estim - combined stimulation and sand control (english)

8/18/2019 ESTIM - Combined Stimulation and Sand Control (English)

1/18

30 Oilfield Review

Combined Stimulation and Sand Control

Specialized fracturing treatments in conjunction with gravel packing create highly conductive propped fractures that yield

sustained production increases and control fines migration in weakly consolidated reservoirs. This “frac-packing” method,

which became increasingly popular in the past 10 years, bypasses formation damage and eliminates many productivity

impairments that are common in conventional cased-hole gravel packs.

Frac pack

60%

Viscous-slurry pack

12%

High-ratewater pack

0

5

10

15

Dimensionlessskin

20

25

30

Gravelack

Frac pack

llllllllllllllllllll

1992

2002

> Fracturing for sand control. Initial frac-pack results in the early 1990s indicated improved productivity compared with conventional gravel packing (left ).As a result, frac packing now represents more than 60% of the US sand-control market (top right ), and companies providing stimulation services investheavily in research and development. These investments included construction of purpose-built vessels with high-volume mixing equipment, high-pressurepumps and sophisticated monitoring systems, such as the Schlumberger Galaxy stimulation vessel (center ).


2/18

Syed Ali

David Norman

David Wagner

ChevronTexaco

Houston, Texas, USA

Joseph Ayoub

Jean Desroches

Sugar Land, Texas

Hugo MoralesHouston, Texas

Paul Price

Rosharon, Texas

Don Shepherd

Saudi Aramco

Abqaiq, Saudi Arabia

Ezio Toffanin

Beijing, China

Juan Troncoso

Repsol YPFMadrid, Spain

Shelby White

Ocean Energy

Lafayette, Louisiana, USA

For help in preparation of this article, thanks to ErnieBrown and Leo Burdylo, Sugar Land, Texas, USA; MehmetParlar and Colin Price-Smith, Rosharon, Texas; PedroSaldungaray, Jakarta, Indonesia; and Ray Tibbles, KualaLumpur, Malaysia.

ClearFRAC, CoilFRAC, DataFRAC, FIV (Formation IsolationValve), MudSOLV, PropNET, QUANTUM, SandCADE andScalePROP are marks of Schlumberger. Alternate Path,AllPAC and AllFRAC are marks of ExxonMobil; this technol-

ogy is licensed exclusively to Schlumberger.

Summer 2002 3

Hydraulic fracturing in high-permeability reser-

voirs to stop sand production is an accepted well-

completion technique. Today, one of the first

decisions during development planning for fields

that produce sand is whether or not to frac

pack—a combination of fracture stimulation and

gravel packing. More than a decade of success

proves that compared with conventional gravel

packing, this technique significantly improves

well productivity (previous page).

Frac packing as a percentage of sand-control

treatments and in terms of total jobs is growing

steadily. Use of this technique increased tenfold—

from fewer than 100 jobs per year during the early

1990s to a current rate of almost 1000 each year. In

West Africa, about 5% of sand-control treatments

are frac packs, and operators frac pack at least 3%

of the wells in Latin America.

Advances in stimulation design, well-

completion equipment, treatment fluids and

proppants continue to differentiate frac packing

from conventional gravel packing and fracturing.

US operators now apply this sand-control methodto complete more than 60% of offshore wells.

Shell used the term frac pack as early as 1960

to describe well completions in Germany that

were hydraulically fractured prior to gravel

packing.1 In current usage, frac packing refers to

tip-screenout (TSO) fracturing treatments that

create short, wide fractures and gravel packing

of sand-exclusion screens, both in a single oper-

ation. The resulting highly conductive propped

fractures bypass formation damage and alleviate

fines migration by reducing near-wellbore pres-

sure drop and flow velocity.

An early application of frac packing occurredduring 1963 in Venezuela, where producing com-

panies performed small fracturing treatments

using sand and viscous crude oil and then circu-

lated screens into place downhole through sand

remaining inside the casing.2 This approach

proved successful, but was not applied in other

areas until almost 30 years later.

In the ensuing years, operators used various

fracturing techniques to address drilling and

completion damage that often extends deep into

high-permeability reservoirs. These small “slug

fracture,” or “microfracture,” treatments were

designed to address formation damage that acids

or solvents would not remove and that reperfo

rating could not bypass, especially when perfora

tion-tunnel stability was questionable in weakly

consolidated sands.

Interest in frac packing resurfaced in the early

1980s when operators began to fracture high

permeability formations using TSO techniques.

Wider propped fractures yielded sustained pro

duction increases in the Prudhoe Bay and Kuparuk

fields on the North Slope of Alaska, USA, and in

chalk formations of the North Sea. These results

attracted the attention of producers in othe

areas and prompted evaluation of TSO fracturing

for sand control.

Interest in frac packing increased after 1985

driven by activity in the Gulf of Mexico, where

many conventional gravel packs do not achieveadequate productivity. Induced formation dam

age from drilling or completion fluids, cement fil

trate, overbalanced perforating and fines

migration contribute to unsatisfactory results, as

does mechanical skin damage created by the

redistribution of stresses after drilling.

Formation collapse and sand influx as a result o

incomplete gravel packing around screens o

unpacked perforations also restrict production.

Frac packing reduces pressure drops caused

by formation damage and completion restric

tions, which commonly are represented by a

dimensionless value called skin.5 Unlike gravepacking, frac-pack skin decreases as wells pro

duce and treatment fluids are recovered, and

productivity tends to improve over time

Consequently, the trend among operators is to

apply this technique in most of the wells that

require sand control.

1. McLarty JM and DeBonis V: “Gulf Coast Section SPEProduction Operations Study Group—TechnicalHighlights from a Series of Frac Pack TreatmentSymposiums,” paper SPE 30471, presented at the SPEAnnual Technical Conference and Exhibition, Dallas,Texas, USA, October 22–25, 1995.

2. Liebach RE and Cirigliano J: “Gravel Packing inVenezuela,” presented at the Seventh World PetroleumConference, Mexico City, Mexico, 1967, ProceedingsSection III: 407–418.

3. Smith MB, Miller WK and Haga J: “Tip ScreenoutFracturing: A Technique for Soft, Unstable Formations,”paper SPE 13273, presented at the SPE Annual TechnicalConference and Exhibition, Houston, Texas, USA,September 16–19, 1984; also in SPE Production Engineering 2, no. 2 (May 1987): 95–103.

Hannah RR and Walker EJ: “Fracturing a High-Permeability Oil Well at Prudhoe Bay, Alaska,” paperSPE 14372, presented at the SPE Annual TechnicalConference and Exhibition, Las Vegas, Nevada, USA,September 22–25, 1985.

Martins JP, Leung KH, Jackson MR, Stewart DR andCarr AH: “Tip Screenout Fracturing Applied to theRavenspurn South Gas Field Development,” paperSPE 19766, presented at the SPE Annual TechnicalConference and Exhibition, San Antonio, Texas, USA,October 8–11, 1989; also in SPE Production Engineering 7, no. 3 (August 1992): 252–258.

Reimers DR and Clausen RA: “High-PermeabilityFracturing at Prudhoe Bay, Alaska,” paper SPE 22835,presented at the SPE Annual Technical Conference andExhibition, Dallas, Texas, USA, October 6–9, 1981.

Martins PJ, Bartel PA, Kelly RT, Ibe OE and Collins PJ:“Small, Highly Conductive Hydraulic Fractures Near

Reservoir Fluid Contacts: Applications to Prudhoe Bay,”paper SPE 24856, presented at the SPE Annual TechnicalConference and Exhibition, Washington, DC, USA,October 4–7, 1992.

Martins JP, Abel JC, Dyke CG, Michel CM and Stewart G:“Deviated Well Fracturing and Proppant ProductionControl in Prudhoe Bay Field,” paper SPE 24858, pre-sented at the SPE Annual Technical Conference andExhibition, Washington, DC, USA, October 4–7, 1992.

4. Carlson J, Gurley D, King G, Price-Smith C and Walters F:“Sand Control: Why and How?” Oilfield Review 4, no. 4(October 1992): 41–53.

Mechanical skin consists of localized formation damageresulting from redistribution of in-situ stresses after theremoval of rock during the drilling process, especially inextremely permeable reservoirs. Formation stressesoriginally supported by drilled material concentrate near the borehole wall, compressing or crushing the rock

matrix within a cylindrical ring around the wellbore. Thiseffect restricts pore throats and reduces near-wellborepermeability, potentially trapping fine particles thatmigrate toward the well during production.

For more about mechanical skin damage: Morales RH,Brown E, Norman WD, BeBonis V, Mathews MJ, Park EIand Brown R: “Mechanical Skin Damage in Wells,”paper SPE 30459, presented at the SPE Annual TechnicaConference and Exhibition, Dallas, Texas, USA, October22–25, 1995; also in SPE Journal (September 1996): 275–281

5. Negative skin indicates stimulation; positive skin indi-cates damage.


3/18

In the Gulf of Mexico, frac packing became

increasingly popular beginning in the late 1980s.

Amoco, now BP, performed five frac-pack com-

pletions in the Ewing Bank area during 1989 and

1990 by batch mixing up to 6 pounds of proppant

added (ppa) per gallon of treatment fluid.6 In

1991, ARCO, now BP, performed frac packing in

the South Pass area.7 Pennzoil, now Devon

Energy, used this technique in the Eugene Island

area.8 At about the same time, Shell began frac

packing inland wells from barges in Turtle

Bayou field, Louisiana, USA. Later, Shell

expanded the use of this technique in the North

Sea and to offshore wells in Borneo, and also to

onshore wells in Colombia, South America and

northwest Europe.9

Frac-packing success led to increased use,

and this technique soon became the preferred

sand-control method in the Gulf of Mexico,

where several thousand oil and gas leases lie in

water deeper than 3000 ft [914 m]. During 1992,

BP completed frac packs in Mississippi Canyon

Block 109, where water depths range from 850 to1500 ft [260 to 460 m].10 A few years later, Shell

and Chevron used frac packing to develop fields

in water up to 3000 ft deep.

Technology transfer and frac-packing success

in other areas, such as Indonesia, the North Sea,

the Middle East, West Africa and Brazil, are further

expanding the worldwide application of this tech-

nique. Operators plan to frac pack Gulf of Mexico

wells in more than 4000 ft [1220 m] of water, and

in the North Sea and offshore Brazil, intend to push

the frontier of frac packing into water as deep as

6000 ft [1830 m]. Fracture stimulation and frac

packing in high-permeability reservoirs now repre-

sent 20% of the fracturing market.

This article reviews the evolution of frac

packing and discusses developments in stimula-

tion fluids, proppants, downhole equipment,

design simulation, job execution and post-

stimulation evaluation. Case histories illustrate

application of this technique to enhance well

productivity while preventing proppant flowback

and sand production.

Tip-Screenout Fracturing

Gravel packs typically have some degree of

damage—positive skin—and rarely achieve low

skin values consistently. Frac-pack completions,

on the other hand, often result in higher produc-

tivity wells than gravel packs performed below or

above fracture-initiation pressure, either by

slurry packing or high-rate water packing

(HRWP).11

Evaluations of wells completed duringthe past 10 years with these sand-control tech-

niques show the dramatic impact of frac packing

on total completion skin (below).12

The permeability contrast between forma-

tions and propped fractures determines required

fracture length for optimal reservoir stimulation.

In lower permeability reservoirs, there is a large

permeability contrast, and therefore, greater

relative fracture conductivity.13 In high-permeability

reservoirs, there is less contrast, and the relative

conductivity of a narrow fracture is reduced by

several orders of magnitude. This negates the

value of fracture extension away from a well and

underscores the need for wide fractures because

conductivity is also directly proportional to

propped width.

Short, wide fractures stimulate well produc-

tivity even in high-permeability formations.

These highly conductive fractures alleviate sand

production associated with high flow rates, per-

foration collapse in weakly consolidated forma-

tions and fines migration in formations with

poorly sorted grain sizes by reducing near-

wellbore pressure drop and flow velocity. These

factors also defer critical stress conditions that

crush formation grains until a lower reservoir

pressure is reached.

Hydraulic fracturing in low-permeability, or

tight, formations creates narrow propped frac-

tures about 0.1 in. [2.5 mm] wide, extending

1000 ft [300 m] or more from a wellbore (next page,

top left).14

A TSO treatment generates proppedfractures with widths up to 1 in. [2.5 cm] or more

in soft-rock formations and wellbore-to-tip half-

lengths of about 50 ft [15 m], depending on for-

mation characteristics.15 For conventional

treatments, final proppant concentrations in

terms of fracture surface area are less than

2 lbm/ft2 [10 kg/m2]. This contrasts with 5 to

10 lbm/ft2 [24 to 49 kg/m2] concentrations for

TSO designs.

A propped fracture increases completion

radius and area open to flow. Compared with

radial inflow, the resulting bilinear flow pattern

reduces flow convergence and turbulence at theperforations, which enhances productivity. For

example, a propped fracture with a 50-ft half-

length and height of 22 ft [7 m] has 4000 sq ft

[372 m2] of surface area; a gravel-pack comple-

tion in a 9-in. borehole has a maximum surface

area of about 50 sq ft [5 m2] open to radial flow.

The effective completion radius for each of these

hypothetical frac-pack and gravel-pack comple-

tions is 50 ft and 4.5 in. [11.4 cm], respectively.

The tip of a hydraulic fracture is the final area

packed by proppant during conventional fractur-

ing of low-permeability, hard-rock formations. In

contrast, TSO designs limit fracture length, orextension, by achieving fluid-leakoff rates that

dehydrate the proppant slurry early in a treat-

ment. This dehydration causes proppant to pack

off near the peripheral edge, or tip, of a dynamic

fracture. The hydraulic fracture inflates like a bal-

loon as additional proppant-laden fluid is injected,

creating a wider, more conductive pathway as prop-

pant packs toward the well (next page, top right).

32 Oilfield Review

14

12

Frac pack High-ratewater pack

Viscous-slurrygravel pack

10

8

6

Dimensionlessskin

4

2

0

Picture to be added later

> Completion damage. Evaluation of sand-control completions performed in the Gulf of Mexico during the past 10 years show the dramatic impact of fracpacking on dimensionless skin, and therefore, well productivity and ultimatehydrocarbon recovery. Operators report average skins of 12 and 8 for gravel-pack completions performed by viscous-slurry and high-rate water pack(HRWP) techniques, respectively. Frac packing consistently delivers loweraverage skins, typically about 3.


4/18

Summer 2002 33

6. McLarty and DeBonis, reference 1.

7. Hainey BW and Troncoso JC: “Frac-Pack: An InnovativeStimulation and Sand Control Method,” paper SPE 23777,presented at the SPE International Symposium onFormation Damage Control, Lafayette, Louisiana, USA,February 26–27, 1992.

8. Monus FL, Broussard FW, Ayoub JA and Norman WD:“Fracturing Unconsolidated Sand Formations OffshoreGulf of Mexico,” paper SPE 24844, presented at theSPE Annual Technical Conference and Exhibition,Washington, DC, USA, October 4–7, 1992.

Mullen ME, Stewart BR and Norman WD: “Evaluation ofBottom Hole Pressures in 40 Soft Rock Frac-PackCompletions in the Gulf of Mexico,” paper SPE 28532,presented at the SPE Annual Technical Conference andExhibition, New Orleans, Louisiana, USA, September25–28, 1994.

9. Wong GK, Fors RR, Casassa JS, Hite RH andShlyapobersky J: “Design, Execution, and Evaluation ofFrac and Pack (F&P) Treatments in Unconsolidated SandFormations in the Gulf of Mexico,” paper SPE 26563, pre-sented at the SPE Annual Technical Conference andExhibition, Houston, Texas, USA, October 3–6, 1993.

Roodhart LP, Fokker PA, Davies DR, Shlyapobersky J andWong GK: “Frac and Pack Stimulation: Application,Design, and Field Experience From the Gulf of Mexico

to Borneo,” paper SPE 26564, presented at the SPEAnnual Technical Conference and Exhibition, Houston,Texas, USA, October 3–6, 1993.

10. Hannah RR, Park EI, Walsh RE, Porter DA, Black JW andWaters F: “A Field Study of a Combination Fracturing/Gravel Packing Completion Technique on the Amberjack,Mississippi Canyon 109 Field,” paper SPE 26562, pre-sented at the SPE Annual Technical Conference andExhibition, Houston, Texas, USA, October 3–6, 1993; alsoin SPE Production & Facilities 9, no. 4 (November 1994):262–266.

11. Slurry-packing techniques use viscous polymer-basefluids to place high gravel concentrations, while HRWP techniques use lower gravel concentrations transportedin a less viscous fluid, usually brine.

Dynamic fracture

Fracture inflation

Annular opening

Cement

Perforation

Screen

Casing

Propped fracture

“External” proppant pack

Tip screenout

Proppant

> Tip-screenout (TSO) fracturing. In high-permeability reservoirs, fracturestimulations require fluid systems that leak off early in a treatment. Dehydra- tion of the slurry causes proppant to pack off at the fracture tip, halting further propagation, or extension (top ). As additional slurry is pumped, biwingfractures inflate and proppant packs toward the wellbore (middle ). A TSO treatment ensures wider fractures and improves conductivity by promoting

grain-to-grain contact in the proppant pack. This technique also generatesenough formation displacement to create an annular opening betweencement and formation that becomes packed with proppant. This “external”pack connects all perforations and further reduces near-wellbore pressuredrop (bottom ).

Low-Permeability Formations Bilinear Flow

Proppant Pack

Fracture with viscous fluid

Fracture with viscous fluid

Fracture with water

Fracture with water

High-Permeability Formations

Formation

Proppant embedment

> Fracture geometry. In low-permeability formations, viscous fracturingfluids generate long, narrow fractures; less viscous fluids, such as water,leak off quickly and create shorter fractures ( top left ). Hydraulic fractur-ing increases effective completion radius by establishing linear flow intopropped fractures and dominant bilinear flow to a wellbore (top right ). Inhigh-permeability formations, fracturing treatments create short, wide

propped fractures that provide some reservoir stimulation and mitigatesand production by reducing near-wellbore pressure drop and flowvelocity (bottom left ). In low-strength, or soft, formations, proppant con-centration after fracture closure must exceed 2 lbm/ft2 [10 kg/m2] toovercome proppant embedment in fracture walls (bottom right ).

12. Mullen ME, Norman WD and Granger JC: “ProductivityComparison of Sand Control Techniques Used forCompletions in the Vermilion 331 Field,” paper SPE27361, presented at the SPE International Symposium onFormation Damage Control, Lafayette, Louisiana, USA,February 7–10, 1994.

Monus et al, reference 8.

Fletcher PA, Montgomery CT, Ramos GG, Miller ME and

Rich DA: “Using Fracturing as a Technique forControlling Formation Failure,” paper SPE 27899, pre-sented at the SPE Western Regional Meeting, LongBeach, California, USA, March 23–25, 1994; also in SPE Production & Facilities 11, no. 2 (May 1996): 117–121.

Hannah et al, reference 10.

Papinczak A and Miller WK: “Fracture Treatment Design to Overcome Severe Near-Wellbore Damage in aModerate Permeability Reservoir, Mereenie Field,Australia,” paper SPE 25379, presented at the SPE AsiaPacific Oil & Gas Conference and Exhibition, Singapore,February 8–10, 1993.

Stewart BR, Mullen ME, Howard WJ and Norman WD:“Use of a Solids-Free Viscous Carrying Fluid in FracturingApplications: An Economic and Productivity Comparisonin Shallow Completions,” paper SPE 30114, presented at the SPE European Formation Damage Conference, TheHague, The Netherlands, May 15–16, 1995.

13. Fracture conductivity is a measure of how easily producedor injected fluids flow within a propped hydraulic fracture.

14. Hydraulic fracturing begins with injection of a proppant-free fluid stage, or pad, at pressures above formationbreakdown stress to initiate a crack in the rock andcool-down the near-wellbore region. This pad stage cre-ates two fracture “wings” 180 degrees apart that propa-gate in the preferred fracture plane (PFP). The PFP liesin the direction of maximum horizontal stress, perpendicular to the least horizontal rock stress. Proppant-ladenfluid stages follow to generate a required geometry—height, width and length—and pack the biwing fracturewith proppant. Proppants ensure that a conductive pathway remains open after fluid injection stops anddynamic fractures close.

15. Hanna B, Ayoub J and Cooper B: “Rewriting the Rulesfor High-Permeability Stimulation,” Oilfield Review 4,no. 4 (October 1992): 18–23.


5/18

Fracture conductivity and reservoir stimula-

tion do not account for all of the resulting pro-

ductivity increase. Another factor is elimination

of flow restrictions through the perforations.

Aggressive frac packing opens a dynamic frac-

ture up to 2 in. [5 cm] wide across all or most of

the completion interval. The principles of rock

mechanics dictate that the amount of subsurface

movement required to generate wide TSO

fractures also must create an annular opening

outside of the cement sheath. This opening then

is packed with proppant to form a ring, or “halo,”

around the wellbore.

This “external” pack provides a more effec-

tive hydraulic connection between propped frac-

tures and all the perforations, which further

reduces pressure drop across completion inter-

vals. Computer simulations indicate that perfora-

tions that are not aligned with propped fracture

can contribute up to 50% of fluid flow into a well-

bore in high-permeability formations (below).16

The proppant halo is a key factor in frac-pack

success and the basis for screenless completionsthat control sand without mechanical screens

and internal gravel packs (see “Emerging

Technologies,” page 45 ).

Frac packing is a frontline defense against

sand production, and properly designed TSO

fracturing treatments are vital to the success of

this important well-completion technique.

Conventional cased-hole gravel packs often

experience progressive loss of productivity, but

production from properly designed and executed

frac packs tends to improve over time as treat-

ment fluids are recovered and wells clean up.17

Treatment Execution

Initially, operators performed frac packing in mul-

tiple steps—a TSO fracturing treatment followed

by wellbore cleanout, installation of sand-exclu-

sion screens and separate gravel-packing opera-

tions.18 However, high positive skins and limited

productivity indicated damage between the

propped fracture and internal gravel pack. Frac

packing was simplified into a single operation to

further improve well production and reduce oper-

ational costs.19 The TSO fracturing treatment now

is pumped with screens in place. Gravel packing

of screen assemblies is accomplished at the end

of a treatment.

Like conventional gravel packing, fluids and

proppants for frac packing are injected through

tubing and a gravel-pack packer with a service

tool in squeeze or circulating configuration (right).

However, to withstand higher pressures during

TSO fracturing, service companies adapted stan-dard gravel-packing assemblies. Modifications

include increased metal hardness, larger cross-

sectional flow areas and minimizing sudden

changes in flow direction to reduce metal erosion

by fluids and proppants.

Squeeze configuration is used for most frac-

pack treatments, especially in wells with produc-

tion casing that cannot handle high pressures.

Circulating position provides a path for fluid

returns to surface through the tubing-casing

annulus, or communication—a “live” annulus—

to monitor pressure at surface independent of

friction in wellbore tubulars, depending on

whether the annular surface valve is open or

closed. Friction pressures generated by pumping

proppant-laden slurry through tubing and com-

pletion equipment often mask true downhole

pressure responses when monitoring treating

pressure on the tubing.

34 Oilfield Review

100

Flow

throughfracture,%

0100 1000

Permeability, mD

10,000

20

40

60

80

Unaligned perforations

Propped fracture

> Perforation contributions. Inflow is not limited to the propped fracture cross-sectional area and perforations aligned, or connected, with the fracture wings.Computer simulations indicate that unaligned perforations contribute almost50% of the inflow from high-permeability formations, underscoring the impor- tance of TSO fracturing and creation of an external pack.

QUANTUMgravel-pack packer

Mechanical fluid-losscontrol device

Washpipe

Screens

Perforations

Screens

Perforations

Ball seat

Ball valve

Fluid flow

Crossover ports

Service tool

Annular BOP

Annulus surface valveand pressure gauge

Circulating ports

Temperature andpressure gauge

Bottom packer

> Downhole tools. In gravel packing and fracpacking, a service tool directs fluid flow througha gravel-pack packer and around the screenassembly. Squeeze configuration is establishedby closing the annular blowout preventer (BOP)and the tubing-casing annulus surface valve

(left ), or by closing the ball valve downhole(right ). Shutting in the annulus with the downholeball valve open allows bottomhole pressure to bemonitored independent of friction in the tubing.Closing the downhole valve prevents fluid returns to surface and protects weak casing from highpressures; pressure also can be applied to theannulus to offset high pressure in the tubing.Mechanical devices such as flapper valves or the FIV Formation Isolation Valve system preventexcess fluid loss into formations after the service tool is retrieved.


6/18

Summer 2002 35

Early service tools used a conventional check

valve, which prohibited pressure declines from

being observed after fracturing. More recent

designs of QUANTUM gravel-pack packer tools

eliminate the check valve, replacing it with an

improved downhole ball valve that allows pres-

sure fluctuations to be monitored in real time dur-

ing treatments when the ball valve is open. A live

annulus allows more accurate evaluation

of treatments.20

Frac packing usually begins in squeeze con-

figuration. After tip screenout occurs, establish-

ing circulating configuration ensures complete

packing of the screens and grain-to-grain prop-

pant contact. The service tool then is shifted to

clean out excess slurry by pumping fluid down

the annulus and up the tubing. The amount of

upward movement required to shift some service

tools pulls reservoir fluids into a wellbore. This

swabbing effect can bring formation sand into

perforation tunnels before a fracture is com-

pletely packed or reduce conductivity between

fractures and the internal gravel pack, which canlimit frac-pack productivity.

Set-down service tools, such as the

QUANTUM gravel-packing system, close the

downhole ball valve and shift tool configuration

with upward movement. This type of tool also is

used for deep completions and treatments con-

ducted from floating rigs or drillships.

In addition to a variety of reservoir conditions

and of fracturing and gravel-packing requirements,

treatment execution must address the complexity

of completing multiple zones and long intervals.

Even the best frac-pack designs end in failure if

excess fluid loss into formation causes proppantbridges to form between screens and casing,

restricting or blocking annular flow. Annular prop-

pant packoff, or bridging, results in early treatment

termination, low fracture conductivity and an

incomplete gravel pack around screens.

Placing proppant with sand-exclusion screens

in place requires close attention to annular clear-

ances. As frictional pressure increases, there is

potential for fluid from slurry in the screen-casing

annulus to pass through the screens into the

washpipe-screen annulus. Fluid bypass worsens

as the slurry dehydrates, and proppant concentra-

tion increases to an unpumpable state, causingproppant to bridge in the screen-casing annulus.

Annular blockage near the top of a comple-

tion interval prevents continued fracturing of

deeper zones or zones with higher in-situ stress

and inhibits subsequent packing of the screens.

Even a partial flow restriction in the annulus

increases frictional pressure drop, restricts rate

distribution and limits fracture-height growth

across the remainder of the completion interval.

Annular voids below a proppant bridge increase

the likelihood of screen failure from erosion by

produced fluids and fine formation sand.

For homogeneous reservoirs where pay inter-

vals are less than 60 ft [18 m] thick, fracture-

height growth typically covers the entire zone. In

longer intervals, the probability of complete frac-

ture coverage decreases, and risk of proppantbridging increases dramatically. Long intervals

can be split into stages and treated separately.

This requires more downhole equipment, such as

two stacked frac-packing assemblies, and addi-

tional installation time, but increases frac-pack-

ing effectiveness (see “Conventional and

Alternate Path Screens,” next page ).

Alternate Path technology is also available to

gravel pack and frac pack longer intervals (right).

AllFRAC screens use hollow rectangular tubes, or

shunts, welded on the outside of screens to pro-

vide additional flow paths for slurry. Exit ports

with carbide-strengthened nozzles located alongthe shunt tubes then allow fluids and proppant to

exit below annular restrictions, which allows

fracturing and annular packing to continue after

restrictions form in the screen-casing annulus.

AllFRAC screens for frac packing use slightly

Screen

Basepipe

Fracture

Shunttubes

Nozzle

Casing

Shunt tubes

Perforations Screens

Annularproppant bridge

Void

Nozzle

> Alternate Path technology. Proppant bridges, ornodes, that form in the screen-casing annulus,commonly as a result of slurry dehydration orpremature fracture screenout in zones withlower in-situ stress, cause early treatment termi-nation. In wells with conventional sand-exclusionscreens, this limits fracture height and frac-packefficiency. Alternate Path technology uses shunt tubes with strategically located exit nozzleswelded on the outside of conventional screens(top and middle ). Shunt tubes provide a flow pathfor slurry that bypasses annular restrictions to

allow continued treating of lower intervals andpacking of voids around the screens (bottom ).

16. Burton RC, Rester S and Davis ER: “Comparison ofNumerical and Analytical Inflow Performance Modellingof Gravelpacked and Frac-Packed Wells,” paper SPE31102, presented at the SPE International Symposium onFormation Damage Control, Lafayette, Louisiana, USA,February 14–15, 1996.

Guinot F, Zhao J, James S and d’Huteau E: “ScreenlessCompletions: The Development, Application and FieldValidation of a Simplified Model for Improved Reliability

of Fracturing for Sand Control Treatments,” paper SPE68934, presented at the SPE European FormationDamage Conference, The Hague, The Netherlands,May 21–22, 2001.

17. Stewart BR, Mullen ME, Ellis RC, Norman WD and MillerWK: “Economic Justification for Fracturing Moderate to High Permeability Formations in Sand ControlEnvironments,” paper SPE 30470, presented at the SPEAnnual Technical Conference and Exhibition, Dallas,Texas, USA, October 22–25, 1995.

18. Monus et al, reference 8.

19. Hannah et al, reference 10.

20. Mullen et al, reference 8.


7/18

larger shunt tubes than AllPAC screens for gravel

packing to accommodate higher injection rates

for fracturing.

Shunt tubes provide conduits for slurry to

bypass collapsed hole and external zonal isola-

tion packers as well as annular proppant gravel

bridges at the top of intervals or adjacent to

higher permeability zones with high fluid leakoff.

If annular restrictions form, injection pressure

increases and slurry diverts into the shunt tubes,

the only open flow path. This ensures fracture

coverage and complete gravel packing around

screens across an entire perforated interval.

Conventional and Alternate Path Screens

In the late 1990s, Saudi Aramco chose frac pack-

ing to control sand in oil wells about 200 km

[124 miles] southeast of Riyadh, Saudi Arabia

(below).21 This new field in the Central Province

encompassed two heterogeneous Permian-age

reservoirs comprising high-permeability sand-

stones at 8700 to 9000 ft [2650 to 2740 m] that

are interbedded with shale and siltstone.The deeper B reservoir was a high-quality

sandstone interbedded with thin, low-permeabil-

ity siltstone. Reservoir thickness varied from 20

to 65 ft [6 to 20 m]. Well tests indicated perme-

ability from 0.5 to 2 darcies; air-derived core per-

meabilities were 3 to 4 darcies. The A reservoir

was a sequence of slightly more heterogeneous

individual sandstones between lower permeabil-

ity siltstone strata. This overlying reservoir was

up to 200 ft [61 m] thick with net pay up to 75 ft

[23 m]. Permeabilities from well tests were 0.1 to

2.5 darcies; air-derived core permeabilities were

about 2 darcies.

A well completed without sand-control mea-

sures produced for less than six months before

sand influx and suspected perforation collapse

stopped production. If completion practices

induced a significant pressure drop downhole, it

would be difficult to control sand at oil rates and

wellhead pressures that met production targets

while allowing wells to flow naturally into exist-

ing facilities 50 km [31 miles] away. Frac packing

satisfied well-completion requirements for both

the A and B reservoirs.

Long completion intervals necessitated differ-

ent frac-packing techniques for each reservoir(next page, top). Saudi Aramco used conventional

screens in the B reservoir where pay zones were

less than 60 ft thick. For longer perforated inter-

vals in the A reservoir, the operator chose

AllFRAC Alternate Path screens with three shunt

tubes, each designed for 6 bbl/min [1 m3 /min], to

achieve required injection rates (next page, bottom).

Wells with an oil-water contact near the bot-

tom perforations required close control of frac-

ture height to avoid early water breakthrough. In

other wells, perforations extended over long

intervals and individual zones were spaced far

apart. Engineers selected a stacked-screen com-

pletion to meet frac-packing objectives in these

wells. Dividing the productive interval into two

sections allowed Saudi Aramco to optimize treat-

ment designs for each zone and avoid fracturing

into water-bearing zones.

Typically, these frac-packing treatments

included the pad, an initial low-concentration

stage with 0.5 lbm/gal [0.06 kg/L], or pounds of

proppant added (ppa) per gallon of fracturing

fluid, and additional proppant stages ramped up

to 3, 6 or 9 ppa [0.36, 0.72 or 1 kg/L]. In a few

wells, 9-ppa stages were pumped successfully.

Higher proppant concentrations were difficult to

place in more permeable zones, but placing 3 and6 ppa in the formation yielded good results.

Saudi Aramco and Schlumberger modified ini-

tial fracturing designs based on minifracture

analysis using the Schlumberger DataFRAC frac-

ture data determination service (see “Design and

Implementation,” page 38 ). Fracture-closure

stress, fluid-leakoff coefficient and fracture

height from these pretreatment injectivity tests

helped ensure that the main treatments achieved

a tip screenout. The operator adjusted pad and

proppant stages as needed and accounted for

high fluid leakoff in the B sand by increasing

the maximum injection rate to 18 bbl/min[2.9 m3 /min]. Engineers also restricted pump

rates to 16 bbl/min [2.5 m3 /min] for the A-reser-

voir Alternate Path completions to limit friction

pressure in the shunt tubes.

The operator performed post-treatment

hydrochloric [HCl] acid jobs on some wells to

reduce cleanup time. Other wells cleaned up

within two months without acid treatments.

Overall well productivity continued to improve as

treatment fluids were recovered. Experience from

the first wells helped optimize frac-packing pro-

cedures. Saudi Aramco reduced polymer concen-

trations in treatment fluids and includedslow-release encapsulated breakers to optimize

fracture placement and post-treatment cleanup.

36 Oilfield Review

21. Shepherd D and Toffanin E: “Frac Packing UsingConventional and Alternative Path Technology,” paperSPE 39478, presented at the SPE InternationalSymposium on Formation Damage Control, Lafayette,Louisiana, USA, February 18–19, 1998.

Riyadh

Welllocations

Saudi Arabia

EUROPE

AFRICA

SAUDI ARABIA

IRAN

IRAQ

SUDAN

ERITREA

EGYPT

YEMEN

OMAN

UAER

e

d

S e a

T h e

G u

l f

A r a b i a

n S

e a

0

0 300 600 900 km

200 400 600 miles

> Sand-control completions onshore. In the late 1990s, Saudi Aramco beganfrac packing new oil-well completions in central Saudi Arabia about 200 km[124 miles] southeast of Riyadh. These frac packs controlled sand influx andreduced downhole pressure drop, allowing the wells to flow naturally intofacilities 50 km [31 miles] away under existing intake-pressure conditions.


8/18

Summer 2002 37

0

Gamma Ray, API Depth, ft

8800

8850

8900

8950

9000

9050

Resistivity, ohm-m

200 2 3

Perforations

Flapper valve

Washpipe

Tubing

Bottompacker

Conventional

Gravel-packpacker

< B-sand completions. A typical well log indi-cates a maximum pay interval of about 65 ft withclosely spaced perforations in the B reservoir(left ). Relatively short perforated intervals allowedSaudi Aramco to install a single completion andfrac pack these lower sands using standardscreens (right ).

0

Gamma Ray, API Depth, ft

8800

8850

8750

8900

8950

9000

Resistivity, ohm-m

100 2 3

Perforations

Tubing

Gravel-packpacker

Flapper valve

Gravel-packpacker

Flapper valve

packer

< A-sand completions. A typical well log showsperforations across a 180-ft [55-m] interval of theA reservoir (left ). Saudi Aramco performed twoseparate treatments using a stacked-screenassembly to frac pack these longer intervals(right ). Standard screens were used for the low-est zone, which was shorter. AllFRAC screenswith shunt tubes were installed to complete the

longer upper zone.


9/18

The initial group of wells included five com-

pletions with conventional screens in the B

reservoir and five completions in the A reservoir

with AllFRAC screens that successfully treated

intervals up to 200 ft. Two completions in the A

reservoir used stacked-screen assemblies.

Treatments were pumped through 9000 ft of 3-in.

outside diameter (OD) tubing at surface injection

pressures below 10,000 psi [69 MPa] and pump

rates of 14 bbl/min [2.2 m3 /min] to 18 bbl/min.

The operator performed well tests by flowing

through surface facilities or with downhole pro-

duction logging tools to evaluate 12 frac-packing

treatments on the first 10 completions in this

field (left). Except for two, these frac-pack

completions yielded low completion skin and

good sand control in formations with up to

3-darcy permeability.

A positive frac-pack skin is caused by inade-

quate connectivity between propped fractures

and wellbores, incomplete zone coverage or fail-

ure to achieve a high-conductivity TSO fracture. If

these conditions produce a completion skin of 8or more, well productivity may be no better than

that of a conventional gravel pack. Achieving

optimal frac-pack performance, as in the case of

these Saudi Aramco wells, requires detailed pre-

liminary designs, careful proppant and fluid

selection, accurate pretreatment injectivity test-

ing and treatment optimization as needed.

Design and Implementation

During the initial design of frac-packing treat-

ments, completion engineers determine required

fracture geometry based on reservoir conditions,

rock properties and barriers to fracture-heightgrowth. Fracture length and, more importantly for

high-permeability formations, fracture width

enhance well productivity. Operators select an

optimal TSO fracture design by maximizing the

net present value (NPV) from enhanced well pro-

ductivity (left).22

Proppant selection —The type of proppant

chosen to keep fractures open and form a granular

filter is an important design consideration. Frac-

packing success is due, in part, to larger proppant

sizes than those commonly used in gravel packing.

High concentrations of large, spherical proppants

minimize embedment and offset the effects of tur-bulent flow in propped fractures.

Operators use various grain sizes and

proppant types, including natural sand, custom-

sieved sand, resin-coated sand and intermediate

or high-strength man-made ceramic proppants,

depending on formation stress and fracture-

closure pressure. Proppants for frac packing

should accomplish four fracturing objectives:

38 Oilfield Review

20

Well

Dimensionlessskin

0

-5

10

15

5

1

11

3

0

2

20

5

3 3

1

-1

2 3 4 5 6 7 8 9 10

> Frac-pack performance. Saudi Aramco has frac packed 23 wells in the Cen- tral Province of Saudia Arabia and published results from an initial group of10 wells completed with 12 frac packs. In Wells 6 and 7, the operator usedstacked-screen assemblies and divided the completion intervals into twostages to optimize treatment designs and avoid fracturing into an underlyingwater-bearing zone. Premature screenout and early treatment termination inWells 2 and 5 contributed to high skins and poor productivity, confirming thatfracture conductivity and connectivity with the wellbore are critical factors.Eight wells had skins below those expected in conventional gravel packs and typical frac packs.

3.520

2030

4050

6070 6

7

8

9

Propp

ant c

oncen

tratio

n, lbm

/ft2

Fracture half-length, ft

Optimal

10

11

3.53

3.54

3.55

3.56

Netpresentvalue(NPV),millionsofdollar

s

3.57

3.58

3.59

3.60

> Frac-pack economics. Optimal fracture half-length and width maximize netpresent value (NPV). In this example, the optimal fracture length and width, orproppant concentration, are 30 ft [9 m] and 7 lbm/ft2 [34 kg/m2], respectively.Discounted operating costs and incremental income are expressed in pre-sent-day value. Completion and stimulation investments and discounted oper-ating costs are subtracted from discounted incremental income to arrive at the NPV for a treatment. Incremental income increases for longer and widerfractures, but at some point increasing costs for larger treatments yielddiminishing returns.


10/18

Summer 2002 39

• provide an effective permeability contrast

• control sand influx and fines migration

• minimize proppant embedment in soft rock

•maintain fracture conductivity without

proppant crushing.

In the past, gravel-packing considerations

dominated proppant selection.23 Gravel packs

require gravel, or sand, sized to prevent forma-

tion particles and fines from invading the annular

pack. The widely accepted Saucier rule dictates

that sand, or gravel, particles be five to six times

the mean particle diameter of formation grains.24

Fracture permeability and conductivity improve

as proppant sizes become larger, but production

of formation sand grains and fine particles that

reduce pack permeability also increases. Frac

packs require proppants sized to optimize frac-

ture permeability.

In the early 1990s, operators began evaluat-

ing larger sizes of stronger proppants to increase

fracture permeability and relative conductivity in

high-permeability reservoirs.25 For example,

larger 20/40-mesh proppants were used used forfrac packing instead of smaller 40/60-mesh prop-

pants often required for gravel packing. 26

Experience indicated that proppant sizes dictated

by gravel-packing criteria could be increased to

next larger size for frac packing.

Saucier criteria for sizing proppants in relation

to formation grain size were relaxed in frac-pack

designs because the large flow area of hydraulic

fractures mitigates formation failure and sand

influx. Balancing the mechanisms of sand produc-

tion—flow velocity, proppant particle sizes and

fluid properties—allows operators to increase

fracture conductivity and improve frac-packperformance by using larger proppant sizes.

Completing deeper wells with high fracture-

closure stresses led operators to use more man-

made ceramic proppants because they are

stronger and their consistent spherical shape

reduces embedment, which also increases frac-

ture conductivity (above right). The majority of

frac packs use ceramic 20/40-mesh intermedi-

ate-strength proppant (ISP) when reservoirs have

good pressure support and closure stresses are

not excessive.

Fluid selection —After evaluating reservoir

characteristics, engineers choose an optimalfluid for combined stimulation and gravel pack-

ing. The polymer-based hydroxyethylcellulose

(HEC) fluids used in gravel packing, hydroxypropyl

guar (HPG) fracturing fluids with a borate

crosslinker for additional viscosity, and more

recently, viscoelastic surfactant (VES) fracturing

fluids, all are applicable. Frac-packing fluids must

have a variety of properties.27

Fluid selection depends primarily on TSO frac-

turing criteria. Unlike massive hydraulic fracture

stimulations in low-permeability formations, a

low leakoff rate, or high fluid efficiency, is less

desirable for frac packing. In fact, a somewhat

inefficient fluid helps achieve tip screenout andpromote grain-to-grain proppant contact from

fracture tip to wellbore.

However, frac-packing fluids also must

maintain sufficient viscosity to create wide

dynamic fractures and place high proppant con-

centrations that ensure adequate conductivity

after fracture closure.28 After tip screenout, fluid

systems transport proppant in the low-shear

environment of a wide dynamic fracture, but also

must suspend proppant under higher shear rates

in tubing, around screen assemblies, through

the perforations and during fracture initiation

and propagation.Fluid viscosity should break easily to minimize

formation and proppant-pack damage after treat-

ments. Optimal fluids need to be compatible with

formations and chemicals such as polymer break-

ers; they must also produce low friction and clean

up quickly during post-treatment flowback. To max-

imize retained fracture conductivity, operators

exercise great care with viscosity breakers o

acid treatments after frac packing to optimize

post-treatment cleanup for maximum productiv

ity and hydrocarbon recovery. Finally, frac-pack

ing fluids should be safe, cost-effective and easy

to mix, especially in offshore applications.

22. Morales RH, Norman WD, Ali S and Castille C: “OptimumFractures in High Permeability Formations,” paper SPE36417, presented at the SPE Annual TechnicalConference and Exhibition, Denver, Colorado, USA,October 6–9, 1996; also in SPE Production & Facilities 15no. 2 (May 2000): 69–75.


24. Saucier RJ: “Considerations in Gravel Pack Design,”Journal of Petroleum Technology 26, no. 2 (February1974): 205–212.

25. Hainey BW and Troncoso JC: “Frac-Pack: An InnovativeStimulation and Sand Control Technique,” paper SPE23777, presented at the SPE International Symposium onFormation Damage Control, Lafayette, Louisiana, USA,February 26–27, 1992.

26. Naturally occurring sand and man-made ceramic prop-

pants are specified according to sieve analysis based onparticle-size distributions and percentage of particlesretained by screens with standard U.S. mesh sizes.

27. Hainey and Troncoso, reference 25.

28. Morales RH, Gadiyar BR, Bowman MD, Wallace C andNorman WD: “Fluid Characterization for Placing anEffective Frac/Pack,” paper SPE 71658, presented at theSPE Annual Technical Conference and Exhibition, NewOrleans, Louisiana, USA, September 30–October 3, 2001.

1000

Permeab

ility,darcies

0 2 4 6 8 10 12

Closure stress, 1000 psi

100

10

30/50 Ceramic

20/40 Natural sand

40/60 Natural sand

20/40 ISP ceramic

20/40 Ceramic

> Proppant specifications. In the mid-1990s, operators began using larger,stronger and more conductive proppants in frac-pack completions. Man-made ceramic materials have since become the proppant of choice in the USGulf of Mexico to maintain fracture conductivity at higher closure stress indeeper formations. For example, switching from smaller 40/60-mesh sand(green) to a larger intermediate-strength 20/40-mesh ceramic proppant (yel-

low) increases proppant permeability and fracture conductivity by a factor ofsix in laboratory tests at 2000 psi [13.8 MPa] of closure pressure (inset ). Anintermediate-strength proppant (ISP) is priced competitively with custom-sieved natural sands.


11/18

Fluids based on HEC have many preferred

frac-packing characteristics, but also several

drawbacks. Systems based on HEC exhibit

increased friction pressures compared with

delayed crosslinked HPG or VES fluids, and fric-

tional losses become significant in deeper wells

or smaller diameter tubulars. In addition, prop-

pant transport characteristics for HEC fluids are

not as good as those of crosslinked HPG or VES

fluids. High temperatures cause HEC fluids to thin,

and viscosity is not as high at low shear rates.

High-viscosity crosslinked HPG systems leave

some polymer residue, but maximize fracture-

height growth in moderate- to high-permeability

formations. They also perform well in longer

intervals and transport higher proppant concen-

trations for greater fracture conductivity.

Pumping pressures increase with HPG systems,

but service companies can used a delayed

crosslinker to reduce tubular friction.

Delayed-crosslink HPG fluids start at a lower

viscosity and require less hydraulic horsepower

to pump downhole. Prior to reaching the perfora-tions, temperature in the wellbore and fluid pH

cause the viscosity of these fluids to increase in

order to achieve low fluid-leakoff rates. The

majority of frac packs are pumped with

crosslinked or delayed-crosslink HPG fluids.

Viscoelastic ClearFRAC polymer-free fracturing

fluids, introduced in the mid-1990s, use a VES liq-

uid-gelling agent to develop viscosity in light

brines. This type of fluid provides low friction pres-

sures while pumping, enough viscosity at low

shear rates for good proppant transport, adequate

leakoff rates to ensure tip screenout and high

retained permeability for better fracture conduc-tivity. Field data also indicate that fracture con-

finement using VES fluids is better than with

conventional fracturing fluids, which is an advan-

tage when frac-packing near water-bearing zones.

These VES systems mix easily and do not

require additives such as bactericides, breakers,

demulsifiers, crosslinkers, chemical buffers or

delayed-crosslink agents. Systems based on VES

also are not susceptible to bacterial attack. If

wells must be shut in for extended periods before

flowback and cleanup, solids-free ClearFRAC flu-

ids are recommended to avoid precipitation of

damaging polymer materials.Fluids based on HEC and VES systems mini-

mize formation damage in zones with low to mod-

erate permeability, but high leakoff rates and

deeper invasion often result in slower recovery of

treatment fluids.29 Adding enzyme or oxidizing

breakers to frac-packing fluids reduces formation

damage and improves well cleanup. Slow-release

40 Oilfield Review

Shearrate,sec

1

Viscosity,cp

0 10 20 30 40100

1000

10,000

45 ppt

40 ppt

35 ppt

Fluid shear

Time, min

Fracture extension Tip screenout0.1

1

10

100

1000

> Fluid viscosity versus typical shear rate (blue) in laboratory tests.Under frac-packing conditions of fracture extension and tip screenoutin an Amoco, now BP, Matagorda Island field in the US Gulf of Mexico,a 35-ppt crosslinked HPG fracturing fluid (green) exhibited adequateviscosity behavior, while 40- and 45-ppt systems (red and gold, respectively) had unneccessarily high viscosities.

70

60

50

40

30

Gasrate,MMscf/D

20

10

0

1 2 3 4Well

5 6 7

50 ppt 35 ppt

> Improving frac-pack productivity. Production from frac-pack completions ina field of the Gulf of Mexico Matagorda Island area doubled after Amoco,now BP, began using an optimized 35-ppt crosslinked HPG fluid (Wells 5–7)instead of an initial fluid system with 50-ppt polymer concentration (Wells 1–4).Well 7 also had a high productivity ratio, but output was limited by small pro-duction tubing.


12/18

Summer 2002 4

encapsulated breakers deposited in the proppant

pack allow higher breaker concentrations to be

used without sacrificing fluid efficiency.

In addition to fluid leakoff and friction pres-

sure considerations, shear rate and temperature

are critical in selecting frac-packing fluids and

polymer concentrations.30 The first frac-pack

treatments were performed using the same HEC

fluid systems as gravel-packing operations. Later,

a shift to more conventional fracturing fluids

occurred because of increasing temperature

requirements and the need to maximize fracture

conductivity in high-permeability formations.

Initially, selection criteria for these fluids

were similar to those of conventional fracturing

applications in which narrow hydraulic fractures

in consolidated, low-permeability formations cre-

ate high shear rates with low fluid-leakoff rates.

These factors result in breakdown of fluid viscos-

ity and less cooling of formations, and greater

polymer concentrations are required to maintain

viscosity throughout a treatment. The use of

higher polymer concentrations carried over intofracturing and frac-packing designs for high-

permeability reservoirs.

In frac packing, however, fractures are wider

with lower fluid velocities and shear rates.

Pretreatment fluid injection also decreases for-

mation temperature near the wellbore. Pumping

large volumes of treatment fluid decreases heat

transfer from a reservoir, resulting in cooler tem-

peratures inside a fracture. Failure to consider

these effects results in use of higher polymer

concentrations than actually required. This

increases the potential for formation damage and

decreases the likelihood of a tip screenout.For example, because of differences in shear

rate, a crosslinked fluid with a polymer loading

of 20 lbm/1000 gal (ppt) [2.4 kg/m3] of base

fluid can have the same viscosity in a high-

permeability formation as a 40-ppt [4.8-kg/m3]

fluid in a low-permeability formation. Proper fluid

selection and specification dramatically increase

frac-packing efficiency and well productivity.

In 1996, Amoco, now BP, completed four

Matagorda Island wells in the western Gulf of

Mexico by frac packing.31 The reservoir tempera-

ture was 300ºF [150ºC], so the operator chose a

high-viscosity 50-ppt [6-kg/m3] crosslinked HPGfluid that was also used in fracture-stimulation

treatments for low-permeability reservoirs.

Production from these frac-pack completions was

comparable to that of gravel-packed wells. The

operator attributed the relatively poor perfor-

mance to lack of tip screenout because of

improper fluid design.

The operator and Schlumberger evaluated the

effects of shear rate on fluid properties in orderto remedy poor performance (previous page,

top).32 Based on the results of this investigation,

frac-pack treatments on the next three wells

used a 35-ppt [4.2-kg/m3] fluid. Fluid efficiency

decreased because of lower viscosity, allowing

better slurry dehydration, which achieved desired

TSO results. Average daily production from these

wells doubled compared with the initial four

wells (previous page, bottom).

Pretreatment testing —Laboratory testing

and history matching of previous treatments pro-

vide insight into stress profiles and the perfor-

mance of treatment fluids, but in-situ formationproperties vary significantly in high-permeability

unconsolidated reservoirs. After developing

preliminary stimulation designs, engineers per-

form a pretreatment evaluation, or minifracture,

to quantify five critical parameters, including

fracture-propagation pressure, fracture-closure

pressure, fracture geometry, fluid efficiency

and leakoff.33

This procedure consists of two tests, a stress

test and a calibration test, performed prior to the

main treatment to determine specific reservoir

properties and establish the performance charac-

teristics of actual treatment fluids in the payzone. A stress, or closure, test determines mini-

mum in-situ rock stress, which is a critical refer-

ence pressure for frac-pack analysis and

proppant selection (above).

A calibration test involves injecting actual frac-

turing fluid without proppant at the design

treatment rate to determine formation-specific

fluid efficiency and fluid-loss coefficientsFracture-height growth can be estimated by tag

ging proppants with radioactive tracers and

running a post-treatment gamma ray log

A pressure-decline analysis confirms rock prop

erties and provides data on fluid loss and

fluid efficiency.

An integral part of pretreatment testing is live

annulus monitoring and real-time measurements

from downhole quartz gauges to obtain pressure

responses independent of frictional pumping pres

sures. Accurate analysis using DataFRAC services

ensures that the current frac-pack design and sub

sequent treatments achieve wide fractures with atip screenout for optimal results.

Surface data from pretreatment tests com

bined with bottomhole injection pressures are

history matched using a computer simulator

such as the SandCADE gravel-pack design and

evaluation software, to calibrate the fracturing

model and finalize treatment design. Calibrated

data from DataFRAC analysis are also used to

assess stimulation effectiveness during post

treatment evaluations.


30. Morales et al, reference 28.

31. Norman WD, Mukherjee H, Morales HR, Attong D,Webb TR and Tatarski AM: “Optimized Fracpack DesignResults in Production Increase in the Matagorda IslandArea,” paper SPE 49045, prepared for presentation at

the SPE Annual Technical Conference and Exhibition,New Orleans, Louisiana, USA, September 27–30, 1998.

32. Morales et al, reference 28.


Bottomholepressure

Time

Increasinginjection

rate

Constantinjection

rate

Constantflowback

Shut in Constantinjection rate

Net pressure

Fracture closure pressure

Instantaneous shut-inpressure (ISIP)

Fracture-extensionpressure

Reboundpressure

Pressurefalloff

> Pretreatment minifracture testing. Stress, or closure, tests involve injecting low-viscosity, nondam-aging fluid at increasing rates to initiate a fracture and determine the pressure required to propagate,or extend, fracture length. Fracture-closure pressure is determined by monitoring pressure declineduring a slow, constant-rate flowback.


13/18

Treatment design, particularly TSO fracture

stimulation, is critically important to successful

frac packing. If premature screenout or failure to

achieve a tip screenout results in insufficient

fracture width to overcome proppant embedment

in the formation, well productivity may, at best,

be equivalent to that of a conventional gravel

pack. Standard frac-packing practice is to

redesign treatments on-site after minifracture

testing and analysis are complete.

Treatment design —Previously, frac packs,

which sometimes failed because of premature

fracture screenout or early annular packoff,

were designed solely using hydraulic-fracturing

simulators that neglected gravel-packing and

completion equipment inside the wellbore, such

as crossover ports in gravel-pack packers, blank

pipe, screens and washpipe. With the SandCADE

simulator, engineers now specify tip-screenout

designs and simulate frac-packing treatments

using coupled wellbore and hydraulic-fracturing

simulators.34 This software also simulates slurry

flow including the effects of well inclination,gravel setting and bridging around screens, and

fluid flow through packer and screen assemblies.

The fracture simulator supports tip-screenout

designs in high-permeability formations. Inducing

gravel packoff in wellbores by deliberately reduc-

ing pump rate or shifting service tools to circulate

at the end of a treatment also can be modeled.

The SandCADE simulator also models fracturing

of multiple layers and shunt-tube flow (below).

Well-Completion Applications

Fracturing designs based on TSO technology,

larger proppant particle sizes, advances in frac-

turing fluids and improved treatment evaluation

combined with more robust and versatile pump-

ing equipment and downhole tools make frac

packs a viable completion alternative in many

wells. Experience from more than 4000 Gulf of

Mexico frac packs in formations with permeabil-

ities ranging from 3 mD to 3 darcies helps oil and

gas producers identify frac-pack candidate wells

(next page, left). Frac-packing well-completion

applications include the following:

• wells prone to fines migration and sanding

• high-permeability, easily damaged formations

• high-rate gas wells

• low-permeability zones requiring stimulation• laminated sand-shale sequences

• heterogeneous pay zones

• low-pressure and depleted reservoirs.35

Today, operators select sand-control methods

by determining first if conditions justify frac

packing. There are 11 major advantages to

frac packing:

• bypass formation damage

•increase completion radius and flow area

•reduce pressure drop and fluid velocity

•connect individual laminated zones

•re-stress borehole relaxation after drilling

• alleviate fines migration and sand production

•improve well productivity

•achieve consistent low-skin completions

•sustain increased production

•maintain completion longevity

•reduce the likelihood of sand-control failure.

Most wells requiring sand control benefit

from frac packing. Exceptions include locations

where high-pressure pumping equipment is

unavailable, wells with casing sizes that are less

than 5 in., wells with weak casing when there is

a risk of failure and loss of wellbore integrity or

completions with a possibility of fracture-height

growth into water or gas zones. Frac packing alsomay not be economical for low-rate wells,

water-source or injection wells that do not pro-

duce revenue directly, and reservoirs with

42 Oilfield Review

35603580360036203640366036803700

37803800382038403860388039003920

388039003920394039603980

40004020-0.1 0

Fracture widthat wellbore, in.

Fracture half-length, ft Fracture widthat wellbore, in.

Fracture half-length, ft

0.1 0 10

Depth,ft

Depth,ft

20 30 40 50 60 70 80 90

35603580360036203640366036803700

37803800382038403860388039003920

38803900392039403960398040004020

-0.1 0 0.1 0 10 20 30 40 50 60 70

Proppantconcentration,lbm/ft2

0–2

2–4

4–6

6–8

8–

1010–12

12–14

>14

> Modeling frac-packing treatments. The Schlumberger SandCADE simulator is the only commercially available software that accounts for completion andgravel-packing equipment. A hydraulic-fracturing simulator that calculates fracture geometry, proppant distribution in fractures, and two-dimensional fluidflow as boundary conditions is coupled to wellbore simulator, which models fluid and slurry flow in the screen-casing annulus and Alternate Path shunt tubes. A special feature simulates fracturing of multiple layers with or without shunt tubes. This example illustrates simultaneous frac packing of threezones. Without shunt tubes, the treatment places most of proppant in the middle zone (left ). Alternate Path screens ensure treatment of the entire completion interval as well as more uniform fracture lengths and widths (right ).


14/18

Summer 2002 43

limited reserves or homogeneous thick zones

where horizontal gravel packing in openhole is

more appropriate.36

In more prolific reservoirs, flow turbulence

associated with perforated casing restricts pro-

duction, so operators often drill and complete

openhole horizontal wells to optimize productiv-ity. Stand-alone screens, openhole gravel packs

or expandable screens are sand-control options

in these settings, especially for thick reservoir

sections. Frac packing in openhole completions is

the next logical step to provide long-term sand

control without sacrificing productivity.

Openhole Frac Packing

Widuri field, operated by Repsol YPF, lies off

shore in the Java Sea, Indonesia (above). Drilled

in an area scheduled for waterflooding, Well B

28 targeted a thin sandstone of the Talang Aka

formation at 3500 to 3600 ft [1067 to 1097 m

with 29% porosity and 1- to 2-darcy permeability.37 Original reservoir pressure was

1350 psi [9.3 MPa], but solution-gas drive and

weak aquifer support resulted in rapid depletion

to 600 psi [4 MPa]. Moderate consolidation and a

tendency to produce sand mandated sand-contro

completions. Initially, wells were completed as

cased-hole gravel packs. Because of lower reser

voir pressure, the operator planned cased-hole

frac packs for new wells.

34. Sherlock-Willis T, Romero J and Rajan S: “A CoupledWellbore-Hydraulic Fracture Simulator for RigorousAnalysis of Frac-Pack Applications,” paper SPE 39477,

presented at the SPE International Symposium onFormation Damage Control, Lafayette, Louisiana, USA,February 18–19, 1998.

35. Hannah et al, reference 10.

Ayoub JA, Kirksey JM, Malone BP and Norman WD:“Hydraulic Fracturing of Soft Formations in the GulfCoast,” paper SPE 23805, presented at the SPEInternational Symposium on Formation Damage Control,Lafayette, Louisiana, USA, February 26–27, 1992.

DeBonis VM, Rudolph DA and Kennedy RD:“Experiences Gained in the Use of Frac Packs inUltralow BHP Wells, U.S. Gulf of Mexico,” paper SPE

High-Permeability Reservoirs

Low-Permeability Reservoirs

Laminated Sand-Shale Sequences

Wellbore

Shale

Sand

Proppant

> Frac-pack applications. Frac packing is aviable completion alternative for many wells inreservoirs with sand-production tendencies.

In reservoirs with moderate to high permeabil-ity that are susceptible to drilling and completiondamage that extends deep into the formation,frac packing and wide TSO fractures connectreservoirs and wellbores more effectively.

For reservoirs with heterogeneous pay or lami-nated sand-shale sequences, frac packing pro-vides an effective hydraulic connection acrossmost of a completion interval. When perforated-interval length is limited, frac packing connectsmore pay with fewer perforations.

In low-permeability reservoirs, fracture exten-sion increases the wellbore drainage radius, andbilinear flow stimulates well productivity. In for-mations with low bottomhole pressures, fractur-ing bypasses perforating debris and damage,

mitigating the impact of overbalanced perforat-ing. Frac-pack completions also improve hydro-carbon recovery from low-pressure anddepleted reservoirs by minimizing completionskin across the pay interval, thus reducing draw-down and ultimate abandonment pressure.

MALAYSIA

Singapore

Jakarta

INDONESIA

BORNEOA n d a m a n S e a

T i m o r S ea

Widuri field

0

0 300 600 900 km

200 400 600 miles

ASIA

AUSTRALIA

Indonesia

> Openhole frac packing. Repsol YPF chose to frac pack an openhole com-

pletion located north of Jakarta, Indonesia, in the offshore Widuri field tomaximize well productivity.

27379, presented at the SPE International Symposium onFormation Damage Control, Lafayette, Louisiana, USA,February 7–10, 1994.

36. Ali S, Dickerson R, Bennett C, Bixenman P, Parlar M,Price-Smith C, Cooper S, Desroches J, Foxenberg B,Godwin K, McPike T, Pitoni E, Ripa G, Steven B, Tiffin Dand Troncoso J: “High-Productivity Horizontal GravelPacks,” Oilfield Review 13, no. 2 (Summer 2001): 52–73.

37. Saldungaray PM, Troncoso J, Sofyan M, Santoso BT,Parlar M, Price-Smith C, Hurst G and Bailey W: “Frac-Packing Openhole Completions: An Industry Milestone,”paper SPE 73757, presented at the SPE InternationalSymposium and Exhibition on Formation DamageControl, Lafayette, Louisiana, USA, February 20–21, 2002


15/18

An unexpected low bottomhole pressure of

390 psi [2.7 MPa] resulted in complete fluid loss

while drilling the B-28 well. A high-pressure,

reactive shale above the pay zone required that

the operator set 7-in. casing to isolate this poten-

tially unstable section. Hole collapse caused this

casing string to be set high, leaving 70 ft [21 m]

of shale exposed after drilling deeper. Repsol YPF

suspended the well temporarily after attempts to

run a screen assembly failed.

After five months of waterflooding, reservoir

pressure increased enough to hold a water col-

umn and maintain hole stability. Repsol YPF

decided to attempt an openhole frac pack

because running 5-in. casing was too restrictive

for an internal gravel pack. This approach

presented several challenges, including openhole

stability, screen deployment, fracturing a long

high-permeability section, proppant slurry con-

tamination in exposed shale and annular packing

efficiency in a 70º inclined wellbore. Incomplete

packing and completion failures on other comple-

tions raised concerns about frac-packing effec-tiveness in high-angle wells.

Repsol YPF chose an innovative combination

of Alternate Path screens and a multizone (MZ)

isolation packer to avoid fluid contamination,

facilitate effective fracturing and ensure com-

plete packing of the long openhole section (left).

Two shunt tubes designed for pumping

15 bbl/min [2.4 m3 /min] extended through the

packer, across the reactive shale section and

the entire pay interval. The design incorporated

an inner washpipe that conveyed drilling fluid to

a drilling motor. This motor could rotate a bit

on the bottom of the assembly if required todeploy the completion equipment. An outer

shroud with holes protected screens from

damage in the openhole.

Elastomer cups on the MZ packer prevented

annular flow and diverted fluid to the shunt

tubes. Exit nozzles on the shunt tubes did not

begin until just above the screens to avoid any

injection across the shale. Slurry bypassed the

shale section, exiting through nozzles along the

screens to fill gaps in the pack below proppant

bridges that might form. This configuration

preserved fracture and proppant conductivity

by eliminating fluid contamination from the reac-tive shale.

Frac-pack execution went smoothly despite

concerns about the high-angle wellbore, multiple

competing fractures and excessive fluid leakoff

through 225 ft [69 m] of openhole interval with

47 ft [14 m] of high-permeability net sand.

44 Oilfield Review

Drilling bit

Drilling motor

AllFRAC screenswith nozzles

AllFRAC blank pipewithout nozzles

Reactive shale

MZ isolationpacker with

bypass shunts

Shunttubes

QUANTUMgravel-pack

packer

QUANTUMservice tool

Washpipe

> Widuri Well B-28 completion. Run as part of the completion assembly, a

multizone (MZ) isolation packer was located below the QUANTUM gravel-pack packer inside the 7-in. casing. Two large shunt tubes extending through the packer bypassed the reactive shale section. A protective shroud covered the AllFRAC screens and shunt tubes to prevent mechanical damage fromhole instability or assembly rotation to reach total depth. The shroud alsocentralized the screens for a more complete annular pack. To reach bottom, this assembly could ream and clean out the hole if necessary using a positive-displacement motor and bit at the end of the screen assembly. An innerwashpipe conveyed fluid to the motor.


16/18

Summer 2002 45

Treatment simulation indicated a final fracture

half-length of 18 ft [5.5 m] with a propped width

of 1 in.

Initial production of 2000 B/D [318 m3 /d] total

fluid with 500 B/D [79 m3 /d] of oil from an elec-

trical submersible pump exceeded operator

expectations. Post-treatment skin was not mea-

sured by pressure-buildup analysis, but a sensor

on the electrical submersible pump monitored

downhole flowing pressures, which indicated a

small pressure drop at the completion face.

Well performance was evaluated by calculat-

ing a productivity index (PI) based on net reser-

voir thickness from flowing bottomhole

pressures, reservoir pressure and production

rates (below). Well B-28 outperformed most

wells in the field and compared favorably with

wells completed by openhole horizontal gravel

packing. Considering the excessive fluid losses

during drilling, this level of productivity demon-

strates the feasibility of openhole frac packing as

a sand-control alternative in extremely perme-

able reservoirs with high mobility ratios.

Emerging Technologies

New developments continue in all aspects of frac

packing, from improved sand prediction and

treatment modeling, to new fluids that reduce

damage in both propped fractures and annular

packs. New placement techniques improve frac

packing by applying new downhole equipment or

eliminating subsurface hardware entirely. Fluid

additives that are in testing promise to minimize

production declines by reducing fines migration

and preventing scale deposition.

Frac-pack placement data generally indicate

creation of a fracture and subsequent tip screen-

out, but post-treatment pressure data often

indicate positive skin values and some remaining

damage, raising questions about the effective-

ness of propped fractures. More realistic models

have been developed to resolve discrepancies

between geophysical evaluations, well-log

interpretation, fracturing data from frac-packing

treatments and well-test pressure analysis

(right).38 Generating consistent solutions and

resolving discrepancies require measurement of

multiple parameters within a discipline, and inte-

gration across disciplines.

Differential stresses make uniform hydraulic

fracture diversion and complete coverage diffi-

cult in long intervals of heterogeneous forma-tions, even when Alternate Path technology is

used. This is particularly true if stress profiles

vary significantly when high-permeability zones

with lower stresses occur at the top of a long

interval. Preferential propagation of fractures in

zones with lower in-situ stresses results in sub-

optimal reservoir stimulation. Ocean Energy used an innovative technique in

the Gulf of Mexico to ensure uniform stimulation

and annular packing across long intervals in a

field in the Eugene Island area.39 The operato

pumped more than one pad-slurry sequence dur

ing a treatment without shutting down to

sequentially increase the resistance to fractureextension, or fracture stiffness, in each zone from

lowest to highest in-situ stress. As proppan

packed toward a well, fractures became more

difficult to propagate, and the next pad-slurry

sequence diverted into other zones of long

heterogeneous intervals.

In this application, AllFRAC screens improved

frac-pack treatment diversion across long

intervals. Multiple temperature gauges with

electronic memory placed strategically in the

washpipe monitored slurry diversion through

38. Ayoub JA, Barree RD and Chu WC: “Evaluation of Fracand Pack Completions and Future Outlook,” paper SPE38184, presented at the SPE European FormationDamage Conference, The Hague, The Netherlands,June 2–3, 1997; also in SPE Production & Facilities 15,no. 3 (August 2000): 137–143.

39. White WS, Morales RH and Riordan HG: “ImprovedFrac-Packing Method for Long HeterogeneousIntervals,” paper SPE 58765, presented at the SPEInternational Symposium on Formation Damage Control,Lafayette, Louisiana, USA, February 23–24, 2000.

Cased-hole frac-packaverage = 0.17

Specificproductivity

index(PI),B/D/psi/ft

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

1 2 3 4 5 B-28 6

Well

7 8 9 10 11 12

> Openhole frac-pack performance. Inflow-performance calculations yieldedaverage and maximum specific productivity index (PI) values of 0.17 and 0.39B/D/psi/ft [0.013 to 0.03 m3 /d/kPa/m] for 12 Widuri field cased-hole frac packs,respectively. Only two cased-hole wells in this field performed equal to orbetter than the Well B-28 openhole frac pack, which had a specific PI of 0.28B/D/psi/ft [0.021 m3 /d/kPa/m].

Welltest (T)

T1

T2

T3

F1

F2

L1 G1

L1

G1

F1

G1

L1 G1

F2

G2

Fractureplacement (F)

Interpretationand Simulation

Models

Model Results andMatching Solutions

Well logs (L)

Geology andGeophysics (G)

> Post-treatment evaluations. Solutions from geo-logical and geophysical modeling, well-log inter-pretation, fracture-placement evaluation and well test analysis are not unique. Various combinationsof input data—length of pay interval, reservoirpressure, porosity, permeability, and fracturelength, width or height—often generate multiplesolutions and different results from each model.Better simulation and interpretation software helpestablish a match among all of the different modelssuch as T2, F1, L1 and G1 in this depiction.


17/18

shunt tubes to other zones (above). Temperature

decreases indicated fluid flow past a gauge, and

temperature increases corresponded to reduced

flow or no fluid movement at a gauge.

Temperature responses at the gauges confirmed

complete interval coverage and diversion of

treatment fluids into individual target zones. Net

pressure developed during the treatment indi-

cated a tip screenout.

Individually frac packing multiple zones in a

single wellbore is time-consuming and expensive.An alternative to stacked frac-pack completions

uses Alternate Path screens and MZ isolation

packers with bypass shunt tubes to complete

more than one zone in a single pumping operation

with the same gravel-pack packer (next page,

left). This diversion technique uses pressure drop

in the shunt tubes to control fluid flow. Changing

the number and length of shunt tubes going to

each zone controls pressure drop. Engineers vary

shunt-tube configurations to achieve the desired

rate distribution. This system potentially allows

up to three zones to be completed at reduced cost

and improved profitability.Operators often bypass many marginal, or

secondary, pay intervals. These zones may never

be exploited because of the mechanical risk of

extending the frac-pack interval length up or

down and the cost of mobilizing a rig to recom-

plete the well, especially offshore where the vast

majority of frac packs are performed. Recently,

new technologies have been introduced that

promise to speed routine application of

“rigless” completions.

Coiled tubing-conveyed fracturing technology,

including CoilFRAC stimulation through coiled

tubing service, is rapidly becoming a viable tool

for exploiting bypassed pay.40 This new technol-

ogy has been applied successfully onshore in

multilayered, low-permeability reservoirs, but

the next step is to take this technology offshore.

Accessing offshore wellbores in a workover envi-

ronment and placing a frac pack or screenlesscompletion in a new zone without using a costly

conventional drilling or workover rig opens up

countless fu

estim - combined stimulation and sand control (english)

Documents