4 Oilfield Review

Open-Channel Fracturing— A Fast Track to Production

The goal of a hydraulic fracturing treatment is to improve well productivity by

creating a flow path from the formation to the wellbore. Conventional fracturing

treatments fill the fracture completely with proppant, which holds the fracture open

to preserve the production pathway. A new hydraulic fracturing technique creates

a network of open channels throughout the proppant pack, improving fracture

conductivity by orders of magnitude. Performing this technique has significantly

improved the economic viability of wells in several producing fields.

Emmanuel d’HuteauYPF, S.A.Neuquén, Argentina

Matt GillardMatt MillerAlejandro PeñaSugar Land, Texas, USA

Jeff JohnsonMark TurnerEncana Oil and Gas (USA), Inc.Denver, Colorado, USA

Oleg MedvedevEdmonton, Alberta, Canada

Tom RheinPetrohawk Energy CorporationCorpus Christi, Texas

Dean WillbergSalt Lake City, Utah, USA

Oilfield Review Autumn 2011: 23, no. 3. Copyright © 2011 Schlumberger.ClearFRAC and HiWAY are marks of Schlumberger.

In 1947, Stanolind Oil & Gas performed the first experimental hydraulic fracturing treatment in the Hugoton field in southwestern Kansas, USA. Since then, E&P companies have employed this reservoir stimulation technique extensively to enhance or prolong well productivity. Indeed, many fields pro-ducing today would not be economically viable with-out the benefits provided by hydraulic fracturing.

During a fracturing treatment, specialized equipment pumps fluid into a well faster than it can escape into the formation. Pressure on the forma-tion rises until it breaks down, or fractures (below). Continued pumping causes the fracture to propa-gate away from the wellbore, increasing the forma-tion surface area from which hydrocarbons can flow into the wellbore. This helps the well achieve a

> Idealized cross-sectional view of a propagating fracture. Continued pumping of fluid (dashed arrows) causes the fracture to extend bilaterally along the plane of minimum stress, forming structures called wings.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 1ORAUT11-HWY 1

Fracture wing

43223schD4R1.indd 1 11/17/11 6:36 PM

Autumn 2011 5

43223schD4R1.indd 2 11/17/11 6:36 PM

6 Oilfield Review

higher production rate. As a result, operators recover their well development costs more quickly, and the ultimate amount of produced hydrocarbons increases dramatically (below).

During hydraulic fracturing, two principal substances—proppants and fracturing fluids—are pumped into a well.1 Proppants are particles that hold the fractures open and preserve the newly formed pathways to facilitate hydrocarbon production. The particles are carefully sorted for size and sphericity to form an efficient conduit, or proppant pack, which enables fluids to flow from the reservoir to the wellbore. Some proppants also feature a resin coating that binds the parti-cles together after the proppant is placed in the well, thereby improving pack stability. Generally, larger and more spherical proppants provide more-permeable proppant packs or, in industry vernacular, packs with higher conductivity.

Fracturing treatments consist of two main fluid stages. The first stage, or pad stage, does not contain proppant. Fluid is pumped through cas-ing perforations at a rate and pressure sufficient to break down the formation and create a frac-ture.2 The second stage, or proppant slurry stage, transports proppant through the perforations into the open fracture. The fracture closes onto the proppant when pumping ceases, holding the proppant in place both while the fracturing fluid flows back into the wellbore as well as during hydrocarbon production.

Fracturing fluids must be sufficiently viscous to create and propagate a fracture as well as trans-port the proppant down the wellbore and into the fracture. Once the treatment is completed, the viscosity must decrease sufficiently to promote rapid and efficient evacuation of the fracturing fluid from the well. Ideally, the proppant pack should also be free of fluid residue, which may impair conductivity and hydrocarbon production.

For six decades, chemists and engineers have worked to develop proppants and fracturing flu-ids that produce the ideal propped fracture. As a result, the chemical and physical nature of these materials has changed significantly over time. Proppants have evolved from crude materials, such as nut shells, to naturally occurring sands and to high-strength spheres manufactured from ceramics or bauxite. Fracturing fluids have pro-gressed from gelled oils to linear- and cross-linked-polymer solutions. Chemical breakers were introduced to decompose the polymer, reduce the amount of polymer residue in the frac-ture and improve conductivity (right). In the late 1990s, Schlumberger introduced an essentially residue-free system, ClearFRAC polymer-free fracturing fluid.3 The proppant pack conductivity in wells treated with ClearFRAC fluid nearly equaled the theoretical prediction.

Having maximized proppant pack conductiv-ity, the industry began to investigate ways to further improve hydraulic fracturing results. Engineers found the answer when they focused

> Effect of hydraulic fracturing on well productivity. This graph shows how fracture size affects one-year cumulative gas production from three hypothetical wells with different formation permeabilities. The fracture half-length is the distance that one of the fracture wings extends from the wellbore. The productivity benefit of hydraulic fracturing increases proportionally with decreasing formation permeability.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 2ORAUT11-HWY 2

Cum

ulat

ive

gas

prod

uctio

n, M

Mcf

Effective fracture half-length, ft

100

75

25

50

00 200 400 600 800 1,000

Wellbore pressure = 3.5 MPa [500 psi]Fracture height = 6.1 m [20 ft]

Formation permeability1 mD0.1 mD0.01 mD

> Fracturing-fluid evolution. Early fracturing treatments employed hydrocarbon-base fluids. Operators frequently added gelling agents to oil from the same producing formation. Water-base fluids, such as linear-polymer solutions, were introduced in the 1960s. However, as wells became deeper and hotter, these fluids were no longer sufficiently viscous. To improve thermal stability, chemists added metal salts, causing crosslinking reactions that increased the effective polymer molecular weight by several orders of magnitude. Today, crosslinked-polymer fluids are routinely used at well temperatures up to about 232°C [450°F]. Posttreatment fluid recovery required the addition of strong oxidizing agents, or breakers, to decompose the polymer and reduce fluid viscosity. Encapsulated breakers were eventually developed, which allowed higher oxidizer concentrations and reduced the amount of polymer residue in the proppant pack. Fluid foaming allowed lower polymer concentrations, further improving proppant pack cleanup. Including fibers improved the fluids’ ability to transport proppant, permitting further polymer concentration reductions. The most recent generation of fracturing fluids employs nonpolymeric, low-molecular-weight viscoelastic surfactants. Fluid viscosity arises from the formation of rod-shaped micelles. When the fluid contacts hydrocarbons downhole, its viscosity decreases substantially, promoting efficient recovery and virtually residue-free proppant packs.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 3ORAUT11-HWY 3

1950

1960

Hydrocarbons

Gelled oils

Linear-polymer gels

• Guar derivatives• Hydroxyethyl cellulose

Crosslinked-polymer gels

• Boron, titanium and zirconium salts

Foamed fluids

Viscoelasticsurfactant gels

Firs

t Gen

erat

ion

Seco

nd G

ener

atio

n

Breakers

Encapsulated breakers

Fiber-assistedtransport

Thir

d G

ener

atio

n

FluidSystems

SupportingTechnologies

1970

1980

1990

2000

2010

Year

43223schD4R1.indd 3 11/17/11 6:36 PM

Autumn 2011 7

on the proppant pack in a different way. Since the advent of hydraulic fracturing, engineers strove to fill the fracture completely with prop-pant—in other words, create a continuous prop-pant pack. What if it were possible to fill the

fracture with a discontinuous proppant pack consisting of discrete proppant columns sur-rounded by open channels? This approach would separate the load-bearing task of the proppant pack from that of providing a fluid

pathway. Engineers speculated that, if the prop-pant pack were properly designed, the fracture conductivity would be orders of magnitude higher than that of the cleanest conventional proppant pack (above).

> Continuous and discontinuous proppant packs. In a conventional proppant pack (left), all of the proppant particles are in mutual contact. Fluid flow is confined to the interstices between the proppant grains. A discontinuous proppant pack (right) consists of proppant agglomerations or columns, creating a network of discrete open channels through which fluids may flow.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 4ORAUT11-HWY 4

1. For more on fracturing fluids and proppants: Gulbis J and Hodge RM: “Fracturing Fluid Chemistry and Proppants,” in Economides MJ and Nolte KG (eds): Reservoir Stimulation, 3rd ed. Chichester, West Sussex, England: John Wiley & Sons, Ltd. (2000): 7-1–7-23.

2. Perforations are holes created in the casing after it has been cemented in place. The most common perforating method employs guns equipped with explosive shaped charges. Detonation creates short tunnels through the casing and the cement sheath, thereby providing hydraulic communication between the wellbore and the producing reservoir.

3. Chase B, Chmilowski W, Marcinew R, Mitchell C, Dang Y, Krauss K, Nelson E, Lantz T, Parham C and Plummer J: “Clear Fracturing Fluids for Increased Well Productivity,” Oilfield Review 9, no. 3 (Autumn 1997): 20–33.

43223schD4R1.indd 4 11/17/11 6:36 PM

8 Oilfield Review

After several years of research and develop-ment, Schlumberger scientists have achieved this goal. The fruit of their efforts, the HiWAY flow-channel hydraulic fracturing technique, is a fundamental advance in the science of reservoir stimulation. This article describes how the HiWAY technique was developed in the labora-tory and introduced to the oil field. Case histories from Argentina and the US demonstrate the well productivity improvements that have been achieved by applying this technique.

Redesigning the Proppant PackScientists at the Novosibirsk Technology Center in Russia began their quest for a discontinuous proppant pack with an ambitious experimental program to confirm its feasibility and to develop the means by which such a technology could be applied. The scale of the experiments increased gradually from small benchtop labo-ratory simulators to full-scale testing with standard field equipment.4

The first task was to validate the theoretical conductivity benefit expected from discontinu-ous proppant packs. Employing a standard American Petroleum Institute (API) test method, engineers placed a proppant pack in a fracture simulator. The simulator applies a closure stress representing overburden pressure and measures the force necessary to pump a single-phase fluid through the pack at various flow rates (left).5 The engineers then used Darcy’s law and the Navier-Stokes equations to calculate the proppant pack permeability.6 The measured permeabilities of the discontinuous proppant packs were consis-tent with the theoretical model prediction—1.5 to 2.5 orders of magnitude higher than the con-tinuous packs.

Having verified the conductivity benefit of discontinuous proppant packs through experi-ments, scientists turned their attention to meth-ods by which proppant columns could be created downhole in an actual fracture, withstand the stresses associated with fluid flow and fracture closure and maintain open flow channels. By performing modeling studies and experiments, engineers evaluated several concepts for creat-ing proppant columns in a fracture. These included adding memory-alloy fibers around which proppant grains would congregate, plac-ing encapsulated breakers in localized areas and heating the proppant slurry in a discontinuous manner. In the most promising method, engi-neers changed the manner by which proppant is delivered downhole.

>Measuring the conductivity of a proppant column network. A standard API conductivity cell (top) features two steel platens, driven by a hydraulic press to apply closure stress. The proppant pack is placed between two sandstone slabs (usually Berea Sandstone), and the resulting “sandwich” is placed between the two platens of the hydraulic load frame. After installing the platen assembly inside an enclosure equipped with flowlines, technicians pump a single-phase fluid (usually water or brine) through the proppant pack at 1 to 10 mL/min, measure the resulting pressure drops and calculate the proppant-pack permeability. The enclosure may also be heated to simulate reservoir temperature. Technicians created a discontinuous proppant pack by placing four proppant columns between the two slabs of sandstone (middle). Conductivity measurements were performed at closure stresses between 1,000 and 6,000 psi [6.9 and 41.4 MPa] (bottom). The permeabilities of continuous proppant packs prepared with 20/40-mesh sand (blue diamonds) and 20/40-mesh ceramic proppant (green triangles) were less than 1,000 D. Permeability generally decreased with closure stress due to proppant crushing and proppant embedment into the sandstone. The discontinuous proppant packs were formed from 20/40-mesh sand and, in agreement with the theoretical prediction (red line), the measured permeabilities (black squares) were orders of magnitude higher.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 5ORAUT11-HWY 5

Perm

eabi

lity,

D

Stress, psi

0100

1,000

10,000

100,000

1,000 2,000 3,000 4,000 5,000 6,000 7,000

ExperimentModel20/40-mesh ceramic 20/40-mesh sand

Mass flowcontroller

Waterchamber

Pressuretransmitters

Dataacquisition and

processing

Force

Force

Heating jacketVent

Backpressureregulator

Valve

Check valve

Hydraulic load frame platen

N2

Sandstone slabs

Hydraulic load frame platen

Side piston

Sandstone slabs

43223schD4R1.indd 5 11/17/11 6:36 PM

Autumn 2011 9

4. Gillard M, Medvedev O, Peña A, Medvedev A, Peñacorada F and d’Huteau E: “A New Approach to Generating Fracture Conductivity,” paper SPE 135034, presented at the SPE Annual Technical Conference and Exhibition, Florence, Italy, September 20–22, 2010.

5. American Petroleum Institute: API RP 61: Recommended Practices for Evaluating Short Term Proppant Pack Permeability, Washington, DC: American Petroleum Institute, 1989.

6. Darcy’s law may be represented by the following equation: , where q is the flow rate, kf is the

proppant pack permeability, w is the proppant pack width, μ is the fluid viscosity and Δp/L is the pressure drop per unit of proppant pack length. The Navier-Stokes equations are a set of coupled differential equations that describe how the velocity, pressure, temperature and density of a moving fluid are related. For more information: Zimmerman RW and Bodvarsson GS: “Hydraulic Conductivity of Rock Fractures,” Transport in Porous Media 23, no. 1 (1996): 1–30.

7. One may question how well the circular pipe geometry represents the slot geometry of an actual fracture. Since the relative cross-sectional area of a pipe is larger than that of a slot, destabilizing influences are more pronounced. Therefore, the pipe geometry provides a more conservative stability assessment.

In the conventional method, proppant is pres-ent throughout the entire proppant slurry volume. However, if the proppant slurry stage consisted of alternating fluid pulses—with and without prop-pant—a series of proppant slugs could settle in the fracture and form columns (right).

For the pulsing method to succeed, it was essential that the proppant slugs not disperse during their journey down the tubulars, through the perforations and into the fracture. In the first experiments to test this concept, engineers observed the static settling behavior of proppant slugs in a transparent slot filled with a fracturing fluid. After injecting a sample of proppant-laden fluid into the top of the slot, engineers could visu-ally assess the settling behavior over time. Scientists quickly learned that proppant slugs prepared with conventional fracturing fluids readily dispersed as they traveled down the slot. Eventually, they discovered that proppant slug stability could be dramatically improved by inte-grating fibers (below right).

The next series of experiments evaluated the dynamic stability of proppant slugs. The appara-tus featured up to 33 m [108 ft] of 0.78-in. [2-cm] ID pipe—a geometry that allowed scientists to test flow rates, fluid velocities and proppant con-centrations consistent with flow through a frac-ture.7 An X-ray registration system measured proppant slug stability. The X-ray absorbance across the pipe diameter is linearly proportional to the proppant concentration; therefore, stabil-ity measurements could be obtained by recording

> Comparing the HiWAY technique with a conventional fracturing treatment. During the proppant stage of a conventional fracturing treatment (red line, top), all of the slurry contains proppant, and operators usually increase the proppant concentration stepwise. The proppant stage of a HiWAY fracturing treatment (green line) features alternating pulses of proppant-laden (dirty) fluid and clean fluid. The proppant concentration in the pulses may also be increased stepwise. Engineers monitor proppant pulses during actual fracturing treatments (bottom). Proppant concentrations are commonly expressed in pounds per gallon added, or ppa. One ppa means that one pound of proppant is added to each gallon of fracturing fluid. It must not be confused with the more common pounds per gallon, or lbm/galUS. During hydraulic fracturing treatments, ppa better reflects field practice. There is no recognized SI equivalent to ppa.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 6ORAUT11-HWY 6

Sand

con

cent

ratio

nPr

oppa

nt c

once

ntra

tion,

ppa

Time

Tail-in stage

Clean fluid pulseProppant (dirty) pulse

Pad

Cycle

8

7

6

5

4

3

2

1,800 2,000 2,200

Pump time, s

Design Versus Actual Proppant Concentration

2,400 2,600 2,800

1

0

Conventional treatmentHiWAY treatment

Proppant concentrationdesign, ppaProppant concentrationfrom densitometer, ppa

> Initial proppant slug settling experiments. The initial proppant slug position in a slot filled with a fracturing fluid was at the top of the slot (left). The proppant slug fragmented within 30 min when fibers were not present in the slug or the fluid (center). When fibers were used, the slug remained largely intact after two hours (right).

Oilfield ReviewAUTUMN 11 HiWAY Fig. 7ORAUT11-HWY 7

Initial Position Without Fibers With Fibers

q = kfw Δpµ L

43223schD4R1.indd 6 11/23/11 12:23 AM

10 Oilfield Review

X-ray absorbance before and after the proppant slugs traveled through the pipe. The results con-firmed that fibers enhance proppant slug stabil-ity (left).

Following the promising laboratory experi-ments, full-scale tests were performed at the Schlumberger Kellyville Learning Center (KLC) in Oklahoma, USA. These experiments were designed to test the stability of proppant slugs traveling through surface lines, wellbore tubulars and perforations at flow rates similar to those experienced during an actual fracturing treat-ment. The layout included a field blender and approximately 198 m [650 ft] of 7.6-cm [3-in.] treating line connected to the blender discharge (below). Five sets of perforations were arranged along the treating line. Each set consisted of ten 0.95-cm [0.374-in.] holes—five at 0° phasing and five directly opposite at 180° phasing. Fluid escaping from the perforations was collected in ten 1.04-m3 [275-galUS] tote tanks. Two densi-tometers—one at the blender discharge, the

> Influence of fibers on proppant slug dispersion during flow through a tubular body. Proppant concentrations were measured before and after proppant slugs flowed through 32.9 m [108 ft] of pipe. The proppant concentration profile from the fluid containing fibers (blue) was significantly less dispersed than that from the fluid containing no fibers (red).

Oilfield ReviewAUTUMN 11 HiWAY Fig. 8ORAUT11-HWY 8

Prop

pant

con

cent

ratio

n, p

pa

Time of experiment, s

Before pipe flow After pipe flow

With fibersWithout fibers

0

10

8

6

4

2

0

50 100 150 200

> Yard-scale testing of the HiWAY concept. At the Schlumberger Kellyville Learning Center (KLC), engineers arranged five sets of perforations along a treating line, and two 275-galUS tote tanks collected fluid passing through each perforation set (top and bottom left). Each perforation set (bottom right) consisted of 10 holes—five at 0° phasing and five at 180° phasing. A field blender mixed and pumped the fluids.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 9ORAUT11-HWY 9

Wastetank

Blender

Pressuretransducer

Flowmeter

Tote 1 Tote 2 Tote 3 Tote 4 Tote 5

Tote 10 Tote 9

10 ft

Tote 8 Tote 7 Tote 6

43223schD4R1.indd 7 11/17/11 6:36 PM

Autumn 2011 11

other at the end of the treating line—measured proppant concentrations and provided an indica-tion of slug stability. A unique feature of the Schlumberger blender is a programmable mixer that precisely controls the proppant concentra-tion in the fracturing fluid, which is coupled with an array of dry- and liquid-additive feeders and a unique fiber-additive system.

The first set of KLC experiments, performed with the perforations closed, measured the sta-bility of proppant slugs traveling through the treating line at 11.6 m/s [38 ft/s]. This velocity corresponded to a pump rate of 2.7 m3/min [17 bbl/min]. The carrier fluid was a borate-crosslinked guar system, with a guar concentra-tion of 3.6 kg/m3 [30 lbm/1,000 galUS], and a fiber concentration of 5.0 kg/m3 [42 lbm/1,000 galUS]. The proppant concentration in the slugs was 10 ppa, and the fiber concentration was 10.0 kg/m3 [84 lbm/1,000 galUS]. The results showed that the proppant slugs were stable after passing through the treating line (above right).

The second set of KLC experiments evaluated the ability of proppant slugs to pass through the perforations and remain intact. In addition, the scientists wanted to verify that the proppant slugs could fractionate and distribute themselves among all of the perforations. During each test, technicians measured the accumulated fluid vol-umes in each of the tote tanks connected to the five sets of perforations. The first perforation set was equipped with a densitometer that continu-ously recorded fluid density, and fluid samples were manually collected from the last perfora-tion set. When fibers were present in both the clean-fluid and proppant-laden pulses, the fluid volume distribution among the tote tanks was uniform. The fluid density variations measured by the densitometer and by manual sampling also matched, further confirming the feasibility of the proppant slug method for creating a discontinu-ous proppant pack (right).

Last, engineers attempted the HiWAY place-ment technique in an experimental well. Under these circumstances, it was impossible to directly observe the behavior of the fluid pulses. Instead, engineers recorded surface- and downhole- pressure measurements during pumping and employed a mathematical model to reconstruct

> KLC yard-test results. The plot presents proppant slurry density profiles at the field blender (red line) and at the end of the treating line (blue line). At 11.6 m/s, the fluid transit time through the treating line was 33 s. As indicated by the five sets of red and blue peaks, the slurry density profiles at the treating line discharge were nearly identical to those at the blender, indicating that the proppant slugs withstood the journey.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 10ORAUT11-HWY 10

Slur

ry d

ensi

ty, l

bm/g

alUS

13.5

13.0

11.5

12.0

12.5

10.5

11.0

10.0

9.0

8.080 130 230 330180

Time, s280 380

9.5

8.5

> KLC perforation tests. Technicians pumped proppant pulses through the treating line and monitored how well the pulses remained intact during their journey down the line and through the perforations. The fluid volumes collected from each of the five perforation clusters were practically uniform (top), indicating that the pulses could fractionate and distribute themselves among the perforation clusters. For further confirmation, the slurry densities (bottom) during and between pulses were continuously measured by a densitometer at the beginning of the treating line (blue line) at Tote 1 and measured manually at the end of the line at Tote 6 (red lines). The slurry densities at the beginning and at the end also matched, confirming that the proppant slugs distributed themselves among the perforation clusters and remained stable.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 11ORAUT11-HWY 11

Dens

ity, l

bm/g

alUS

Volu

me,

gal

US

12

11

10

300

250

200

100

150

50

0

9

8

Perforation cluster

Time, min:s08:52 09:36 10:19 11:02 11:45 11:28

1 2 3 4 5

43223schD4R1.indd 8 11/17/11 6:36 PM

12 Oilfield Review

the proppant concentration profile as the pulses passed through the perforations (left). The pres-sure gauge data indicated that the proppant-laden pulses survived the journey downhole and successfully entered the perforations.

Evaluating Proppant Column DurabilityHaving shown that proppant columns can be con-structed using available blending and pumping equipment, the scientists turned their attention to the postplacement stability of discontinuous proppant packs. Following a fracturing treat-ment, the columns must be strong enough to withstand formation closure pressure as well as erosive forces arising from fluid flow during well cleanup and production.

To investigate the effects of closure stress, technicians manually prepared proppant columns and placed them in a hydraulic press equipped with sensors for monitoring the distance between the press rams. The apparatus subjected the prop-pant columns to compressive loads as high as about 228 MPa [33,000 psi]. Measured parameters included column height and diameter and prop-pant particle size distribution.

As expected, proppant column height decreased with closure stress; however, it is nota-ble that more than 80% of the column shrinkage occurred during the first 6.9 MPa [1,000 psi] of compression. As loads increased, further shrink-age was minimal. Inspection of the proppant col-umns revealed that the initial shrinkage resulted from carrier fluid leakoff and proppant consolida-tion. Additional column shrinkage at higher pres-sures was due to proppant compaction and crushing. The median size of the proppant parti-cles decreased with increasing pressure. Ultimately, at 228 MPa closure stress, sufficient proppant column height remained for efficient fluid flow (middle left). This closure stress is approximately two times higher than that which occurs in the deepest oil or gas wells, indicating that the HiWAY technique would not have a depth limitation owing to pressure.

Proppant column diameter, or footprint, increased with closure stress. However, closure test results showed that with each increase in the proppant column diameter, the relative footprint increase became less pronounced (left). The labo-ratory tests investigated column diameters of sev-eral centimeters. However, column diameters in an actual fracture would be of the order of several meters; therefore, engineers did not expect that the conductive pathways between columns would be lost as a result of closure stress.

> Concentration profile of a proppant pulse in an experimental well. The plot represents the behavior of a 15-s, 8.3-ppa proppant pulse, pumped downhole through 2,700 m [8,960 ft] of tubing at 3.99 m3/min [25 bbl/min]. Engineers used surface- and downhole-pressure data to calculate the proppant concentrations at the bottom of the tubing string. The duration of the proppant slug increased to about 20 s, and the proppant concentration fell to about 6.5 ppa. Although the initial proppant concentration was 8.3 ppa, this is not a significant dispersion because the slug duration and concentration profile were preserved.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 12ORAUT11-HWY 12

Prop

pant

con

cent

ratio

n, p

pa 6

7

5

5 10 15 20 25

4

3

2

1

00

Time, s

> The effect of fracture closure stress on proppant column height. Technicians tested fiber-laden columns fabricated from 20/40-mesh sand (blue) and two 20/40-mesh ceramic proppants (red and green). The initial column height was 6 mm [0.24 in.]. Most proppant column shrinkage occurred at closure pressures less than about 1,000 psi, and further shrinkage took place at a slower rate as the closure pressure continued to rise. At closure pressures up to 33,000 psi—far higher than those experienced in the deepest wells—sufficient column height remained to support fluid flow.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 13ORAUT11-HWY 13

Colu

mn

heig

ht, m

m

Stress, psi

20/40-mesh sand20/40-mesh ceramic20/40-mesh ceramic

1.0

00 5,000 10,000 15,000 20,000 25,000 30,000 35,000

2.0

3.0

4.0

5.0

6.0

> Laboratory measurements of proppant-footprint enlargement under closure stress. Technicians tested fiber-laden columns fabricated from 20/40-mesh ceramic proppant. The closure stress was 20.7 MPa [3,000 psi]. The results show that the relative amount of enlargement decreased with increasing column diameter.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 14ORAUT11-HWY 14

Diam

eter

incr

ease

, %

Proppant column diameter, cm

90

80

70

60

50

40

30

20

10

01 3 5 7 9 11

43223schD4R1.indd 9 11/17/11 6:36 PM

Autumn 2011 13

Fluid flow during cleanup and production, other potentially destabilizing events, also required investigation. To evaluate this risk, the scientists constructed an erosion cell that could be inserted between the rams of a hydraulic press (above). The cell accommodated two sandstone cores, between which a proppant col-umn could be placed. While the hydraulic press exerted closure pressure, technicians pumped a fluid past the column at various flow rates cor-responding to and exceeding normal well pro-duction. They assessed erosion visually and by measuring column weight loss (right).

The results demonstrated that nearly all proppant erosion occurred within the first few minutes of proppant exposure to fluid flow. In addition, the amount of erosion decreased with increasing closure stress, particularly above about 69 MPa [10,000 psi]. Visual analysis revealed that all of the erosion took place along the sides of the columns, not at the surfaces directly facing fluid flow.

Following these experiments, the scientists were confident that proppant columns placed in a hydraulic fracture would survive the rigors of subsequent well operations.

,Measuring proppant column erosion arising from fluid flow. Engineers fabricated an erosion cell that could be inserted between the rams of a hydraulic press (left). The apparatus allows simultaneous closure stress and fluid flow within the cell. The proppant column is inserted between two sandstone cores (bottom right). Fluid flows through the apparatus at two velocities. V1 is the fluid velocity facing the proppant column, and V2 is a higher fluid velocity resulting from constriction as the fluid passes by the column. A filter at the erosion cell outlet (not shown) collects eroded proppant particles.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 15ORAUT11-HWY 15

Fluid flow

Sandstone cores

Proppant column

V1

V2Sandstone core

Oilfield ReviewAUTUMN 11 HiWAY Fig. 15ORAUT11-HWY 15

Fluid flow

Sandstone cores

Proppant column

V1

V2Sandstone core

> Effects of flow rate and closure stress on proppant column erosion. The initial, uneroded proppant columns were 1.38 mm [0.05 in.] high and 36 to 42 mm [1.42 to 1.65 in.] in diameter, and the closure stress was 16.6 MPa [2,400 psi]. Technicians increased the fluid flow rate incrementally and measured the amount of proppant collected in the filter at the erosion cell outlet (top). Most of the erosion occurred during the first few minutes, at lower flow rates (blue line). Proppant column erosion decreased significantly at higher closure stresses (bottom).

Oilfield ReviewAUTUMN 11 HiWAY Fig. 16ORAUT11-HWY 16

Prop

pant

mas

s er

osio

n, %

Rate, L/min

Effective closure stress, psi

Prop

pant

mas

s er

osio

n, g

/cm

2

70

0.6

0.4

0.2

0

60

50

40

30

20

10

00 5 10

5,000 10,000 15,000 20,000

5 min

10 min

15 min 20 min 25 min 30 to 60 min

25,000 30,000

15 20 25

43223schD4R1.indd 10 11/17/11 6:36 PM

14 Oilfield Review

Selecting Field Testing CandidatesEngineers designed a conservative field testing approach that considered several important for-mation and well design parameters. At first, they decided to limit the technique to vertical wells and, to ensure adequate proppant column sepa-ration in the fracture, they initially arranged the perforations in clusters rather than in the con-ventional, even-spaced configuration (below).

Engineers also needed to consider the nature of the producing formation. How would the frac-ture walls respond to the presence of void areas in a discontinuous proppant pack? If the rock is too soft or flexible, the walls might bend or flow into the voids, compromising the fracture con-ductivity. For guidance, the scientists turned to a related discipline—mining engineering. Proppant conglomerates are analogous to columns in an underground mine, and mining engineers must consider the relationships between the columns and the overlying rock.

> Conventional and HiWAY perforating schemes. Arranging perforations in clusters enhances the separation between proppant slugs entering the fracture and ensures an optimal conductive pathway from the fracture to the wellbore.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 17ORAUT11-HWY 17

HiWAYperforating

Conventionalperforating

> HiWAY field study results. The Loma La Lata field, operated by YPF, S.A., is located in southwest Argentina (top). The field’s wells produce oil (green dots) and gas (red dots). The initial (30-day) average production rate from gas wells stimulated by the HiWAY technique (bottom, blue) was 8.2 MMcf/d, while those stimulated conventionally (orange) averaged 5.4 MMcf/d. The wells are grouped on the plot according to their proximity and reservoir property similarity.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 19ORAUT11-HWY 19

ARGENTINASantiago

40°

35°

CHILE

Initi

al g

as p

rodu

ctio

n ra

te, M

Mcf

/d

Well

0 km

0

0

2

N-1 N-2 N-3 N-4 N-5 N-6 N-7C-1 C-2 C-3 C-4 C-5 C-6 C-7 C-8

4

6

8

10

12

14

mi 200

200

HiWAY technique

Conventional technique

43223schD4R1.indd 11 11/17/11 6:36 PM

Autumn 2011 15

The principal parameters governing the sta-bility of an underground mine are column strength, the overburden pressure and the Young’s moduli of the columns and the overlying rock.8 The scientists judged that the initial appli-cations of the HiWAY technique should be per-formed in fields where the ratio between the formation Young’s modulus and the fracture clo-sure stress is higher than about 1,000. Such for-mations tend to be hard and inflexible. Once HiWAY technology was validated in these forma-tions, engineers would consider lowering the limit incrementally.

Scientists and engineers spent many years preparing theoretical models and conducting tests to reach this point in the development of HiWAY technology. Now, with the well candidate selection guidelines in mind, they were ready to leave the experimental world and apply their new technique in producing oil fields.

Improving Gas Production in ArgentinaThe Loma La Lata field, operated by YPF, S.A., is located in southwest Argentina. With more than 300 wells, this field produces 26% of the country’s natural gas. Three producing zones lie in fine- to coarse-grained sandstone of the Sierras Blancas Formation in the Neuquén basin, at depths between 2,896 and 3,200 m [9,500 and 10,500 ft]. The bottomhole temperature and pressure vary between 113°C and 118°C [235°F and 245°F], and 24.1 and 31.0 MPa [3,500 and 4,500 psi]. Reservoir permeabilities and porosities are mod-erate—between 0.08 and 5 mD, and 12% and 17%, respectively. The Young’s modulus of the formation is between 27,600 and 48,300 MPa [4 and 7 million psi]. Closure stresses vary between 27.6 and 41.4 MPa [4,000 and 6,000 psi].

Despite ongoing drilling and fracturing activ-ity, gas production for the field had recently begun to decline. This situation, combined with the increasing energy demand in Argentina, prompted the operator to consider novel well-stimulation methods. Engineers conducted a 15-well field study wherein seven wells received the HiWAY treatment, and the remaining offsets were stimulated conventionally.9 To ensure an optimal comparison, all of the wells received the same fracturing fluid and proppant. The initial 30-day production rate from the wells stimulated by the HiWAY channel-fracturing technique exceeded that from the offsets by 53% (previous page, right).

Engineers continued to monitor production from some of these wells for two years (above). Cumulative gas production from the HiWAY well was 29% higher than that from the offsets, equivalent to about US$ 4.8 million additional revenue at today’s prices. Thus, the HiWAY wells not only delivered higher initial production rates but also sustained significant production gains over time. The results also showed that the flow channels inside the discontinuous prop-pant pack remained in place for an extended period of time. For these reasons, YPF, S.A., is continuing to incorporate the HiWAY technique in its well development activities.

The Argentina study demonstrated that the HiWAY technique is effective in moderately per-meable reservoirs. Following this success, engi-neers decided to execute the new technology in low-permeability, gas-bearing formations.

Channel Fracturing in Tight-Gas ReservoirsEncana Oil and Gas (USA), Inc. operates the Jonah field in Wyoming, USA. Most production originates from the Lance Formation, which con-sists of fluvial sand intervals with permeabilities between 0.005 and 0.05 mD, and gas saturations of 33% to 55%. Bottomhole temperatures range from 79°C to 118°C [175°F to 245°F], and the for-mation Young’s modulus fluctuates between 24,130 and 41,400 MPa [3.5 and 6.0 million psi]. Closure stresses range between 35.9 and 49.6 MPa [5,200 and 7,190 psi]. Because the sand-interval thicknesses vary between about 3 and 61 m [10 and 200 ft], at depths between 2,290

and 4,115 m [7,500 and 13,500 ft], and are inter-bedded with siltstones and shales, stimulation treatments must be performed in multiple stages.

Typically, the wells consist of 20 to 50 stacked sand sequences, and Encana usually divides them into 10 to 14 stages. Each stage requires its own perforating and fracturing treatment. Engineers begin with the deepest stage and allow several days of fluid flowback and cleanup before continuing upward to the next stage. After engi-neers perforate and stimulate the last zone, well production may proceed.

Implementation of the HiWAY fracturing technique commenced with a 12-stage well con-taining 191 m [626 ft] of net pay.10 A nearby 12-stage offset well with 204 m [669 ft] of net pay received conventional fracturing treatments. Engineers pumped the same borate-crosslinked guar fracturing fluid and 20/40-mesh sand prop-pant into both wells at proppant concentrations between 4 and 6 ppa. Because of the formation of open channels, 44% less proppant was necessary

8. Young’s modulus, E, is an elastic constant that indicates how a material deforms when subjected to stress. Resistance of a material to deformation increases with the value of E.

Sneddon IN: “The Distribution of Stress in the Neighbourhood of a Crack in an Elastic Solid,” Proceedings of the Royal Society of London A 187, no. 1009 (October 22, 1946): 229–260.

9. Gillard et al, reference 4.10. Johnson J, Turner M, Weinstock C, Peña A, Laggan M,

Rondon J and Lyapunov K: “Channel Fracturing—A Paradigm Shift in Tight Gas Stimulation,” paper SPE 140549, presented at the SPE Hydraulic Fracturing Technology Conference and Exhibition, The Woodlands, Texas, USA, January 24–26, 2011.

> Loma La Lata field production improvement. By applying the HiWAY fracturing technique in the Loma La Lata field, Argentina, the operator saw a production improvement over the conventional wells in the first year with a substantial increase in the second. In the second year of production, the HiWAY well (blue) produced 29% more gas than the conventionally stimulated offset wells (orange).

Oilfield ReviewAUTUMN 11 HiWAY Fig. 20ORAUT11-HWY 20

Tota

l gas

, Bcf

1-year results 2-year results1.5

2.0

3.0

4.0

2.5

3.5

4.5

5.0

HiWAY well

Offset wells

43223schD4R1.indd 12 11/17/11 6:36 PM

16 Oilfield Review

in the HiWAY well. The operator tracked produc-tion from both wells for 180 days (above). Cumulative production from the HiWAY well was 26% higher than that of the offset well. This encouraging result led Encana to embark on a more ambitious comparative well study.

Thirteen wells were stimulated in the same section of the field—five by the HiWAY process and the other eight by conventional means. As before, all wells received the same borate-cross-

linked guar fluid and sand proppant. The pro-gram consisted of 135 fracturing stages. Because of the open fracture channels, cleanup occurred more quickly, and the recovered-fluid volumes were 48% higher than those of the offset wells. After 30 days, the normalized production from the wells treated by the HiWAY technique was 23% higher than that from the offsets. Models predict that, after two years, the cumulative production will be 17% higher than that of the offsets.

Stimulating the Eagle Ford ShaleThe Eagle Ford Shale in the US is of significant importance because of its ability to produce both gas and relatively large volumes of oil and con-densate. The shale has a high carbonate content, making it brittle and amenable to fracturing treatments. The formation extends from Mexico northeast into east Texas, and is roughly 80 km [50 mi] wide and 644 km [400 mi] long. The average thickness is 76 m [250 ft] at a depth of approximately 1,220 to 3,660 m [4,000 to 12,000 ft] (next page, top left). Engineers at Petrohawk Energy Corporation, which operates the Hawkville field near Cotulla, Texas, investi-gated whether the HiWAY fracturing technique could improve production of both gas and condensate.11

The formation is extremely tight, with perme-abilities between 100 and 600 nD and porosities between 7% and 10%. The bottomhole tempera-tures and pressures are also elevated—132°C to 166°C [270°F to 330°F] and 48.3 to 69.0 MPa [7,000 to 10,000 psi]. Young’s moduli are lower than those described earlier—between 13,800 and 31,050 MPa [2.0 and 4.5 million psi]. These are challenging conditions for the successful execution of fracturing treatments, regardless of the technique.

Wells in this section of the Eagle Ford forma-tion are usually horizontal, presenting additional challenges for the HiWAY fracturing method. Schlumberger scientists had not yet performed extensive modeling studies of open-channel for-mation in deviated wellbores; nevertheless, Petrohawk decided to try the new technique. The operator selected two wells for an initial assess-ment: Well 1, located in a gas-producing region, and Well 2, in a condensate sector. Offset wells were available to provide a valid comparison (next page, bottom left). The operator had stimulated the offsets with fracturing fluids consisting of either slickwater or a hybrid system that employed slickwater during the pad stage and a crosslinked-polymer fluid during the proppant stage.12 Schlumberger engineers chose a borate-cross-linked guar fluid to stimulate the HiWAY wells.

For Well 1, the initial gas production rate was 411,000 m3/d [14.5 MMcf/d]—37% higher than in the best comparable offset well. After 180 days, the cumulative gas production was 76% higher than for the same offset well. Ultimate gas pro-duction from this well is expected to be 252 mil-lion m3 [8.9 Bcf]. Well 2 initially produced 130 m3 [820 bbl] of condensate per day—32% higher than the best comparable offset. After 180 days, the cumulative condensate production was 54% higher than that from the comparable offset well.

> Six-month production comparison between tight-gas wells stimulated conventionally and by the HiWAY technique. The wells, operated by Encana, are located in the Jonah field (top). The data were normalized to account for lithologic and reservoir-quality differences between the two wells (bottom). Cumulative production from the HiWAY well was 26% higher than that from the conventionally stimulated wells.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 21ORAUT11-HWY 21

Nor

mal

ized

cum

ulat

ive

prod

uctio

n, M

cf

Time after last fracturing treatment, days

12,000

10,000

8,000

6,000

4,000

2,000

00 4020 8060 100 160140120 180

HiWAY techniqueConventional technique

41°

42°

43°

Wyoming

Jonah field

Utah

Colo

rado0 km

0 mi 50

50

Wind River Mountains

Great Divide basin

Rawlins

uplift

Wyo

min

g-Id

aho

thru

st b

elt

Park Range

Bridger basin Washakie

basin

Sand WashbasinUinta Mountains

ver insountains

U S A

11. Rhein T, Loayza M, Kirkham B, Oussoltsev D, Altman R, Viswanathan A, Peña A, Indriati S, Grant D, Hanzik C, Pittenger J, Tabor L, Markarychev-Mikhailov S and Mikhaylov A: “Channel Fracturing in Horizontal Wellbores: The New Edge of Stimulation Techniques in the Eagle Ford Formation,” paper SPE 145403, presented at the SPE Annual Technical Conference and Exhibition, Denver, October 30–November 2, 2011.

12. Slickwater fracturing fluids are composed of water and a polymer (usually polyacrylamide) for lowering the friction pressure when pumping the fluid through tubulars. The pump rate during slickwater fracturing treatments is high—up to 15.9 m3/min [100 bbl/min]. Consequently, large amounts of water are required to stimulate a well.

43223schD4R1.indd 13 11/22/11 10:42 PM

Autumn 2011 17

Based on these results, Petrohawk has increased its utilization of HiWAY technology. Ten additional wells were completed using the new technique, showing production trends consistent with the initial test wells. The technique has increased gas production from the Eagle Ford Shale by 51% and condensate production by 46%.

Expanding the Scope of HiWAY ServiceAs of this writing, more than 2,600 HiWAY fractur-ing treatments have been performed in eight countries, with a placement-success rate higher than 99.8%. This statistic confirms that the years of theoretical and experimental work clearly resulted in robust design and execution criteria.

Creating discontinuous proppant packs signif-icantly reduces the cost and the environmental footprint of fracturing treatments. Engineers esti-mate that, on a global basis, operators have saved more than 86,180,000 kg [190 million lbm] of proppant compared with what would be used in conventional treatments. This corresponds to about 7,000 fewer road trips and about 900 fewer railroad trips hauling proppant to the wellsites. Accordingly, diesel fuel consumption declined by about 283.9 m3 [75,000 galUS], and CO2 emissions dropped by about 725,750 kg [1.6 million lbm].

The scope of the HiWAY service is still expand-ing. For example, engineers are looking for ways to broaden the types of reservoirs for which the technique is applicable. Field experience has demonstrated that the initial guideline for the ratio between the formation Young’s modulus and the closure stress was too conservative. The limit has been reduced from 1,000 to 350, thereby opening HiWAY service to a wider range of forma-tions, particularly shales.

The initial success with horizontal wells in the Eagle Ford formation led scientists to per-form additional modeling and experimental work to tailor the HiWAY technique to a deviated well environment. As a result, horizontal wells account for 69% of the treatments performed to date. Work has also begun to extend HiWAY ser-vice from a cased hole to an openhole environ-ment. As operators perform the technique successfully in a wider range of well types, it is conceivable that discontinuous proppant packs will become standard industry practice. —EBN

> The Eagle Ford Shale formation. The formation produces oil, condensate and gas. Petrohawk operates the Hawkville field in south Texas.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 22ORAUT11-HWY 22

0 km

0 mi 100

100

U S A

MEXICO

Texas

T e x a s

Oil window

Dry gas windowHawkville field

Wet gas–condensate window

Eagle Ford Shale formation

Gulf of Mexico

> Six-month gas and oil production from horizontal wells that received the HiWAY treatment in the Hawkville field, Texas. Production from gas wells (top) and oil wells (bottom) that received the HiWAY treatment (blue) was significantly higher than that observed from the offset wells.

Oilfield ReviewAUTUMN 11 HiWAY Fig. 23ORAUT11-HWY 23

Cum

ulat

ive

oil p

rodu

ctio

n, b

bl

1,600,000

120,000

100,000

80,000

60,000

40,000

20,000

0

1,200,000

800,000

400,000

00 30 60

Time, d

HiWAY Well 1OffsetOffsetOffset

HiWAY Well 2Offset

Time, d

90 120 150 180

0 30 60 90 120 150 180

Cum

ulat

ive

gas

prod

uctio

n, M

cf

43223schD4R1.indd 14 11/17/11 6:37 PM