such clays. Adding up to 50% alcohol will further reduce potential clay swelling. Alcohol also lowers the surface tension of the liquid and has a higher vapor pressure to aid in producing back the frac fluid. Maximum protection against formation damage can be realized by using a hydrocarbon foam. Suitable oils for N2 foam fracturing would include diesel and condensates. Lease crude oils should be laboratory tested for foamability prior to field usage. Foam generators may be desirable when fracturing with hydrocarbon foam or any of the gelled systems.

Foam Rheology The viscosity of a fracturing fluid is important because of its influence in creating fracture geometry and in transporting proppant. Adding linear polymers or crosslinked polymers to water increases its viscosity. Viscosity of the fluid mixture is also increased by adding N2 or CO2 gas to create an internal phase (gas bubbles), when a stabilizing surfactant (foaming agent) is present.


High viscosity foam fluids can be prepared size distribution, plays an important but using low amounts of water and gelling lesser role in determining foam viscosity. agents, thereby minimizing the liquid load Foams exposed to shear for a sufficient time placed on a formation. will equilibrate to a bubble size distribution Foam rheology has been described by a that is characteristic of that shear rate. Texnumber of investigators using various fluid ture is also influenced by the surfactant that models. Work by Mitchell,5 using ungelled must be present in sufficient concentration to water foamed with N2, characterized foam as stabilize the foam under dynamic condia Bingham plastic fluid, having a positive tions.7,8 yield stress at zero shear rate. Foams containing polymers have been described by several models,6 including power law and yieldCrosslinked Foams pseudoplastic. Foam viscosity depends on a number of Foams containing polymers that have variables, including quality, viscosity of the been crosslinked are more viscous than foams external phase, and texture. The most imporwithout crosslinking. An example is given in tant parameter is foam qualitythe percent Fig. 5-1 of CO2 foams containing 0.48% of a volume occupied by the internal gas phase. guar derivative. The foam containing polySince gas volume is a function of temperature mer crosslinked with zirconium has approxiand pressure, downhole conditions must be mately twice the viscosity of the known. As quality increases, foam viscosity noncrosslinked foam. Crosslinked N2 foams increases. In addition, the yield point characcan be generated with any of the typically teristics of foams are an exponential function used polymer-crosslinking agent combinaof quality. tions since N2 is considered chemically inert Higher quality foams have better transport properties, particularly at very low shear rates, because of high yield points. The viscous character of the external liquid phase is also a major parameter. Flow of high-quality foam may be visualized as gas bubbles sliding past one another on thin films of the liquid external phase. If the liquid film contains a viscosifying agent, then the bubbles will undergo greater drag forces because of the viscous thin films, and flow will be more difficult, resulting in higher bulk viscosity. Fig. 5-1: Viscosity of gelled foams—0.48% linear and crosslinked polymers. Texture, or the bubble


5-5

and does not interfere with crosslinking chemistry. Combinations would include guar and guar derivatives with aluminate, borate, titanate, and zirconate crosslinking agents. Crosslinked CO2 foams must be formed with polymer-crosslinking agent combinations that are active in the pH range of about 3 to 5 because of the strongly acidic pH effect of CO2 on the aqueous external phase. For example, borate crosslinked foams cannot be made with CO2 since a pH above 8 is required to crosslink guar with borate. Several differences exist between the type of fractures created by crosslinked and noncrosslinked foams. Crosslinked foams have higher proppant carrying capacity than noncrosslinked foams because of their higher viscosity. Proppant is easier to transport into the fracture since a wider fracture is created by the more viscous crosslinked foam. Since a wider fracture is created, the fracture will be shorter for a given volume of fluid pumped. A shorter, wider fracture has less total fracture area created, meaning less surface area exposed to fluid leakoff. The fluid-loss coefficients for crosslinked and noncrosslinked foams are similar for the same leakoff area, so the total leakoff with crosslinked foams is less. Lower overall leakoff, coupled with wider fractures, means that proppant placement with crosslinked foams is easier to accomplish than with noncrosslinked foams. Gel filtercakes generated with crosslinked foams are about as thin as noncrosslinked foams, 0.004 in. [0.10 mm]. Although such thin filtercakes cause minimal occlusion of proppant pack conductivity, the chemical character of the residue is still crosslinked and is harder to remove than noncrosslinked linear polymers. Compressible foams are structured, twophase fluids that are formed when a large internal-phase volume (typically 55 to 95%) is dispersed as small, discrete entities through a continuous liquid phase. Under typical formation temperatures of 90°F [32.2°C]

5-6

encountered in stimulation work, the internal phases exist as gas and hence are properly termed foams in their end-use application. At typical surface conditions of 75°F [23.9°C] and 900 psi [6,205 kPa], N2 is a gas, but CO2 is a liquid. A liquid/liquid two-phase structured fluid is classically called an emulsion. The end-use application of the two-phase fluid, however, normally is above the critical temperature of CO2. Evidence shows the similarity of two-phase structured fluids independent of the state of the internal phase. The liquid phase typically contains a surfactant and/or other stabilizers to minimize phase separation (bubble coalescence). These dispersions of an internal phase within a liquid can be treated as homogeneous fluids, provided bubble size is small in comparison to flow geometry dimensions. Volume percent of the internal phase within a foam is its quality. The degree of internalphase dispersion is its texture. At a fixed quality, foams are commonly referred to as either fine or coarse textured. Fine texture denotes a high level of dispersion characterized by many small bubbles with a narrow size distribution and a high specific surface area, and coarse texture denotes larger bubbles with a broad size distribution and a lower specific surface area. Because foams exhibit shear-rate-dependent viscosities in laminar flow, they are classified as non-Newtonian fluids. In addition to shear rate, their apparent viscosities also appear to depend on quality, texture, and liquid-phase rheological properties. Measured laminar-flow apparent viscosities generally are larger than those of either constituent phase at all shear rates. When the liquid phase is thickened by adding solids, soluble high-molecular-weight polymers, or other viscosifying agents, even larger foam viscosities are produced. While laminar flow is characterized by strictly viscous energy dissipation, turbulent flow is characterized more by kinetic than


viscous energy dissipation. Density and velocity are the factors that establish kinetic energy, and reduced foam density may outweigh an increased viscosity contribution and produce a turbulent-flow friction loss less than liquid-phase friction loss. Soluble highmolecular-weight polymers produce a form of turbulent drag reduction that is analogous to that which occurs in a nonfoamed liquid. In this case, a substantial drag reduction effect is evident when one compares the turbulent flow friction loss of foams with and without a gelled liquid phase.

Foam Fluid Loss

fluids that contain gelling agents, but the filter cakes are much thinner. Filter cakes deposited in the laboratory from linear gel foams and crosslinked foams typically ranged between 0.0016 to 0.006 in. [0.04 to 0.15 mm].10 Even though the deposited filter cake is less thick, the overall fluid leakoff rate in matrix with foams is still less than with nonfoamed fluids. The reason for the lower leakoff rate is that bubbles of gas from the foam enter the formation matrix and impede the loss of liquid. Two-phase flow in porous media is slower than single phase flow. A plot of overall fluid-loss coefficient vs. matrix permeability is shown in Fig. 5-2. A thinner filter cake represents less gel mass to block produced fluid flow after the treatment has been completed. Regained permeability tests on 0.1 to 0.3 md cores indicated 87 to 95% of the original matrix permeability was regained after exposure to foamed fluids. Fracture conductivity studies of proppant packs indicated that 80 to 100% of baseline conductivity was measured after treatment with linear gelled foamed fluids.11

Good fluid-loss control is important in creating fracture geometry and transporting proppant into the fracture. Fluid loss from nonfoamed gelled fracturing fluids may be understood as the loss of water into formation capillaries at an initial rate determined by the permeability of the rock matrix. As water is lost from the gelled fluid, polymer (filter cake) gradually forms on the formation face. Once a gel filter cake has been deposited, the permeability of the filter cake is lower than the permeability of the formation, so the filter cake controls further loss of water to the formation. Linear gel and crosslinked gelled fluids typically deposit filter cakes about 0.03 to 0.04 in. [ 0.75 to 1.0 mm] thick under dynamic conditions in the laboratory.9 Gel filtercakes are also Fig. 5-2: Foam fluid-loss coefficient for gelled external phase. deposited from foamed


5-7

Crosslinking the gelling agent in a foam reduced regained conductivity compared to linear foam, but crosslinked foams still had higher conductivity than nonfoamed crosslinked fluids. The capability of both formation matrix and proppant pack to flow fluids at rates near their undamaged capacities is a measure of the clean character of foamed fluids.

Fluid-loss Coeffecients Foam has been established as a successful fracturing fluid for several years. Claims about its efficiency in fluid leakoff control have ranged from excellent to virtually no leakoff at all. Yet treatment experience has indicated that foam fractures occasionally do screen out.2 Because excessive fluid leakoff is one potential cause of a premature job termination, an adequate knowledge of fluid-loss coefficients is essential for proper design of stimulation treatments. Foam has been described previously as a nonwall-building fluid.12 Such a fluid should have leakoff properties described by Howard and Fast13 as

CI = 0. 0469 k∆ pφ / µ , .......................... (5-1) where k is permeability (darcies), ∆p is pressure drop (psi) across the matrix, φ is fractional porosity, and µ is viscosity (cp). There is a problem with calculating CI for foam, however. Because foam is a two-phase structured fluid and because some of the bubbles may be able to enter the pores of the rock matrix, the rheology of foam in porous media is not well defined. The expansion of bubbles in the foam caused by pressure drop can be significant. A fluid-loss coefficient can be determined empirically without knowledge of the fluid viscosity by the equation from Howard and Fast:13

5-8

CIII = 0.0164 m/Ac , ...................................... (5-2) where m is the slope of an experimental plot of fluid volume vs. the square root of time and Ac is the cross-sectional area of the filter medium in square centimeters. CIII is useful for wall-building fluids, but CI is intended for nonwall-building fluids. For nonwall-building fluids, the slope of the experimental plot of filtrate volume will be linear with time, rather than with the square root of time. The majority of foam fracturing treatments performed contain a viscosifier to stabilize the liquid phase, typically hydroxypropyl guar (HPG), that has wallbuilding character.

Test Results of Factors Affecting Foam Performance Foam fracturing fluid may experience many conditions during a fracturing treatment. Both foam-fluid and rock-matrix variables may affect fluid-loss characteristics. A series of experiments were conducted to examine how several variables may effect foam performance. Some results of these experiments are presented and discussed. Permeability of the rock matrix had a significant effect on fluid-loss characteristics, as indicated in Fig. 5-3. The overall effect showed CIII for foam increases one order of magnitude for an increase of two orders of magnitude in permeability. Some variations, dependent on foam texture and rock pore structure, might be expected from this trend. For example, a given bubble size may be of proper size to plug small holes in low-permeability rock but may be too small to block flow into much larger pores in high-permeability rock. Foam texture, which was not examined as a variable, was controlled only by the generation technique used in this experiment.


Fig. 5-3: Foam fluid-loss coefficient at 75°F with liquid phases containing 20 lb HPG/1,000 gal.

of differential pressure was minimized with gel in the foam. However, with no gel present, foam fluid-loss increased significantly with increases in differential pressure (Table 5-2). A moderate increase in fluid loss was observed when the core temperature was increased from 75 to 200°F [24 to 93°C]. The effective increase probably was caused by the thinning of the liquid phase of the foam by temperature (Fig. 5-4). No significant effects were observed when

The amount of gel present in the liquid Table 5-1: Effect of Core Length on Fluid-Loss Coefficient phase also had a significant influence on CIII (Fig. 5-4). Previous studies concluded that Liquid phase contains 20 lb HPG/1,000 gal foam is not a wall-building fluid.12 This study C III also found that foam containing no gel in the Length (in.) liquid phase is not a wall-building fluid. 0.15 0.00128 However, foam containing 10 lb HPG/1,000 gal [1,198 g/m3] or more in the liquid phase 0.30 0.00122 does form a wall. The gel layer could be 0.45 0.00111 observed visually on the core face after test0.62 0.00122 ing. The semipermeable barrier controlled a number of the variables investigated. 1.87 0.00118 There was a slight effect of core length on fluid loss with no gel present. Yet when 20 lb HPG/1,000 gal [2,397 g/ Table 5-2: Effect of Pressure on Fluid-Loss Coefficient m3] was present in the liquid phase of the foam, HPG concentration High/Low Pressure varying core length from (psi) 0 lb/1,000 gal 20 lb/1,000 gal 0.15 to 1.87 in. [0.38 to 4.75 cm] produced no 700/200 0.00274 0.00122 effect on fluid loss (Table 1,200/200 0.00518 0.00134 5-1). In addition, the effect 1,200/700


--

0.00136

5-9

external liquid phase and a discontinuous, internal gas phase moving discretely or by rupture of the gas cells.14 This mechanism would be sensitive to the viscosity of the liquid phase. A viscous liquid phase should flow slowly and exhibit less liquid loss through porous media relative to a thin liquid phase. The effect of gellingagent concentration on effluent quality is shown in Fig. 5-5. Effluent volumes for gas and liquid Fig. 5-4: Effect of core temperature and gel concentration on foam fluid-loss coefficient in 0.3-md sandstone. were averaged and qualities calculated for foams with 0, 20, and 40 lb foaming agents were exchanged in waterHPG/1,000 gal [0, 2,397, and 4,793 g/m3]. based foams. The differences in fluid-loss The Ohio sandstone curve represents more characteristics between anionic and nonionic data and should be more reliable than the surfactants were insignificant. The addition of other samples. The higher-permeability up to 50% methanol in the liquid phase also Bandera sandstone showed a similar trend did not change fluid-loss properties. Oileven though fewer data points were availbased foam with no gel was comparable to a able. In general, the liquid-loss relative to gas water-based foam with no gel. Therefore, in a foam fluid was greater when low-viscosfluid loss is controlled more by the texture of ity liquid phase were used. In all cases, the the foam and viscosity of the liquid phase effluent quality was lower than the original than by whether the liquid phase is water, foam quality. The numerical value of effluent alcohol, or oil. Most of the foams were tested at 75 quality. In one experimental series, quality Table 5-3: Effect of Foam Quality on was varied at 65, 70, 75, and 80 in a foam Fluid-Loss Coefficient containing 20 lb HPG/1,000 gal [2,397 g/m3] Liquid phase contains 20 lb HPG/1,000 gal (Table 5-3). No effect was observed when quality was changed, indicating that the gel C III Foam Quality layer was controlling leakoff. 65 0.00126 The effluent discharged from the cores did not have the same composition as the 70 0.00119 foam impinging on the cores. The mechanism 75 0.00123 of foam flow through porous media described by Holm is of a continuous, moving, 80 0.00119

5-10


indicate that thinner filter cakes formed from foams than from linear gel or crosslinked gel fluids tested under comparable dynamic conditions.10 Because foams contain a gelling agent only in the liquid phase, there is less gelling agent available for deposition than from nonfoamed gelled fluids. The values reported here have been used to design successful foam fracturing treatments since 1981. Typical foam fracturing fluids have fluid-loss control compaFig. 5-5: Quality of fluid passing through Ohio and Bandera cores vs. gel rable to gelled or concentration of liquid phase. complexed fluids for permeabilities near 1 md. quality was probably influenced by volume Based on field experience, the screenout expansion of gas bubbles passing from high potential appears similar for foams and to lower pressure through the core. This crosslinked-gel fluids. expansion ratio is probably greater in the In core permeabilities higher than 1 md, laboratory test core than in an actual stimulafoam fluids still exhibit fluid-loss control, tion. although solid additives may be helpful to Because foams containing polymers are achieve adequate leakoff control. Typical wall-building fluids, the potential exists for foam fracturing fluids have improved fluiddamage to original matrix permeability. A loss control in core permeabilities lower than number of sandstone cores were tested for 1 md. permeability to nitrogen gas regained after the tests with foam. Table 5-4 lists regained gas permeability ranging from 87 to 95% of the original gas permeability after 4 hours of flowback. The data show no significant difference between foams containing 0 to 50 lb Table 5-4: Gas Permeability Regained in 4 Hours HPG/1,000 gal [5,991 g/cm3] in the liquid phase. This Original Permeability Foam Type Regain (%) (md) indicates minimal permeability damage caused by foam Water 0.3 91 to 93 fracturing fluids. Measure50% Methanol 0.3 92 to 95 ments of filter-cake deposition reported elsewhere 50 lb HPG/1,000 gal 0.13 87 to 91


5-11

Test Conclusions 1. Foam fracturing fluids that contain HPG in the aqueous phase are wallbuilding fluids. 2. The fluid-loss coefficient, CIII , is dependent on permeability, gel concentration in the liquid phase, and temperature of the core. 3. For foam fluids containing gel, little effect on CIII was observed by the variation of core length, foam quality, foam agent, or type of liquid phase. The effect of differential pressure was minimal. 4. For foam fluids with no gel, fluid loss is dependent on core length and differential pressure across the core. 5. The composition of the fluid passing through the core differs from the foam impinging on the core. Effluent composition is enriched in the liquid phase and is dependent on the concentration of gel in the foam. 6. Regained-gas-permeability tests indicate minimal damage caused by foam fracturing fluids.

Fracture Conductivity

kf ∝ d502φ5 , ................................................... (5-3) Notice the importance of the proppantpack porosity. In addition, a wider grain-size distribution of a given d50 reduces the permeabilityhence the modern tendency to market narrow sieve fractions, with a bigger mean grain size within a given nominal mesh range. For gas wells, non-Darcy (turbulence) flow effects in the propped fracture result in an extra pressure drop, ∆pt:

 ρν ∆p t ∝   µ

 β  ν , ....................................... (5-4) 

where v is the fluid flow velocity and β the non-Darcy flow factor, which is dependent on kf. Non-Darcy flow effects calculated from Guppy et al.15 for typical hydraulically fractured gas wells can reduce the effective fracture conductivity by more than a factor of 3.

Effective Closure Stress

Proppant Pack Permeability Many factors influence the effective proppant-pack permeability (kf): proppant grain size, effective closure stress acting on the proppant pack and formation face, multiphase flow effects, and fracturing-fluid residue damage.

Proppant Grain Size The permeability of a lightly stressed proppant pack is a function of the porosity of

5-12

the pack, φ, and the mean diameter of the proppant grains, d50:

The fracture conductivity dependence on effective closure stress (minimum in-situ stress minus pore pressure) cannot be assessed theoretically; empirical relations based on extensive proppant conductivity measurements over a wide range of conditions are required. The majority of these measurements have been carried out on the most popular proppant size (20/40 mesh) with low liquid flow rates (negligible non-Darcy flow effects) at low temperature and short measurement times. Limited data are available for measurements with gas. The most recent measurements were conducted at reservoir


temperature with long measurement times.16,17 Significantly lower proppant conductivities are measured for realistic reservoir conditions than reported for the early shortterm conductivity tests. Lack of reproducibility of absolute-permeability measurements between the various laboratories plagues this area of research because standard procedures for preparation-in particular pack porosityhave not been agreed on. The high closure stresses encountered in deeper wells require the use of artificial (intermediate- or high-strength) proppants to improve fracture conductivity. The recent long-term measurements discussed show a technical need for these stronger (and therefore more conductive) proppants at lower closure stresses (shallower depth). This is also true if coarser proppant sand grades are used in an attempt to increase fracture conductivity because crushing occurs at lower closure stress.

Multiphase Flow Effects Proppant-pack conductivity is normally measured with single-phase flow. Adding a second or third phase reduces the effective proppant-pack permeability to the original phase significantly. A proppant-pack permeability decreases by more than a factor of 5 if water-saturated gas (two phases) flows through the pack.

based fluids, emulsions, and foams. Fracturing-fluid residue in the proppant pack and filter-cake buildup at the rock surface reduce fracture conductivity. Crosslinked fracturing fluids result in more residue than polymer emulsion fluids. Laboratory tests with the latter fluids yield loose proppant grain pack that is virtually residue-free.18 By contrast, use of crosslinked fluids produces a proppant pack containing a lot of fibrous material between the grains, which are then glued together.

Filtercake Buildup During the fracturing operation, as highpressure fracturing fluid leaks away into the formation, a polymer and fluid-loss additive filter cake is formed. The filter-cake thickness is determined by the particular fracturing fluid used, the formation characteristics, the fracture-to-reservoir pressure difference, and the erosional effects caused by slurry being pumped along the fracture faces. During fracture closure, the proppant is embedded into the filter cake, making it difficult to remove the cake during production. A typical filter-cake thickness of 0.13 in. [0.5 mm] on each fracture wall will completely block a thin fracture propped with two layers of 20/ 40-mesh proppant. Such filter cakes occur, for example, when crosslinked fluids are used with diesel added as a fluid-loss agent.

Fracturing-Fluid Residue Damage The fracturing fluid is an essential part of a hydraulic fracturing treatment. It creates the fracture in the reservoir and transports the proppant. The fluid is very viscous and shows a controlled (restricted) formation leakoff to ensure an efficient fracturing operation. Fracturing fluids often include a breaker to reduce their high viscosity to low values to facilitate cleanup. Types of fracturing fluids include water-based fluids, oil-

Table 5-5: Global Ranges for ProppantPack-Conductivity Retention Factors Type

Range

Foams

>80

Polymer emulsion fluids

65 to 85

Gelled oils

45 to 70

Linear gels

45 to 55

Crosslinked HPG

10 to 50


Best

Worst

5-13

Polymer emulsion fluids do not give significant filter-cake buildup. Table 5-5 shows the global ranges that have been published for proppant-packconductivity retention factors with gas as the cleanup fluid.16,18 The most important variables that influence these ranges are the specific fracturing fluid, the fracture width (in particular when thick filter cakes are formed), and the cleanup (reservoir) fluid. Proppant permeability damage by crosslinked fluids substantially depends on the filter-cake buildup. It is minimized by use of an effective viscosity breaker; this is particularly necessary in shallow (low-temperature) formations. An inadequate breaker leads to virtually complete loss of the proppant-pack conductivity. On the other hand, aggressive breaker schedules (short break times) can provide high retention factors in excess of 80%. Such aggressive breaker schedules, however, can result in excessive proppant settling in the fracture before closure. A proper balance is required. Polymer emulsion fracturing fluids help because conductivity recovery is less sensitive to breaker efficiency. However, surfactants used in these fluids sometimes make the proppant pack oil-wet. Retained effective permeability to water (2% KCl brine is often used in tests) is much lower in such cases, as low as 30% in some cases. Proppant-pack conductivity increases the non-Darcy flow (turbulence) factor. This turbulence factor needs to be accounted for when optimum hydraulic fracture stimulations for gas wells are designed. Foam fluids contain only one-third to one-fourth the amount of water as a nonfoamed fracturing fluid. Even though this lesser amount of water represents less potential damage to the formation, the water still needs to be removed to minimize damage to the formation. Controlled flowback procedures are important for any fracturing treat-

5-14

ment, and they are especially important for foam fluids. The common practice for flowback of foam fluids has been to wait from 30 minutes to 4 hours before opening the wellhead valves to a small production choke. More recent techniques include opening the well immediately at a low rate. A properly stabilized foam fluid structure will remain intact with high viscosity after 4 hours under downhole conditions. Common enzyme or oxidizing breakers reduce only the viscosity of gelling agents and do not directly attack the stabilizing surfactants. Reduction of pressure at the wellbore will cause some migration of fluid, carrying proppant back towards the wellbore. Experience of most foam flowbacks has been that little proppant is produced if the flowback rate is kept low. The fact that so little proppant is produced indicates that the formation has closed near the wellbore, trapping the proppant and forming a bridge to prevent further production of proppant from the fracture. If high flowback rates are used, a proppant bridge may not be formed or else be eroded, and significant amounts of proppant can be produced with the potential to harm the formation, wellhead equipment, and personnel.

Treating Pressure Response The pumping pressure experienced at the wellhead during a stimulation treatment is the result of several factors: pw = pbht + ppf - ph ,....................................... (5-5) where pw is wellhead pumping pressure (psi), pbht is bottomhole treating pressure (psi), pf is fluid friction pressure in tubular goods (psi), ph is hydrostatic pressure (psi/ft), and ppf is perforation friction (psi).


Table 5-6: Fluid Velocity Change for 1-bbl/min Fluid Rate Change Tubular Configuration

Velocity Change (ft/sec)

4.892-in. ID casing

0.72

4.00-in. ID casing

1.07

2.992-in. tubing

1.92

2.441-in. tubing

2.88

2.375-in OD x 4.892-in. ID tubing/casing annulus

0.94

did not use limited-entry design, so ppf will be considered negligible. The hydrostatic weight of the fluid column helps reduce the surface pumping pressure required to fracture the formation. N2 foam fluids always have a lower density than water. The density of CO2 foams will be significantly higher than that of N2 foams and similar to that of water. The addition of proppant has a large effect on pH, especially at latter proppant stages. Table 5-7 shows the effect of sand on hydrostatic pressure.

Constant Internal Phase The bottomhole treating pressure (BHTP) The observation of high friction pressures is a function primarily of formation stresses for foams pumped down small tubing replus pore pressure. BHTP may increase quired re-examination of the structure of during a treatment as a result of laminar-flow foam fracturing fluids. For nonfoamed fracfriction within the fracture. turing fluids, when proppant is added to the Fluid friction in turbulent flow down the fluid, the proppant causes no major change in tubular is a function of flow rate, tubular the viscous character of the fluid. Foam, diameter, fluid density, and fluid viscosity. A however, is a two-phase structured fluid, change in flow rate of 1 bbl/min has a relaconsisting of a gaseous internal phase and tively small effect of velocity in casing, but it a liquid external phase. Discrete gaseous has a larger effect in small tubing (Table 5-6). bubbles are surrounded by a continuous, thin The use of high pumping rates, small-diamliquid coating. The viscosity of the foam fluid eter tubing, high sand concentrations, highis a function of the foam quality, as shown in quality foams, and high gel concentrations Fig. 5-6. Quality is the ratio of gas volume for high foam viscosity all increase pf. to gas-plus liquid volume at a specific temReidenbach et al.19 gave a correlation for perature and pressure. A similar viscosity turbulent-flow friction pressure of foams. They Table 5-7: Hydrostatic Pressure of 70 Quality N2 and CO2 provide a relationship Foam Fluids at 100°F, 5,000 psi, Containing Sand developed from N2 foam data, but the equation Sand (lb/gal) N2 Foam (psi/ft) CO2 Foam (psi/ft) works satisfactorily with 0 0.226 0.411 CO2 or proppant if the proper density is used. 2 0.302 0.472 Perforation friction is 4 0.367 0.523 important when the 6 0.422 0.567 number of perforations is limited to restrict fluid 8 0.470 0.606 flow to certain zones. The 10 0.512 0.639 field examples cited here 12


0.548

0.667

5-15

viscosity slightly, adding it to a high-quality foam will cause a larger increase in viscosity. For example, addition of 1 lb/gal [120 g/m3] sand to a 40 lb/1,000 gal [4,793 g/m3] linear gel will increase the viscosity by 5 cp at 100 seconds-1. Addition of 1 lb/gal [120 g/m3] sand to a 70quality, foamed, 40 lb/ 1,000-gal [4,793-g/m3] gel will increase the viscosity by 14 cp at 100 seconds-1. To maintain a constant viscosity fracturing fluid, the balance between the Fig. 5-6: Apparent viscosity of CO2/water two-phase fluid as a function of CO2 internal and external quality. phases must be kept constant, hence the term relationship exists for two-phase liquid/ constant internal phase. liquid emulsions. As the percentage of interFig. 5-7 illustrates the concept of constant nal phase increases in a two-phase fluid, the internal phase. Fluid A is a conventional fluid viscosity increases. foam pad fluid (no proppant) containing a When a solid proppant particle is added fixed volume of gas and liquid. Fluid B is a to a two-phase foam, it is readily apparent proppant-laden fluid with solid added while that a solid particle cannot become part of the gas and liquid volumes are held constant. continuous liquid phase. Rather, it Conventional Constant-Internal-Phase must exist as a discrete entity, alongside the gas Solid bubbles, Because Solid Solid the solid particles occupy volume, they produce the Gas Gas Gas Gas Gas effect of increasing the quality and hence the viscosity. Although the Liquid Liquid Liquid Liquid Liquid addition of propA B C D E pant to a nonfoamed gelled Fig. 5-7: Diagram of fluid ratios for conventional and constant-internal-phase design. fluid may increase

5-16


During a fracturing treatment, these volumes are pumped in a given time, so the ratios also relate to pumping rates. The volume of internal phase (gas plus solid) in Fluid B is greater than that of Fluid A, although the liquid is constant, and would result in higher viscosity and a higher downstream rate. This condition has often led to excessive friction losses, higher wellhead pressures, and premature job termination. An attempt to reduce solid, liquid, and gas rates proportionally to make the downstream rate the same as the pad does not solve the overall problem. Although the ratios in Fluid C are the same as in Fluid B, the internal phase ratio of Fluid C is higher than that of Fluid A, so the viscosity of Fluid C is higher than that of Fluid A and will give higher friction pressure. In addition, adjusting all three ratios increases operational difficulty. An example of the viscosity increase caused by proppant addition may be calculated.5 Addition of 5 lb/gal [599 g/m3] sand to a 70-quality foam containing 40 lb/1,000gal [4,793 g/m3] base gel will increase the internal-phase fraction to 75.6%. The apparent viscosity of the fluid will increase from 325 to 445 cp at 40 seconds-1. A solution was proposed to keep both downstream flow rate and viscosity constant. When solid proppant is added, a constant liquid rate should be maintained but the gas flow rate should be decreased sufficiently to equal the absolute solid flow rate. Application of the constant-internal-phase concept has allowed much better control of foam fracturing treatments down small tubing, especially with CO2.

liquid phase in the usual manner at a blender unit. The resulting slurry goes through highpressure pumps and any additional surface equipment and approaches the wellhead. High-pressure N2 or CO2 is added just before the wellhead and dilutes the sand concentration by several-fold. Standard field blending equipment can routinely handle proppant concentrations of 20 lb/gal [2,397 g/m3] at low fluid rates, and higher concentrations are possible for short periods. Following conventional foam design, where proppant is not considered part of the foam, a 67-quality foam would be limited to about 7 lb/gal [839 g/m3] sand downhole. However, use of the constant internal-phase design decreases N2 or CO2 as proppant concentration is increased. Therefore, the dilution effect of the gas is less, and higher downhole proppant concentrations may be reached. For example, starting with a 70-quality foam pad, proppant concentrations of 12 lb/gal [1,438 g/m3] have been placed successfully during foam stimulation treatments. A potential disadvantage of constantinternal-phase design is that the fluid pumped last contains less gas to assist in fluid return. Even in nongas-assist fluids, however, the easiest fluid to recover is the fluid pumped last. The fluid in greatest need of gas assist is the first fluid pumped. Therefore, the constant-internal-phase design loses very little potential for fluid recovery by reducing gas during the latter proppant stages.

Increased Proppant Concentration

Field simulation treatments considered here were pumped down one of three tubular configurations: casing, annulus, or tubing. Surface pressure responses for foam fluids may differ for each configuration. Table 5-8 lists examples of treatments according to tubular configuration. The numbers were

The constant-internal-phase design has allowed higher proppant concentrations to be pumped than conventional foam designs. In a foam stimulation treatment, sand is not added directly to the foam fluid but to the

Field Treatment Results of Constant Internal Phase


5-17

of 4.5-in. casing. The smaller diameter casing and Fluid Volume Proppant Depth Tubular ID Rate Pressure higher foam rate Fig. Type* (quality) (1,000 gal) (lb x 1,000) (ft) (in.) (bbl/min) (psi) should show more 8 75/67 N2 Conv 41 71 6,200 4.892 15 3,800 pronounced fric9 70 N2 CIP 130 306 10,000 4.000 20 4,800 tion effects than the previous 10 70 CO2 CIP 67 72 9,800 4.892 25 5,000 example. During 2.375 x 11 76/67 N2 Conv 33 70 5,600 13 4,000 4.892** treatment, the downhole foam 12 75 N2 Conv 76 128 10,000 2.992 14 7,500 rate including sand 13 70 CO2 Conv 86 144 10,100 2.992 15 5,600 was increased by 14 70 CO2 CIP 57 104 9,500 2.992 15 6,300 10% from the * Conv= conventional; CIP= constant internal phase designed 10 bbl/ ** Annular space between 2.375-in. tubing and 4.892-in. casing min. As sand concentrations increased from 1 to reproduced from actual treatment pressures 7.5 lb/gal [120 to 899 g/m3], the pumping and rates collected by data-acquisition compressure dropped stepwise from 6,000 to puters on location. 4,500 psi. When sand feed was stopped and The first example in Fig. 5-8 is a conventhe foam slurry was flushed from the casing, tional N2 foam treatment down 6,200 ft of 5.5 wellhead pressure increased by 2,000 psi in. casing. A 75-quality foam pad was initially because of the loss of hydrostatic pressure of pumped, followed by 67% quality sand-laden fluid. The wellhead pressure increased throughout the pad but declined as sand was added, increasing the hydrostatic weight of the foam column. Once the foam column stabilized in the wellbore, pumping pressure remained stable until the sand addition was stopped. Pumping pressure then increased from 3,400 psi for sandladen fluid to 3,900 psi for neat foam owing to loss of hydrostatic weight. Fig. 5-9 shows an example of a constantFig. 5-8: Conventional 75/67-quality, N 2 foam pumped down 6,200 ft of 5 1/2-in internal-phase N2 foam casing. treatment down 10,000 ft Table 5-8: Treating Information for Conventional and Constant-Internal-Phase N2 and CO2 Foams

5-18


proppant increased the hydrostatic weight of the fluid column. Fig. 5-11 shows an example of a conventional N2 foam fracturing treatment down 5,600 ft of 2.375 x 4.891 in. annular space. Wellhead-pressure rise during a 75-quality foam pad was followed by a pressure decline upon switching to sandladen 67-quality foam. The slight decline was caused by the combined effect of slightly lower foam viscosity and increased hydrostatic Fig. 5-9: Constant-internal-phase, 70-quality, N2 foam pumped down 10,000 ft of 4 1/2-in casing. pressure with added sand. During later sand stages, wellhead pumpsand. The small differences in fluid friction ing pressure increased in spite of increased pressure between Figs. 5-8 and 5-9 are overhydrostatic weight. Fluid friction pressure, as shadowed by hydrostatic effects resulting from the proppant. Fig. 5-10 gives an example of constant internal-phase CO2 foam treatment down 9,800 ft of 5.5-in. casing. This example shows the control of clean gel, liquid CO2, and proppant rates to give a constant downhole foam slurry rate. The clean-gel rate remained constant and the CO2 rate decreased as the proppant rate increased. The pumping pressure rose to a maximum by the end of the pad stage but steadily Fig. 5-10: Constant-internal-phase, 70-quality, CO2 foam pumped down 9,800 ft decreased as the added of 5 1/2-in. casing.


5-19

psi. From the upward trend in treating pressure, an imminent sandout might have been expected but did not occur. If a lower maximum allowable treating pressure had been set for this well (e.g., if lighter wellhead equipment or lighter tubing had been used), the stimulation treatment would have been stopped prematurely and not completed. Fig. 5-13 shows an example similar to that in Fig. 5-12. In this case a conventional CO2 foam Fig. 5-11: Conventional 75/67-quality, N2 foam pumped down 5,600 ft of 2 3/8 x 5 1/2-in. annulus. treatment was pumped down 10,100 ft of 3.5-in. tubing. The pumping calculated from Eq. 1 increased from 666 psi pressure began to rise as soon as sand was in the pad to 776 psi in the 3-lb/gal proppant added to the foam. The friction pressure of stage to 984 psi in the 4.5lb/gal proppant stage. This treatment ended in a sandout. Fig. 5-12 gives an example of conventional N2 foam treatment down 10,000 ft of 3.5-in. tubing. A pumping pressure of 7,500 psi was established during the pad and continued into the early sand stages. During the later sand stages, the hydrostatic weight increase caused by the additional sand did not offset the increase in foam-slurry friction pressure, so wellhead Fig. 5-12: Conventional 75-quality, N 2 foam pumped down 10,000 ft of 3 1/2-in. treating pressure (WHTP) tubing. rose to more than 9,000

5-20


while the slurry rate increased and the CO2 rate decreased. The WHTP was very well behaved. Both wells in this formation had a nearly constant pressure response during injection of the CO2 foam pad. Once sand addition began, the treating pressure of the conventional treatment of Fig. 5-13 rose at a substantial rate with increasing proppant addition, whereas the treating pressure for the constant internal-phase treatment of Fig. 5-14 Fig. 5-13: Conventional 70-quality, CO2 foam pumped down 10,100 ft of 3 1/2in. tubing. actually declined with proppant addition. Although theses are sand-laden 70-quality CO2 foam has often only a few examples, they are consistent with been reported as being higher than that of N2 field experience that WHTPs are more foam, and the 3,500 psi rise in pressure compared with the previous case tends to confirm this report. One might suspect an imminent sandout in Fig. 5-13, but such was not the case. Fig. 5-14 demonstrates the corrective action that a constant-internal-phase design can have over conventional foam design. A 70% constant-internalphase CO2 foam treatment was pumped down 9,500 ft of 3.5-in. OD tubing. Table 5-9 gives the pumping schedule for the treatment of Fig. 5-14. Fig. 5-14: Constant-internal-phase, 70-quality, CO2 foam pumped down 9,500 ft Note that the clean-gel of 3 1/2-in. tubing. rate remained constant,


5-21

The fracture conductivity may be inFoam Volume Proppant Clean-Gel Slurry Rate Liquid CO2 creased by Stage (gal) (lb/gal) Rate (bbl/min) (bbl/min) Rate (bbl/min) enlarging the 1 25,000 0 4.50 3.50 10.56 propped fracture 2 5,000 1 4.50 5.15 9.93 width, bf, by application of 3 5,000 2 4.50 5.75 9.34 high proppant 4 7,500 3 4.50 6.29 8.80 concentration. 5 10,000 4 4.50 6.80 8.31 This has become popular during 6 5,000 5 4.50 7.27 7.84 the last few years. A dimenevenly controlled with a constant-internalsionless fracture conductivity (CfD) of 15 is a phase design, especially in treatments at high proper design value for (pseudo-) steadyflow rates down small tubing. state flow conditions. This value is often not achieved in practice. Moreover, the fracture Conclusions conductivity found from production-test interpretation on hydraulically fractured The use of the constant-internal-phase wells is often an order of magnitude smaller design has proved successful for foam treatthan expected. ments. The design technique provides for a Tight reservoirs with high initial transient decreasing N2 or CO2 rate as proppant rate is production rates require higher dimensionincreased. Because all internal phases are less fracture conductivities than indicated considered to be the same, higher proppant above because these transient rates can last 3 concentrations of up to 12 lb/gal [1,438 g/m ] for more than one year and significantly have been placed successfully and with better contribute to the economic success of the control of wellhead pumping pressure than fracturing treatment. More sophisticated in conventional designs. tools, such as type curves or reservoir simulators, are required to assess optimum fracture conductivity in these cases. Table 5-9: Constant-Internal-Phase Design for CO2 Foam Injected Down 9,500 ft of 2.992-in. ID Tubing

Fluid Recovery The relationships between the productivity improvement factor, Fp, obtained by hydraulic fracture stimulation and the dimensionless fracture conductivity, CfD, of the propped fracture have been published by Prats.20 CfD is proportional to proppant-pack permeability, kf, and fracture width bf: CfD ∝ kfbf, ........................................................ (4)

5-22

Treatment Designs for Hydraulic Fracturing PROP Hydraulic Fracture Design Program Halliburtons premier hydraulic fracture design program, PROP, has many features and options that allow good engineering design of most stimulation processes. For


stant internal-phase fraction, constant clean quality, etc.). Note: because the calculations allow qualities or IPFs to be Proppant's Volume isspecified separately for each treatment stage, the program Added to is flexible enough to allow design using almost any Added to method of proportioning the Included in components (gas, liquid, and proppant); some techAdded to niques are simply more automatic. Foam friction and hydrostatic calculations are made along the length of the wellbore so as to (1) convert specified downhole rates and qualities into component rates at surface or standard conditions (i.e., pressures and temperatures), and (2) help determine fluid temperature at the perforations. Up to five tubing/casing strings can be considered. Proppant is considered in the friction as well as hydrostatic calculations. Perforation friction is considered. The user is allowed to specify the perforation discharge coefficient as well as the number of (open) perforations. Gas requirements for the treatment are calculated. N2 foam and CO2 treatments may be designed. For proppant settling calculations, a settling velocity correlation is used. One consequence of this calculation is that if the gravitational forces on the proppant particles are sufficient to overcome the yield stress of the foam, settling is predicted to occur; otherwise, it is not. Foam rheology is modeled using the three-parameter Herschel-Bulkley model. The values of n, K, and τ0 are determined from the correlations

Table 5-10: PROP Foam Fracture Treatment Design Methods

Option

Total Rate

0

Varies

Liquid Rate

Clean Quality

Constant Constant

Internal Phase Fraction Varies

1

Constant Constant

Varies

Constant

2

Constant Constant

Varies

Constant

3

Constant

Constant

Varies

Varies

example, PROP offers four methods of designing a foam fracture treatment by direct data entry and output, as shown in Table 510. Modifications, such as binary foam and variations on the above standard options, are readily accommodated by PROP. The output of PROP assists designers, customers, and operators in completing a successful job by including functional information such as rates for components and proppant schedules as a function of time on the job. Options include choosing outputs to present foam quality at bottomhole static temperature as specified or calculated. Other features include the following: Options to specify foam rates and qualities (internal-phase fractions) (1) in the fracture adjacent to the perforations (at calculated perforation temperatures and BHTP), (2) at the fracture tip (at BHST and BHTP), (3) at an estimated average fracture temperature, or (4) at a user-selected temperature. Allows differing qualities or internalphase fractions (IPF) from stage to stage. Allows differing injection rates from stage to stage. Design calculations may be made by a variety of different techniques (con-


5-23

developed by Reidenbach19 et al. that take pi into considerPressure ation the gelphase rheology and the com∆p position of the (flow convergence) internal phase (N2 or CO2). ∆ pskin The rheological parameters can change not rs xf only from stage to stage, Radius but with time and temperature as the Fig. 5-15: The pressure drop near the wellbore, due to radial convergence and damage, can initiate formation failure. FracPac II technology focuses on minimizing properties of this pressure drop for a given flow rate. the external phase (i.e., base gel) change. FracPac takes advantage of both parameters. Notes: A few of the mentioned features Low values of Youngs Modulus allow for are not available at the moment, but relatively wide fracture widths, as compared will be available in the next major to more stiff, or higher modulus, rocks. release of the PROP program (now in Combining tip screenout with a low Youngs preliminary testing). Modulus helps create maximum fracture width. High permeability allows significant FracPac II fluid leakoff during the screenout mode, resulting in an increased concentration of A Halliburton FracPac II treatment is proppant in the fracture at the end of the job. designed to create a short, wide, and highly The result is a maximum amount of proppant conductive fracture that will enhance hydroplaced per square foot of fracture area. This carbon production in poorly consolidated counteracts the effect of permeability damage formations. FracPac II can help alleviate and improves sand control. permeability damage and sand migration Candidate Selection production barriers. FracPac II also offers advantages over conventional gravel-pack The FracPac II process can be applied to treatments by avoiding near-wellbore damreservoirs where the rock is anticipated to age and providing longer term, more successfail, leading to sand production. Assessment ful, sand/fines migration control. FracPac II of the failure mechanisms for a given reserstresses the use of modern technology for voir will supply information critical for a design and job execution. successful design. Core samples and pressure Two reservoir properties commonly analysis, along with drilling and/or compleassociated with poorly consolidated rocks are tions records, should be analyzed. With low Youngs Modulus and high permeability.

5-24


Gravel-packed wells that have lost productivity due to pore collapse (Fig. 516) Normal decline Poorly consoliFracPac II treatment dated reservoirs Apparent gravelexhibiting permeabilProductivity pack failure ity damage from drilling/completions fluids Regravel pack Factors that will adversely affect selection of a candidate well are the location of oil/water Time contact or gas/oil contact. These should Fig. 5-16: Productivity decline, which may appear to be due from gravel pack failure, may actually result from pore collapse. A FracPac treatment would be the only way be considered when of reaching past this type of formation damage. treating a specific zone. Low stress adequate information, the hydraulic fracture contrast of boundary layers may result in too that is necessary to prevent failure can be much height growth. The manner in which a designed. wellbore is perforated may have a negative Reservoir candidates should have a impact upon a treatment. Wellbore tubulars permeability of sufficient magnitude (normust be of sufficient strength to withstand mally > 5 mD) so that fracture conductivity is the execution of the job. As with any complemore important than fracture length. The tion, the quality of the cement job, both in following list represents conditions that bonding quality and TOC, should be examwould make a well suitable for a FracPac II ined for possible inter-zonal communication. treatment: Wireline Logging Reservoir rock that fails due to high pressure drawdown developed near Knowing key rock properties is essential the wellbore (Fig. 5-15) for a successful design. Wireline logging data Reservoir rock that fails due to pore should be used to obtain this data if laboracollapse tory analysis is not available. The optimum Reservoirs that have a history of sand situation would occur when logging informaproduction tion is available to correlate with laboratory Reservoirs that require restricted tests. The recommended suite of logs follows: production rates to prevent sand migration Open Hole Logging Reservoirs that are overpressured, Full wave sonic log/Dipole sonic resulting in the sand being poorly Density log compacted Formation tester Formations that tend to have water Cased Hole Logging coning problems


5-25

Full wave sonic cased hole tool/ Dipole sonic (the quality of the cement bond will affect the efficiency of the tool in this case) The post processing of this information provides the following logs:

Formation Strength This log calculates the drawdown necessary to cause rock failure and sand production. The Mohr-Coulomb failure model is used. Information from the sonic and density logs are used in this analysis. For cased hole logging, values from a previously run density log must be used, or estimated from other sources.

Fracpressure Log The Fracpressure log calculates the least principal horizontal stress. In addition to the stress profile, a complete listing of all the critical rock properties are presented. These include Youngs Modulus, lithology, water saturation, average interval pressures, and fracture barrier identification. The formation tester, used to measure pore pressure, increases the accuracy of stress calculations. For cased-hole logs, the value of pore pressure must be estimated. Stress data, from the Fracpressure log, can be input directly into the FracPac II design simulator.

Perforating The manner that a well is perforated may affect a FracPac treatment. The phasing of the perforations and wellbore deviation through the pay are two of the most critical factors. If the option is open as how to perforate the reservoir for a FracPac candidate, this portion of the design should be carefully studied. For vertical completions, multiple phase perforating (other than 0/180° phasing) will

5-26

most likely result in a high percentage of holes taking little or no sand during the treatment. At the least, this will necessitate a gravel pack to be incorporated following the FracPac. If the gravel pack fails to exclude fines from entering the wellbore from untreated holes, the success of the FracPac may be obscured. The recommended approach for vertical wells is to perforate with 0/180° phasing. Preferably, the phases would be in line with the direction of fracture orientation. For deviated and horizontal wellbores, attention must be given to the fact that multiple fractures can be formed (Fig. 5-17). Unless the axis of the wellbore is closely aligned with the maximum principal horizontal stress, multiple fractures will likely occur. The result of this is premature screenouts. The recommended approach, for wells drilled in the direction of the minimum horizontal stress, is to cluster the perforations within a 1- or 2-ft section to increase the likelihood that a one-fracture system will develop.

Fracture Design To obtain a successful FracPac II design, the following parameters must be considered: Fracture geometry Tip screenout Fluid loss Injection rate Proppant selection Proppant concentration Proppant embedment Fluid viscosity Concerning stimulation, fracture length is not as important as the permeability contrast between the fracture and formation. Fracture length should be adequate to extend beyond near wellbore damage and the area where radial convergent flow occurs. In many cases, a fracture length of 30 to 50 ft [10 to 16.4 m] may be quite adequate for successful results. There may be circumstances where increased fracture length is required to obtain


Fig. 5-17: For deviated wellbores, especially those that are horizontal, the relationship of the principal stresses will have a major impact upon fracture initiation.

sufficient fracture width for proppant placement. Also, if pore collapse is anticipated as pressure is depleted (causing permeability reduction), fracture length should be extended. Once fracture length becomes important for stimulation purposes, and can be economically justified from production improvement, the process should be considered as a fracturing job rather than a FracPac. Fracture width needs to be maximized. The goal is to place the highest possible amount of proppant per square foot of fracture area. Fluid viscosity and pressure increase from tip screenout are the factors

that will govern fracture width. The propped fracture width should be close to the created width. It would be best for the fracture height to be limited within the zone of interest. Accurate stress logs, used with a simulator, will serve to estimate the effect of boundary layers on height growth. In the absence of stress barriers, a penny-shaped fracture can be expected. In this case, treatment volume would control height growth. Fracture length and height would maintain a constant proportion.


5-27

Fluid loss at the fracture wall can make it difficult to maintain fracture extension. For permeable formations, a fluid efficiency of 10 to 20% is quite common. As the area of the fracture grows, the total leakoff may actually increase to the point that it equals injection rate. Increasing the injection rate, maximizing fluid viscosity, and using fluid-loss additives will help improve the fluid efficiency. Of these options, changing injection rate will have the greatest impact on fluid efficiency. However, this option may also result in undesirable height growth and increase the job cost. High fluid loss will provide the benefit of being able to pack the fracture with sand during the screenout mode. Proppant selection should focus upon maximizing the permeability of the propped fracture, especially near the wellbore. For a given production rate, drawdown will decrease as flow capacity in the fracture increases (Fig. 5-15). For permeable formations, it is very difficult to obtain sufficient flow capacity to change the radial pattern of fluid flow to the wellbore. To do so requires significant concentrations of large proppant. When sand/fines migration is initiated at reduced drawdown, the proppant will need to be selected based upon the sieve analysis of the formation. Normally a proppant size is determined by multiplying the mean diameter of proppant grains (d50) of the formation sieve analysis by five or six. This will result in a flow capacity that is less than optimal. However, the surface area over which the formation is screened will be much larger than with a gravel pack. The job should be designed to reach high proppant concentrations early in the job. Maximum concentration will need to be based on numerous factors including rate, fluid type, and field experience. High proppant concentrations will minimize the volume of fluid lost to the formation to obtain a packed fracture. If for some reason screenout

5-28

does not occur, the existing proppant concentration will offer significant benefits. The use of PropLok coating system is highly recommended. This curable resin, added on-the-fly at the blender tub, will alleviate flowback problems associated with high proppant concentrations and low closure stresses. PropLok may also reduce proppant embedment and provide an additional way of controlling sand migration. Proppant embedment will reduce the propped fracture width significantly in many instances. Narrow propped frac widths may actually allow the fracture to heal. This is an additional reason for maximizing frac width and placing as much proppant as possible per unit area of fracture. The fluid used for a FracPac will need adequate viscosity to create a wide fracture and place the proppant. The 60 to 80 lb/1,000 gal [7.2 to 9.6 kg/m3] linear HEC gels popular for gravel packing will work for FracPac purposes. Other fluid systems, such as BoraGel/Hybor Gel, PUR-GEL, and Kleen Gel II, will offer superior viscosity. Often, this is accomplished at a reduced cost. As formation permeability increases, the deeper the fracturing fluid invades into the fracture wall. Also, stabilized fluid-loss control is reached earlier in time as permeability increases. There are indications that complexed gel systems may aid fluid-loss control and reduce the depth of invasion at the fracture face. The use of N2 or CO2 foam should also be considered for FracPac treatments. Foam fluids may act to control excessive leakoff without the aid of additional fluid-loss additivies. Less liquid is available to cause permeability damage at the fracture face. Improved flow capacity, in the propped fracture, can be expected due to less polymer usage. The gas phase of the foam may also act to aid in fluid recovery.


3-D Fracture Design Simulator The FracPac 3-D Fracture Design Simulator program uses planned screenout as a way to place large amounts of proppant per square foot of fracture area. Injecting proppant-laden fluid into the formation will continue after tip screenout has begun. The job will continue until a predetermined increase in bottomhole pressure is reached. The simulator tracks changes in fracture geometry and proppant placement. Formation characteristics and pumping schedule are required to be input by the user. A net increase in bottomhole pressure must also be entered. The program provides for three stress options. The first option is to enter an aver-

age stress for the upper boundary and one for the lower boundary. The second option is to enter stress vs. depth pairs for the boundary layers. The third option is for the simulator to directly read stresses from digitized full-wave sonic log data files.

Prefracture Testing Prefracture testing incorporates a series of pumping jobs, prior to the FracPac, that yield valuable information about the target reservoir. For extremely short fracture lengths, these tests may not prove to be economical. However, as job volumes increase and designs call for extended fracture lengths, these tests will prove beneficial to overall success.

Fig. 5-18: Example of a FracPac 3D Fracture Design Simulator output.


5-29

Of the prefracture testing techniques, the minifrac yields the most useful information but will cost the most. The most valuable piece of data obtained from the minifrac is the fluid-loss coefficient. A temperature survey or radioactive tracer log may be run in conjunction with the minifrac for determining fracture height. A dual minifrac may be used for determining both the Cw and spurt loss.

assembly is set, the four gravel-pack positions are obtained by reciprocating the work string. No rotation is required throughout the entire operation. Before performing a FracPac II treatment, the effect of pressure and temperature on the tubulars should be checked by calculating expected tubing contraction.

Example Procedure

Downhole Tools Ideally, FracPac II treatments are performed using open-ended tubing that allows monitoring bottomhole treating pressure from the annulus. Niether screens nor gravel packing would be required as a part of the completion system. Where a treating packer is required, it is useful to incorporate a downhole pressure gauge to record bottomhole treating pressure. This data can then be analyzed after the job for trends in pressure that are not affected by tubular friction. This would be especially beneficial for the first jobs performed in a field. Gravel packing may be required in very poorly consolidated reservoirs, in instances where perforation phasing and density would require a gravel pack, and in deviated wellbores where significant footage is perforated. Special tools will be required. Otis Sand Control provides a complete line of sand control screens manufactured by Howard Smith Screen Company. These include all-welded wire wrap and Sinter-Pak screens. The Sinter-Pak screen design excels in resisting bending and compression stresses encountered in deviated or horizontal holes. Such screens are also more efficiently cleaned with acid than other designs. Multi-position gravel-pack systems are designed to provide a variety of operating positions. Multi-position tools allow the Versa-Trieve gravel-pack assembly to be run and set hydraulically. Once the packer and

5-30

The following is an example of a recommended procedure for executing a FracPac II design on location. This includes downhole equipment for gravel packing. 1. If required, pull existing gravel pack from well. 2. Go in hole with tubing and circulate well clean with filtered completion fluid. Pull out of hole. 3. Pick up gravel-pack assembly (screen, packer, and multi-position tool). 4. Go in hole slowly until two stands off depth. Make slow pick up and slack off and record weight indicator readings. 5. If applicable, tag sump packer and verify position. 6. Rig up Halliburton Services pump equipment. 7. Test surface lines to necessary maximum. 8. Break circulation by pumping down workstring. 9. Hydraulically set packer at required depth. 10. Calculate differential pressure between slurry weight of final proppant concentration and annular fluid over depth of workstring. Test tubing/ casing annulus to 500 to 1,000 psi above this value. 11. Reverse one tubing volume with filtered completion fluid. 12. Hold prejob safety meeting to cover sequence of job events.


13. Break circulation to establish lower, upper, and reverse positions; record all rates, volumes, and pressures. 14. Locate multi-position tool to reverse position, and pump a tubing cleaning treatment (pickle treatment). 15. Place tool in reverse position, and spot minifrac fluid to end of tubing. 16. Perform first portion of dual minifrac with an initial volume equal to 20% of the FracPac treatment volume, or 100 gal per gross foot minimum. Use identical fluid and pump rate as planned for the FracPac. Incorporate a radioactive isotope. 17. Monitor shut-in pressure for two to four times pump time. 18. Execute second minifrac with approximately 13 to 15% of the FracPac volume. Monitor shut-in as with first job. 19. Rig up and run gamma ray and temperature log. 20. Calculate closure pressure and fluidloss coefficient from minifrac and log data. 21. Adjust FracPac procedure, if necessary. 22. Execute FracPac: Monitor treating pressure for BHTP trends. If screenout mode is reached, reduce rate to stay below maximum wellhead treating pressure. Have additional fluid and proppant available so that job can be extended, in case screenout does not occur. If screenout has not occurred when proppant slurry left to inject equals the gravel-pack requirement, slow rate to 2 bbl/min to force screenout. Do not overflush. 23. If screenout does not occur, conduct gravel-pack operation.

Minifractures "Minifracture" treatments, or prestimulation injection tests, have been used to estimate fluid-loss characteristics since 1979. This technology has only recently been extended to foams. Meaningful minifracture analyses require fluids with similar or identical properties to the actual stimulation fluids. Conventional aqueous fracturing fluids are inappropriate for estimating the fluid-loss behavior of gasified fluids. Foams exhibit great compressibility and thermal effects during shut-in that can mask actual fluid-loss behavior. The capability to account for these effects and properly analyze the pressure response would be very beneficial in optimizing stimulation treatments using fluids foamed with N2 or CO2.

Minifracture Analysis Technique Analysis Theory Minifracture analysis techniques, mostly centered on the determination of fluid efficiency and alternate fracture geometries, involve prediction of volume loss from pressure decline data following fracture extension. The relationship is based on the fracturing fluids being isothermal and incompressible. In practice, thermal and compressibility effects of fluids in the wellbore and fracture may become significant. Both of these effects may cause significant underestimation of fluid loss if the observed pressure decline is analyzed using conventional methods. Works by Soliman21 and Tan et al.22 show the need to correct pressure declines from waterbased fracturing fluids in high bottomhole temperature (usually above 250°F) wells. This technique generates an effective pres-


5-31

sure decline for analysis. The magnitude of the correction increases as fluid loss decreases or bottomhole temperature increases. A technique similar to that presented by Soliman is used to estimate the effective pressure decline. The equation is presented below:

(

)

∆Ρeff = 1 + C p Pavg ∆Ρ +

Ct dΤ Ρ(t ) dt , (5-7) ∫ βs dt

and volume fraction would be unity. The βs term in Eq. 5-7 is defined as the ratio of average to net wellbore pressure. For various geometry models, βs was derived as follows:23

( 2n' + 2) / (2 n' +3 + a ) for Pk  β s =  0.9 for Cz 2 3π / 32 for Radial 

   , (5-8)  

The above equation was used by Soliman where n´ is the power-law exponent for to obtain his effective pressure decline; fluid and a is the viscosity constant. however, the pressure term was changed The viscosity constant, a, can range from 0 from the difference between bottomhole for uniform viscosity fluid to 2 for fluids that pressure and fracture-closure pressure to the strongly degrade with temperature. difference between the bottomhole pressure The correction technique in Eq. 5-7 requires and the reservoir pressure. The change is a thermal recovery profile during shut-in to significant, since the magnitude of the correcdetermine the thermal effect on the pressure tion is based on this pressure drop. Since the decline. The thermal recovery profile has a modification, excellent agreement between significant effect on the correction of pressure effective pressure decline analysis and treatdecline. Bottomhole gauges are recommended ment simulations has occurred.22 For a for monitoring the actual bottomhole temperaconstant observed pressure decline this ture recovery profile. Mathematical temperachange makes the effective pressure decline ture simulators can be used to estimate the extrapolate to reservoir pressure rather than wellbore and fracture profiles. The current to fracture closure pressure. method assumes that the change in temperaCp, compressibility coefficient, and Ct , thermal expansion coefficient, have been calculated using the volumetric average of the fluid components. These coefficients are corrected with a volume fraction which is defined as the ratio of total volume (wellbore volume and fracture volume) to the fracture volume. If the wellbore is flushed with incompressible fluids, the wellbore volume is neglected in the correction Fig. 5-19: Pressure decline for pump-in/shut-in Test 2.

5-32


Minifracture Fluids Pump-in/Shut-in Test 1 40 lb CMHPG/Mgal in 4% KCl with 50 lb degradable particulate fluid-loss additive/Mgal Pump-in/Shut-in Test 2 70-quality CO2 foam with 40 lb CMHPG/Mgal delayed crosslinked fluid and 15 lb degradable particulate fluid-loss additive/Mgal foam Fig. 5-20: Observed and corrected pressure declines for pump-in/shut-in Test 2.

Treatment Fluids

ture within the wellbore and fracture length can be approximated by the thermal response at the perforations. A technique presented by Lee24 was used to obtain the fracture geometry and fluid loss coefficient once a corrected pressure decline was obtained. This technique makes use of an energy balance equation instead of the pressure difference between the ISIP and Pc to determine the fracture geometry. This technique gives more applicable values than previous methods.

Same as pump-in/shut-in Test 2.

Minifracture Test Results

The pressure decline for pump-in/shut-in Test 1 is presented in Fig. 5-19. A fluid-loss coefficient of 0.0055 ft/sqrt (min) was calculated with a closure pressure of 5,100 psi. The pressure decline for pump-in/shut-in Test 2 is presented in Fig. 5-20. A fluid-loss coefficient of 0.0032 ft/sqrt (min) was calculated with the observed data and a fluid-loss coefficient of 0.0050 ft/sqrt (min) was calculated with corrected data. The fluid-loss coefficient increased about 35% as a result of the correction. It is noteworthy that the foam yielded a lower fluid-loss coefficient than the base gel despite having a lower concentration of degradable particulate fluid loss additive.

Minifracture tests were performed in Webb County, Texas. Following are the results. 25

Well Data Reservoir Temperature: 215°F Reservoir Pressure: 4,200 psi Permeability: 1-10 md

Proposed Treatment Schedule

Actual Treatment Schedule

3,000 gal pre-pad

3,000 gal pre-pad

15,000 gal pad

10,000 gal pad

13,500 gal

5,000 gal

@ 2 to 8 lb/gal 16/20 ISP

@ 2 to 8 lb/gal 16/20 ISP

1,500 gal

2,000 gal

@ 8 lb/gal 16/20 ISP*

@ 8 lb/gal 16/20 ISP*

1,570 gal flush

1,100 gal flush *resin-coated


5-33

The tracer and temperature surveys indicated that gross fracture height was about one-half of what was expected. The treatment design was modified using the fluid-loss coefficient of 0.0050 ft/sqrt (min) and the smaller gross fracture height. The treatment screened out after about 70% of the flush was pumped. Simulation of the screenout with a two-dimensional fracture model yielded a fluid-loss coefficient of 0.0058 ft/sqrt (min). This is even higher than the value calculated with the corrected minifracture data. The treatment would probably have screened out much sooner if the uncorrected fluid loss coefficient had been used to modify the design.

Conclusions Foam fluids have established their value as low damage fracturing fluids. Foams have good inherent fluid-loss control characteristics. Foams containing polymers leave a much thinner gel filtercake residue than nonfoamed fluids. The proppant bed regains a high percentage of conductivity after treatment. Foams have a low water content, so there is less aqueous fluid to recover from the formation after the fracturing treatment. Gas in the foams expands to assist in recovery of treatment fluids. The rheology of foams has been characterized. Crosslinked foams provide easier placement of proppant in a formation than noncrosslinked-gel foams. Constant-internalphase designs provide higher proppant concentrations downhole. The positive benefits of clean foam fluids are partially offset by the slightly higher cost of a foam fracturing treatment. But where formation damage is a major factor in selecting a fracturing fluid, foams are the fluid of choice.

5-34

References 1. Warnock, W.E., Harris, P.C., and King, D.S.: Successful Field Applications of CO2-Foam Fracturing Fluids in the Arkansas-Louisiana-Texas Region, JPT (Jan 1985) 80-88. 2. Bleakley, W.B.: Mitchell Energy Foam Fracs Tight Gas Zones, Pet. Engr. Intl. (Dec 1980) 24-26. 3. Harris, P.C., Klebenow, D.E., and Kundert, D.P.: Constant-Internal-Phase Design Improves Stimulation Results, SPEPE (Feb. 1991) 15-19. 4. Harris, P.C.: A Comparison of Mixed Gas Foams With N2 and CO2 Foam Fracturing Fluids on a Flow Loop Viscometer, paper SPE 20642 presented at the 1990 SPE Annual Technical Conference and Exhibition, New Orleans, Sept. 23-26. 5. Mitchell, B.J.: Viscosity of Foam, PhD dissertation, Univ. of Oklahoma (1970). 6. J.L. Gidley, et al., Ed.: Recent Advances in Hydraulic Fracturing, , Monograph Series, SPE, (1989) 12 198-209. 7. Borchardt, J.K., et al.:Surfactants for CO 2 Foam Flooding, paper SPE 14394 presented at the 1985 SPE Annual Technical Conference and Exhibition, Las Vegas, Sept. 22-25. 8. Nikolov, A.D., et al.: The Effect of Oil on Foam Stability: Mechanisms and Implications for Oil Displacement by Foam in Porous Media, paper SPE 15443 presented at the 1986 SPE Annual Technical Conference and Exhibition, New Orleans, Oct. 5-8. 9. Norman, L.R., Hollenbeak, K.H., and Harris, P.C.: Fracture Conductivity Impairment Removal, paper SPE 19732 presented at the 1989 SPE Annual Technical Conference and Exhibition, San Antonio, Oct. 8-11.


10. Harris, P.C.: Dynamic Fluid-Loss Characteristics of CO2-Foam Fracturing Fluids, SPEPE (May 1987) 89-94. 11. Davies, D.R. and Kulper, T.O.H.: Fracture Conductivity in Hydraulic Fracture Stimulation, JPT (May 1988) 550-552. 12. Blauer, R.E. and Kohlhaas, C.A.: Formation Fracturing with Foam, paper SPE 5003 presented at the 1974 SPE Annual Meeting, Houston, Oct. 6-9. 13. Howard, G.C., and Fast, C.R.: Hydraulic Fracturing, Monograph Series, SPE, Richardson, TX (1970) 2, 36. 14. Holm, L.W.: The Mechansim of Gas and Liquid Flow Through Porous Media in the Presence of Foam,: SPEJ (Dec. 1968) 359-69. 15. Guppy, K.H. et al.: Non-Darcy Flow in Wells with Finite-Conductivity Vertical Fractures, SPEJ (Oct. 1982) 681-98. 16. Much, M., and Penny, G.S.: Long-Term Performance of Proppants Under Simulated Reservoir Conditions, paper SPE 16415 presented at the 1987 SPE/DOE Low-Permeability Reservoirs Symposium, Denver, May 18-19. 17. McDaniel, B.W.: Conductivity Testing of Proppants at High Temperature and Stress, paper SPE 15067 presented at the 1986 SPE California Regional Meeting, Oakland, April 2-4. 18. Roodhart, L.P., Kuiper, T.O.H., and Davies, D.R.: Proppant Pack and Formation Impairment During Gas Well Hydraulic Fracturing, paper SPE 15629 presented at the 1986 SPE Annual Technical Conference and Exhibition, New Orleans, Oct. 5-8. 19. Reidenbach, V.G., Harris, P.C., Lee, Y.N., and Lord, D.L.: Rheological Study of Foam Fracturing Fluids Using Nitrogen and Carbon Dioxide, SPEPE (Jan. 1986) 31-41. 20. Prats, M.: Effect of Vertical Fractures on Reservoir Behavior; Incompressible Fluid

21.

22.

23.

24.

25.

Case, SPEJ (June 1961) 105-17; Trans., AIME (1961) 222. Soliman, M.Y.: "Technique for Considering Fluid Compressibility and Temperature Changes in Minifrac Analysis," paper SPE 15370 presented at 1986 SPE Annual Technical Conference and Exhibition, New Orleans, Oct. 5-8. Tan, H.C., McGowen, J.M., and Soliman, M.Y.: "Field Application of Minifrac Analysis to Improve Fracturing Treatment Design," SPE Production Engineering (May 1990) 125-132. Nolte, K.G.: "A General Analysis of Fracturing Pressure Decline with Application to Three Models," SPE Formation Evaluation, (Dec. 1986) 571-583. Lee, W.S.: "Study of the Effects of Fluid Rheology on Minifrac Analysis," paper SPE 16916 presented at the 1987 SPE Annual Technical Conference and Exhibition, Dallas, Sept. 17-30. Juranek, T.A., et al.: "Minifracture Analyses and Stimulation Treatment Results for CO2-Energized Fracturing Fluids in South Texas Gas Reservoirs," paper SPE 20706 presented at the 1990 SPE Annual Technical Conference and Exhibition, New Orleans, Sept. 23-26.

Additional References Biot, M.A., Masse, L., and Medlin, W.L.: Temperature Analysis in Hydraulic Fracturing, JPT (Nov. 1987) 1389-1397. Craighead, M.S., Hossaini, M., and Freeman, E.R.: Foam Fracturing Utilizing Delayed Crosslinked Gels, paper SPE 14437 presented at the 1985 SPE Annual Technical Conference and Exhibition, Las Vegas, Sept. 22-25. Ely, J.W., Arnold, W.T., and Holditch, S.A.: New Techniques and Quality Control Find Success in Enhancing Productivity


5-35

and Minimizing Proppant Flowback, paper SPE 20708 presented at the 1990 SPE Annual Technical Conference and Exhibition, New Orleans, Sept. 23-26. Harris, P.C.: Dynamic Fluid-Loss Characteristics of Nitrogen Foam Fracturing Fluids, JPT (Oct. 1985) 1847-1852. Harris, P.C., Haynes, R.J., and Egger, J.P.: The Use of CO2-Based Fracturing Fluids in the Red Fork Formation in the Anadarko Basin, Oklahoma, JPT (June 1984) 1003-1008. Harris, P.C.: Effects of Texture on Rheology of Foam Fracturing Fluids, SPEPE (Aug. 1989) 249-257. Robinson, B.M., Holditch, S.A., and Whitehead, W.S.: Minimizing Damage to a Propped Fracture by Controlling Flowback Procedures, JPT (June 1988) 753-759. Stim-Lab, Inc, Consortium: Preliminary Report on the Investigation of the Effects of Fracturing Fluids upon the Conductivity of Proppants, June 22, 1989. Watkins, E.K., Wendorff, C.L., and Ainley, B.R.: A New Crosslinked Foamed Fracturing Fluid, paper SPE 12027 presented at the 1983 SPE Annual Technical Conference and Exhibition, San Francisco, Oct. 58.

5-36


Section 6 Foam Cementing Contents Introduction .............................................................................................6-3 Foam Generation ....................................................................................6-4 Stabilizing Additives .....................................................................................6-4 Strength Development ................................................................................. 6-5 Gas Injection ................................................................................................ 6-5

Downhole Behavior .................................................................................6-7 Constant Gas Rate Foam Cement .............................................................. 6-7 Constant Density Foam Cement.................................................................. 6-7

Cement and Additives .............................................................................6-9 Job Considerations ............................................................................... 6-10 Primary Cementing .................................................................................... 6-10 Squeeze Cementing ..................................................................................6-11

Design Considerations ......................................................................... 6-12 Prejob Checklist ......................................................................................... 6-12 Operator ...............................................................................................6-12 Service Company ................................................................................. 6-12 Drilling Contractor ................................................................................ 6-13 Using a Reactive Flush .............................................................................. 6-13 Cement Rheology ...................................................................................... 6-13

Evaluating Foam Cementing Results.................................................... 6-14

Foam Cementing

6-1

6-2

Foam Cementing

Foam Cementing Introduction There have always been areas in which weak zones can support only a limited height of a normal-density (11 to 18 lb/gal) cement column without breaking down. Foam cement provides a means of preparing 4 to 15 lb/gal cementing slurries that develop relatively high compressive strengths in a minimum period of time, even at low formation temperatures. The use of foamed cement offers a lowdensity slurry that develops relatively high compressive strengths and low permeabilities

protects water-sensitive clay, shale, and salt formations can control high-volume water flow in weak formations, when mixed as a quick-set formula enhances protection against annular gas invasion is economically competitive can be used from 28 to 600°F. Halliburton Foam Cement is a stabilized system consisting of cement with carefully chosen additives, a foam stabilizer, a gas (usually nitrogen), and water. Success of foam cement comes from the ability to maintain cement slurry density below the hydrostatic breakdown of sensitive formations,

Fig. 6-1: Equipment needed in the field to mix and monitor foam cements is very similar to that used in conventional jobs. The major exceptions are the foam generator inserted into the slurry discharge and the nitrogen unit.

Foam Cementing

6-3

which prevents lost circulation and fallback problems. This density control flexibility allows a wide latitude in designing the overall job before it is actually run in the field. Appropriate computer-assisted programs are used for prejob planning. If necessary, one can choose to change the density as the pressure and circulation events vary during job execution. To prepare a stable foam cement, the slurry should be conveyed through an effective mechanical foam generating device that imparts sufficient energy and mixing action with pressurized gas to prepare uniform gas bubbles of the correct size. In nearly all respects, regular cementing equipment is set up as for an ordinary cementing job. The foam generator is inserted in the cement slurry discharge line that is connected to the well head, and the nitrogen unit is connected to the foam generator. The cement slurry is mixed in a normal fashion, and foaming surfactants and stabilizers are injected into the slurry as it is picked up by the displacement pump unit. Fig. 6-1 on the previous page depicts a typical field job equipment layout.

Foam Generation Stabilizing Additives Foam cement requires that a stable foam be created in which the entrained gas is trapped in discrete bubbles that are uniformly dispersed throughout the slurry. If gas bubbles are not discrete and within a certain size range, the foam may be unstable, and the set cement will have high permeability and low compressive strength. Retained stability at high foam quality is important for foam cements with densities less than 9 lb/ gal. Small, fine foam bubbles are believed to promote stronger cement walls around the bubbles and provide a set cement of increased integrity. Stabilization is achieved by using an efficient foaming surfactant in addition to an effective chemical foam stabilizer. Halliburton Foam Stabilizer is recommended for circulating temperatures of 50 to 110°F and HC-2 for temperatures greater than 110°F. The normal surfactant requirement is 1.5% foam additive and 0.75% foam stabilizer, with both quantities based on the volume of mixing water that the cement slurry is mixed with. Both additives are mixed together, stirred to insure uniformity, and then injected as one solution with an injection pump. This results in

Table 6-1: Effect of Water Ratio on Foam Cement Strengths Water Ratio

0.72

0.60

0.46

0.38

Surface Density of Cement Slurry (lb/gal)

13.6*

14.5**

15.6**

16.4**

Compressive Strength (psi) Curing Time

Density of Foam (lb/gal)

24 hr

72 hr

24 hr

72 hr

24 hr

72 hr

24 hr

72 hr

8

224

230

260

518

395

665

825

1070

6

84

128

131

168

163

288

235

208

4

43

57

38

82

18

56

20

60

Samples cured at atmospheric pressure and 100°F. All samples contained 1.5% surfactant + 0.75% stabilizer by volume of water. * Class H + 2% solids stabilizer + 2% CaCl2 ** Class H + 3% CaCl2

6-4

Foam Cementing

Table 6-2: Permeability of Set Foam Cement, K (air) Density (lb/gal) Surface slurry = Class H + 2% CaCl2, w/c = 0.38

4 Temp

6

8

10

Permeability (md)

65°F

129

28

1.3

1.5

85°F

159

111

6.7

2.3

Density (lb/gal) Surface slurry = Class C + 2% CaCl2, w/c = 0.56

4 Temp

6

8

Permeability (md)

65°F

--

15.2

1.32

1.12

85°F

--

846*

0.42

0.11

approximately 0.4 gal of foam additive and 0.2 gal of foam stabilizer being injected per barrel of surface cement slurry.

Strength Development

1,400

Compressive Strength (psi)

As with ordinary slurries, the water to cement ratio (w/c) of a foam cement slurry strongly affects the strength of the set solid. This is illustrated by the results in Table 6-1. The chemical and physical properties of the cement also affect strength development as shown in Fig. 6-2. Permeability of set foam cement varies as a function of both entrained gas volume and curing temperature. Table 6-2 lists typical permeability data. To those familiar with the lack of strength development of ordinary lowdensity oil well cements (10 to 11.5 lb/ gal), the ability of foam cement to achieve strengths in excess of 500 psi with air permeabilities less than 20 md at cool temperature conditions seems remarkable. Foam cement achieves higher strengths than water-extended cements primarily because of the very low density of gas versus the density

Gas Injection

The required gas injection rate per barrel of cement slurry should be determined by entering the job data in a foam cement computer program (such as FMCEM). Fig. 6-3 on the next page shows typical nitrogen requirements for an 8.5 lb/gal foam cement. Foam jobs can be designed using a constant nitro-

* Sample most likely had a microcrack present

Foam Cementing

10

of water. As a result, it takes fewer volumes of gas per volume of cement to achieve the same density reduction. Absence of these additional dilution volumes in foam cement results in much stronger, competent cement. Table 6-3 on the next page presents some typical compressive strengths of foam cement.

1,200

Class C + 2% CaCl2 w/c = 0.56

1,000 800 600

Class A + 2% CaCl2 w/c = 0.46

400 200

Class H + 2% CaCl2 w/c = 0.38

4

6

8

10

Density (lb/gal) Fig. 6-2: This graph of results from a 24-hour compressive strength development test run at 100°F illustrates how physical and chemical properties can affect foam cement strength development.

6-5

Table 6-3: Compressive Strengths of Foam Cement Curing Temperature

65°F

Curing Time

12 hr

24 hr

100°F 72 hr

Density of Foam (lb/gal) Surface Slurry: 15.6 lb/gal (Class A, 2.0% CaCl2)

Surface Slurry: 14.8 lb/gal (Class C, 2.0% CaCl2)

Surface Slurry: 16.4 lb/gal (Class H, 2.0% CaCl2)

12 hr

24 hr

140°F 72 hr

12 hr

24 hr

72 hr

Compressive Strength (psi)

10

130

220

490

370

630

1,040

510

870

1,100

8

70

190

210

230

530

720

250

430

680

6

40

100

210

150

230

300

160

340

200

4

10

60

38

60

110

140

70

110

60

10

50

410

1,130

260

1,280

1,280

650

1,250

1,390

8

70

240

320

270

350

780

260

650

530

6

50

120

200

150

180

310

120

150

140

4

10

30

110

60

80

150

50

70

80

10

60

160

290

130

400

440

150

570

500

8

40

80

160

110

200

350

120

200

190

6

20

50

100

90

90

180

50

90

90

4

10

20

30

10

30

50

10

30

30

Nitrogen Requirements (scf/bbl)

gen (N2) rate or a constant downhole density. small, the density variations will also be With a constant N2 rate, if the cemented small. interval is long or back to the surface, draUnfortunately, constant density designs matic density changes can occur because of have problems also. When the first stages the decreasing hydrostatic pressure as the slurry rises 4,000 in the annulus. This problem can be avoided by initiating the 8.5 lb/gal 3,000 job with a low N2 rate and increasing the rate as the job proceeds. To inject the 2,000 ever-changing quantity of 10.5 lb/gal N2 required to produce a 1,000 column of constant final 12.5 lb/gal density is not operationally feasible. Instead, foam jobs based on constant density 2,000 4,000 6,000 8,000 10,000 can be successfully perDownhole Hydrostatic Pressure (psi) formed with the N2 rate varied incrementally. If the Fig. 6-3: Nitrogen requirements for preparing an 8.5 lb/gal foam cement. increments are sufficiently Values are in standard cubic feet of N per barrel of 14.8 lb/gal cement. 2

6-6

Foam Cementing

that have the lower N2 concentrations turn the corner under maximum pressure, the density will be much greater than the design density and can cause breakdown. To avoid this situation, the job can be run with constant N2. If the interval is long, two or three constant rates may be chosen, and a neat cap cement can be pumped down the annulus after the plug is bumped, if cement is brought back to surface. The FMCEM program can be operated in four different modes. These options allow the user a choice between a constant density and a constant N2 format with or without a set of job calculations. These job calculations contain information about N2 foamer, mix water, and cement volumes and rates to be used during the job. The program may be run with interactive data input or with the file input. Physical stabilization results when the gas is introduced into the cement slurry with sufficient energy to create microscopic, discrete gas cells. This is accomplished by using a foam generator equipped with 3/16in. or 1/4-in. jets. Foam cement is stable, unlike nitrified cement or drilling fluid. The entrained gas will not coalesce from the cement slurry if the slurry remains under the designed temperature and pressure conditions.

Downhole Behavior Foam cement applications can be divided into two types: constant gas rate and constant slurry density. These two designations represent the two extremes and are normally greatly modified to arrive at a practical job design.

Constant Gas Rate Foam Cement The constant gas rate technique can be used to remedy lost circulation problems, within certain limitations. Fig. 6-4 on the next page shows the difficulties in attempting to use a constant gas rate foam cement and circulate it back to surface. This example shows a foam cement with 30 sv/vus (standard volumes of N2 per unit volume of unfoamed slurry, which is 168 scf/bbl). With no backpressure on the annulus at the surface (curves labeled 0/0), the pressure gradient (PG) is below the fluid entry gradient to about 7,000 ft, and cement above 2,000 ft would not be dense enough to provide low enough permeability for casing protection. If N2 content is reduced, density at the shallow depths can be corrected, but the maximum pressure gradient easily can be exceeded at the greater depths. This profile can be partly corrected by holding backpressure at the surface. The 500/ 0 curves in Fig. 6-4 show the effect of holding 500 psi backpressure. However, this method runs the risk of breaking down weak, shallow formations unless intermediate or deep surface casing has been set to about 1,000 ft. A better approach to using a constant rate foam cement is to use a nonfoamed cap of either mud or regular lightweight cement ahead of the foam cement. Fig. 6-5 on the next page shows the results of using a 3,000 ft cap of 9.9 lb/gal mud (curves A) and a 12.9 lb/gal regular lightweight cement (curves B). Even with a lighter 9.9 lb/gal mud cap, the foam slurry density is never less than 9.2 lb/ gal, which provides low permeability and sufficient compressive strength, and the pressure gradient profile falls well within the maximum and minimum limits.

Constant Density Foam Cement Theoretically, constant density can be maintained throughout a foam cement

Foam Cementing

6-7

Fig. 6-4: Graph illustrates problems involved with using a constant gas rate foam cement to remedy lost circulation problems. Two options are shown—one using no backpressure on the annulus during circulation and another in which 500 psi is held at the surface.

Fig. 6-5: To overcome problems illustrated by Fig. 6-4, a cap of mud or heavier cement can be used with constant gas rate foam cements.

column by continuously adjusting the gas ratio. In practice, incremental adjustments are used, but the increments are designed to cause only minor, acceptable density variation throughout the column. The results of changing the N2 ratio for every 1,000 ft of slurry at shallow depths and every 2,000 ft at the greater depths are shown in Fig 6-6. The initial ratio was 8.5 sv/vus (47.7 scf/bbl) for the slurry to be placed near the surface, and this increased to 123 sv/vus (690 scf/bbl) for the slurry at 12,000 ft. The 8.5 sv/vus requires only 191 scf N2 if the unfoamed slurry is pumped at 4 bbl/min. This rate is too low to make accurate delivery with most N2 pumps currently used in oilwell servicing. The properties of foam cement with only 8.5 sv/vus in the top 500 ft (3 to 8 lb/gal) are marginal for competent

cement. Unless intermediate casing has been set or unless poor quality cement in the upper 500 to 1,000 ft can be tolerated, placement of a nonfoamed slurry cap is recommended followed by foam cement prepared by incrementally adjusting the N2 ratio. Results of using only 200 ft of a neat Class C slurry cap or lead slurry are shown in Fig. 6-7. The minimum foam slurry density is 9.8 lb/gal, and the pressure gradient still does not exceed the breakdown pressure at 8,000 ft. Actual applications of foam cement have shown that a blending of fixed gas rate and constant foam slurry density procedures will provide the most practical method in field operations. The following suggestions are offered:

6-8

Foam Cementing

Fig. 6-6: To obtain a constant density foam cement, the N2 ratio must be adjusted with depth. In this example, the ratio is changed every 1,000 ft at shallower depths and at every 2,000 ft at greater depths to obtain an average slurry density of 10.5 lb/ gal.

Use constant N2 ratios only for jobs in which a nonfoamed cap equal to 10 to 30% of the total depth can be used or when poor cement and low hydrostatic pressure can be tolerated in the top 25% of the column. Use incremental N2 ratio adjustment if a constant N2 ratio results in unacceptable strength and permeability in the upper part of the foam slurry. Limit incremental adjustments to a maximum interval of 1,000 ft for depths less than 6,000 ft and to a maximum interval of 2,000 ft for depths greater than 6,000 ft.

Foam Cementing

Fig. 6-7: In this example, the N2 ratio was adjusted with depth, but a 200-ft 14.1 lb/gal cement cap was also used.

Cement and Additives Cement slurries using many conventional cement additives are generally batch-mixed before being foamed. Certain additives are not recommended for use with foam cements because they will destabilize the foam cells. Any additives that act as defoamers or dispersants should be avoided (e.g., NF-1, DAIR, CFR-1, CFR-2, HR-12, HALAD(R)® -9, HALAD® -14, HALAD® -22A, etc.). Most additives that promote gel strength usually are beneficial (e.g., THIX-SET A and B, THIXSET 31A and 31B, LA-2 latex, Diacel LWL, WG-17, bentonite, ECONOLITE, etc.). To achieve extended pumping times, it is best to

6-9

use Diacel LWL and/or WG-8 as Table 6-4: Properties of 550°F High Temperature Cycling set retarders and foam stabilizers Foam Cement (for steam injection conditions) whenever possible. Surface Slurry: 15.4 lb/gal (Class G, 40% SSA-1, 3% Lime) Foam cement can also be Foam Cement Density made salt-tolerant. Halliburton's Properties CFA-S foaming agent permits 10 lb/gal 11.5 lb/gal 13 lb/gal generation of foam cement using Compressive strength after 1,210 psi 1,680 psi 2,260 psi fresh water, NaCl concentrations 20 days at 550°F up to saturation, KCl concentraCompressive strength after 1,630 psi 1,550 psi 2,440 psi tions up to 5%, or seawater. 100 days at 550°F* Studies show foamed salt cement Compressive strength after 1,240 psi 2,020 psi 2,430 psi provides improved bonding to 160 days at 550°F** salt zones and other freshwater Air permeability after 100 2.4 md 1.0 md 0.9 md sensitive formations. days The use of SUPER FLUSH Porosity 75 68 64 (liquid or powder) spacer as a K-value (BTU/hr-ft-°F) 0.14 0.18 --preflush is highly recommended for use with foam cement to * Cycled to 100°F twice ** Cycled to 100°F three times further promote bonding and displacement. ing pressure throughout the job at total depth As with conventional slurries, adding fine and any other zone of interest. If the initial silica flour (SSA-1) to foamed slurries helps design fractures the well, this program can be prevent strength retrogression when temused to determine ways to modify the design peratures in excess of 230°F will be encounto prevent fracturing. Such modifications tered. Geothermal foam cement inherently could include varying the rates at different has several attractive properties such as low points in the job, changing fluid density, density, good strength, temperature stability, running more or less spacer, or foaming a and excellent heat insulation properties. fluid ahead of the cement. Table 6-4 shows properties of 550°F highA section is included in the program for temperature cycling foam cement. N2 concentration. This can be for foamed cement and spacer. The N2 section will usually be based on the output from the Job Considerations FMCEM program. To accomplish primary cementing with Primary Cementing foam cement, the wellhead should be equipped with annular pressure-containing As with most conventional cementing devices. If foam cement is to be circulated to operations, foam cement jobs are initially the surface, the presence of this equipment is designed based on static density. It is posnearly mandatory. When pressure-containing sible that when frictional pressures are devices are not feasible, an unfoamed cement considered, a job that has a safe final hydrocap should be run ahead of the foam cement. static pressure might actually fracture the The unfoamed cap interval is tailored for well during the job. A computer cement job each specific job and has a minimum interval simulator program (CJOBSIM) allows the of 200 ft. user to simulate the actual downhole circulat-

6-10

Foam Cementing

For safety and cleanup ease, the return relief lines should be carefully staked and chained to discharge in an acceptable waste area, such as a sump pit. Foam cement under pressure will greatly expand if released at atmospheric pressure.

When a known potential exists for going on vacuum on a squeeze job, foamed cement can be used to help reduce the pressures exerted on the weak formation, thereby providing a means of obtaining surface pressure indication during the squeeze. A key element in optimizing job design variSqueeze Cementing ables such as cement density, circulating pressure, etc., is a computer squeeze job Since squeeze cementing is frequently simulator (SQZSIM2) program. remedial in nature, the volume of cement A critical factor in using the squeeze job actually required to perform the job is often simulator has been its capability to assist in quite small. Excess cement blend is often planning and performing foam cement preparedwhether or not it is actually squeeze jobs. Probably the most critical job mixedfor an average squeeze job because parameters that can be computed using the of the uncertainty regarding the endpoint of computerized simulation for foam cement the job and/or to ensure that an adequate squeeze jobs are (1) the N2 injection volume required to obtain a specific foam cement volume of cement is available in case formadensity and (2) the base slurry volume (to tion breakdown should occur. When small yield the requested foam volume), taking into cement volumes are employed, loss of ceaccount the effects of hydrostatic, circulating, ment slurry to a created fracture can result in and frictional pressures on the downhole inadequate coverage of the zone of interest, density of the compressible foam cement thus requiring another squeeze job. system. As an example of these applicaTable 6-5: Squeeze Job Parameters tions, two existing producing wells were to be drilled to a deeper producCase 1 Case 2 ing formation (Table 6-5). The formaCurrent Casing 5 1/2 in. 5 1/2 in. tion that was being produced had a Tubing Size 2 7/8 in. 2 7/8 in. low fracture gradient that would break Tubing Depth 5,640 ft 5,496 ft down if the well was loaded with conventional drilling fluid. Safely Tool 5 1/2 in. retainer 5 1/2 in. retainer drilling to this zone would require the Injection depth 5,800 ft 5,800 ft use of foam air drilling techniques to Injection Pressure 3,240 psi 3,310 psi help maintain circulation while drilling to the new producing interval. A Well Fluid 8.9 lb/gal brine 8.9 lb/gal brine foam cement squeeze job was deJob Sequence signed to eliminate the need for this Spacer(s) 10 bbl water 10 bbl water costly technology and to prevent 27.3 bbl of Class C 33 bbl of Class C invasion of drilling fluids and fines (foamed), 10.8 (foamed), 10.8 Lead Cement* into the older producing formation. lb/gal lb/gal Following the squeeze job, the foam8 bbl of Class C, 2.3 bbl of Class squeezed interval would be drilled out Tail Cement 14.8 lb/gal C, 14.8 lb/gal to allow the wells to be completed Displacement 31 bbl water 31 bbl water with liners. * volume of unfoamed slurry

Foam Cementing

6-11

A prejob simulation was performed to obtain the desired foam cement density (10.8 lb/gal) and to predict wellhead pressures for each job. Squeeze pressure was not obtained in Case 1. In Case 2, however, additional pressure was applied at the end of the job, increasing the density of the foam cement. Actual wellhead pressures from both jobs followed very closely those predicted by the computer simulation (Figs. 6-8 and 6-9). The Actual pressure Simulated pressure graphs show that, alFig. 6-8: Comparison of actual vs. simulated wellhead pressure (Case History 1). though the time at which certain pressures were Reliable equipment should also be planned obtained varied slightly from the times for. A summary of important design considpredicted through the simulations, the maxierations is offered below. mum and minimum pressure values themselves are consistent with those predicted by Prejob Checklist the simulator. Both wells were successfully drilled to Operator target depth without lost circulation problems, and production liners were set. Result1. Depth of well ing production data for the original zones 2. Location of lost circulation zone(s) showed decreased water-to-oil ratios (WOR) 3. Breakdown gradients of fragile zones without a decrease in oil production. 4. Circulating and static temperatures 5. Ultimate formation temperatures 6. Fallback history? (if yes, design Design Considerations thixotropic cement blend) 7. Electric, fluid, and mechanical disTo successfully circulate and retain foam placement hole volume cement at the correct density in a well that 8. Backpressure requirement on annulus has severe lost circulation problems, an 9. Adequate spacer/flush operator must carefully design the job and follow the correct procedures. Prejob planning and calculation from accurate data provided by the operator is as important as on-the-job timing and execution of the plans.

6-12

Foam Cementing

5. Chained diverter return lines: sump disposal area 6. Annulus choke system 7. Casing placement precautions and control 8. Minimum pipe movement

Using a Reactive Flush Exceptionally bad lost circulation problems can be lessened or remedied by pumping a chemically reactive flush Actual pressure Simulated pressure system, such as SUPER Fig. 6-8: Comparison of actual vs. simulated wellhead pressure (Case History 1). FLUSH, behind the drilling fluid and ahead of the cement slurry. This technique was Service Company developed for use with conventional slurries as early as 1971. The technique remains 1. Pump truck unit- cement slurry highly recommended for use with foam density and slurry rate instrumentacement practices. tion The reactive flush is designed to coat both 2. Nitrogen service unit- gas rate, gas the formation and pipe with chemically active pressure, and control equipment material. Larger quantities of the material 3. Check valves may tend to preferentially enter areas that are 4. Bulk cement equipment accepting fluids under pressure. When 5. Surfactant, surfactant pump, surfaccontacted by the foam cement slurry, a rapid tant rate meter chemical reaction occurs that tends to create a 6. Foam generator- metering jets, mulnearly immobile precipitate that can seal the tiple connection nipples, and adapters zones that have accepted the flush. Postjob evaluations have indicated that the use of the Drilling Contractor reactive flush promotes improved bond logs and improved mud-displacement efficiency 1. Freshwater source as well as sealing lost circulation zones. 2. Displacement fluid Foaming the reactive flush, using the same 3. Location layout- sufficient equipment technique as with the foam cement, before space pumping can improve its effectiveness. 4. Reliable well head annulus pressure control

Foam Cementing

6-13

Cement Rheology

Table 6-6: Foam Cement Viscosity Behavior

Class G + 40% Silica Flour; 15.8 lb/gal surface density Foam cement rheology Slurry Apparent Plastic places some limits on job design. Yield Point Density Viscosity Viscosity n' k' The general nature of foam (lb/100 sq ft) (lb/gal) (cp) (cp) cement is that its viscosity 15.8 0.67 0.02460 34 125 113 increases as the density is lowered (i.e., the gas content in12.0 0.65 0.03149 58 141 112 creases). Foam cement slurries 10.0 0.59 0.04373 68 130 96 are usually more viscous than 8.0 0.57 0.05022 74 129 92 the surface slurry from which 6.0 0.48 0.06713 68 90 56 they are prepared. Mathematically, this behavior is reflected as a steady increase of K´ value as During the job, such things as interrupted more gas is stabilized in the foam cement. circulation, sudden or unexpected increases The n´ value for a typical slurry slightly or decreases in surface pressure, and infordecreases, indicating the shear thinning mation about mixing and measuring proceability of the foamed fluid remains approxidures should be carefully recorded. Experimately constant. ence indicates that certain guidelines should Table 6-6 lists typical data derived from be followed to help assure the best possible Fann viscometer readings for Class G cement results from a foam job. The following basic plus 40% silica flour. Steady decrease of items are important for a successful foam plastic viscosity as the cement density is cement job: lowered correlates to lower solids per given A means of mixing an unfoamed volume, and the general increase is yield surface cement slurry at a specified point indicates greater solids carrying capacair-free density with a reasonable ity for the lighter foam cement slurries. The accuracy (e.g., ± 0.1 lb/gal) practical significance of this foam cement Method of measuring the unfoamed behavior is to make placement of foam slurry pump rate and total volume cement in turbulent flow unlikely in wells with an accuracy of ± 5% or better that already suffer from severe lost circula Techniques for introducing foam tion problems. Consideration of the frictional stabilizing chemicals into the backpressure generated when placing foam unfoamed slurry or N2 stream with an cement with viscosity also suggests that low accuracy of ± 10 % cement displacement rates should be ob Facilities for measuring and controlserved, especially after the foam cement ling gas injection rate based on mass clears the pipe and enters the annulus. or standard volume Injection of the gas into the unfoamed slurry stream with sufficient energy to Evaluating Foam Cementing obtain maximum stabilization Results An inline mixing device to add stability and uniformity to the foam slurry. Success of foam cement jobs can be Up to a point, higher energy provides measured in two waysfactors noticed greater stability. Inappropriate field during the job and post-job evaluation. sampling methods can easily lead to

6-14

Foam Cementing

false conclusions regarding the stability of field-mixed foam cement. Post treatment measurements using bond logs and temperature surveys have been used to evaluate foam cement jobs. Bond logs indicate the presence of foam cement primarily through attenuation of the amplitude curve and the micro-seismogram. The amplitude curve responds to differing densities of both the conventional cements present and the foam cements, and is helpful in locating the interface between the two. To get the most information possible, the amplitude should be set as high as possible to provide a greater range, and therefore better resolution, on the curve, which will better show changes in density. The micro-seismogram may not show an apparent bond through the foam that is as good as shown through normal density slurries, but arrival of formation signals along with each free pipe signal indicates that bonding has occurred. Correlation of the micro-seismogram with gamma ray or density logs for verification of formation signals has been found to be a helpful tool in evaluating bond quality. Temperature surveys, run 8 to 24 hours after completion of the job, have proven valuable in locating the top of the different intervals of cement. Cap and tail-in cements will show a temperature gradient greater than normal background, while the foamed interval will be about the same as background profiles or may even show a lessthan-normal temperature gradient. Evaluation of results should not be left to one graph, one chart, or one log, but rather as much information as possible should be gathered and correlated.

Foam Cementing

6-15

6-16

Foam Cementing

Section 7 Other Nitrogen Applications Contents Sand Washing ........................................................................................7-3 Incompressible Wash Fluids ........................................................................ 7-3 Compressible Wash Fluids .......................................................................... 7-4 Sand Washing with Foam............................................................................ 7-4 Operations.................................................................................................... 7-5 Wash Penetration Rate ......................................................................... 7-5 Produced Fluids .....................................................................................7-6 Wash Tools ............................................................................................ 7-6 Conventional Circulation ........................................................................ 7-6 Job Procedure .............................................................................................. 7-6

Unloading Wells .....................................................................................7-7 Unloading Design Considerations ............................................................... 7-7 Unloading Concerns .................................................................................... 7-8 Nitrogen-Assisted Unloading ....................................................................... 7-8 Nitrogen Behavior .................................................................................. 7-8 Nitrogen Unloading Methods .................................................................7-9

Gas Displacement ................................................................................ 7-10 Pressurizing Medium ............................................................................ 7-11 Drillstem Test Cushion ............................................................................... 7-11 Perforating Technique ............................................................................... 7-11 Gas Lift Medium ......................................................................................... 7-12

Commingled Gas .................................................................................. 7-12 Reduce Mud Weight .................................................................................. 7-12 Remove Differentially Stuck Pipe .............................................................. 7-12 Perform Hydrojetting .................................................................................. 7-12

Sand Consolidation............................................................................... 7-13 Operations..................................................................................................7-13 The Resin .................................................................................................. 7-16

Leak Detection Service ......................................................................... 7-17 Advantages ................................................................................................ 7-17 Procedure...................................................................................................7-17

References ...........................................................................................7-19

Other Nitrogen Applications

7-1

7-2


Other Nitrogen Applications Sand Washing One of the most common problems associated with producing oil and gas wells is sand production. This problem occurs in wells completed in unconsolidated or loosely consolidated sandstone reservoirs or in wells recently subjected to fracture stimulation. Coil tubing can be used to clean out (wash) sand from a wellbore in much the same manner as with a conventional workstring; fluid is circulated at rates sufficient to wash the sand back to the surface. The tubing is lowered as the sand is displaced from the wellbore. The obvious advantage of continuous coil tubing over jointed tubing for sand washing is the ability to maintain circulation while going in or coming out of the hole, or when washing sand under pressure. When using jointed tubing, the circulation must be interrupted while making up or breaking down joint connections. Water is commonly used for sand washing, but many wells require using lighter, compressible fluids. Compressible wash fluids used in coil tubing service are dry N2 and foams (aqueous or oilbased). These

lighter fluid sytems offer many benefits for low-pressure or liquid-sensitive reservoirs. Wash-fluid density is designed to minimize fluid losses to the reservoir, to minimize pressures in the wash string, and to minimize the pressure drop at the surface returns side. Incompressible fluids are commonly used when velocity is the major fluid criteria, and compressible fluids are commonly used when solids-carrying capability is the major fluid criteria.

Incompressible Wash Fluids Both water and hydrocarbon liquids are commonly used as incompressible wash fluids. They can be used when sufficient formation pressure exists to allow circulation of these fluids in the well. The design of these wash fluids is based on formation compatibility, well deviation, required

Casing Tubing Coiled Tubing Packer

Fig. 7-1: Sand washing using coiled tubing.


7-3

solids-transport capability, rheology, and surface logistics. For wellbores deviated above 30°, increasing the wash fluid viscosity reduces cleanout efficiency, and cleanout becomes most difficult around 60° deviation. For deviations less than 30°, increasing circulating fluid viscosity results in more efficient sand cleanout. Maximizing annular velocity and wash string centralization improves cleanout efficiency. A given wash fluid will carry solids out of a well if the annular velocity exceeds the terminal particle settling velocity. Due to the need to prevent holdup of solids in the annulus and unknowns about the nature of the solids being removed, designs typically aim to have the fluid velocity twice the particle settling velocity. When annular velocity is less than 100 ft/min, adding a viscosifier may be necessary. It is recognized that most combinations of coiled tubing and casing diameters will not allow wash fluids (slurries) to achieve turbulence in the annular space even when water is used.

Compressible Wash Fluids Compressible wash fluids are composed of gas and either water-based or oil-based liquids and surfactants. The liquid phase is chosen on the same basis as noted above. Compressible fluids are more difficult to design and use than incompressible fluids. Compressible wash fluids are composed of varying gas fractions and are used to compensate for low bottomhole pressure (BHP) formations or to lift solids when annular liquid velocities are low. Since fluid volumes change with temperature and pressure in a compressible system, wash-fluid returns will not travel at the same rates throughout the annulus.

7-4

Sand Washing with Foam In some wells, the maximum velocity that can be achieved with incompressible fluids is insufficient to carry the sand from the wellbore to the surface. This may be due to the extreme depth, the production tubing being large, the formation pressure being too low, or a combination of these and other factors. In such cases, a compressible fluid such as foam is required. Foam can be generated in hydrostatic pressure gradients ranging from 0.350 to 0.057 psi/ft, depending on wellbore pressures and temperatures. Stable foam rheology most closely resembles Bingham plastic fluids, where yield stress must be overcome to initiate fluid movement. The greater sand-carrying capacity of foam allows sand to be washed from deep, large diameter holes with limited pump rates and low velocities. This makes the use of coil tubing possible in wells that might otherwise require a workover unit. Foam is a gas-in-liquid emulsion consisting of 52 to 96% gas, ideally N2. For this application, the liquid can be aqueous or oilbased. Surfactants are mixed with the liquid phase in concentrations ranging from 1 to 5% by volume to reduce surface tension. The wet liquid phase is then commingled with N2 in a foam-generating tee. Turbulence created by N2 and wet liquid mixing provides sufficient dispersion to form a homogeneous, emulsified fluid. Foam is generated by pumping a mixture of 99% water and 1% surfactant through an atomizer tee where it is mixed with N2 gas. Because foam is comprised mostly of gas, changes in pressure, temperature, and solids loading affects the foam quality. As such, compressible fluids have constantly changing rheology. It is well understood that the compressible fluid has maximum carrying capacity when the foam quality is maintained at 65 to 90.


Operations Predicting annular velocities and solids removal capability requires complex calculations. Sand-washing foam fluids are generally designed using a computer program like FOAMUP. The sand washing rate can be established within the need to maintain a wash fluid quality of 65 to 90. To calculate the N2-to-fluid ratio, a bottomhole treating pressure must be assumed. This pressure should be less than the reservoir pressure to sustain circulation without losing fluid in the formation. After the bottomhole circulating pressure to be used is determined, foam calculations should be performed using the stable foam calculating sheet or a computer program. Because a circulating system is being used, a surface backpressure value equal to the bottomhole pressure less the hydrostatic weight and friction pressure loss of the column of foam must be maintained to control the foam quality at depth. A surface choke system is needed to maintain the backpressure on the system consistent with maintaining the foam quality in the well system. Managing this choke system and handling the returning foam and solids need careful consideration. Preparation must be made for foam disposal if necessary. The foam liquid may be recirculated if no hydrocarbons are mixed with the foam and all sand is first removed. The coil tubing and all surface equipment should be tested before going in the hole. Circulation of foam should be started at the surface to displace any fluids in the hole as the tubing is lowered and to be sure there is circulation on reaching the sand fill. The washing operation should not be performed too fast; the sand carrying capabilities of foam, although excellent, could be exceeded. Care should be taken not to wash down into the solids too fast. Surface observation and measurement of solids washed is gener-


ally required. Often, a settling tank is used. Solids that are entrained in wash fluid must be continually removed from the well. The operator's experience should be used to determine the best sand-washing rate along with the pipe size, fluid rate, and N2 rate since the nature of the solids entrained are often only the subject of speculation prior to the job. Once circulation is established in a compressible wash program, unit volumes of wash fluid are pumped down the coiled tubing at pressures needed to overcome friction pressure losses. In this condition, wash fluid is under high pressure and occupies minimal volume. As a unit volume of compressible fluid exits the coiled tubing, decreasing hydrostatic head in the annulus and reduced friction pressure allow gas in the wash fluid to expand. This expansion and subsequent increases in wash fluid velocity create high frictional pressure losses. After washing the sand to the desired depth, circulation should be maintained until the returns are clean. The bottom should be tagged several times to ensure that all sand has been removed. After the well is cleaned out, it may be jetted in or filled with fluid by stopping either the water or N2.

Wash Penetration Rate Coil tubing rate of penetration into packed solids, coupled with annular fluid velocity, determines the solids concentration in fluid returns. Dispersion of solids in wash media causes an increase in effective weight of annular fluid returns. As a result, the hydrostatic pressure differential between clean wash fluids in the coiled tubing and dirty fluids in the annulus increases. It is not uncommon to run 1 1/4-in. OD coiled tubing in 2 7/8-in. OD production tubing at 60 ft/min when washing sand. If wash fluid is circulated at 0.50 bbl/min, annular fluid velocity is about 2 ft/sec. The

7-5

unobstructed production tubing volume is 0.0325 ft3/ft and the annular volume is 0.0240 ft3/ft. If there is greater than 60 ft of loose sand above a bridge, the coiled tubing can penetrate 60 ft in one minute. At an annular velocity of 2 ft/sec, 180 ft of annulus is displaced by dirty fluids. Penetration of loose sand packs are generally not indicated at the surface and several sand bridges may be encountered when washing deep production tubing. If sufficient circulation time for solids to reach the surface is not allowed, significant hydrostatic pressure increases could develop in the annulus due to entrained solids. If this occurs, increased hydrostatic pressure from dirty annular fluids may force some fluid into the formation. If, as a result, annular velocity is reduced below the rate required to keep particles suspended, solids will fall back and could stick the coiled tubing.

Produced Fluids Formation fluid types can also determine how a wash program will proceed. In a liquid-producing wellbore (oil and water), fluids are essentially incompressible and can support a piston displacement of solids up the annulus. If produced fluid is gas, be prepared for gas kicks or lost returns when breaking through sand bridges. In addition, the difference between gas and liquid densities can allow gas to override wash fluids. This results in loss of wash fluid to the formation, regardless of BHP. When washing low BHP oil wells with aqueous foam, be prepared for foam degradation when it commingles with oil. Oil will destabilize foam regimes at the contact interface, which breaks down into a gassified, oil-water emulsion. As the foam degenerates and moves up the annulus, the sand-laden returns become compromised and solids fallback can occur.

7-6

Wash Tools Wash tool selection should be governed by wash program hydrodynamic requirements. Wash tools should only be used if additional turbulent action is needed downhole. Several tools are available for ported hydraulic jetting on packed solids or mechanical action to break up bridges. These wash tools can often be constructed to serve as bypass mandrels, further extending their use. Depending on wash port number, size and wash fluid system selected, frictional pressure losses can range from 50 to 2,500 psi.

Conventional Circulation Pumping fluid down tubing and taking returns up the annulus is the most common coiled tubing technique for washing solids. In addition to wash fluid system criteria, maximum tensile loads on coiled tubing strings should be estimated to ensure that stress does not approach minimum tube yield. Both compressible and incompressible fluids can be used with conventional circulation. Selection of an appropriate size of coiled tubing depends on minimum pump rates, total circulation system pressure losses, and minimum load rating required to safely wash and retrieve pipe from the well. Use of downhole safety check valves and ported wash tools does not limit conventional circulation wash programs.

Job Procedure 1. Rig up equipment according to standard operating practices. Check flowlines to make sure a minimum of ells are installed and that flow lines are properly secured. 2. Install a flow tee under the BOP's to direct returns out of the well so that


3.

4.

5.

6.

7.

8.

9.

10.

cutting out the Christmas tree is avoided. Install an adjustable choke in the flowline (particularly applicable to foam washing) and have a replacement stem on location. Pressure test coil tubing unit and tree to 110% of maximum expected working pressure or minimum anticipated surface pressure, whichever is greater, before going in the hole. Start the tubing in the hole at 30 to 40 ft/min if top of fill is unknown while circulating fluid at a slow rate. If top of fill has been located, do not exceed 60 ft/min. Check tubing drag every 1,000 to 1,500 ft to prevent sticking the wash string. Have coiled tubing representatives identify tubing sections that have been cycled extensively and avoid conducting periodic drag tests in these intervals. Wash the sand slowly. When breaking through bridges, allow sufficient time to circulate solids from the well before continuing downhole. String sand out in case of lost circulation to prevent sticking the wash pipe. (When washing with foam, fluid can be lost in the formation and still show returns on surface; therefore, the pressure recorder chart should be monitored for pressure loss.) Before entering open casing, always circulate a minimum volume of fluid that would fill the tubing string twice. (Run wash tool in open casing). Maintain returns throughout the wash program. If circulation is lost during the operation, immediately pull the tubing up the hole approximately 2,000 ft; hold there and work until circulation is regained. Maintain circulation until tubing is pulled completely out of wellbore.


11. Keep pipe moving at all times while jetting.

Unloading Wells During the life of oil and gas wells, well control practices during completion or workover can create hydrostatic overbalance, which can reduce inflow performance and may cause the well to stop producing. This overbalance results from the pressure of fluids in the wellbore exceeding the producing formation pressures. Temporary and lengthy shut-in periods can also create hydrostatic overbalance when the once active wellbore loads up with fluid. If no other damage exists, wells can often be returned to production by reducing the hydrostatic pressure of the fluid column. Once an underbalance is created, the well can flow again.

Unloading Design Considerations Before an unloading program is designed and started, the wells flow and production potential must be determined. Sas-Joworsky4 provides equations for estimating reservoir fluid production rates and well flow rates in his article titled Coiled Tubing...Operation and Services, Part 5Unloading Wells with Lighter Fluids. He also explains the following mechanical considerations: completion type wellbore tubular sizes workover service tubing size required operating system pressure for surface flowlines and separation equipment These mechanical parameters are used to predict backpressure, which is system pressure losses. Backpressure decreases effective formation drawdown and reduces fluid production to surface.

7-7

The tubular sizes are the most important mechanical factor for unloading wells. Production tubing ID determines flowing liquid head and frictional pressure loss for a given production flow rate and gas-liquid-ratio (GLR) when producing fluids to the surface. As tubing ID increases, fluid velocity and frictional pressure losses in the flowing fluid regime decrease. However, as fluid velocity decreases, slippage and flowing pressure gradient increase. The effects of pressure loss in the tubing become critical when trying to flow fluids to the surface with coiled tubing concentric to production tubing.

Unloading Concerns Numerous coiled tubing service techniques can be used to reduce wellbore hydrostatic pressure, thereby achieving an underbalance and unloading wellbores. The object of these techniques is to initiate flow from the formation without creating adverse pressure shocks downhole. In many cases, varying degrees of skin damage in the completion interval clean up as the wellbore unloading program progresses. However, apparent skin damage could also be due to relative permeability changes near the wellbore, perforation plugging, or damage deposited during completion or workover. Once a maximum recommended pressure drawdown has been selected, it should not be exceeded during unloading programs. If produced fluid volumes remain significantly below projected flow rates for the applied drawdown, it is most likely due to inaccurate parameters in flow potential calculations, but actually, something may be wrong downhole. A common response to low surface flow rates is to increase drawdown and hope that downhole flow restrictions dissipate. Unfortunately, this reaction generally causes formation integrity failure in unconsolidated formations, resulting in perforation tunnel collapse and damaged flow potential. Con-

7-8

solidated formations are somewhat more forgiving and may not be damaged as much from downhole pressure shocks. Unloading programs should be designed to create the minimum pressure drawdown needed to initiate flow. Once stable flow is established, formation damage can be properly assessed and corrective steps taken.

Nitrogen-Assisted Unloading The most common method used to unload wells is nitrogen (N2) circulation through coiled tubing run to a predetermined depth below the static fluid level. Although this technique is commonly called jetting or jet-lifting, it starts flow by reducing wellbore hydrostatic pressure through aeration and not by jetting the fluids to surface. N2 is most commonly used for unloading programs because it is chemically inert and only slightly soluble in liquids. Coiled tubing conveyed gas circulation offers greater use than conventional single-point gas-lift operations because the gas injection point can be moved up and down the wellbore to optimize fluid withdrawal rates.

Nitrogen Behavior When using N2 to unload wellbores and initiate flow, it is important to recognize the effects of lifting high GLR fluids within the annulus between the coiled tubing OD and the production tubing ID. As the annulus area decreases, annulus pressure losses increase exponentially. Also, the length of concentric coiled tubing inside the production tubing significantly affects annular friction pressure loss and flowing fluid head. Nitrogen pumped down the coiled tubing is compressed to overcome the annulus fluid gradient. As the N2 injection point is lowered further into the well, the increased pressure gradient compresses the N2 more. When N2 exits the coiled tubing and starts to rise in the


annulus, it expands. This expansion of N2 dispersed in the annular liquid increases apparent fluid velocity, resulting in a further decrease in flowing pressure gradient. Nitrogen expands dramatically as it continues to flow to surface. Fluid velocity and frictional pressure losses in the annulus increase significantly relative to velocity and frictional pressure losses at the downhole N2 injection point. Turbulence from the expanding gas increases frictional pressure loss. At some depth in the well, based on well parameters, frictional pressure losses will overcome reduced pressures from flowing fluid gradients in the tubing annulus. Also, decreased annular cross-sectional areas greatly increase frictional pressure losses for equivalent N2 and liquid circulation rates. A higher N2 circulation rate may actually yield lower production due to reduced annular cross-sectional area and the exponential increase in system frictional pressure loss. If this fluid production drop is not interpreted correctly, the injection point may be run deeper into the wellbore and the N2 injection increased. This reaction could be disastrous by creating greater frictional pressure loss in the annulus and, in some cases, causing liquid flow to cease. To increase flow from the completion, N2 circulation must be cut back very slowly while pulling the coiled tubing back up the wellbore. If a high circulation rate is interrupted while deep in the tubing, the rapid decrease in annular friction pressure loss may cause a pressure shock at the formation. This pressure shock can be greater than recommended pressure drawdowns for optimum well performance and induce sudden, uncontrolled flow rates that can damage the completion. For these reasons, using the smallest coiled tubing size available and performing unloading procedures with the lowest possible N2 circulation rates is recommended.


Nitrogen Unloading Methods Continuous injection- The most effective method for achieving an underbalanced hydrostatic head with N2 is to run coiled tubing into the wellbore while slowly circulating nitrogen. This technique allows N2 in the fluid column to disperse in the wellbore, thereby aerating annulus liquids slowly and initiating production from the formation in a controlled manner. In initiating an unloading program with N2, coiled tubing is run in the well at about 40 to 60 ft/min. Low N2 circulation rates, generally from 150 to 250 scf/min, are initiated when the end of coiled tubing is just above the fluid level to minimize waste. Coiled tubing is then lowered to a predetermined depth in the well to assist fluid lifting until the completion can sustain production. As coiled tubing is run into the wellbore, the fluid column is aerated, creating an underbalanced effect. Intermittent injection- Another technique used to lighten fluid columns is intermittent N2 injection. This is accomplished by running coiled tubing to a predetermined depth below the fluid level in a wellbore before starting N2 pumping. In this case, N2 pump pressure must be greater than the fluid column hydrostatic pressure at the injection point. Once N2 injection pressure overcomes fluid column hydrostatic pressure, N2 enters the annulus and initiates a single-point gas lift operation. As wellbore pressure above the N2 circulation point decreases, gas expansion in the coiled tubing accelerates, causing an effect similar to an N2 circulation rate increase. This may cause undesirable and erratic wellbore pressure drops, which can destabilize pressure drawdown at the formation.

7-9

Gas Displacement Nitrogen is most widely used to displace well fluids from the tubing string or from the well system to begin oil or gas production. When the well is completed with a single string of tubing without a packer, or with a circulating valve, the fluid is displaced down the tubing and out of the annulus (Fig. 7-5). On a single completion with a packer, the fluid may be displaced out of the tubing into the annulus before the packer is set. By using N2 for gas displacement, you can save rig time and eliminate the danger of lost swabs. Not only can N2 reduce the danger of damaging internal pipe coatings, but it can displace fluid in multiple-diameter tubing strings. You can control the return fluid rate during gas displacement by reducing N2 pressure. Nitrogen is chemically compatible with all completion fluids and formation waters. After treating the formation with frac, acid, or other chemicals to control corrosion, scale, and paraffin, you can use nitrogen to displace the treating fluid from the tubing into the formation (Fig. 7-6). Nitrogen gas is chemically compatible with all completion fluids, does not damage the formation, and allows faster cleanup and

Completion fluid, gas, or gas and sand N2

Casing

Fig. 7-6: Fluid displaced from the annulus into the tubing.

flowback without swabbing, because it has low hydrostaticity. Another way to return reusable completion fluids, remove sand from the tubing, or provide annulus insulation is to displace fluid from the annulus into the tubing (Fig. 77). Fluid can be displaced around open-ended tubing, around an unseated packer, or through a circulating valve. This technique is successful because N2 is noncorrosive to tubing and casing. It reduces heat transfer because of the low specific heat and thermal conductivity. Because N2 has a

N2 gas at wellhead pressure (WHP)

N2 gas at wellhead pressure (WHP)

Casing

Casing

Packer Tubing Bottomhole pressure (BHP)

Fig. 7-5: Fluid displaced down tubing and out the annulus.

7-10

Fig. 7-7: Fluid displaced from tubing into the formation with nitrogen.


low density, it also allows circulation with low bottomhole pressure.

Pressurizing Medium When used as a pressurizing medium, N2 can be used for the following: drillstem test cushion perforating technique gas lift medium

Drillstem Test Cushion A water or N2 cushion is sometimes used on a drill stem test (DST) for five purposes: (1) protect the drill pipe or tubing from collapse, (2) protect unconsolidated sands from caving in when the tester valve opens, (3) help control high pressure and high volume wells by bringing-in the well slowly, (4) help prevent dehydration of the mud or salt water when the tester valve opens on high temperature (above 375°F) wells, and (5) help relieve the sudden differential pressure across the packer seat. While a cushion may help in these ways, it may also hinder obtaining a good drill stem test. When testing hard-rock formations, considerable rig time can be saved by using minimum cushions or no cushion at all. Wells that flow to the surface must first get rid of the cushion prior to really starting to obtain information on the test. Gas wells clean up faster with minimum cushion. Minimum cushion puts maximum differential on the formation and packers when the tester valve opens rather than during the middle of the test when the cushion has been removed. This is important when using a nitrogen cushion. If the packer or packer seat is going to fail, it will do so immediately when the tester valve opens rather than several hours later after bleeding of the N2 cushion.


Too much water cushion could result in an indication of a dry test when actually it was caused by the hydrostatic head of the cushion being greater than formation pressure. Conversely, unconsolidated formations must be protected by a cushion. In this instance, the cushion is beneficial because it reduces the differential pressure suddenly applied across the face of the formation when the tester valve opens. Too much differential across the face of the formation will many times cause an unconsolidated formation to cave in or produce extreme quantities of sand, plugging the tools and possibly causing a stuck string. Maximum differential pressure across the face of the formation will result in highest production rate; therefore, when the drill pipe, the formation, and safety permit, miminum cushion should be used. To use N2 as a DST cushion, perform the steps below: 1. Before a packer is set, pressure the drillpipe with N2 gas to check for leaks. You can do this with or without a water cushion, depending on the need for collapse protection. 2. After the packer is set and the tool opened, bleed the pressure at the surface to increase the pressure differential into the tubular string.

Perforating Technique Using a nitrogen cushion in perforation reduces fluid damage to the reservoir and perforation damage to the formation. It controls production rate by controlling surface pressure reduction. To perforate with N2, perform the following steps: 1. Displace well fluid from the tubing string to the needed depth. 2. Set the packer and secure the wellhead.

7-11

3. After the perforating gun is in position, adjust the pressure within the tubing string with nitrogen. 4. Perforate with positive pressure differential into the wellbore. 5. Retrieve perforating tools and vent the tubing pressure to start production.

Gas Lift Medium Using nitrogen as a gas lift allows sampling and gauging reservoir fluids. It also can inexpensively remove fluid from stimulation or zone communication. To use N2 as a gas lift, the following steps must be performed: 1. Displace the well fluid from the tubing out through the continuous tubing/production tubing annulus. 2. Continue to circulate N2 gas to carry formation fluids to the surface. 3. Continue injection until the well will flow naturally. Note: production data is sometimes provided for sizing permanent gas lift equipment.

Commingled Gas Nitrogen can be commingled with various well treating fluids to

Nitrogen eliminates fluid-loss materials usually needed for balanced or under-pressure drilling. It is chemically compatible with all mud systems and increases cement circulation height.

Remove Differentially Stuck Pipe Nitrogen can be commingled with mud or used to displace mud in techniques that lower bottomhole pressure. Reducing wellbore pressure until it equals the pressure surrounding the pipe will allow it to be removed. Another technique used to remove pipe is to spot a bubble of N2 gas over the differentially stuck zone. The low viscosity and high leakoff rate of the N2 will tend to equalize pressure around the pipe. Using N2 for this purpose is beneficial because it does not damage the mud system.

Perform Hydrojetting Hydrojetting involves mixing N2 with the gel and sand to increase the penetration rate. As expanded gas passes through, the nozzle accelerates sand speed. When used this way, N2 increases the penetration rate, maintains circulation past lower pressure formations, and reduces the risk of fracturing the formation being jetted.

reduce mud weight remove differentially stuck pipe perform hydrojetting

Reduce Mud Weight Nitrogen is commingled with drilling fluids to reduce mud weight and combat lost circulation. This can be done while drilling or performing primary cement jobs. Nitrogen quickly dissipates from the drilling mud on return to the surface, allowing quick return to heavier weights.

7-12


Sand Consolidation In gas production operations in the Gulf of Mexico, a gravel-pack screen was salvaged by applying furan resin for near-wellbore sand consolidation. This technique differs from traditional sand consolidation methods in that coiled tubing is reciprocated across the gravel-pack screen interval throughout the treatment. Additionally, nitrogen is commingled with all injected liquids to create short-lived foams that expand the volume of the liquids, provide some diverting effect, and help ensure maintenance of permeability during the treatment. The procedure was done at about one-fifteenth the expected cost to replace the gravel pack, which was estimated at more than one million dollars.2 The work was needed after an upper gas zone was accessed with a jet punch. Gas production after the tubing was punched was more than 14 MMcfd, however, buildup tests indicated a flow potential of 18 MMcfd. An acid stimulation treatment was performed, after which the well began to produce sand and had to be choked back to 2.5 MMcfd to avoid sand production. After the resin repair, flow recovered to 13 MMcfd with no sand production. Originally, the completion was a selective alternate, or stack pack, where two reservoirs were gravel packed by installing two screens (separated by packers) on one string of production tubing. The lower zone watered out and was shut off by setting a plug in a landing nipple in the isolation string (Fig. 7-2).

Table 7-1: Well Data Casing

7 in., 3.5 lb/ft

Tubing

3.5 in., 12.95 lb/ft

Perforations

Casing- 11,465 to 11,497 ft Isolation Tubing- 11,467 to 11,494 ft

Gravel-pack assembly

4 in., 0.007 gauge x 42 ft

Gravel-pack packer

11,351 ft

Bottomhole temperature

210°F

Bottomhole pressure (est.)

2,570 psi

Plug back TD (PX plug)

11,499 ft

Deviation

22°

Well Type

Gas

Operations

Fig. 7-2: An isolation string (extension of the production tubing) with a landing nipple, was run across the upper gravel pack during initial completion. A plug was set in the landing nipple to shut off watered-out lower zone. A “tubing punch” was performed to penetrate the isolation tubing.

Once sand production began, a remedial operation to replace the gravel-pack screen was estimated at $1.3 million, or 15 times the estimates for repair by resin. The excessive cost for a rig workover and the fact that

additional completions existed in the same reservoir made sand consolidation the logical choice. A 200-ft class liftboat with coiled tubing and nitrogen units aboard was preloaded and


7-13

moved into position. The outline of the planned procedure was as follows: 1. Rig up N2 and 1-in. coiled tubing with nozzle. 2. Test all surface equipment to 5,000 psi. 3. Pickle coiled tubing with 250 gallons 15% HCl with acid inhibitor and displace with 2% KCl (12 bbl) containing 35 lb pH neutralizer. 4. Go in hole with coiled tubing and wash well down to plug back TD at 11,499 ft. Use filtered 2% KCl water throughout washing operations. Foam, if necessary. 5. With end of coil tubing positioned across perforated interval, spot the following treatment and squeeze away at 1/4 to 1/2 bbl/min: 2,750 gal 15% sodium chloride (NaCl) water containing 0.25 % surfactant and 600 scf/bbl N2 1,060 gal externally catalyzed furan resin containing 600 scf/bbl N2 1,000 gal 15% NaCl water containing 0.25% surfactant and 600 scf/bbl N2 3,500 gal 10% HCl with catalyst mixed in NaCl water containing 0.25% surfactant, 0.3% acid inhibitor, and 600 scf/bbl N2 Displace coiled tubing with filtered 2% KCl water (12 bbl) 6. Shut well in for 8 to 12 hours while resin cures. 7. Resume production at 4 MMcf/D until load water is recovered. However, the coiled tubing (furnished by a third-party vendor) parted after placing the resin. The well was killed with brine, and the parted tubing was fished from the well. Since the resin used was externally catalyzed, the 40-hour delay resulting from the fishing job had no effect on the resin. When repair operations resumed, the catalyst was overflushed through the gravel pack and

7-14

consolidation occurred as planned. The actual job was as follows: Day 1: Preloaded jackup rig arrived on location. Rig jacked up in place. Coiled tubing rigged up. Started pressure test. Day 2: Pressure tested to 5,000 psi. Ran in hole to 9,810 ft; stripper on injection head blew out. Stripper replaced; continued in hole. Pumped 2,750 gal 15% NaCl2 water with 0.25% surfactant and 600 scf/bbl N2. Pumped 1,060 gal resin containing 600 scf/bbl N2. Followed with 1,000 gal NaCl2 with 0.25% surfactant and 600 scf/bbl N2. Pulled up off bottom. Discovered coiled tubing had parted. Went in hole and found the tubing below the stripper. Day 3: Mixed and pumped 65 bbl NaCl water at 1/4 bbl/min and 200 psi down coiled tubing to kill well. After well was dead, went in hole and caught fish at 28 ft. Removed injector head and made cut. Well came in, everything secure. Pulled out of hole. Pulled coiled tubing out of head and swapped out reels. Rigged up new coil. Went in hole to pump acid. Day 4: Pressure tested new coiled tubing. Ran coiled tubing in hole to bottom. Pumped 3,500 gal 10% HCl with catalyst. Displaced with 2% KCl water. Pulled out of hole. Rigged down.


Fig. 7-3: The brine preflush has entered the screen and exited the casing perforations. This salt water will help prepare the sand surfaces for adsorption of the resin.

Figs. 7-3 through 7-5 show the stages of the treatment. N2 has been commingled with all the injected fluids in this treatment. Nonreactive and immiscible with the other fluids formed, the N2 inclusion forms of a short-lived, volume-expanding foam. The N2 in the foam acts as a diverter during the multiple-phase flow to help maintain permeability. In Fig. 7-3, the saltwater preflush has entered the screen and exited casing perforations to contact formation sand. In Fig. 7-4, resin has been pumped and is coating the sand. The coiled tubing has been reciprocated across the screen interval to direct the fluid at the borehole wall (on longer intervals, a special nozzle can be attached to the end of the coil optimize fluid flow direction). The saltwater spacer slug and the catalyst have been pumped in Fig. 7-5, and the gravel pack repaired by the resin coating in-situ sand is hardening and forming a permeable but solid sand filter. A final brine flush is injected to enhance displacement of the acid catalyst.


Fig. 7-4: Resin has been pumped into the well and sprayed laterally against the walls of the borehole along the damaged interval. The resin is coating the sand and will be further dispersed into the formation by the brine spacer before the catalyst is injected.

Fig. 7-5: The catalyst overflush has been injected, and the resin is hardening. After curing is complete, the gravel pack repaired by this in-situ resin coating will form a permeable but solid sand filter.

7-15

Typically, such sand packs have 85 to 90% of the original permeability. The set resin is resistant to acids (except HF), brines, alkalies, and other common well treating fluids. Laboratory tests on the durability of sand consolidation resin systems show that furan resins remain stable and retain high strength when subjected to damaging fluids. Brine, considered to be more damaging than oil to the stability of resinconsolidated sand, was selected as the test fluid. Field experience with furan resin overflush treatments indicate no resin consolidation impairment where brine has been produced.3

The Resin The resin system used in the screen repair was a water-compatible, furan resin catalyzed by overflushing with acid. Long noted for high temperature stability, the furan resins have been widely used as a foundry core resin binder.4 As described in job procedures, NaCl water is placed ahead of the resin and pumped between the resin and catalyst. The saltwater lead is placed to help prepare sand surfaces for the chemical reaction needed for the resin to adsorb on the sand. A notable benefit of externally catalyzed resin in sand-packing treatments has been the fact that it has been possible to reverse circulate excess coated sand from the wellbore. The well can then be repacked before injecting the catalyst because the resin does not set until contacted by acid.4 This proved especially beneficial when the fishing job required a 40-hour delay before catalyzing. The tail-in load of salt water (the spacer) separates resin from catalyst so that no partial reaction is started until resin is properly placed. The salt water also begins removal of excess resin from pore spaces, flushing the resin further into the formation. Traditionally, furan resin has been used in sand control treatments both to consolidate

7-16

formation sand in-situ and to precoat sand at the surface before pumping downhole. Extra resin has normally been injected after washing out the excess pack sand and before pumping the catalyst to consolidate a portion of the formation sand adjacent to the pack sand. This new treatment is unusual in that large grained pack sand is being treated in place. An average radius of over 3 ft may be expected from in-situ consolidations of this kind.3,4 Resin-coated sand has also been used to repair damaged slotted liner gravel packs in situations where productivity of the well, environmental impact, or a severely damaged liner would preclude the expense of complete liner workovers. The use of resin-coated sand in such a repair job is relatively economical and has been marginally successful. Where the original liner was in place and could be treated directly, the success ratio has been 90% or more, but in cases where a wirewrapped screen has been used as an inner liner, the results have not been as satisfactory.5 Because the damaged screen must be completely cleared of sand on both the inside and outside, so that resin coated sand can be squeezed through the damage area and form the patch, the above alternative approach was rejected. There was no way to ensure removal of both the pack sand and the invasive formation sand in the well.


Leak Detection Service Essential to the operation of oil and gas processing systems is the knowledge that the plant is in a safe operating condition. With the ever increasing demands and complexity of our often aging processing plants, the handling of toxic gases at high pressures requires the need for stringent safety standards. To comply with these standards, we have to be able to locate and repair all leaks. Halliburton has developed an efficient and cost effective technique for purging and leak detecting a processing plant. Halliburton has the capability and expertise to sweep out the dangerous gases. Halliburtons N2 converter pump is a self-contained, flameless twin pump that provides high pressure, high rates, and/or low rates to cover the numerous types of jobs demanded. Conventional methods such as hydrostatically testing, visual inspection, or, on gas plants, a soap solution for detection of bubbles, are all unrepresentative and limited. However, for accuracy and reliability, Helium Leak Detection gives results at a level not previously known in the oil and gas industry.

Advantages Halliburtons Helium Leak Detection advantages include the following: Simulates live gas conditions Tests conducted at operating pressures Gases are safe and inert. No need to remove instrumentation Forms an integral part of the hook-up program Uses reliable and robust technology Helium is rare in the atmosphere, therefore, sensitivity cannot be compromised by other gases. The process saves time and costs by permitting leak testing to be carried


out before start up and by removing the need for live gas detection. By giving a quantifiable leak rate, it is possible to monitor leaks over a period of time to determine any deterioration of the joint. Using standards derived from the US Navy in testing their submarine nuclear reactor compartments, Halliburton has the capability of detecting leaks of 1 scf/yr to 100,000 scf/yr using the same mass spectrometer. Using our Zone II* pump units and our certified N2 tanks, we will safely, under controlled conditions, bring plant sections up to working pressure. Operators can then proceed with leak detecting.

Procedure A vessel or section is first isolated. Halliburton then ties in with a high-pressure 3/4 in. hose and pumps 99% N2 gas and 1% helium gas into the vessel until the operating pressure is reached. All flanges, valves, valve stems, etc. are taped or bagged using duct tape or plastic bags. This will contain the possible leak. Upon pressuring the vessel, the leak detector will pierce the tape or bag and draw a sample through the previously calibrated mass spectrometer. Should more than 5 scf/yr gas leak be monitored and recorded, then the flange is a fail and would require a retest after further maintenance. Less than 1 minute per test is required.

7-17

Fig. 7-4: Helium leak detection schematic.

7-18


References 1. Sas-Jaworsky II, A.: Coiled Tubing... Operation and Services, Part 5Unloading Wells with Lighter Fluids, World Oil (April 1992) 59-66. 2. McInturff, C., et. al.: "Resin Salvages Gravel Pack in Offshore Well," Oil & Gas Journal (Sept. 30, 1991) 94-96. 3. Rensvold, R.F.: Sand Consolidation Resins - Their Stability in Hot Brine, paper SPE 10653 presented at the 1982 SPE Formation Damage Control Symposium, Lafayette, Mar. 24-25. 4. Murphey, J.R., Bila, V.J., and Totty, K.: Sand Consolidation Systems Placed with Water, paper SPE 5031 presented at the 1974 SPE/AIME Annual Fall Meeting, Houston, Oct. 6-9. 5. Murphey, J.R., Roll, D.L., and Wong, L.: Resin-Coated Sand Slurries for Repair of Damaged Liners, paper SPE 13649 presented at the 1985 SPE California Regional Meeting, Bakersfield, Mar. 2729.


7-19

7-20


Foam Applications Manual

Recommend Documents