Oilfield Review

January 2016
PERU
BOLIVIA

PARAGUAY
BRAZIL

URUGUAY
CHILE ARGENTINA

Neuquén
basin

Assessing Heavy Oil Reservoirs

Cement Sheath Evaluation
Mud Removal for Cementing
Shale Production in Argentina
Rod Pumps in Unconventionals
 For more on shale, see “Shale Gas Revolution”

Knowledge at your fingertips.

from the Autumn 2011 Oilfield Review, available
online or on the app.

http://www.slb.com/oilfieldreview
Since 1989, Oilfield Review has been a reference for E&P professionals
worldwide and encompasses:
· More than 500 in-depth, highly referenced articles across
E&P technology and innovation
· The “E&P Defining Series,” a quick introduction to fundamentals.
The Oilfield Glossary has more than 5,000 oilfield terms, supplementing
and supporting the journal.

http://www.glossary.oilfield.slb.com

Download the app by visiting the Apple or Google Play online store
and searching for Schlumberger Oilfield Review.
Oilfield Review
Authoritative. Relevant. Informative.
EDITORIAL
Oilfield Review: Changing with the Industry
Successfully managing change is a skill that
those engaged in E&P activities should hone,
whether because of new responsibilities,
technology development and innovation or an
industry downturn—or preferably an upturn.
Oilfield Review is changing, too, both in print and
online. Before undertaking any change, however,
clarifying the foundations and principles that must
not change is essential.
The Technical Review. The publication changed its name to Oilfield
Review in 1989. As this last cover of the Technical Review illustrates,
despite the advances in technology in the intervening decades, in
particular in characterizing reservoirs, having a conversation in front
of real rock helps those just starting in the industry—as well as
seasoned scientists and engineers—connect theory to reality.
Schlumberger experts and technology have
provided the foundation for Oilfield Review in-depth
articles since the journal’s 1989 origin. These heavily
researched articles remain the pillars of the journal.
They will continue to be thoroughly referenced, an
aspect often cited by readers as providing as much
value as the article itself. Oilfield Review will remain
a source of authoritative, relevant and informative
content to E&P professionals across the industry.
Various readers have commented on the benefit
of having the articles summarized in a single place,
informing their engagement. Looking at this issue’s
five feature articles on diverse topics illustrates the
point. Article summaries now appear on a single
page opposite a refreshed table of contents, designed
to help readers quickly decide whether to engage
with each topic.
Based on a clear demand from online readers,
we will continue to develop the defining series:
concise, authoritative two-page summaries of key
industry disciplines and technologies. These short
articles are more popular online (measured by digital
download) by a factor of 10 over the regular technical
articles. This download factor reflects the wide
audience interested in learning about an area away
from their core expertise or one of new responsibility.
Students and those new to the industry form a par-
ticularly large audience for the defining articles.
Geomechanics, mud logging and gas lift are the
topics of this issue’s defining series. Defining series
topics that you will find online but do not appear in
print are rod lift pumping, reservoir engineering
and HPHT.
Also new to the print edition, occupying the last
page, is the first in a series of oilfield history articles
contributed by Mark Mau, the Schlumberger histo-
rian, under the heading Looking Back. Mark, together
with Henry Edmundson, the Oilfield Review founding
editor, recently published Groundbreakers: The
Story of Oilfield Technology and the People Who
Made It Happen. In this issue, Mark looks back on
the origins of artificial lift.
The online content in the Oilfield Review section
of the Schlumberger website, slb.com, has also
been updated. The ever-growing defining series is
now an independent section within the existing
Oilfield Review web pages. Online video content
associated with technical articles can now be
accessed on each issue’s summary page—as well as
in the Oilfield Review app—to enrich reader expe-
rience and understanding. Various readers com-
mented on preferring to watch a video and then
decide whether to dive into the associated article.
Oilfield Review now has a Facebook page. Of
course, this in itself is nothing groundbreaking.
Facebook is mature and mainstream—most journals
have been on the social media platform for some
time. For Oilfield Review, this presence will improve
the connection to millennials, who grew up with
social media and remain frequent, if evolved-in-
purpose, users and now form a new E&P industry
generation. The Pew Research Center reported in
August 2015 that 72% of all US adults who go online
use Facebook, including 82% of those ages 18 to 29.
As well as engaging new readers, an Oilfield Review
Facebook page will enable existing readers to keep
up to date with new content as it is published and
to link to such content to read when interested.
The simple conclusion for Oilfield Review is to
change while remaining the same—to leverage the
increased importance of online engagement, develop
ways to engage and expose a wider audience to the
vibrant new technology and innovation in the oil
field and continue to produce the deep technical
articles that remain the journal’s foundation.
Your comments are welcome. Please email
oilfieldreview@slb.com.
Charles Cosad
Executive Editor, Oilfield Review
1
 Oilfield Review
January 2016
Volume 28, Number 1
ISSN 0923-1730
Articles
4Movability
www.slb.com/oilfieldreview

Executive Editor
Quicklook for Heavy Oils
Charlie Cosad To distinguish between movable oil- and
freshwater-bearing formations, operators may
Senior Editors
Tony Smithson turn to dielectric tools when assessing heavy
Matt Varhaug oil reservoirs.
Rick von Flatern

Editors
Irene Færgestad
Richard Nolen-Hoeksema
10
An Integrated Approach to Cement
Contributing Editors Evaluation
Erik Nelson Engineers have developed a technique to
Ginger Oppenheimer improve cement bond log interpretation by
Design/Production allowing logging personnel and operators to
Herring Design easily view and analyze previous well events.
Illustration 
Chris Lockwood
Mike Messinger 20
Departments
George Stewart Mud Removal—Clearing the Way
for Effective Cementing
1
Printing Editorial
RR Donnelley—Wetmore Plant Newly developed spacers are helping operators
Oilfield Review: Changing with the Industry
ensure zonal isolation and well integrity from
Advisory Panel
Hani Elshahawi
cement jobs. 47
Shell Exploration and Production Contributors
Houston, Texas, USA
Gretchen M. Gillis
26
Vaca Muerta Shale—Taming a Giant
49
Coming in Oilfield Review
Aramco Services Company
Houston, Texas Through collaboration of engineers at YPF, SA
Roland Hamp and Schlumberger, Argentina has been added 50
Woodside Energy Ltd. to the growing global list of unconventional The Defining Series
Perth, Australia
resource producers. Geomechanics, Mud Logging and Gas Lift
Dilip M. Kale
ONGC Energy Centre 56
Delhi, India
George King
40
Rod Pumps in Unconventional Resource Wells
Looking Back
Looking Back on Artificial Lift
Apache Corporation
Houston, Texas Because rod pump systems perform efficiently at
Michael Oristaglio low production rates, operators eventually deploy
Yale Climate & Energy Institute them on most shale wells.
New Haven, Connecticut, USA
John Thorogood
Drilling Global Consultant LLP
Aberdeenshire, UK On the cover:
In the shadow of the Andes Mountains, YPF, SA is developing an unconventional resource
Publishing play in the Neuquén basin of South America. The production of commercial quantities of
Oilfield Review is published and hydrocarbons from the Vaca Muerta Shale is one of the first such successful developments
printed in the USA. outside the US. The Argentina-based company has drilled more than 400 wells and has
initiated production of both oil and gas.
© 2016 Schlumberger.
All rights reserved.

2
 Article Summaries
Caliper critical to establishing and Zonal isolation is accomplished maximize production and reduce
arm
R A4

maintaining zonal isolation in the through cementing operations that the development cycle of the play,
R A3

well, supporting the casing and can only be effective if drilling flu- engineers at YPF are using fracture
R A2

Mudcake
R A1

Articulated
probes
pad
preventing external casing corrosion. ids, or mud, are fully displaced from simulators, predictive numerical
TA

the casing-borehole annulus ahead models and field-specific workflows

TB

Evaluation of the cement sheath

R B1

is critical and may be enhanced by of the cement. The challenge for that are created using information
R B2

Movability Quicklook for

R B3

an integrated approach in which chemists and fluids experts is to derived during an extensive evalua-
R B4

Heavy Oils engineers consider cement design a spacer that prevents the tion program. Page 26.
Central to developing an optimal evaluation logs in the context of cement slurry and drilling fluids
field production strategy is an events that took place during being
PERUcleaned from the well from
Position, %

assessment of the reservoir rocks drilling and subsequent cement mixing. Optimally performing spac-
BOLIVIA BRAZIL10
0 20 30 40 50 60 70 80 90 100
0

22,000 100

20,000 B 80
C

and the fluids they contain. slurry placement. Schlumberger ers also clean the wellbore ahead of 18,000
16,000
60
40

Because resistivity tools cannot dis- engineers have developed a technique the cement slurry, ensuring a good
14,000 A 20

Load capacity, %
12,000 0

Load, lbf
D –20
10,000

tinguish between oil- and freshwater- by which logging personnel and cement bond between PARAGUAY
the casingParaná basin 8,000 –40
B C
Pump fill set point –60
6,000
4,000 –80

bearing formations, a conventional operators may easily view and analyze and the wellbore along the length of 2,000
0 A
–100
–120

logging suite may be ineffective in previous well events, thereby the cemented interval. Page 20.
–2,000 D –140
0 10 20 30 40 50 60 70 80 90 100 110 120 130

Rod Pumps in Unconventional

Position, in.

heavy oil reservoirs. Assessment of improving log interpretation. Chaco basin

these reservoirs may also be further This comprehensive examination

Resource Wells
complicated by difficulty in distin- of the entire well construction Operators use rod pump systems to
URUGUAY
CHILE ARGENTINA
guishing producible oil from oil that is process can help engineers confirm enhance production
SOUTH AMERICA from shale wells.

not movable. zonal isolation or determine why a Neuquén These wells, which may experience
Therefore operators may turn to primary cementing operation failed basin
decline rates of up to 70% per year,
dielectric tools that measure dielec- to achieve isolation. Page 10. require artificial lift early in life.
tric permittivity to distinguish San Jorge Downhole conditions change rap-
between oil and low-salinity water.
Vaca Muerta Shale— basin
idly as the well is produced, and as a
In addition, using the quicklook
Taming a Giant result, operators must switch artifi-
1 2 3 4 1 2 3 4 1 2 3 4
analysis process for a multifrequency
Casing Standoff Cement Displacement
Simulation
Acoustic Impedance
Map
Acoustic Impedance Casing Standoff Cement Displacement
Simulation
Acoustic Impedance
Map
Acoustic Impedance
Unconventional resource
Casing StandoffAustral- plays have
Cement Displacement
Simulation
cial lift system types from high-rate
Acoustic Impedance
Map
Acoustic Impedance

Magallanes Prospective basin

dielectric dispersion service, opera- revolutionized oil and basin gas produc- 500to low-rate
1,000 km devices. Because rod
AI_MD Minimum AI_MD Minimum AI_MD Minimum
–1 Mrayl 9 –1 Mrayl 9 0 –1 Mrayl 9

tion in North America but have been 300pumps perform efficiently at low
–2.00 8.00 –2.00 8.00 –2.00 8.00
tors can better characterize oil mov-
At Centralizers Average At Centralizers Average At Centralizers Average
58 % 100 –2e+03 Mrayl 56 –1 Mrayl 9 58 % 100 –2e+03 Mrayl 56 –1 Mrayl 9 58 % 100 0 600 mi
–2e+03 Mrayl 56 –1 Mrayl 9

production rates, they are often the

Image Orientation° Image Orientation°
slow to develop in other parts of the
Depth, Between Centralizers Maximum Depth, Between Centralizers Maximum Depth, Between Centralizers Image Orientation° Maximum
ability. Page 4.
ft 0 % 100 0 % 5 0 90 180 270 360 –1 Mrayl 9 ft 0 % 100 0 % 5 0 90 180 270 360 –1 Mrayl 9 ft 0 % 100 0 % 5 0 90 180 270 360 –1 Mrayl 9

world. That is changing, however, final artificial lift system on a well.

500
Mud Removal—Clearing the
500 500

and now the Vaca Muerta formation However, some shale well operators
1,000
Way for Effective Cementing
1,000 1,000

in Argentina has been added to the are deviating from replacing systems
1,500 1,500 1,500

Spacer design and composition have growing list of unconventional over time, and are, instead, install-
2,000
not kept up with the increasing
2,000
resource producers.
2,000
ing rod pump systems early in the
2,500
complexity of drilling fluid chemis-
2,500
Success in the Vaca Muerta for-
2,500
well’s life. Page 40.
3,000
try. Recent industry efforts are now
3,000
mation has been the result of an
3,000

3,500
providing enhanced and optimized
3,500 integrated approach that includes
3,500

An Integrated Approach to field- and formation-specific work-

spacer solutions that more effi-
Cement Evaluation
4,000 4,000 4,000

4,500 ciently clean the wellbore, helping

4,500
flows and the application of a
4,500

The cement sheath created during

5,000
ensure zonal isolation and well
5,000
dynamic fracture model to optimize
5,000

primary cementing operations is

5,500
integrity from cementing operations.
5,500
the stimulation program. Now, to
5,500

6,000 6,000 6,000

About Oilfield Review Oilfield Review online Oilfield Review app Correspondence
6,500 6,500 6,500

Oilfield Review is the Schlumberger Visit www.slb.com/oilfieldreview for the Download the free app by visiting the Oilfield Review
flagship journal of technology, innovation current and all previous editions, videos Apple or Google Play online store and 5599 San Felipe
and the science of E&P. Contributors to associated with certain articles, the searching for Schlumberger Oilfield Houston, TX 77056
articles are industry professionals and defining series, as well as links to the Review. Previous issues and video United States
experts from around the world. Those apps and Oilfield Glossary. content are available. (1) 713-513-3760
listed with only geographic locations are E-mail: oilfieldreview@slb.com
employees of Schlumberger. www.slb.com/oilfieldreview

Reproductions without permission are

strictly prohibited.
3

Steve Grayson
Petrophysics Consultant
Camarillo, California, USA
James Hemingway
Sugar Land, Texas, USA
Oilfield Review 28, no. 1 (January 2016).
Copyright © 2016 Schlumberger.
For help in preparation of this article, thanks to Martin
Isaacs, Houston, Texas.
Dielectric Scanner and EPT are marks of Schlumberger.
1. Head IM, Gray ND and Larter SR: “Life in the Slow Lane;
Biogeochemistry of Biodegraded Petroleum Containing
Reservoirs and Implications for Energy Recovery and
Carbon Management,” Frontiers in Microbiology
(November 11, 2014), http://journal.frontiersin.org/
article/10.3389/fmicb.2014.00566/full (accessed
November 15, 2015).
2. Connan J: “Biodegradation of Crude Oils in Reservoirs,”
in Brooks J and Welte D (eds): Advances in Petroleum
Geochemistry, vol 1. London: Academic Press (1984):
299–335.
4 Oilfield Review
Movability Quicklook for Heavy Oils
Log analysis of heavy oil reservoirs may be hampered by difficulties in
distinguishing heavy oil from freshwater. A wellsite quicklook evaluation
process that incorporates a multifrequency dielectric measurement is helping
operators overcome this challenge.
Vast reserves of heavy oil have been discovered
in more than 30 countries around the world.
These resources, estimated at 890 billion m3
[5.6 trillion bbl] of heavy oil and bitumen, make
up a dominant share of the world’s petroleum
inventory.1 Though challenging to produce,
heavy oils will become increasingly important as
reserves of lighter oil inevitably decline.
Most heavy oil deposits originated as light-
to medium-gravity oils formed at depth. These
oils migrated upward and became trapped in
shallower formations, where they were later
degraded, chiefly through microbial action. Much
of this degradation occurs near the oil/water
contact; bacteria live within the aqueous region
but do not thrive in oil.2 The bacteria metabolize
paraffinic, naphthenic and aromatic hydrocar-
bons into heavier molecules. Chemical or physical
processes may also be involved. These processes
strip away lower molecular-weight hydrocarbons
through preferential migration of lighter compo-
nents, devolatilization and evaporation beneath
leaky seals and washing by formation waters. As
the lighter fractions are removed, only complex
hydrocarbon compounds are retained, and heavy
oil is formed.
Heavy oil has been defined variously as oil
whose gas-free viscosity is between 100 cP and
10,000 cP [100 mPa.s and 10,000 mPa.s] at reser-
voir temperature or oil that is slightly less dense
than water whose gravity ranges between 10 and
22.3 degrees API. Heavy oils are characterized
by long, complex hydrocarbon molecules and are
low in hydrogen; they are often high in carbon,
nitrogen, sulfur or heavy metals.
The long hydrocarbon molecules impart
high internal friction that results in increased
viscosity. Because of its inherent resistance to
flow, heavy oil requires specialized methods to
produce it. These methods include waterflood-
ing and steam injection to mobilize the oil or
chemical treatments that alter wettability, lower
viscosity and reduce interfacial tension. Central
to developing an optimal recovery strategy is an
assessment of the reservoir rock and the fluids
contained within.
These assessments usually begin with a suite
of logs, which the operator uses first to identify
oil-bearing formations and then to ascertain
which intervals are producible. The conventional
approach to formation evaluation is based on
computing water saturation from resistivity and
porosity logs. The presence of hydrocarbons is
generally indicated by high resistivity measure-
ments in porous zones. By measuring the poros-
ity of the rock and determining the percentage
of pore volume saturated by water, geoscientists
can then infer the remaining percentage of pore
space that is saturated by hydrocarbons.
However, this technique is ineffective for
determining hydrocarbon saturation in res-
ervoirs containing formation waters of low or
unknown salinity; heavy oil reservoirs, typically
found at shallow depths, tend to contain rela-
tively freshwater. Freshwater, like oil, is a poor
conductor of electricity. Conventional resistivity
tools exhibit high resistivity through oil zones
and through low-salinity water zones, causing
oil-bearing intervals to be indistinguishable
from those containing freshwater. Assessment of
heavy oil reservoirs may be further complicated
by difficulty in distinguishing producible oil from
oil that is not movable. Conventional resistivity
and porosity logging suites do not measure the
movability of heavy oil within a formation. This
article describes how operators in California,
USA, use Dielectric Scanner measurements in a
wellsite quicklook evaluation to identify intervals
containing movable heavy oil.

A Different Approach
Heavy oil reservoirs often require operators to
look beyond conventional logging suites for
determining oil saturations and movability.
Dielectric permittivity, the ability of an electro-
magnetic field to permeate a formation, has been
measured by various logging devices since the
1970s. A dielectric measurement can distinguish
between oil and water of low or indeterminate
salinity using the contrasting dielectric proper-
ties of oil and water. Dielectric tools have a shal-
low depth of investigation and thus analyze the
flushed zone surrounding a wellbore—that part
of the formation invaded by drilling fluid filtrates, Caliper
arm

R A4
which have partially or completely displaced the

R A3
original formation fluids. Because the movable

R A2
fluids have been flushed, the dielectric measure-
Mudcake

R A1
Articulated
ment evaluates the filtrate plus any residual flu- pad probes
ids that have not been flushed by the invading

TA
filtrate. In wellbore intervals that experience

TB
very little invasion, the oil saturations of the

R B1
flushed zone and deeper uninvaded zones are

R B2
nearly equal, thus making the shallow dielectric

R B3
measurement useful for inferring saturation

values of the uncontaminated zone within heavy
oil reservoirs.
The Dielectric Scanner multifrequency dielec-
tric dispersion service uses multiple frequencies,
polarizations and transmitter-receiver spacings
to acquire dielectric permittivity and conductiv-
ity measurements at four depths of investigation.
This logging technology isn’t new—it has been
used extensively since its introduction in 2010 as
an improvement over the EPT electromagnetic
propagation tool. However, the development of a
versatile quicklook analysis process takes advan-
tage of multiple depths of investigation to help
operators evaluate heavy oil reservoirs.
Through rapid analysis of logging data at the
wellsite, operators obtain information needed to
determine whether to complete an interval, con-
demn it or investigate further using sidewall cor-
ing or other techniques. In some instances, this
quicklook analysis has identified where to shoot
sidewall cores that later confirmed the presence
of producible oil.
R B4
Figure 1. The Dielectric Scanner tool. This tool uses a caliper arm to push
the articulated pad against the formation. The pad’s curvature also helps
improve contact with the borehole wall. The transmitters (TA and TB) and
antenna sets (RA1 to RA4 and RB1 to RB4) are collocated crossdipoles and can
operate simultaneously in transverse (red arrow) and longitudinal (blue arrow)
polarization modes. Two open-ended coaxial cable probes measure mudcake
properties and provide quality control. For accurate fluid property input, the
tool measures temperature and pressure at the point of measurement. Depth
of investigation is approximately 2.5 cm to 10 cm [1 in. to 4 in.] depending on
transmitter-to-receiver spacing and formation fluid properties.
Sensors and Measurements include phase shift and attenuation. Unlike some
Dielectric measurement technology has evolved dielectric tools, the Dielectric Scanner tool can be
substantially since the introduction of the EPT run in both water-base and oil-base mud systems.
electromagnetic propagation tool, which has The tool uses an articulated pad that can
only a single depth of investigation and single move freely in three dimensions, which enhances
frequency. The Dielectric Scanner tool makes borehole contact for measurements in rugose
nine separate autocalibrated, borehole-com- hole conditions (Figure 1). Two electromagnetic
pensated measurements. These measurements

January 2016 5
 Microresistivity Dielectric Water-Filled Porosity Neutron Porosity
1 ohm.m 1,000 50 % 0
Water 40 % 10
Saturation
Depth, Deep Resistivity Total Porosity
Density Porosity
0 % 100 ft 1 ohm.m 1,000 50 % 0
40 % 10
X,000
T-R1 Porosity
40 % 10
T-R2 Porosity

X,100 40 % 10
T-R3 Porosity
Heavy oil
40 % 10
T-R4 Porosity
X,200 40 % 10
Density to T-R1
saturation profile

X,300 Density to T-R2

Water zone
saturation profile

Figure 2. Oil crossover response. In the presence of oil, the water-filled Density to T-R3
porosity curve of the dielectric measurement (Track 3, blue) will read lower
than the total porosity curve derived from the density-neutron measurements saturation profile
(black). This separation creates a readily identifiable crossover response for
Density to T-R4
distinguishing oil from water. (Adapted from Hizem et al, reference 3.)
saturation profile

transmitters in the middle of the pad are brack- The T-R3 and T-R4 curves
eted by two arrays of four receivers in a sym- nearly overlie, indicating
metrical configuration. Each transmitter and flushing shallower than the
receiver consists of orthogonal crossed dipole maximum DOI of 10 cm.

antennae. The transmitters broadcast along two

axial orientations at four frequencies ranging from
20 MHz to 1 GHz; the lower frequencies propa-
gate deeper into the formation. The receiver
arrays take advantage of four transmitter-to-
The T-R3 and T-R4 curves
receiver (T-R) spacings to record dielectric per- separate, indicating flushing
mittivity measurements. In some cases, vertical to the maximum DOI of 10 cm.
resolution of 2.5 cm [1.0 in.] can be achieved for
evaluating thin formation laminations.

Quicklook Evaluation Figure 3. Porosity track of the quicklook presentation. Variations in the
In most wells, the first suite of logs run in the hole saturation profile are generated by comparing total porosity from the
density measurement with water-filled porosity measurements from the
helps operators assess the formation and deter-
Dielectric Scanner tool. Separation between density porosity and four
mine whether further evaluations are warranted. dielectric measurements (T-R1 to T-R4) obtained at depth of investigations
In addition to gamma ray, resistivity, spontane- (DOIs) of approximately 2.5 cm, 5 cm, 7.5 cm and 10 cm characterize the
ous potential and caliper tools, this suite often depth of invasion and, by extension, the movable oil saturation in the
formation (top). Dielectric curve separations indicate depth of invasion and
includes neutron porosity and bulk density tools. show how oil saturation varies within the 10-cm maximum DOI region of the
The neutron porosity and bulk density measure- tool (bottom). If the T-R3 and T-R4 curves read the same but are separated
ments are useful in determining total porosity. from T-R1 and T-R2 curves (yellow oval), the invasion depth is less than
In heavy oil wells, this initial suite is sometimes 10 cm. Where the T-R3 and T-R4 curves separate, invasion has flushed oil
within the 10-cm DOI and possibly farther. If all four dielectric curves are
run in conjunction with the Dielectric Scanner separated from each other (green oval), the mud filtrate has swept deeper
tool, which measures the water-filled porosity of than 7.5 cm and probably deeper than 10 cm. In the event that all four
reservoir rocks. By comparing the total porosity curves overlie, the oil saturation is unvarying within the 10-cm DOI. This
measurements with water-filled porosity mea- overlap indicates that no flushing has occurred and the oil in place has not
yielded to mud filtrate invasion. An alternate interpretation when the four
surements, the operator can calculate water curves overlie is of an oil-bearing zone that has been fully flushed and is at
saturation in the flushed zone of the formation. residual oil saturation. (Adapted from Grayson and Hemingway, reference 4.)

6 Oilfield Review
 A wellsite quicklook application uses this Neutron Porosity
porosity comparison to differentiate between 85 % 0
intervals containing water or heavy oil.3 At its Density Porosity
foundation is an analysis process originally used 85 % 0
in conjunction with the EPT tool, which also com- T-R1 Porosity
pared total porosity with water-filled porosity. 85 % 0

Across water-filled intervals, the neutron-density Gamma Ray T-R2 Porosity

total porosity curve will overlie the dielectric
water-filled porosity curve; conversely, in inter-
vals containing oil, these curves will diverge. The
dielectric tool measurement is insensitive to oil,
so as the curves separate, the water-filled poros-
ity curve will read lower than the total porosity
curve. This response produces an “oil crossover”
indicator, allowing for easy identification of pay
at the wellsite, regardless of formation water
salinity (Figure 2). The amount of separation is
proportional to the oil saturation in the flushed
zone of the near-wellbore region.4
In heavy oil and low-permeability reservoirs,
if oil is not flushed by mud filtrate while drilling,
the porosity comparison provides a quantifica-
tion of water saturation (Sw) deeper within the
formation in the undisturbed zone surrounding
the wellbore. When oil movement has occurred,
however, the operator must be able to assess the
depth and extent of flushing. This capability was
not provided by tools that had only a single depth
of investigation.
By measuring across four depths of investiga-
tion, the Dielectric Scanner tool is able to cre-
ate a saturation profile across the flushed zone.5
The resulting quicklook presentation shows a
total porosity curve derived from the bulk den-
sity measurement along with four dielectric
measurements of water-filled porosity taken at
1-in. [2.5-cm] intervals spanning approximately
1 to 4 in. [2.5 to 10 cm] into the formation.
The oil saturation profile reflects oil move-
ment resulting from the invasion process. By
examining the relative change of oil saturation
in the flushed zone, analysts can evaluate oil
movability (Figure 3). In zones characterized
by low oil movability, the saturation profile
curves will overlie one another. Conversely, if
oil is seen to move within the flushed zone, the
curves will separate.
Movability in a formation is influenced by a
variety of factors that affect the capability of an
invading fluid to displace oil within a formation.
The differential between the formation pres-
sure and the pressure exerted by the mud col-
umn drives the mud filtrate into the formation.
Formation porosity and permeability impact the
quantity of filtrate and the maximum flow rate
that can be achieved through the rock. Mud prop-
erties affect the development of mudcake, which
0 gAPI 150 85 % 0
Spontaneous Potential Resistivity, Deep T-R3 Porosity
–80 mV 20 0.2 ohm.m 200 85 % 0
Oil Saturation: T-R1 Resistivity, Medium T-R4 Porosity
100 % 0 0.2 ohm.m 200 85 % 0
Oil Saturation: T-R4 Depth, Resistivity, Shallow Pad Temperature
100 % 0 ft 0.2 ohm.m 200 110 degree, F 140
X90
Y00
Y10
Y20
Y30
Figure 4. Heavy oil in a diatomite reservoir. The Dielectric Scanner quicklook presentation shows
dielectric water-filled porosity curves taken from four depths of investigation. Water-filled porosity
decreases with depth of investigation in the highlighted zone (yellow oval), indicating strong oil flushing.
There, the T-R1 curve approaches the density porosity, typical of near complete flushing of the oil. The
deeper reading curves indicate considerably more oil at T-R2 and additional increases at the T-R3 and
T-R4 measurements. (Adapted from Grayson and Hemingway, reference 5.)
eventually forms a barrier to prevent further fil- by a neighboring operator in an adjacent section
trate invasion. Formation characteristics, such of the field. Given that the salinity of injected
as pore size, the wettability of the rock face and water usually differs from that of native waters
the viscosity of the oil, influence how easily the in the reservoir, the operator had concerns about
oil will be displaced by the filtrate. The quicklook how injection water might affect the resistiv-
shows the effects of these factors. ity response in the usual triple-combo logging
suite.6 Therefore, the operator also obtained a
Field Applications Dielectric Scanner log to supplement the forma-
Diatoms, microscopic algae that have silica-rich tion evaluation program (Figure 4).
cell walls, have existed since the Cretaceous
3. Hizem M, Budan H, Devillé B, Faivre O, Mossé L, Simon
Period. In some basins, the remains of ancient M: “Dielectric Dispersion: A New Wireline Petrophysical
diatoms form a sedimentary rock called diato- Measurement,” paper SPE 116130, presented at the
SPE Annual Technical Conference and Exhibition,
mite, which can make an excellent reservoir Denver, Colorado, USA, September 21–24, 2008.
rock. In California, cyclic steam injection is being 4. Grayson S and Hemingway J: “A Heavy Oil Mobility
Quicklook Using Dielectric Measurements at Four Depths
used to produce 13 degree API gravity heavy oil of Investigation,” paper SPE 166329, presented at the
from a diatomite reservoir. At one lease, a well SPE Annual Technical Conference and Exhibition,
New Orleans, September 30–October 2, 2013.
was drilled prior to initiation of steam injection
5. Grayson ST and Hemingway JL: “Application of an
in this portion of the field. However, because the Oil-Movability Quicklook Technique Using Dielectric
well is situated close to a lease line, the opera- Measurements at Four Depths of Investigation,”
Petrophysics 55, no. 5 (October 2014): 461–469.
tor wanted to determine whether the well would 6. Grayson and Hemingway, reference 4.
be affected by thermal stimulation carried out

January 2016 7
 Mud-filled borehole Sidewall core analysis, a routine part of many
Drilling fluid Oil well evaluations, may also serve to emphasize the
impact that heavy oil movability has on oil satu-
Dielectric measurements ration measurements. Oil movability will affect
DOI DOI DOI DOI Porosity
~2.5 cm ~5 cm ~7.5 cm ~10 cm 50 % 10
the capability of filtrates to flush oil from the
pore space of rock in the near-wellbore region
and can create a measurable effect on the satu-
ration of the rock. In low-movability zones, oil
saturation in sidewall cores (SWCs) is relatively
unaffected by flushing. In high-movability zones,
flushing reduces the oil saturation measured in
SWC bullet SWC samples.
Percussion sidewall cores are typically
acquired in heavy oil reservoirs and used by
geologists to estimate porosity, oil saturation
and oil gravity. Depending on rock strength and
bullet design, most SWC bullets will penetrate
the formation to a depth of about 2.5 cm [1 in.].
After the core bullets are retrieved to the sur-
face, they are sent to a laboratory for analysis.
Figure 5. Sidewall core in a zone of low movability. The 2.5-cm depth of penetration typical of most
sidewall cores falls well within the region investigated by the Dielectric Scanner tool. As the quicklook
These analyses may provide input for calculat-
display (right) shows, through zones characterized by low oil movability, the dielectric saturation ing field reserves.
profile curves tend to overlie. (Adapted from Grayson and Hemingway, reference 5.) Because the dielectric oil-movability quick-
look provides a wellsite indication of oil flush-
ing, it can be a valuable tool for selecting depths
The separation between the dielectric curves of enhanced oil movability. This coincidence sup- from which to obtain SWC samples that are more
illustrates a drastic change in oil movement over ports the idea that the steam stimulation by the representative of actual saturations. Where oil
a very short interval at Y15 ft. Where significant offset operator has contributed significantly to movability is low and no flushing is seen on the
flushing of the oil is seen, data from the tem- enhanced production. Other wells drilled in the dielectric quicklook log, SWC saturations should
perature sensor in the tool pad show that the field, located farther from the lease line, do not accurately reflect the oil saturation in the zone
maximum temperature coincides with the zone show these responses. (Figure 5).
By contrast, in zones of higher movability, the
oil will be flushed away from the near-wellbore
Mud-filled borehole region, where an SWC bullet acquires the rock
Drilling fluid Oil sample. In this case, the oil saturation derived
from the SWC analysis will be lower than the
Dielectric measurements actual oil saturation of the zone and the 2.5-cm
DOI DOI DOI DOI Porosity and 10-cm dielectric curves of the quicklook pre-
~2.5 cm ~5 cm ~7.5 cm ~10 cm 50 % 10
sentation will be separated. The oil saturation
obtained from the 2.5-cm dielectric curve should
reflect the saturation seen in the core; however,
the saturation from the 10-cm curve will reflect
the actual oil saturation of that interval if the
invasion has not extended to 10 cm into the for-
mation (Figure 6).
SWC bullet
Using the oil saturation invasion profile
through the 10-cm depth of investigation, the
operator can place sidewall cores in optimal
locations where minimal flushing has occurred.
But where cores have already been obtained,
the core-derived oil saturations may be comple-
mented or sometimes replaced by oil saturation
values derived from the combination of dielectric
Figure 6. Sidewall core in a zone of high movability. Flushing (brown, grading into green) is observed and triple-combo log measurements. Log-derived
in the saturation profile as moved oil causes the dielecric curves (far right) to separate. In this
oil saturations, measured continuously over the
heavy oil zone, water-filled porosity decreases, but oil saturation increases with increasing depth
of investigation. In zones affected by flushing, core-derived oil saturations are suspect and may not entire zone of interest, have an advantage over
represent true oil saturation. (Adapted from Grayson and Hemingway, reference 5.)

8 Oilfield Review
 Neutron Porosity
75 % 15
Density Porosity
75 % 15
T-R1 Porosity
75 % 15
Gamma Ray T-R2 Porosity
0 gAPI 150 75 % 15
Spontaneous Potential Resistivity, Deep T-R3 Porosity
–80 mV 20 0.2 ohm.m 200 75 % 15
Oil Saturation: T-R1 Resistivity, Medium T-R4 Porosity
0 % 100 0.2 ohm.m 200 75 % 15
Oil Saturation: T-R4 Depth, Resistivity, Shallow Pad Temperature
0 % 100 ft 0.2 ohm.m 200 110 degree, F 140

A
X80

B T-R1 to T-R4 dielectric

X90 water-filled porosities

Y00
C

Sidewall core oil saturation

Figure 7. High producibilty potential contrary to sidewall core (SWC) analysis. The Dielectric Scanner measurement
results (Track 3) indicate a potentially productive zone that SWC analysis failed to identify. Points A, B and C were
sampled with SWCs, which were then analyzed. Laboratory analysis of SWC oil saturation yielded 39% at Point A,
35% at Point B and 34% at Point C. The quicklook log through this diatomite heavy oil reservoir showed oil flushing at
Point B but no flushing at Points A and C. Oil-saturation curves derived from the T-R1 and T-R4 dielectric measurements
are identified in Track 1. When flushing takes place, as at Point B, higher oil saturations are indicated by the deepest-
reading (T-R4) dielectric curve. Where no flushing is observed, saturations from the T-R1 and T-R4 curves read the
same. The movability of oil observed at Point B indicates a zone having either higher permeability or having oil of lower
viscosity than the adjacent zones. Although SWC oil saturation at Point B was 35%, the T-R1 measurement shows 42%
oil saturation and the T-R4 measurement shows 69%. (Adapted from Grayson and Hemingway, reference 4.)

those obtained at discrete sample points by side-
wall cores. The quicklook log has been used to
identify zones affected by flushing, where core-
derived oil saturations are suspect (Figure 7).
One operator’s pessimistic core evaluations,
previously used in reserve calculations, were
later discounted after the quicklook log showed
the effects of flushing, resulting in an increase in
booked reserves. This experience demonstrated
the importance of using the deepest-reading
measurement for fluid saturation when flush-
ing is identified, especially when calculating
reserves. It also shows the value of identifying
flushing before shooting sidewall cores to ensure
that representative samples are acquired.
The Dielectric Scanner multifrequency
dielectric dispersion service helps operators
readily distinguish hydrocarbon zones from
freshwater zones. The oil-movability quicklook
analysis gives geoscientists broader insight into
variations of permeability, oil saturation and
viscosity in heavy oil reservoirs. The ability to
identify flushing can help operators improve pet-
rophysical evaluations, especially when calculat-
ing reserves. The quicklook oil saturation profile
provides a straightforward observation of oil mov-
ability variations and can help operators optimize
completion strategies by providing information at
the wellsite to show zones of higher movability
within a formation. —MV

January 2016 9
 An Integrated Approach to Cement Evaluation

Gunnar DeBruijn Cement sheath evaluation is enhanced when engineers consider cement evaluation
Sugar Land, Texas, USA
logs in the context of events that took place during drilling and subsequent cement
Anouar Elhancha slurry placement. Such an integrated approach helps engineers confirm the presence
Polina Khalilova
Pavel Shaposhnikov of zonal isolation or determine why isolation has not been achieved. In the latter
Gioconda Tovar
Houston, Texas case, a complete well history analysis offers operators valuable guidance for
improving future cementing operations.
Paul Sheperd
Denver, Colorado, USA

Oilfield Review 28, no. 1 (January 2016).

Copyright © 2016 Schlumberger.
For help in preparation of this article, thanks to Primary cementing operations rank among the prior to running casing. In the absence of a cali-
Eve Le Bastard, Grabels, France; and Chew Yeong Leong, more important events that occur during a well’s per measurement, the cement volume must be
Houston, Texas.
CemSTRESS, Invizion Evaluation, Isolation Scanner, Techlog
lifetime. The cement sheath plays a critical role estimated. When designing the cement job, engi-
and USI are marks of Schlumberger. in establishing and maintaining zonal isolation in neers may also be guided by other formation con-
the well, supporting the casing and preventing siderations, including pore pressures, fracture
1. Nelson EB: “Well Cementing Fundamentals,”
Oilfield Review 24, no. 2 (Summer 2012): 59–60.
external casing corrosion. gradients and the locations of potential lost cir-
2. Standoff is defined by the API as: Well cementing involves a myriad of geologic, culation zones.
w
standoffAPI = r – r
hole pipe
in which r hole is the hole diameter, chemical and mechanical parameters. The oper- As the casing is run in the hole, centralizers
r pipe is the casing diameter and w is the smallest annular ation may be divided into several principal activi- installed on the outside of the pipe establish a
gap. Guidelines concerning recommended standoff
values are also published by the API.
ties—drilling the wellbore, casing the wellbore, standoff between the casing and the wellbore
American Petroleum Institute: API Specification 10D: placing the cement slurry in the casing to well- that provides open flow paths in an annulus. The
Specification for Bow-String Casing Centralizers, bore annulus, allowing the cement to set and general guideline for centralization is to main-
Washington, DC: API (March 2002).
American Petroleum Institute: API Standard 65—Part 2:
then evaluating the quality of the resulting tain a casing standoff sufficient for effective mud
Isolating Potential Flow Zones During Well Construction, cement sheath.1 Secondary considerations may removal and cement sheath coverage.2 When the
Washington, DC: API (December 2010).
include remedial treatments to correct cement- casing is poorly centralized, annular constric-
3. Smith RC: “Successful Primary Cementing Can Be a
Reality,” Journal of Petroleum Technology 36, no. 11 ing problems and the long-term effects of produc- tions can trap drilling fluid between the casing
(November 1984): 1851–1858. tion on cement sheath integrity. and the wellbore, preventing complete cement
Crook RJ, Benge G, Faul R and Jones RR: “Eight Steps
Ensure Successful Cement Jobs,” Oil & Gas Journal 99,
Primary cementing requires the wellbore to coverage. This problem is exacerbated as a well’s
no. 27 (July 2, 2001). be in a condition that is conducive to successful deviation angle increases.
Bittleston S and Guillot D: “Mud Removal: Research cement slurry placement. For example, the bore- For many years, the industry has employed
Improves Traditional Cementing Guidelines,”
Oilfield Review 3, no. 2 (April 1991): 44–54. hole should be free of washed-out, out-of-gauge strategies to promote optimal cement placement
4. Free fluid is the volume of the particle-free liquid that zones. Caused either by soft or unconsolidated results. These strategies, collectively known in
separates from a stationary cement slurry. Free fluid is
measured by pouring a cement slurry into a graduated
formations or as a result of drilling practices, the industry as good cementing practices, dictate
cylinder and allowing the cylinder to remain undisturbed washouts create irregular and enlarged bore- that drilling fluid be conditioned prior to a
for 30 minutes. A technician measures the particle-free
liquid volume, which is expressed as a percentage of the
holes that are difficult to clean up and tend to cement job.3 Conditioning is a process of homog-
total slurry volume. Free fluid is minimized during cement hold gelled or dehydrated drilling fluid that can enization, cuttings removal, gelled mud dispersal
slurry design. For more on free fluid measurement:
American Petroleum Institute: API RP 10B–2:
contaminate the cement slurry. Because they and adjustment of rheological properties to facil-
Recommended Practice for Testing Well Cements, also create voids in the borehole wall that must itate cementing. Many cement placement simu-
Washington, DC: API (July 2005).
be filled with cement, washouts must be factored lators provide recommendations that address
5. For more on cement sheath evaluation: Allouche M,
Guillot D, Hayman AJ, Butsch RJ and Morris CW: into cement volume calculations. To determine drilling fluid conditioning. In the absence of
“Cement Job Evaluation,” in Nelson EB and Guillot D the location and volume of washouts, engineers cement placement simulators, cementing per-
(eds): Well Cementing–2nd ed. Houston: Schlumberger
(2006): 549–612. usually perform openhole caliper measurements sonnel commonly circulate at least one annular

10 Oilfield Review
volume of drilling fluid before pumping the
cement slurry. This precaution is carried out
after casing is run to remove entrained gas and
cuttings, break the fluid’s gel strength and lower
the fluid’s yield stress and plastic viscosity.
However, such broad measures may be question-
able in light of the highly diverse and increas-
ingly complex wells being drilled today.
Chemical washes and spacer fluids are
pumped ahead of the cement slurry to facilitate
removal of drilling fluid and prevent commin-
gling of the drilling fluid and the cement slurry.
The types and volumes of washes and spacers are
selected on the basis of the drilling fluid proper-
ties and hole geometry as well as the physical and
chemical environment downhole.
When feasible, casing movement—in the
form of reciprocation, rotation or both—is per-
formed while circulating drilling fluid and while
pumping preflushes and cement slurries. Casing
movement helps reduce drilling fluid viscosity
and dislodge gelled drilling fluid from annular
constrictions, thereby providing an improved
environment for cement placement.
The cement slurry is designed to perform
within the parameters of the anticipated well-
bore environment. Accurate temperature and
pressure data are essential inputs to the design
process. Cement parameters that must be opti-
mized include rheological properties, thickening
time, strength development, permeability, free
fluid and long-term durability.4
During job execution, care must be taken to
prepare the preflushes and cement slurries
according to the design and to pump these fluids
at the planned rates to ensure successful cement
placement. The wellhead pressure must also be
examined throughout the operation to verify that
the fluids are being placed properly and that the
well is under control. Modern cement mixing and
pumping equipment is fitted with sensors that
allow engineers to closely monitor and record
these parameters.
After the cement sheath has set and hard-
ened, logging helps engineers ascertain the qual-
ity of zonal isolation.5 Log interpretation can be
improved if logging personnel are informed of
previous drilling and cementing activities as well
as the physical properties of the cement.
Information about hole geometry, potential lost
circulation zones, mud type, spacer fluid and
cement properties, pressure testing and unusual
events that occurred during drilling and cement-
ing may help logging engineers properly calibrate
tools and ensure that sufficient logging data are
obtained for cement sheath evaluation. Building
on established cementing practices, a system for
January 2016
Drilling Cement Placement Cement Sheath Evaluation
Drilling fluid compositions Preflush, spacer and cement Cement bond log
slurry compositions
Drilling fluid rheological properties Variable density log
Preflush, spacer and cement slurry
Rate of penetration Acoustic impedance
rheological properties
Weight on bit Flexural attenuation
Circulating temperatures
Torque, shock and vibration data Third interface echo
Static temperatures
Lost circulation zone locations Solids presence map
Cement slurry thickening times
Washout locations
Cement strength development
Wellbore geometry
Centralizer program
Mechanical earth model
Casing movement during
(wellbore stability)
displacement
Lithology
Cement displacement simulation
Resistivity
Cement acoustic impedance
Casing plan simulation
Pressure and temperature
Figure 1. Well properties and measurements for the Invizion Evaluation service. Well integrity
engineers analyze many types of data to generate an Invizion Evaluation presentation that shows
drilling measurements, cement placement data and cement sheath evaluation logs.
rendering all relevant information into a compat- productivity. The engineers gather the data and
ible format allows a more efficient and thorough arrange the information in a workflow that
analysis of the principal parameters influencing allows offset well histories to be used during
primary cementing and zonal isolation. well planning.
Schlumberger engineers have developed a Within the PTECs, multidisciplinary teams of
technique by which logging personnel and opera- well integrity engineers (WIEs) collaboratively
tors may easily view and analyze previous well analyze many well parameters and arrange the
events, thereby improving log interpretation. information in a workflow that allows straightfor-
This article presents case histories from Alaska, ward visualization of the history and current sta-
Colorado and Texas, USA, and from the Gulf of tus of a well. At key junctures, the WIE teams are
Mexico. These cases illustrate how performing a also able to remotely monitor and analyze the
comprehensive examination of the entire well current status of a well. Because the datasets fre-
construction process can help engineers verify quently come from various sources in a variety of
primary cementing success or diagnose why formats, they must be normalized to allow a
cementing objectives have not been met. In the coherent examination. Engineers accomplish
latter case, the lessons learned can provide guid- this task by entering the data into the Techlog
ance for remediation and improving results in wellbore software platform.
future wells with similar parameters. The platform consists of a comprehensive set
of modules that accommodates the myriad types
Integrated Cement Evaluation of data acquired during a well’s lifetime. An inter-
In the context of well cementing, the concept of active graphical user interface allows engineers
following a cementing operation from the design to evaluate details throughout the entire well
stage through the execution stage and to the construction process.
evaluation stage has existed for some time. The WIEs use the software platform to collab-
Accordingly, Schlumberger has organized dedi- oratively examine formation geology and petro-
cated teams of drilling, geomechanical, logging physics, well geometry, drilling events, drilling
and cementing engineers that perform multidis- and cementing fluids, cement placement events
ciplinary analyses of virtually all parameters rel- and cement sheath evaluation logs (Figure 1).
evant to the lifetime of a well. A community of The information is arranged into a chronological
these engineers is located at Schlumberger account of well events that is used in the newly
PetroTechnical Engineering Centers (PTECs) developed Invizion Evaluation well integrity ser-
around the world. Their principal objective is to vice. Ancillary data such as laboratory test results
provide operators with information to safely and and modeling predictions may be displayed in
efficiently construct wells and maximize well separate windows on the presentation.
11
 The system’s capability to visualize and evalu- [1,200 m]—is 75°F [24°C]. A retarder was added would be discernable by most wireline logging
ate all of the available well data allows the WIEs to the cement slurry to prevent premature setting tools. The delay was expensive in terms of
to perform improved interpretations and deter- at the TD; however, the retarder also presented rig time.
mine why zonal isolation has or has not been risk of slurry overretardation at the TOC depth. To determine whether cementing and logging
achieved. Using the Invizion Evaluation service, Because of the narrow fracture pressure win- could proceed more efficiently while reducing
holistic assessments have been performed on dow, a low fluid density contrast was necessary the waiting-on-cement (WOC) time, the operator
more than 100 casing strings worldwide. to avoid well failure.6 The densities of the drilling elected to use the Invizion Evaluation service.
fluid, spacer fluid and cement slurry were 10.4, The engineers were provided with data gathered
Cementing Depleted Sands in Alaska 11, and 11.5 lbm/galUS [1,250, 1,320 and during drilling and cementing along with the
An operator in Alaska is producing from a reser- 1,380 kg/m3], respectively. State regulations compositions and rheological properties of the
voir characterized by depleted sands, potential require that, before continuing operations, engi- wellbore fluids.
lost circulation zones and a narrow fracture pres- neers must confirm the location of the top of The well deviation created a narrow side of
sure window. The wells have deviation angles up cement and verify the presence of competent the annulus, where cement contamination by
to 60° and the producing interval lies between cement around the circumference of the casing. mud would be more likely. Knowing the rheologi-
9,800 and 10,060 ft [2,990 and 3,070 m]. The bot- Because of the small density contrast between cal properties of the drilling fluid, spacer fluid
tomhole static temperature (BHST) is approxi- the wellbore fluids, the operator typically had to and cement slurry, the engineers performed
mately 150°F [66°C]; the temperature at the wait several days after the cement job for the set cement placement simulations to obtain guid-
anticipated top of cement (TOC)—about 4,000 ft cement to develop an acoustic signature that ance for preventing such cement slurry contami-

1 2 3 4 5 6 7 8 9 10 11 12 13 14
Wellbore Quality Cement Design and Placement Simulation Cement Evaluation

Well Schematic Gamma Ray Deviation Resistivity Porosity Cement Annular Casing Standoff Cement Simulated Mud Displacement Acoustic Impedance Acoustic Impedance Flexural Attenuation Flexural Attenuation Solids, Liquids and
Density Coverage Simulation Map Map Gas Map

Maximum Maximum

Cement Evaluation Zones

0 Mrayl 10 0 dB/m 200
Bulk Density
Between Centralizers Minimum Minimum
1.95 g/cm3 2.95 Mud
0 % 100 –1 Mrayl 9 0 Mrayl 10 50 dB/m 155 0 dB/m 200
Thermal Neutron Cement Annular Tail slurry Gas
Depth, Well Schematic Gamma Ray Deviation A16H Porosity Density At Centralizers CEM_PERC_CEM_COV Lead slurry Image Orientation, ° Average Image Orientation, ° Average Solid
ft 0 gAPI 150 0 degree 90 0.2 ohm.m 2,000 45 % –15 1 g/cm 3
2 0 % 100 0 % 100 Spacer 0 90 180 270 360 0 Mrayl 10 0 90 180 270 360 0 dB/m 200 Liquid

1,000

Mud
2,000

3,000

Spacer
4,000

5,000

6,000

7,000 Lead

8,000

9,000
Tail

10,000

Figure 2. Alaska well presentation. Tracks 1 through 5 display well information tail slurry below 9,150 ft (Track 9, dark gray). Tracks 10 through 14 show
and measurements obtained before cement placement. Tracks 6 through 9 cement evaluation results. Acoustic and ultrasonic logs that were run 27 hrs
present cementing execution information. Track 9 predicted that some after cement placement confirmed adequate cement coverage and zonal
mud contamination (red) could be expected in the lead cement slurry to isolation across both the lead and tail portions of the cement sheath below
a depth of 9,200 ft. However, no contamination was predicted across the 4,000 ft (Track 14, predominantly brown).

12 Oilfield Review
 1 2 3 4 1 2 3 4 1 2 3 4
Casing Standoff Cement Displacement Acoustic Impedance Acoustic Impedance Casing Standoff Cement Displacement Acoustic Impedance Acoustic Impedance Casing Standoff Cement Displacement Acoustic Impedance Acoustic Impedance
Simulation Map Simulation Map Simulation Map
Minimum Minimum Minimum
–1 Mrayl 9 –1 Mrayl 9 –1 Mrayl 9
At Centralizers Average At Centralizers Average At Centralizers Average
Mud Mud Mud
58 % 100 –1 Mrayl 9 –1 Mrayl 9 58 % 100 –1 Mrayl 9 –1 Mrayl 9 58 % 100 –1 Mrayl 9 –1 Mrayl 9
Lead slurry Lead slurry Lead slurry
Depth, Between Centralizers Tail slurry Image Orientation, ° Maximum Depth, Between Centralizers Tail slurry Image Orientation, ° Maximum Depth, Between Centralizers Tail slurry Image Orientation, ° Maximum
ft 0 % 100 Spacer 0 90 180 270 360 –1 Mrayl 9 ft 0 % 100 Spacer 0 90 180 270 360 –1 Mrayl 9 ft 0 % 100 Spacer 0 90 180 270 360 –1 Mrayl 9

500 500 500

1,000 1,000 1,000

1,500 1,500 1,500

2,000 2,000 2,000

2,500 2,500 2,500

3,000 3,000 3,000

3,500 3,500 3,500

4,000 4,000 4,000

4,500 4,500 4,500

5,000 5,000 5,000

5,500 5,500 5,500

6,000 6,000 6,000

6,500 6,500 6,500

Figure 3. Colorado well presentation. The first presentation (left) indicates displacement up to a depth of about 2,000 ft (Track 2, green). Although
numerous areas of poor casing standoff (Track 1, red). The cement cement did not reach the surface, no gas or pressure was observed.
displacement simulation predicted areas of poor mud displacement (Track 2, The third presentation (right ) shows the results of improved rheological
red). The acoustic impedance map also showed regions of poor cement properties of the cement slurry in addition to improved centralization. The
coverage (Track 3, blue) up to the surface—which explained the presence cement reached the surface and the acoustic impedance map shows
of sustained casing pressure at the surface after cementing. The second competent bonding with no contiguous channels (Track 3). Sustained casing
presentation (center) shows the results obtained by improving the casing pressure was not observed using this design.
centralization. The simulator indicated good mud removal and cement

nation and for estimating whether the cement that, despite the contamination, the cement slurry with the log information, illustrated the presence
would cover the entire interval of interest would set within 10 hours; however, the acoustic of good cement across all sections of the cement
(Figure 2).7 The results of the simulation indi- impedance differential was only 0.3 Mrayl—too sheath. The operator has continued to employ
cated a risk of incomplete mud removal and also small to be detected by conventional logging tools.8 this evaluation technique, lowering the WOC
provided an estimate of the volume of cement Three days of curing would be necessary for the time and reducing completion costs.
contaminated by mud. cement to develop a sufficiently high acoustic
In the laboratory, Schlumberger engineers pre- impedance contrast to permit evaluation by con- Solving Sustained Casing Pressure in Colorado
pared cement slurries whose compositions corre- ventional logging methods—confirming the opera- The Niobrara play in the Denver-Julesburg basin
sponded to the mud contamination estimate and tor’s previous experience. is located in a highly populated region along the
cured the samples in an ultrasonic cement ana- The WIEs recommended the use of the Front Range in Colorado, USA. This field pro-
lyzer (UCA) at the well’s BHST. The tests indicated Isolation Scanner cement evaluation service duces approximately 250,000 bbl/d [39,700 m3/d]
because of its ability to acquire more sensitive of oil from the Niobrara formation. One of the
6. A narrow fracture pressure window describes a
condition in which the mud weight required to drill an acoustic impedance measurements.9 The tool principal operators has more than 8,400 active
interval without fluid influx is close to that which would combines the classic pulse-echo techniques of wells in the region. Effective zonal isolation is of
cause the formation to fracture or fail.
7. For more on cement placement simulations: Piot B:
ultrasonic bond logging tools with a flexural particular importance because some wells are
“Primary Cement Job Design,” in Nelson EB and Guillot D imaging technique that provides more effective located immediately adjacent to residential
(eds): Well Cementing 2nd ed. Houston: Schlumberger
(2006): 435–458.
imaging of the annular fill, including reflection areas. Sustained casing pressure caused by gas
Abbas R, Cunningham E, Munk T, Bjelland B, echoes at the cement-formation interface—the migration through inadequate tail cement is a
Chukwueke V, Ferri A, Garrison G, Hollies D, Labat C and third interface echo (TIE). The TIE also allows cause for concern and has been confirmed by
Moussa O: “Solutions for Long-Term Zonal Isolation,”
Oilfield Review 14, no. 3 (Autumn 2002): 16–29. engineers to determine whether material behind cement bond log (CBL) and USI ultrasonic
8. The unit of acoustic impedance is the rayl, normally given the casing is solid, liquid or gas. Correlating data imager logs in certain cases. To mitigate
in Mrayl. One rayl = 1 kg/s/m2.
from the UCA to the capabilities of the gas migration, the operator elected to try the
9. For a description of the Isolation Scanner tool operation:
Bellabarba M, Bulte-Loyer H, Froelich B, Le Roy-Delage S, Isolation Scanner technology, the WIEs deter- Invizion Evaluation service. The operator pro-
van Kuijk R, Zeroug S, Guillot D, Moroni N, Pastor S and mined that logging could commence as early as vided the WIEs with pertinent well data, which
Zanchi A: “Ensuring Zonal Isolation Beyond the Life of the
Well,” Oilfield Review 20, no. 1 (Spring 2008): 18–31. 27 hours after cement placement. The cement were loaded into the Techlog platform to produce
execution data and laboratory data, combined a comprehensive presentation (Figure 3).

January 2016 13
 The analysis indicated that previous wells had ratio and cement compressive strength. The soft- to improve cement sheath flexibility. In addition,
poor casing centralization, resulting in poor mud ware evaluates cement sheath performance in the cement slurry contained an additive to cause
removal that allowed channels to form in the compression, tension or both; it can also identify slight cement sheath expansion after setting,
cement sheath and serve as conduits for annular both inner and outer microannuli and predict thereby improving the bond. The operator has
gas migration to the surface. Based on these find- their size and development over time.11 successfully applied the revised cementing strat-
ings, the operator increased the centralizer den- In wells that experienced gas leakage, egy in wells in the region.
sity by 50%, which led to better cement slurry CemSTRESS analysis showed that the Young’s
displacement and mud removal. Sustained casing modulus of the formation surrounding the tail Mitigating Gas Migration in South Texas
pressure was no longer observed. section of the cement sheath was too low to sup- An operator producing from the Eagle Ford Shale
The investigation also included using the port the mechanical properties of the cement in South Texas, encountered difficulties while
CemSTRESS cement sheath stress analysis soft- compositions that had been pumped. As a result, cementing surface casing.12 The 10 3/4-in. surface
ware to estimate the radial and tangential stresses the cement sheath simulation indicated failure casing was set inside 80 ft [24 m] of 16-in. con-
imposed on casing strings, cement sheaths and under traction, a possible explanation for the ductor pipe. The casing string extended vertically
formations during the life of a well.10 The input poor casing-cement bonding across that interval. from the surface through the conductor pipe and
parameters included Young’s modulus, Poisson’s Engineers altered the cement slurry composition inside a 131/2-in. open hole to 4,525 ft [1,379 m].

1 2 3 4 5 6 7 8 9 10 11 12 13
Wellbore Quality Cement Design and Placement Cement Evaluation

Lithology Porosity Resistivity Caliper 3D Borehole Well Geometry Mud Contamination Cement Simulation Cement Bond Log Variable Density Log Acoustic Impedance Flexural Attenuation Solids, Liquids and
Risk Displacement Map Map Gas Map
90-in. Array
0.2 ohm.m 2,000
60-in. Array
0.2 ohm.m 2,000
30-in. Array Washout

–490.00

51.00
2.06
2.56
3.05
3.54
4.03
4.52
5.02
5.51
0.2 ohm.m 2,000 Undergauge Gas
Standoff Between
Spontaneous Potential Neutron Porosity 20-in. Array Centralizers Solid
High Mud
–240 mV 60 60 % 0 0.2 ohm.m 2,000 20 in. 0 0 % 100 Mrayl 50 dB/m 155 Liquid
Medium Tail slurry
Depth, Zonation Gamma Ray Bulk Density 10-in. Array Standoff at Centralizers Low Lead slurry Image Orientation, ° Image Orientation, ° Image Orientation, °
ft 0 gAPI 150 60 % 0 0.2 ohm.m 2,000 0 in. 20 0 %. 100 None Spacer 0 mV 100 45 135 225 315 45 45 135 225 315 45 45 135 225 315 45
Zone 5
Zone 4

1,000
Zone 3

2,000
Zone 2

3,000

4,000
Zone 1

Figure 4. Overview of an Invizion Evaluation service for an Eagle Ford well. surface during displacement, losses occurred thereafter and the top-of-solids
Engineers divided the Invizion Evaluation presentation into five zones (left). depth was 240 ft (Track 13). Liquid channels existed between 240 and 550 ft.
Key observations included that, although cement slurry had reached the Washouts occurred between 500 and 1,200 ft (Track 4).

14 Oilfield Review
 1 2 3 4 5 6 7 8 9 10 11 12 13
Wellbore Quality Cement Design and Placement Cement Evaluation

Lithology Porosity Resistivity Caliper 3D Borehole Well Geometry Mud Contamination Cement Simulation Cement Bond Log Variable Density Log Acoustic Impedance Flexural Attenuation Solids, Liquids and
Risk Displacement Map Map Gas Map
90-in. Array
0.2 ohm.m 2,000
60-in. Array
0.2 ohm.m 2,000
30-in. Array Washout

–490.00

51.00
2.06
2.56
3.05
3.54
4.03
4.52
5.02
5.51
0.2 ohm.m 2,000 Undergauge Gas
Standoff Between
Spontaneous Potential Neutron Porosity 20-in. Array Centralizers Solid
High Mud
–240 mV 60 60 % 0 0.2 ohm.m 2,000 20 in. 0 0 % 100 Mrayl 50 dB/m 155 Liquid
Medium Tail slurry
Depth, Zonation Gamma Ray Bulk Density 10-in. Array Standoff at Centralizers Low Lead slurry Image Orientation, ° Image Orientation, ° Image Orientation, °
ft 0 gAPI 150 60 % 0 0.2 ohm.m 2,000 0 in. 20 0 %. 100 None Spacer 0 mV 100 45 135 225 315 45 45 135 225 315 45 45 135 225 315 45

600

700

800
Zone 4

900

1,000

1,100

1,200

Figure 5. Invizion Evaluation service interpretation of Zone 4 of an Eagle Ford than 70 gAPI. The resistivity data (Track 3) do not indicate mud invasion
well. Zone 4 extends from 560 to 1,250 ft. The interpretation results indicate except at approximately 650 to 750 ft. Caliper values (Track 4) averaged
well-bonded casing (Track 3, brown) and some apparent liquid patches 16.5 in. [42 cm], indicating washout sections. Casing standoff (Track 6)
(blue), which are also predicted by the cement placement simulation averaged about 60%. The interpretation results indicate bonded solids
(Track 8). The gamma ray curve (Track 1) shows an average value of less around the casing and some liquid patches (Track 13).

The operator faced three principal cementing casing joint to 1,000 ft [305 m]. Then, at lower despite about 10% mud contamination. In Zone 2,
risks: lost circulation, hole washouts and gas depths, where the well was still vertical, every between 4,060 and 2,006 ft [611 m], evidence of
migration from shallow flow zones between fourth casing joint was fitted with a centralizer. mud invasion and lost circulation was apparent;
250 and 600 ft [76 and 183 m] measured depth. To further minimize the risk of gas migration, nevertheless, the interpretation indicated good
An offset well on the same pad developed gas the operator elected to install an external casing bonding within the interval. A discrepancy
migration outside the casing and required reme- packer (ECP) within the casing-conductor between the predicted and measured location of
dial cement squeezes and a casing patch to fix pipe annulus. the interface between the tail slurry and lead
the problem. The casing patch reduced the inner The cementing operation successfully slurry provided evidence that the lead and tail
diameter of the casing to an extent that inter- stopped gas migration; however, after the slurries were mixed during displacement.
fered with further well development plans. The cement slurry reached the surface, lost circula- In Zone 3, between 2,006 and 1,250 ft
operator sought advice from Schlumberger to tion difficulties were encountered, and the TOC [611 and 381 m], mud pockets were observed.
propose measures to improve cementing results fell back into the well before the top plug Log interpretation indicated bonded solids
in future wells. reached the bottom of the casing string. The around the casing and the development of liquid
Updated cementing practices were employed WIEs performed the Invizion Evaluation service channels. Similar results were seen in Zone 4
to prevent gas migration along the surface cas- to investigate these problems as well as verify between 1,250 and 560 ft [381 and 171 m]
ing string. Both the lead and tail cement slurry annular zonal isolation. (Figure 5). Zone 5 extended from 560 ft to the
incorporated a gas migration control additive. The overall presentation indicated that the
10. Bellabarba et al, reference 9.
The tail slurry was also designed to set quickly to TOC was at 240 ft [73 m]. Washouts and liquid 11. A microannulus is a small gap that can form between
further minimize the likelihood of gas migration. channels in the cement slurry were also observed. the casing and the cement sheath or between the
cement sheath and the formation.
The centralizer program was designed to provide To more thoroughly examine the data, the WIEs
12. Surface casing usually consists of a large-diameter pipe
maximum standoff across the shallow gas divided the well into five analysis zones (Figure 4). string to protect freshwater aquifers and provide
zones—engineers installed one centralizer per In Zone 1, extending from TD to 4,060 ft structural strength for additional casing strings,
including the final production casing string.
[1,237 m], good cement bonding was present

January 2016 15
 1 2 3 4 5 6 7 8 9 10 11 12 13
Wellbore Quality Cement Design and Placement Cement Evaluation

Lithology Porosity Resistivity Caliper 3D Borehole Well Geometry Mud Contamination Cement Simulation Cement Bond Log Variable Density Log Acoustic Impedance Flexural Attenuation Solids, Liquids and
Risk Displacement Map Map Gas Map
90-in. Array
0.2 ohm.m 2,000
60-in. Array
0.2 ohm.m 2,000
30-in. Array Washout

–490.00

51.00
2.06
2.56
3.05
3.54
4.03
4.52
5.02
5.51
Crossover 0.2 ohm.m 2,000 Undergauge Gas
Standoff Between
Spontaneous Potential Neutron Porosity 20-in. Array Centralizers Solid
High Mud
–240 mV 60 60 % 0 0.2 ohm.m 2,000 20 in. 0 0 % 100 Mrayl 50 dB/m 155 Liquid
Medium Tail slurry
Depth, Zonation Gamma Ray Bulk Density 10-in. Array Standoff at Centralizers Low Lead slurry Image Orientation, ° Image Orientation, ° Image Orientation, °
ft 0 gAPI 150 60 % 0 0.2 ohm.m 2,000 0 in. 20 0 %. 100 None Spacer 0 mV 100 45 135 225 315 45 45 135 225 315 45 45 135 225 315 45

100

200
Zone 5

300

400

500

Figure 6. Invizion Evaluation service interpretation of an Eagle Ford well, average standoff. Dynamic cementing pressures remained below the
Zone 5. Zone 5 extends from the surface to 560 ft. The interpretation results formation fracture pressure. The cement evaluation log shows the cement
indicate the top of solids were at 240 ft (Tracks 11 through 13) and the top at 240 ft. In the processed solids, liquids and gas (SLG) map (Track 13),
presence of a postcementing liquid channel (Track 13, blue). The gamma liquid channels occur from the top of solids to 560 ft in the wide side of the
ray values vary from 33 to 70 gAPI between 250 and 275 ft (Track 1). Some annulus. The tortuosity of the liquid channels suggests that the channels
mud invasion occurred at approximately 500 ft, where the washout area were formed after the cement was placed. An external casing packer is
was located. The average neutron porosity is 45%, and the density porosity also evident at 75 ft. An offset well required remedial cement squeezes
is 30%. Most of the zone contained washout areas based on the caliper between 500 and 800 ft. No gas was observed in this region of the well.
log (Track 4). One centralizer was installed per joint, resulting in a 65%
1 2 3
Acoustic Impedance Map Bond Index Variable Density Log
Bond Index Level for Zonal Isolation surface, and liquid channels were observed in the
0.000
0.500
1.500
2.500
3.500
4.500
5.500
6.500
7.500

100 % 0 Minimum Amplitude Maximum cement sheath (Figure 6). The presence of the
Bond Index from Discriminated Attenuation
Depth, inflated ECP was evident at 75 ft [23 m]. The tor-
ft Mrayl 100 % 0
tuosity of the channels and their position on the
wide side of the annulus suggested fluid flow into
the annulus after cement slurry placement.
The overall analysis indicated that, despite
the operational problems, the cementing opera-
tion achieved adequate zonal isolation. A possi-
ble explanation for the losses and lowered TOC is
18,700
that during displacement, the pumping pressure
may have initiated ECP inflation, thereby
reducing the effective annular size inside the
conductor pipe. The resulting pumping pressure
increase at shallow depths may have caused for-
mation breakdown. Furthermore, the formation
of channels was a possible result of inadequate
19,100 casing centralization below 1,000 ft [305 m].
Therefore, the recommendation for future offset
wells is to eliminate the ECP from the well design
and to install a centralizer on each casing joint.
Figure 7. Deepwater Gulf of Mexico exploration well log. Engineers relied mainly on the bond index (BI) 13. Bellabarba et al, reference 9.
data (Track 2) to estimate the TOC. The BI falls significantly at 18,640 ft (red line), and engineers chose
that depth as the TOC. This finding is not as evident in the acoustic impedance map (Track 1) or the
variable density log (Track 3).

16 Oilfield Review
 1 2 3 4 5
Wellbore Quality
Gamma Ray Drilling Parameters Deep Resistivity Lithology Well Geometry
Inclination
0 degree 20
Average Rate of Penetration Azimuth
0 ft/hr 50 Inclusions 0 degree 360
Depth, Gamma Ray Mud Weight In Inclusions Dogleg Severity
ft 0 gAPI 50 0 ppg 50 0.2 ohm.m 20,000 Salt 0 degree/100 ft 5

16,000

17,000

18,000

19,000

Figure 8. Openhole LWD measurements from a Gulf of Mexico exploration
well. Spikes on the gamma ray log (Track 1) indicate three zones in which
the salt contains inclusions of other minerals. The contamination is further
confirmed by the resistivity (Track 3) and lithology (Track 4) data. The
resistivity increases across the contaminated zones, and the lithology
presentation shows inclusions of foreign material (blue and tan) within the
salt (magenta). The three zones were identified as potential lost circulation
sites. Drilling dynamics (Track 2) and well geometry data (Track 5) are
also presented.

Well Planning in the Deepwater Gulf of Mexico
An operator developing a deepwater prospect in
the Mississippi Canyon in the Gulf of Mexico
encountered difficulties while completing an
exploration well drilled in about 4,059 ft
[1,237 m] of water. Previously, the operator
drilled to a depth exceeding 20,000 ft [6,100 m]
and cemented a 14-in. casing string inside a
16-in. liner. The length of the overlap between
the casing string and the liner was 8,940 ft
[2,725 m], and the annular clearance within the
liner lap was 0.425 in. [1.08 cm]. The casing
string was designed to isolate a zone composed
entirely of salt. Fluid losses during drilling and
cementing indicated the presence of lost circula-
tion zones.
After the cementing operation, a logging run
was performed to locate the TOC (Figure 7).
Relying mainly on CBL attenuation data and a
computed bond index (BI), the logging engineers
estimated the TOC was located at 18,640 ft
[5,680 m]—3,960 ft [1,210 m] lower than
expected—which meant that the upper portion
of the salt zone was uncemented.
The operator, who plans to drill a similar off-
set well in the future, approached Schlumberger
to provide an Invizion Evaluation service on the
exploratory well and help formulate strategies to
prevent difficulties that might impede successful
zonal isolation in the offset well. The operator
provided archival data from the earlier well. This
information—including openhole logging-while-
drilling (LWD) data; compositions and physical
properties of the drilling fluid, spacer fluid and
cement slurries; and analyses of well fluids that
returned to the surface—was integrated into the
Invizion Evaluation presentation. Engineers also
discovered that, during the previous logging run,
significantly more data were actually acquired
but never examined. The additional data included
flexural attenuation (FA), TIE and a solids, liq-
uids and gas (SLG) map.13
The openhole data revealed the presence of
three intervals along the salt zone that were
contaminated by other minerals (Figure 8).
Engineers had previously assumed that the salt

January 2016 17
 Casing Standoff Displacement Simulation Mud Contamination Risk zone was continuous. The contaminated inter-
vals were identified as possible sites where lost
At Centralizers
circulation took place, and their presence
0 % 100 Mud High
Tail Slurry Medium helped explain why the TOC was below the
Depth, Between Centralizers
Lead Slurry Low intended depth.
ft 0 % 100 Spacer None Equipped with the mud, spacer and cement
slurry property data as well as pressure charts
acquired during the cementing operation, the
WIEs performed fluid placement simulations. The
first simulations confirmed that the pressure
charts were consistent with the occurrence of lost
circulation when the float valves were converted
on the casing prior to cement job execution.14 In
addition, engineers noted that the fluid displace-
ment pressure exceeded the fracture pressure in
16,000
the interval. The elevated pressure likely resulted
from the constricted annular clearance between
the liner and the casing string.
The second set of simulations examined the
capability of the spacer fluid and cement slurry to
displace the drilling fluid (Figure 9). Engineers
noted that centralizers had been installed only at
the bottom of the casing string. Above the cen-
tralizers, the casing standoff was poor, and the
17,000 simulations indicated poor mud removal and a
sizeable channel above about 18,700 ft
[5,700 m]—a depth consistent with what had
been concluded to be the top of cement in the
earlier interpretation.
Integration of the additional logging data
with the previous findings was revelatory
(Figure 10). The acoustic impedance map, cou-
pled with the FA and SLG maps, showed the TOC
to be much higher than previously thought—at
18,000
about 16,000 ft [4,880 m]. Unfortunately, this
depth was still below that required to cover the
entire salt interval. Furthermore, WIEs con-
firmed that the cement sheath quality was good
in the centralized portions of the annulus around
the casing shoe, and the cement slurry that rose
to 16,000 ft had a clear placement channel con-
sistent with computer simulations performed
after the job execution.
19,000 The new information resulting from the
Invizion Evaluation service allowed the operator
to adjust the completion plan for the future. The
operator’s engineers formulated a plan to
improve centralization of tubulars in regions
14. Float equipment, installed near the bottom of a casing
string, comprises valves that allow drilling fluid to enter
the casing interior as the string is lowered into the well.
As a result, the casing is filled with drilling fluid before
Figure 9. Cement placement simulations for an exploratory well in the Gulf of Mexico. Centralizers the cement slurry is pumped. Before the cement job
were installed only in the lower part of the casing string. Above about 19,300 ft [5,880 m], the casing commences, a ball or other activation device is pumped
standoff was poor (Track 1). The displacement simulations (Track 2) predicted that full cement into the casing. When the ball reaches the float
coverage (gray) rose to a depth of about 18,700 ft [5,700 m]. Above, the annulus contained drilling equipment, the valves are converted such that they
permit only one-way flow out of the casing and into the
fluid (brown). The mud contamination risk simulation concurred. Mud contamination (Track 3, red) annulus, thereby preventing reverse circulation, or
began as the casing standoff decreased. Full cement coverage (Track 3, green) was predicted in the U-tubing, of the cement slurry.
centralized region.

18 Oilfield Review
 1 2 3 4 5 6 7
Cement Bond Log Variable Density Log Acoustic Impedance Acoustic Impedance Map Flexural Attenuation Flexural Attenuation Map Solids, Liquids and Gas Map

Values Greater Impedance Values Greater

than Free Pipe than Baseline Gas
Depth, Attenuation
Solid
ft 0 dB/ft 5 0 Mrayl 10 0 Mrayl 8 20 dB/m 70 20 dB/m 100 Liquid

16,000

17,000

18,000

19,000

Figure 10. Complete log information retrieved from an exploratory well in (Track 7) reveals a sharp interface between a contaminated cement sheath
the Gulf of Mexico. Examination of the acoustic impedance map (Track 4), below about 16,000 ft and spacer fluid at shallower depths.
the flexural attenuation map (Track 5) and the solids, liquids and gas map

above the casing shoe. To reduce friction pres-
sure during pumping and minimize lost circula-
tion, the operator decided to eliminate the 14-in.
casing string and extend the 16-in. liner through
the salt zone. Care will also be taken to ensure
that fluid placement pressures remain below for-
mation fracture pressures, further reducing the
probability of lost circulation while allowing the
cement slurry to reach the intended depth.
Expanding the Scope of Holistic Cement
Sheath Evaluations
Engineers and wellsite personnel have only one
chance to achieve a successful primary cement
job for each casing string. Remedial cementing
to solve problems associated with a faulty
cement sheath has a less than stellar success
rate and may even reduce a well’s productivity.
Improved understanding of primary cement-
ing operations can be gained by examining well
histories while performing comprehensive inter-
pretations of log data. Indeed, the added insight
provided by the Invizion Evaluation service can
further enhance the value of integrating all avail-
able information and allowing operators to make
better informed decisions concerning drilling
and cementing practices.
To date, the Invizion Evaluation service has
been aimed at examining current cementing prac-
tices and the objective is to improve them and
troubleshoot less than optimal outcomes. The
longer-term value of the service will be enhanced
as engineers and operators become more proac-
tive during the well cementing process, making
real-time cementing decisions that make use of
all borehole measurements and data obtained
during drilling. For example, the information may
be entered into the Techlog platform as it is
acquired, facilitating close collaboration among
personnel involved in the drilling and cementing
process. Ultimately, engineers may apply the
Invizion Evaluation service at the planning stage,
allowing geologists and geophysicists to collabo-
rate with the drilling and cementing engineers to
further ensure primary cementing success. —EBN

January 2016 19
 Mud Removal—Clearing the Way
for Effective Cementing

Kevin Docherty Drilling mud removal is a key step for attaining successful cementing operations.
Sugar Land, Texas, USA
Although mud removal technology has featured the use of spacers for many years,
Slaheddine Kefi spacer design and composition have not kept up with the increasing complexity of
Ines Khalfallah
Salim Taoutaou drilling fluid chemistry. Recent industry efforts are now providing enhanced and
Clamart, France
optimized spacer solutions that clean the wellbore more efficiently, helping ensure
Matthew Offenbacher zonal isolation and well integrity from cementing operations.
Ray Ravitz
Houston, Texas, USA

Oilfield Review 28, no. 1 (January 2016). Primary cementing is a critically important or being contaminated by fluids in the annulus
Copyright © 2016 Schlumberger.
facet of well construction. To qualify as properly during cement placement.
CemPRIME, MUDPUSH and SMART 3D are marks of
Schlumberger. cemented, a well must have a continuous and Zonal isolation relies on effective mud
1. Kelessidis VC, Guillot DJ, Rafferty R, Borriello G and impermeable hydraulic seal within the annulus removal: the displacement of drilling fluids
Merlo A: “Field Data Demonstrate Improved Mud
Removal Techniques Lead to Successful Cement Jobs,”
that isolates each zone along the wellbore. To and accompanying deposits from the casing–
SPE Advanced Technology Series 4, no. 1 (January 1996): obtain this seal, cementing operations must borehole wall annulus. To achieve optimal mud
53–58.
prevent the cement from bypassing, mixing with removal, the displacement operation must ensure

20 Oilfield Review
that the drilling fluid is thinned and dispersed
and that debris such as pipe dope and scale is
lifted out of the well. In addition, drilling fluid
must be effectively displaced, mudcake thinned
and weakened and uncontaminated cement
allowed to reach a predetermined height in
the annulus.
During cementing operations, a cement slurry
is pumped into a well in which the annulus is
filled with drilling fluid. Typically, the drilling
fluid and cement slurry are incompatible and
may form a highly viscous and unpumpable mix-
ture. To prevent mixing of the two fluids, a fluid
compatible with both the cement slurry and the
drilling fluid—a spacer—is pumped between the
two. The spacer also clears drilling fluid from the
casing and formation walls.
Displacement
Today, operators pump a spacer that has specific
fluid properties behind the drilling fluid and ahead
of the cement slurry. To improve cleaning and dis-
placement efficiency, operators often add surfac-
tant and solvent combinations to the spacer fluid
to lower surface tension and to dissolve and dis-
perse sludge and pipe dope.
Typically about 5% to 10% of the spacer fluid
volume contaminates a portion of the cement
slurry. Because spacer chemistries may affect
January 2016
cement properties, this contamination may alter Laminar flow Turbulent flow
cement thickening, setting time, rheology, com-
pressive strength and compromise zonal isolation.
Ensuring the proper rheological and chemical
properties of displacement and drilling fluids is
essential to fluid design. The cement slurry,
spacer and drilling fluid need to have proper Static mud
layer
rheological properties, or rheological hierarchy,
such that each fluid pumped is more viscous than
the fluid pumped before it. For wells drilled with Flowing
oil-base drilling fluid, the operator is often con- fluid
cerned about leaving the casing and formation in
an oil-wet condition, which weakens cement
bonding. To address this problem, cementers add
surfactants to the spacer to change the wettabil-
ity of the casing and the formation near the well- Borehole
bore from oil-wet to water-wet. wall
The flow regime of the fluids is also impor-
tant (Figure 1). In laminar flow, viscous friction
forces dominate; the maximum velocity is at the
Figure 1. Mud removal flow dynamics. In laminar
center of the borehole, and the velocity reduces flow (left ), flow lines are parallel, and individual
gradually to zero at the wellbore wall. In turbu- particles move in parallel paths. Mud particles
lent flow, the energy of fluid motion dominates; tend to accumulate in static mud layers near the
borehole wall, making complete mud removal
the particles move in an erratic circular motion,
difficult. In turbulent flow (right), the energetic,
and the velocity of the fluids along the walls is swirling eddies entrain more mud particles than
nearly the same as that at the center of the do laminar flow paths before becoming saturated.
borehole. Cementers usually prefer turbulent The turbulent eddies also move surfactants or
dispersants in the spacer fluid throughout the
flow for drilling fluid removal.1 However, in nar-
borehole to deform and remove the static mud
row hole sizes, laminar flow may be preferred layer at the borehole wall.
21
 By using a displacement fluid that is of higher
Casing Cement in density than that of the drilling fluid, engineers
annular space are able to ensure a flat and stable interface
between the two fluids and to avoid fingering—a
condition whereby the interface of two fluids
bypasses sections of the reservoir as it moves
along, creating an uneven, or fingered, profile.2
Because improper friction pressure may also
cause fingering, operators typically ensure that
friction forces created by displacing fluids are
higher than those created by the displaced fluids,
which also helps keep a flat and stable interface
Drilling mud
between the fluids.
Mud traces or residues in the annular space
may cause improper cement setting and the for-
mation of mud channels, leaving conductive
paths in the cement sheath. These can become
fluid migration pathways between zones and
cause hydrocarbons to flow to a low-pressure
Drilling mud zone via the annulus.3 Fluid migration may also
channel in the
cause an excessive pressure difference across the
annular space
annulus, which can lead to casing deformation
Figure 2. Plan view of off-centered casing. When casing is positioned
and ultimately to a loss of well integrity.
off-center in the wellbore, cement flows to the larger pathways of least Fluid migration pathways can be created
resistance, leading to uneven distribution of the cement sheath. This when the casing is off-center in the borehole
distribution can lead to the formation of channels in the annulus between (Figure 2). To avoid this type of mud channeling,
the casing and the wellbore wall; such channels can create opportunities
for flow and containment of nondisplaced mud. centralizers are used to keep the casing central-
ized in the wellbore, which enables even distribu-
tion of cement in the annular space (Figure 3).
Final choice of spacer type designs and solu-
because turbulent flow may increase flow rates require high surface pump pressures, which tions for oil-base and water-base drilling fluid
and friction pressures, which can induce fluid could result in downhole pressures that exceed applications vary widely and are subject to opera-
losses into the formation and well instability; formation fracture pressure. tor preference and service company recommen-
maintaining flow against high friction forces may dations. Although operators may have unique
preferences, most spacer options are specific to
an operational area and considered standard. To
ensure successful mud removal, the spacer must
be optimized for the specific well conditions and
drilling fluid.4 By changing spacer composition,
Figure 4. Displacement of wellbore fluids viscosity and density, or by adding various surfac-
NAF through fluid zones during cementing tants and solvents to the spacers, cementers can
operations. Three fluid zones exist
between the nonaqueous drilling fluid significantly improve mud removal.
NAF film

Zone 1 NAF-spacer (NAF) and the cement slurry: A mixture

of NAF and spacer occurs in Zone 1, Cleaning the Way
where the spacer ensures inversion of Over the past few years, to meet the needs of
the NAF emulsion; Zone 2 consists of
Filtercake

Spacer,
Zone 2 the spacer that has a surfactant, which increasingly complex wells and to ensure long-
including surfactant
ensures that the walls are water-wet; term well integrity, researchers have made
and a mixture of cement and spacer considerable improvements in drilling fluid
makes up Zone 3. All three zones are
Zone 3 Cement-spacer chemistries. Although these fluids can boost
important in evaluating and selecting
optimal spacer composition. drilling performance, because of their increased
complexity, they may be more difficult to remove
Cement slurry
from the well prior to cementing. Developments
in spacer technology have not kept up with those
Annulus of oil-base drilling fluids; the industry is still using
generic surfactants and solvents in viscosified
spacer fluids.

22 Oilfield Review
 Flow out Flow in

Spacer design, selection and testing are per- Flow out Flow in
formed on a case-by-case basis because oil-base
drilling fluid composition and downhole condi-
tions vary significantly. Although standard API
test regimes and other recommended practices
are available for spacer evaluation and testing, Spacer
some exhibit low repeatability and others apply
Spacer
only to fluids at temperatures below 88ºC [190ºF].
Hinged collar
To enhance reliability and repeatability of the
engineered spacer evaluation, Schlumberger Hinged collar
Cement
researchers used improved laboratory proce-
Cement
dures and equipment (Figure 4). For example,
cleaning tests can now be performed under pres- Bowspring
surized conditions and at temperatures exceed- profile
Bowspring
ing 88ºC. In addition, engineers select profile
optimal
spacer chemistry as a function of the base oil, Centralize
salinity and temperature conditions in each
Centralizer
well by applying response surface methodology
(Figure 5).5
A mature development field in Asia, operated
by Mubadala Petroleum, comprises six platforms
and a floating production, storage and offloading
vessel. The field is located in the Gulf of Thailand
and produces oil from layered sandstone reser-
voirs in which bottomhole temperatures can
reach 105ºC [221ºF]. The producing section is
drilled using an 80/20 oil/water ratio low-toxicity
oil-base drilling fluid and cemented with Figure 3. Welded bowspring centralizer. A centralizer (left) is made up of a hinged collar and
extended Class G cement.6 bowsprings. The bowsprings press against the wellbore wall (right) to keep the casing positioned
The operator has drilled more than 100 wells toward the center of the wellbore, and the hinged collars secure the centralizer on the casing or liner.
Centralizers are placed on the casing to keep the casing positioned in the center of the wellbore.
in this field using a conventionally weighted When cement is pumped after the spacer, the centralized casing will ensure that the cement will be
spacer that has additional surfactant for mud evenly distributed in the annulus to yield a perfect seal.
removal. Recent quality assessments of primary
2. Fraser L, Stanger B, Griffin T, Jabri M, Sones G, Parameters used for
Steelman M and Valkó P: “Seamless Fluids Programs: selection of surfactant
A Key to Better Well Construction,” Oilfield Review 8, and solvent package
no. 2 (Summer 1996): 42–56.
3. Tardy PMJ and Bittleston SH: “A Model for Annular
Displacements of Wellbore Completion Fluids Involving
Casing Movement,” Journal of Petroleum Science and
Engineering 126 (February 2015): 105–123.
4. Kefi S, Pershikova E, Docherty K, Barral Q, Droger N,
Bottomhole temperature

Regnault de la Mothe L and Khalfallah I: “Successful

Cementing Based on New Design Methodology for
Displacement of Non-Aqueous Drilling Fluid,” paper
IADC/SPE 167948, presented at the IADC/SPE Drilling
Conference and Exhibition, Fort Worth, Texas, USA,
March 4–6, 2014.
5. Response surface methodology is a statistical
experimental design process. This methodology explores
relationships between several explanatory variables and
one or more response variables, and its main purpose is
to use a sequence of designed experiments to obtain an
optimal response. Response surface methodology is
often used to refine models after important factors have
been determined using factorial designs.
6. Neat cement slurries are prepared from API Class A, C, G
or H cements and typically have densities in excess of NAF base oil type Salinity of the spacer
1800 kg/m3 [15 lbm/galUS]. Any material that has a density
lower than cement can be used as an extender and act to
reduce cement slurry density. Figure 5. Engineered spacer design. Based on tests leading to the development of the
CemPRIME spacer technology, engineers created a surfactant selection diagram.
The diagram gives a recommendation on the surfactant blend based on spacer
salinity, bottomhole temperature and a nonaqueous drilling fluid (NAF) base oil type.
Two surfactants and three solvents make up the five formulas that are efficient in the
majority of cases.

January 2016 23
cement jobs in production sections have indi-
cated the operator has achieved suboptimal
zonal isolation. Engineers suspect that residues
from oil-base drilling fluids have hindered proper
spacer placement and adversely affected the set
properties of the cement.
Even though it had not experienced problems
with nonproductive time or service quality issues,
because zonal isolation is a critical factor for suc-
cess in the planned wells of a nearby field, the
operator sought to assess and improve spacer
performance. Engineers sent samples of the oil-
base drilling fluid to a regional laboratory to com-
pare the efficiency of the CemPRIME engineered
chemistry, which was recently developed by engi-
neers at Schlumberger, with that of the conven-
tional spacer that the operator had used.
Laboratory tests indicated that the performance
of the engineered spacer, when compared with
that of the conventional spacer, provided a two-
fold increase in cleaning efficiency.
During tests, the cement was contaminated
with 10% spacer. Although contamination from
the previously used formulation modified the
cement’s compressive strength by 40% over a
24-hour period, contamination by the engi-
neered spacer had essentially no effect on com-
pressive strength.
On the basis of these test results and other
laboratory tests and evaluations, the operator
opted to test the engineered spacer while
cementing a production zone. Upon completion
of primary cementing, engineers confirmed zonal
isolation through cement bond logs and ultra-
sonic imaging. They also determined that the top
of cement was on target and that implementing
the engineered spacer did not affect job execu-
tion. Following this success, the operator is
implementing the engineered spacer for planned
wells in the same field.7
In some wells that have technically demand-
ing objectives, conventional spacer solutions may
not be the best choice despite the appeal of lower
cost. In 2014, Schlumberger scientists rolled out
a new process for designing and executing the
spacer MUDPUSH Express for cementing. While
its predecessor, MUDPUSH II for cementing, is a
standard industry batch-mixed spacer, this new
spacer features on-the-fly mixing—an automatic
volumetric mixing system. The spacer can be
mixed using traditional dry blend additive, and
the new solution allows continuous mixing of the
mud removal system using a standard cementing
24
unit equipped with a circulating mixer and an
averaging tub. This process provides efficiency
gains during the mud removal process by simpli-
fying mixing and eliminating the need for addi-
tional resources.
A rapid-hydrating polymer that provides
required spacer viscosity facilitates immediate
mixing; the system starts out as a laboratory-
confirmed dry blend containing barite, a rapid-
hydrating polymer and an antifoaming agent.
When combined with water in the recirculating
mixer, the polymer is hydrated and ready for
pumping within one minute. Cement engineers
confirm spacer quality at the wellsite using a fun-
nel viscometer and pressurized mud balance. If
changes in well conditions occur before the
spacer is pumped, the spacer blend can be modi-
fied prior to mixing.
A 3D Point of View
Many operators choose mud displacement solu-
tions based on prior experience or accepted prac-
tices rather than an approach in which the
choice is part of an integrated, engineered opera-
tion. As a consequence, displacements have been
compromised when the various parties focus
exclusively on their respective segments during
the planning phase.
This lack of integration can cause engineer-
ing teams to neglect critical transition elements,
leaving the well poorly prepared for completion
or production operations. Improper wellbore
preparation is a major cause of reservoir damage
and restricted production. Failure to adequately
design and execute drilling fluid displacement
operations may also require additional rig time to
repair damage and recover production. Operators
estimate that 30% of failed completion opera-
tions are caused by inadequate removal of debris
or mud or a combination of these.
A properly designed spacer will facilitate
effective displacement of the drilling fluid,
controlled fluid losses and thinning and weak-
ening of the mudcake and will leave the casing
and the formation near-wellbore water-wet for
better mudcake dispersion and easier liftoff
during production. If a spacer fails to meet any
of these objectives, a well may experience a
cementing failure.
In extreme cases, the selection and use of a
poor spacer solution can have catastrophic con-
sequences. Spacer selection was a component
of the report on the April 2010 Macondo well
blowout. The final incident report states that the
Macondo blowout was a result of a series of risk-
increasing decisions and actions that failed to
mitigate those risks. The report also referenced
the choice of lost circulation material as a spacer
and the possible introduction of risk through clog-
ging of lines that are used for certain well integ-
rity tests. Clogged lines can limit the rig crew’s
ability to conduct an accurate negative pressure
test of the production casing cement job.8
Integrated Solutions
After analyzing more than 40 well operations and
combining findings with the latest generation
chemical, mechanical and hydraulic technolo-
gies, specialists from M-I SWACO have developed
the SMART 3D displacement strategy. This is a
customized engineered displacement solution;
the main focus of this strategy is to mitigate risk
and operational problems that result from debris
and incompatible drilling or completion fluids
remaining in the wellbore (Figure 6).
In 2014, an operator from Romania applied
this SMART 3D strategy for a land well to
increase cleaning efficiency and reduce rig time
related to mud removal. The casing cleanup and
displacement bottomhole assembly (BHA) were
modeled using specialized software and included
casing cleaning tools. Various laboratory tests,
performed in Romania using mud samples from
the field, were conducted to assess cleaning effi-
ciency, and the results helped engineers select
and design suitable chemical displacement pills.
Hydraulics parameters were identified using
a hydraulics software program that provides
estimated pressures and equivalent circulating
densities (ECDs) for each stage of the comple-
tion operation. In addition, software simula-
tions showed engineers the flow regimes for
each fluid, considering flow rates, rheological
properties and hole size. The ability to identify
individual flow regimes is important because
each pill has unique physical characteristics,
and although some pills need to be in laminar
flow, others need to be in turbulent flow to maxi-
mize cleaning performance.
7. Taoutaou S, Goh SH, Vargas Bermea JA, Vinaipanit M
and McClure J: “Achieving Zonal Isolation by Using
New-Generation Mud Removal Chemistry and Design
Methodology to Displace Non-Aqueous Drilling Fluid,”
paper SPE 176061, presented at the SPE/IATMI Asia
Pacific Oil and Gas Conference and Exhibition, Bali,
Indonesia, October 20–22, 2015.
Kefi et al, reference 4.
8. The Bureau of Ocean Energy Management, Regulation
and Enforcement (BOEMRE): “Report Regarding the
Causes of the April 20, 2010 Macondo Well Blowout,”
Washington, DC: US Department of the Interior, BOEMRE,
September 14, 2011.
Oilfield Review
 After evaluating results, the Romanian opera-
tor recognized the benefits of such an approach
and chose to apply the integrated SMART 3D strat-
egy to future displacements in land operations.
Operators in the Gulf of Mexico, Angola, Central
West Africa, continental Europe and in California,
USA, have experienced similar success with the
strategy, achieving enhanced and quicker dis-
placements as well as increased production com-
pared with results in operations in which
traditional spacer solutions have been used.

Planning Ahead
Advances in drilling fluids and cementing tech-
nologies have been driven in large part by increas-
ingly complex wells and operations and must be
paced by advances in mud removal. The industry
is investing resources to meet those challenges by
developing new solutions that take an integrated
approach to the entire wellbore, formation, drill-
ing fluid, cement and hardware systems to ensure
that each stage is optimally engineered and
planned. Similarly, operators and service compa-
nies are pursuing greater integration in engineer-
ing and planning to enable proficient, quality
operations. The goal is to achieve an efficient dis-
placement operation that provides long-term well
integrity enabled by a clean wellbore. —IMF

Figure 6. Integrated displacement technology. The SMART 3D technology

features new-generation tools, chemicals and hydraulic simulations to
efficiently prepare a well for completions or production. Mechanical tools,
such as scrapers (left), brushes (center) and magnets (right), help ensure
a properly engineered mechanical cleanup and represent one of the three
interdependent components of the SMART 3D strategy. The tool setup, spacer
design and hydraulic system can be tailored for each particular well.

January 2016 25
 Vaca Muerta Shale—Taming a Giant

Unconventional resource plays revolutionized oil and gas production in North

America. Because these types of plays provide many challenges to operators, they
have been slow to develop in other parts of the world. However, the Vaca Muerta
formation in the Neuquén basin of Argentina can now be added to the list of
successful producers. The triumph in the Vaca Muerta play resulted from engineers
at YPF, SA and Schlumberger collaborating on an integrated approach that includes
field- and formation-specific workflows and the application of a dynamic unconven-
tional fracture model to optimize hydraulic stimulation programs.

Matías Fernández Badessich Coaxing commercial quantities of hydrocarbons Technically Recoverable Shale Gas
Damián E. Hryb from unconventional reservoirs revolutionized
Others US
Mariano Suarez the oil and gas industry in North America. The 15%
20%
YPF, SA lessons learned there are now being applied in
Buenos Aires, Argentina other parts of the world. One of the latest success Brazil
3% China
stories comes from the giant Neuquén basin of Russia 14%
Laurent Mosse Argentina, where YPF, SA is developing the Vaca 4%
Nuncio Palermo South Africa
Muerta Shale formation for production of oil and 5%
Stéphane Pichon Argentina
natural gas. Australia
Laurence Reynolds 6% 10%
Buenos Aires, Argentina Production from shale reservoirs traces its
roots to activities in the Barnett Shale of north- Mexico Canada Algeria
Oilfield Review 28, no. 1 (January 2016).
central Texas, USA. In 1981, Mitchell Energy & 7% 7% 9%
Copyright © 2016 Schlumberger. Development Corporation drilled a well for the
Mangrove, Petrel and UFM are marks of Schlumberger. express purpose of producing natural gas from
1. US Energy Information Administration (EIA): “Technically
the Barnett Shale; the company then spent Technically Recoverable Shale Oil
Recoverable Shale Oil and Shale Gas Resources: An
Assessment of 137 Shale Formations in 41 Countries 20 years figuring out how to make the play com- Russia US
Outside the United States,” Washington, DC: 22% 14%
US Department of Energy EIA, June 2013.
mercially viable. Success in the Barnett was fol-
lowed by the deployment of new techniques and
China
technologies to find and produce natural gas and 10%
then oil from other basins and formations across
North America. Recent success in the Neuquén Argentina
8%
basin of South American has the potential to Others
19%
revolutionize the oil and gas industry in that Libya
Canada
region in a manner similar to that experienced in Pakistan Australia 8%
3%
North America. 3% Mexico Venezuela 5%
Argentina’s unconventional resource plays have 4% 4%

been recognized by exploration companies for
some time. The Vaca Muerta formation is consid-
ered by many experts to have some of the greatest
potential for hydrocarbon production in the world
(Figure 1). One independent study compared shale
resource potential outside North America and
ranked the Vaca Muerta formation third globally in
technically recoverable gas and fourth in oil.1
Figure 1. Global shale resources. Argentina has
abundant reserves in unconventional reservoirs.
Globally, the country is third in technically
recoverable shale gas reserves (top) and fourth
in technically recoverable shale oil (bottom). The
majority of Argentina’s unconventional reserves
are associated with formations in the Neuquén
basin. (Adapted from the US EIA, reference 1.)

26 Oilfield Review
January 2016 27
 Several operators are evaluating the Vaca
Muerta play; YPF, however, has been the most
active and successful in initiating production to
PERU
BOLIVIA
PARAGUAY
Chaco basin
URUGUAY
CHILE ARGENTINA
Neuquén
basin
San Jorge
basin
Austral-
Magallanes
basin
Figure 2. Neuquén basin. Several basins in South America have potential as unconventional resource
plays. The Neuquén basin, in west-central Argentina (red box), is the major source of oil and gas
production for the country and contains three deepwater marine shale deposits—the Agrio, Vaca
Muerta and Los Molles formations. These shales act as source rocks for oil and gas deposits found in
conventional reservoirs in the basin.
2. Hogg SL: “Geology and Hydrocarbon Potential of the
Neuquén Basin,” Journal of Petroleum Geology 16, no. 4
(October 1993): 383–396.
3. Instituto Argentino del Petróleo y del Gas, http://www.
iapg.org.ar/suplemento/Agosto2015/Produccion%20
por%20cuenca.html (accessed October 30, 2015.)
4. Back-arc basins are geologic features associated with
island arcs and subduction zones. Found at convergent
plate boundaries, they form when one tectonic plate
subducts beneath another, creating a trench. Subduction
zones are frequently associated with formation of
volcanoes and seismically active areas.
5. Hogg, reference 2.
6. Garcia MN, Sorenson F, Bonapace JC, Motta F, Bajuk C
and Stockman H: “Vaca Muerta Shale Reservoir
Characterization and Description: The Starting Point
for Development of a Shale Play with Very Good
Possibilities for a Successful Project,” paper
SPE 168666/URTeC 1508336, presented at the
Unconventional Resources Technology Conference,
Denver, August 12–14, 2013.
7. Howell JA, Schwarz E, Spalletti LA and Veiga GD: “The
Neuquén Basin: An Overview,” in Veiga GD, Spalletti LA,
Howell JA and Schwarz E (eds): The Neuquén Basin,
Argentina: A Case Study in Sequence Stratigraphy and
28
date. The company established early on that the
play has unique characteristics that affect devel-
opment programs. The fact that the Vaca Muerta
BRAZIL
Paraná basin
SOUTH AMERICA
Prospective basin
0 500 1,000 km
0 300 600 mi
Basin Dynamics. London: The Geological Society,
Special Publication 252 (2005): 1–14.
  8. Badessich MF and Berrios V: “Integrated Dynamic Flow
Analysis To Characterize an Unconventional Reservoir in
Argentina: The Loma La Lata Case,” paper SPE 156163,
presented at the SPE Annual Technical Conference and
Exhibition, San Antonio, Texas, USA, October 8–10, 2012.
  9. For more on geochemical characterization of source
rocks: McCarthy K, Rojas K, Niemann M, Palmowski D,
Peters K and Stankiewicz A: “Basic Petroleum
Geochemistry for Source Rock Evaluation,”
Oilfield Review 23, no. 2 (Summer 2011): 32–43.
10. Stinco L and Barredo S: “Vaca Muerta Formation:
An Example of Shale Heterogeneities Controlling
Hydrocarbon’s Accumulations,” paper URTeC 1922563,
presented at the Unconventional Resources Technology
Conference, Denver, August 25–27, 2014.
Badessich and Berrios, reference 8.
11. Ejofodomi EA, Cavazzoli G, Estrada JD and Peano J:
“Investigating the Critical Geological and Completion
Parameters that Impact Production Performance,”
paper SPE 168709/URTeC 1576608, presented at the
Unconventional Resources Technology Conference,
Denver, August 12–14, 2013. .
formation has reservoir-specific characteristics
that differ from other shale plays comes as no
surprise; operators of shale resource plays gener-
ally agree that, to maximize production, each
play must be approached and evaluated based on
its own intrinsic merits. The Vaca Muerta Shale
possesses features that enhance its potential to
deliver financially viable volumes of hydrocarbons.
To maximize production and reduce the
development cycle of the play, engineers at YPF
are now using fracture simulators, predictive
numerical models and field-specific workflows
created from information derived during an
extensive evaluation program. The process is
helping them understand the effects of rock
mechanics on fracture propagation during
hydraulic stimulations. The information gained
from their workflows and simulations is then
used to optimize well spacing, fracture stimula-
tion designs and fracture treatments. The meth-
odology has helped the engineering team reduce
the development time of the play compared to
the time taken by iterative approaches typically
used in other shale developments.
The methodology YPF has taken in developing
the Vaca Muerta formation is driven in part
because the formation has a combination of fac-
tors that differentiates it from other plays: a high
pore pressure, a complex stacking pattern of vary-
ing lithologies and geomechanical complexity
arising from the close proximity of the Neuquén
basin to the Andes Mountains. The engineers at
YPF have also found fracture gradients in some
parts of the basin that are abnormally high, which
greatly impacts completion strategies. A thorough
understanding of the Vaca Muerta formation’s
geomechanics and reservoir heterogeneity is cru-
cial for proper field development. With these
many factors in mind, YPF has started to move
from a traditional, and often costly, trial-and-error
methodology of developing resource plays to a
model-and-optimize methodology.
This article provides a basic overview of the
Vaca Muerta formation and YPF’s work in the
play to date. The article also reviews workflows
devised by the YPF and Schlumberger teams and
the simulation programs that are used to opti-
mize both drilling operations and stimulation
treatments and briefly discusses current produc-
tion and future development plans.
In the Shadow of the Andes
The Neuquén basin lies on the east flank of the
Andes Mountains in west-central Argentina
(Figure 2). Oil was first discovered there in
1918.2 The basin is a major source of oil and gas
Oilfield Review
production from conventional reservoirs—57% of Period Epoch Age Lithostratigraphy
the natural gas and 40% of the total oil produc- Pircala Roca Jaguel formation

Malargüe group
formation formation
tion within Argentina come from it.3 The target Maastrichtian

areas for shale development in the basin are Late

Loncoche formation
Triassic to Early Cenozoic strata, deposited in a Campanian
back-arc tectonic setting.4 Active volcanoes

Late
Rio Colorado formation
located along the present-day Andean Cordillera

Neuquén group
Santonian
Coniacian
produced pyroclastic emissions during various Turonian
Rio Neuquén formation
geologic periods, and the resulting ash beds cover
Cenomanian Rio Limay formation
large areas of the basin. Ash beds are observed
within the shale formations that are targets of

Cretaceous
current exploration. Albian
Rayoso formation

Rayoso group
The basin formed as a result of sea level Lohan Cura
formation
changes and tectonic plate movement that began Aptian Huitrín
in the Early Triassic period, which were followed formation La Amarga
formation
by a succession of transgressive and regressive Barremian

Early
Centenario

formation
Upper member
cycles. Restricted access to the open sea during formation

Agrio
Hauterivian
several periods led to the formation of thick evap- Lower member
orite beds.5 Reactivation of the volcanic arc dur- Bajada Colorada
Valanginian Mulichinco formation
ing the late Jurassic also affected the current formation
Quintuco formation

Mendoza group
basin topography.6 Present-day basin architec- Berriasian
ture resulted from uplift and folding during the Vaca Muerta formation
Picún Leufú
formation
last Tertiary Andean movement. Tithonian
The source rocks for oil and gas found in con-
Tordillo formation Quebrada del
Late

Kimmeridgian
ventional reservoirs of the Neuquén basin are Sapo formation

the Cretaceous Agrio formation, Early Cretaceous Oxfordian

Auquilco formation

Lotena
La Manga formation

group
to Late Jurassic Vaca Muerta formation and the Lotena formation
more mature Middle Jurassic Los Molles forma- Callovian
tion (Figure 3).7 All three formations are deepwa- Tábanos formation Challacó formation

ter marine shales and have potential as resource Bathonian Lajas formation
Middle
Jurassic

plays. The Vaca Muerta formation is the primary

Cuyo group

source for hydrocarbons in conventional reser- Bajocian

voirs and is in the oil window across approxi- Aalenian

Los Molles formation
mately 60% of the basin.
The Vaca Muerta formation was deposited Toarcian

between the Tithonian age—a Late Jurassic Pliensbachian

Chachil formation
Early

transgression—and the Berriasian stage of the

Early Cretaceous epoch.8 The characteristics of
this organic-rich formation make it a prime tar-
get for exploration.9 These include high average
total organic carbon (TOC) levels (1% to 8% and
spikes to 12%), moderate depths of 3,150 m
[10,335 ft] and overpressured conditions that
have a pressure gradient range from 13.6 to
20.4 kPa/m [0.6 to 0.9 psi/ft].10
The formation varies from about 60 m [200 ft]
to 520 m [1,700 ft] in thickness from the embay-
ment area to the basin center. Matrix porosity
varies from 4% to 14%; the average is 9%. Matrix
permeability ranges from nanodarcies to micro-
darcies; the presence of natural fractures con-
tributes to the productive potential of the
formation. Shale maturities based on vitrinite
reflectance, Ro, range from less than 0.5% up to
3%, which represents black oil to dry gas.11 Its
Sinemurian Lapa formation
Pre-Cuyo group
Hettangian
Triassic
Palaeozoic Huechulafquen Formation and Piedra Santa Complex
Continental or volcanic rock Evaporite rock
Volcanic rock Offshore clastic and carbonate rock
Plutonic and metamorphic rock Shallow-marine clastic and carbonate rock
Figure 3. Lithostratigraphy of the Neuquén basin. Three major deepwater
marine shale deposits are in the Neuquén basin. The deep Los Molles
formation is Early to Middle Jurassic in age. The Vaca Muerta formation
straddles Late Jurassic to Early Cretaceous. The Cretaceous Agrio
formation lies above the Vaca Muerta formation. The sediments in
these formations were deposited during successive transgressive and
regressive cycles. Restricted access to open seas and anoxic conditions
that existed during deposition preserved their organic content. (Adapted
from Howell et al, reference 7.)

January 2016 29
 Effects from the early Cenozoic Andes orog-
Neuquén basin eny introduced structural complexities in the
Oil Neuquén basin, especially along the western
Wet gas and condensate front.14 The close proximity of the western Vaca
CHILE
Dry gas
Muerta to the Andes Mountains created stresses
and natural fractures that are less prevalent
along the eastern edge of the basin. The influ-
ence of the Andes on the stress state of the reser-
ARGENTINA voir rocks helps distinguish it from other shale
plays, especially those in North America.
SOUTH AMERICA
Unlocking Potential
Most unconventional plays have unique features
that differentiate them from other plays. In the
nascent period of shale play development, opera-
CHILE tors were looking for the next Barnett Shale.
Although lessons learned from the Barnett Shale
ARGENTINA
and other unconventional plays were crucial for
hastening commerciality and acceptability of
0 125 250 km shales as viable prospects, operators realized
0 75 150 mi early on that each new development required a
specific and often unique approach. Because of
its many distinct characteristics, this is espe-
cially true of the Vaca Muerta Shale. Distinctions
for the formation include the prevalence of ash
Figure 4. Hydrocarbon type in the Vaca Muerta Shale. Hydrocarbon types in the Vaca Muerta Shale beds, complex facies stacking, abnormally high
vary throughout the Neuquén basin (blue outline). The eastern and southern regions contain mostly
oil (green), the west-central section is predominately dry gas (pink) and the region between the two pore pressure and geomechanical stresses result-
yields wet gas and condensate (yellow). (Adapted from the US EIA, reference 1.) ing from the proximity of the Andes Mountains.
To optimize the development program for the
large gross thickness, areal extent and reservoir carbonates and limestone. The middle was Vaca Muerta play, YPF collaborated with other
properties have placed the Vaca Muerta Shale in formed predominantly as a slope deposit and has operators and service companies that had exper-
the ranks of a world-class resource play.12 higher siliciclastic content than the other two tise in shale developments. Working with
The deep marine sediments of the Vaca sections. The upper section returns to a predomi- Schlumberger geologists and engineers, YPF
Muerta Shale consist of fine laminations of black nately carbonate platform.13 engineers devised workflows and simulations to
and gray shales and limey mudstones mixed with The type of hydrocarbon expected in the Vaca efficiently exploit the Vaca Muerta formation
organic material. The anaerobic environment in Muerta Shale generally depends on the location (Figure 5).15 The objective was to avoid the costly
which these organic materials were deposited within the basin. The eastern and southern and time-consuming trial-and-error approach
helped preserve their organic content. The for- regions have potential for oil production, the commonly employed in the development of many
mation comprises three sections—lower, middle western region is predominately dry gas and the shale plays. Creating and using a predictive
and upper. The lower corresponds to an inner area between the oil and dry gas zones is wet gas model rather than relying on a statistical
carbonate platform that consists mostly of marls, and condensate prone (Figure 4). approach was deemed crucial for financial
viability.
Property Source To meet the objective of creating a useful pre-
dictive model, the team members first defined
Seismic and geologic data Horizons, faults and stratigraphy Geologic model
specific known problems for developing the play.
Petrophysical data Water saturation, porosity and Openhole logs and core data
permeability
They established data requirements for each
stage of the process that included seismic, petro-
Geomechanics Pore pressure, stresses and elastic Mud data, openhole logs and
properties formation tests physical and geophysical data (Figure 6). The
existence of 3D seismic data over much of the
Microseismic Event type and location Geophysical interpretations
area was beneficial in creating these models.
Discrete fracture networks (DFN) Azimuth, spacing and length Image logs, image interpretaion Hundreds of legacy wells have also been drilled
and model estimates
for exploring and developing conventional reser-
Completion design Staging perforation cluster and Job data
perforation properties
voirs, although the data acquired in those wells
may not always provide the type of information
Fracture design Pumping schedule, proppant, fluid Job execution data
volumes and pressures needed to define parameters for drilling optimi-
zation, well spacing and fracture design in the
Figure 6. Data requirements for predictive model. unconventional sections. Additional data were

30 Oilfield Review
Reservoir Characterization Automatic Fracture Gridding
Pumping Schedule
3D seismic

Petrophysics

1 2

Complex Hydraulic Fracture Model

3D MEM

3D DFN

Production Evaluation
Sensitivity Analysis 5
Pressure
4

Rates

Production

Figure 5. Unconventional reservoir integrated workflow. In developing
the Vaca Muerta resource play, engineers at YPF and Schlumberger
collaborated on a workflow that reduced the degree of trial and error.
Geologists first characterized the reservoir using 3D seismic data,
petrophysical data, a 3D mechanical earth model and 3D discrete fracture
network models (1). Using these models, they can develop the hydraulic
stimulation program based on automatic fracture gridding (2). A complex
hydraulic fracture model can be compared to stimulated volumes computed
during treatments (3). During the stimulation treatment, fluid rates and
pressures are recorded, and microseismic events may be captured (4). A
sensitivity analysis helps engineers understand the effectiveness of the
changed parameters (5). If the outcome of the stimulation does not match
the model, the empirical data are used to update the simulations. If the
stimulation results match the model, additional information, including core
and fluid properties, is used to predict production, which can eventually be
validated using production logs. (Adapted from Hryb et al, reference 16.)

acquired that were then used to model the forma-
tion and create the simulations.
The engineering teams formulated and
adopted integrated workflows to help them ana-
lyze the old and new data. A 3D geomechanical
model was constructed from well log and 3D seis-
mic data. Petrophysical data were acquired from
pilot wells and used to fine-tune the models and
compute geomechanical properties such as
in situ stresses, Young’s modulus, Poisson’s ratio
and pore pressure. To calibrate the modeled
results, fracture injection tests were performed
As part of their process, the YPF and
Schlumberger engineers used history matching to
validate predicted responses from the simulations
and then compared the model responses with real-
world well results. They were able to determine
the plausible geometry of existing and created
12. US EIA, reference 1.
13. Ejofodomi et al, reference 11.
14. The Andes Mountains were formed by the Cenozoic
tectonic shortening of the South American plate as it
overrode the subducting Nazca plate, an oceanic plate
in the Eastern Pacific Ocean off the west coast of
South America.
fracture networks in near real time from these
models, which allowed them to make well-to-well
optimizations in addition to stage-to-stage modifi-
cations during hydraulic fracture operations.
The methodology adopted for the complex
hydraulic fracture modeling takes geologic models,
15. Lacentre P, Pichon S, Suarez M and Badessich MF:
“Simulación dinámica integrada de fracturamiento
hidráulico y reservorio para pozos horizontales en
Vaca Muerta Shale Oil & Gas,” presented at the
Segundo Jornadas de Simulación, Buenos Aires,
July 7–8, 2015.

in some pilot wells.

January 2016 31
 unconventional plays, reservoir heterogeneity,
Completion Quality (CQ) Reservoir Quality (RQ)
especially regarding geomechanical formation
Mineralogy—mainly clay, carbonate and silica Organic content
properties, must be understood and accounted for
Mechanical properties—Young’s modulus, Poisson’s Thermal maturity in stimulation and drilling programs.
ratio and tensile strength
Effective porosity In shale reservoirs, geologic discontinuities,
Natural fractures—presence, density, orientation and which include faults, natural fractures, bedding
condition (open, closed or cemented) Intrinsic permeability
geometry and changes in mechanical properties,
In situ stress—variations between intervals accounting Fluid saturations—oil, gas, influence stimulation results. Evaluation tech-
for mechanical properties anisotropy condensate and water
niques generally treat mechanical properties as
Organic shale thickness affecting the completion quality (CQ) of the res-
Hydrocarbons in place ervoir rocks (Figure 7). The CQ is a predictive
attribute of the reservoir rocks that depends
Figure 7. Data requirements for fracture modeling. largely on Young’s modulus, Poisson’s ratio, bulk
modulus and rock hardness. Natural fracture
density and orientation, rock anisotropy and pre-
3D seismic models and other data sources to cre- with predictions, then future production fore- vailing in situ stresses are also incorporated in
ate the 3D geomechanical model.16 The geome- casts can be made. When the results are not in determining the CQ of the reservoir rocks.18 The
chanical model is combined with discrete line with predictions, the models are updated or CQ values incorporate information about clay
fracture network (DFN) models to simulate frac- modified to reflect the results. type and volume, which are obtained from spec-
ture treatments. Natural fracture models are troscopy measurements as well as mechanical
based in part on image log data that honor Reservoir Heterogeneity properties. The mechanical properties include
mechanical flow units, fracture azimuth, dip The reservoirs of some early shale development Poisson’s ratio, Young’s modulus and stress pro-
angle distribution and fracture density. projects were treated as if the rocks were isotro- files computed from acoustic and other petro-
A fracture design based on simulator results pic. In the case of organic shales, large variations physical data.
could then be executed with real-time microseis- in formation properties and characteristics can Along with CQ, the reservoir quality (RQ) of
mic monitoring.17 By using actual production exist both laterally and vertically; the extent of the rocks—another predictive property—deter-
data, the engineers are able to evaluate the these variations is often much greater than what is mines the economic viability of shale plays. For
effectiveness of the methodology. If the results encountered in developing conventional reser- organic-rich shales, RQ is a measure of the capac-
indicate that stimulation pressures, fracture voirs. Because hydraulic fracturing and horizontal ity to produce hydrocarbons after hydraulic frac-
propagation and stimulated volumes are in line drilling are basic requirements in the majority of ture stimulation. The RQ property is a product of
porosity, hydrocarbon saturation, mineralogy,
organic content and thermal maturity. Rocks that
Perforation cluster have a combination of superior RQ and CQ
respond to stimulation treatments better than do
rocks with poor RQ and CQ properties.
To accurately characterize unconventional
Engineered Placement of Fracture Stages and Perforation Clusters
resources such as organic shales, the reservoir
evaluation often relies on advanced petrophysics
Fracture stage based on spectroscopy, nuclear magnetic reso-
Rock Quality
Good RQ and good CQ nance and dielectric dispersion measurements.
Bad RQ and bad CQ Once the formation properties are calibrated
Bad RQ and good CQ using advanced petrophysical data, a minimal
Good RQ and bad CQ
dataset may be used to approximate RQ. This
Rock quality minimal set of log measurements includes resis-
tivity, bulk density, porosity, water saturation and
pore pressure.
Stress gradient Sweet spots may be determined from a com-
posite quality index (CQI) that accounts for both
Stress gradient
the RQ and CQ properties. For the completion
Low High design, perforation clusters and fracture stages
are assigned to the best CQI intervals. The
Figure 8. Quality indicators for well completion and design from Mangrove software. Log data are used Mangrove engineered stimulation design used in
to generate reservoir quality (RQ) and completion quality (CQ) indicators. These data are processed the Petrel platform automates the process of
using Mangrove stimulation design software that automatically highlights optimal intervals based on determining the best CQI intervals, and engi-
predetermined cutoff valves. By focusing on intervals that have the best RQ and CQ, and eliminating neers are using completion programs devised
those with poor RQ and CQ, better stimulation results can be obtained compared to those obtained
using a geometrical approach in which intervals are completed and stimulated uniformly without from the Mangrove software recommendations
regard to the formation properties. (Figure 8).

32 Oilfield Review
 Natural fracture

Microseismic events DFN realization Rebranching

Hydraulic
Crossing
fracture

ou UFM
s
ut
tp
FN te
d D bra
Compare to MSM an Cali Well

Prediction fracture geometry

Fracture treatment Fluid

data Slurry
Earth model

Figure 9. Unconventional fracture model. The engineers at YPF adopted a
UFM simulator to characterize Vaca Muerta Shale stimulations. Inputs to
the simulation (left) begin with recorded microseismic (MS) events that
are compared to a microseismic monitoring (MSM) model. The results are
used to calibrate the UFM simulation and the distributed fracture network
(DFN). The DFN model is used to predict fracture geometry created during
hydraulic stimulations. The UFM simulations rely on material balance,
geomechanical properties, DFNs and wellbore geometry to predict fracture
propagation properties (right). The cell-based outputs of the simulations
predict fracture geometry—whether the induced fractures will propagate
through existing fractures, rebranch after crossing fractures or follow
the track of existing fractures (top right). The software computes fracture
heights based on these interactions of natural and induced fractures.
Proppant transport and settling (lower right), which directly impact
production and the stimulated volume predictions are generated with the
UFM simulation software. (Adapted from Cipolla et al, reference 20.)

The CQI is a continuous index based on RQ
and CQ at each depth, although the completion
engineers must account for the interactions that
occur in the formation and around the perfora-
tion clusters during the stimulation process. A
single value of CQ at a given depth can be com-
pared with neighboring CQ values, whereas the
RQ values must be integrated over the fracture’s
estimated vertical propagation length and vol-
ume to represent the capacity of the stimulated
formation to produce hydrocarbons.
Data acquisition in a vertical pilot well is
essential for defining the predicted CQ response,
especially in plays such as the Vaca Muerta,
where vertical heterogeneity is common. The
definition of RQ and CQ for Vaca Muerta play con-
tinues to evolve as the formation characteristics
are better understood.
Unconventional Fracturing Model
To effectively produce from ultralow permeability
shale formations, operators try to generate
closely spaced fractures during stimulations to
increase the reservoir contact and increase
drainage. However, stress effects may oppose the
creation of new fractures in the presence of prop-
agating fractures. To design and properly execute
an effective hydraulic stimulation, engineers
must understand the interactions of fractures in
complex fracture networks typical of those in the
Vaca Muerta play.
One distinctive aspect of the YPF workflow is
the UFM unconventional fracture model software
used to model hydraulic fracture creation and
propagation. The UFM modeling is part of the
Mangrove software.19
Observations from actual jobs show that pre-
dicting fracture propagation in multistage hydrau-
lic fracture operations has not always yielded
accurate results. Most simulation models do not
properly handle complex fracture propagation in
the presence of natural and newly induced frac-
tures. The UFM simulation, first introduced in
2011, tackles this weakness in other simulations
by determining whether a hydraulic fracture will
propagate through an existing fracture, stimulat-
ing fresh formation, or if it will be subjugated by
existing fractures and simply continue along an
existing fracture plane (Figure 9).20 The crossing
model used in the UFM simulation is an essential
component in the predictions.
The UFM software also accounts for stress
shadow effects—the stress exerted on surround-
ing rocks and adjacent fractures by propagating
fractures—in simulating the propagation of mul-
tiple or complex fractures. Stress shadow causes
restrictions in fracture width, which can signifi-
cantly reduce the effectiveness of proppant
placement to the point of screenout.21
When multiple fractures are propagating
simultaneously in close proximity, the transport
of fluid and proppant into the newly formed
fractures may be adversely affected. The stress
shadow results in narrower fracture widths than
static propagation models suggest. The predic-
tions from the UFM simulations more closely
replicate the dynamics of actual treatments
16. Hryb D, Archimio A, Badessich M, Ejofodomi E, Diaz A,
Cavazzoli G, Zalazar F, Lagarrigue E and Pichon S:
“Unlocking the True Potential of the Vaca Muerta Shale
Via an Integrated Completion Optimization Approach,”
paper SPE 170580, presented at the SPE Annual
Technical Conference and Exhibition, Amsterdam,
October 27–29, 2014.
17. For more on real-time microseismic monitoring:
Burch DN, Daniels J, Gillard M, Underhill W, Exler VA,
Favoretti L, Le Calvez J, Lecerf B, Potapenko D,
Maschio L, Morales JA, Samuelson M and Weimann MI:
“Live Hydraulic Fracture Monitoring and Diversion,”
Oilfield Review 21, no. 3 (Autumn 2009): 18–31.
18. Glaser KS, Miller CK, Johnson GM, Toelle B,
Kleinberg RL, Miller P and Pennington WD: “Seeking
the Sweet Spot: Reservoir and Completion Quality
in Organic Shales,” Oilfield Review 25, no. 4
(Winter 2013/2014): 16–29.
19. For more on UFM software processing: Wu R, Kresse O,
Weng X, Cohen C and Gu H: “Modeling of Interaction of
Hydraulic Fractures in Complex Fracture Networks,”
paper SPE 152052, presented at the SPE Hydraulic
Fracturing Technology Conference, The Woodlands,
Texas, February 6–8, 2012.
20. Cipolla C, Weng X, Mack M, Ganguly U, Gu H, Kresse O
and Cohen C: “Integrating Microseismic Mapping and
Complex Fracture Modeling to Characterize Hydraulic
Fracture Complexity,” paper SPE 140185, presented at
the SPE Hydraulic Fracturing Technology Conference
and Exhibition, The Woodlands, Texas, January 24–26,
2011.
21. Screenout occurs when the proppant carried in a
treatment fluid creates a bridge across the perforations
or in fracture networks. The restriction to fluid flow
causes a rapid rise in pump pressure and ends the
placement of additional proppant.

January 2016 33
 Stage 11

Stage 10

Stage 9

Stage 8

Stage 7

Stage 6

Stage 5

Stage 4

Stage 3

Stage 2
Designed
Stages

Composite Quality Index

GG

GG

GG

GG

GG

GG

GG

GG

GG

GG

GG

GG
BG

BG

BG

BG
BG
Good

Good

Good

Good

Good

Good

Good

Good

Good
Bad

Bad

Bad

Bad

Bad

Bad
Bad
Completion Quality Index
Good

Good

Good

Good

Good

Good
Reservoir Quality Index

Minimum Horizontal
Stress Gradient

Missing
data Young’s Modulus

Poisson’s Ratio

Pore Pressure

Clay Volume

Effective Porosity
Water Saturation

Bulk Density

Resistivity

Gamma Ray
3,300

3,400

3,500

3,600

3,700

3,800

3,900

4,000

4,100

Figure 10. Vaca Muerta Shale completion design. The Mangrove software
automatically determines optimal stimulation intervals. The typical inputs
of gamma ray (Track 1), resistivity (Track 2), bulk density (Track 3) and
water saturation (Track 4) are derived from basic well log data. Effective
porosity (Track 4) and clay volumes (Track 5) are computed using data
from advanced logging tools, as are mechanical properties such as pore
4,200
pressure (Track 6), Poisson’s ratio (Track 7), Young’s modulus (Track 8) and
stresses (Track 9). The RQ and CQ indices (Tracks 11 and 12) are combined
to produce a composite quality index (Track 10) from which the stages and
perforation clusters are determined (Track 13). Perforation clusters are
grouped by similar properties to ensure maximal outcomes. (Adapted from
Hryb et al, reference 16.)

because it accounts for fracture geometry,
which ultimately controls fluid transport and
proppant placement.
The UFM software simulates fracture propa-
gation, rock deformation and fluid transport in
the complex environment of stimulation treat-
ments by using material balance, geomechanics,
DFN models and well geometry. The program
delivers fracture height estimates, predicts inter-
actions of natural and induced fractures and
models proppant transport.
Typical Vaca Muerta Well
The workflow and modeling that YPF uses in the
Vaca Muerta development offer considerable flex-
ibility; however, the general process can be out-
lined for a typical well. A subject well, drilled
parallel to another well from the same pad and
vertically separated by approximately 260 ft
[80 m], was drilled laterally for several thousand
feet.22 Engineers used Mangrove software to
determine staging, perforation cluster locations
and completion strategy (Figure 10).
The RQ and CQ outputs from the Mangrove
processing were limited using predetermined
cutoff values initially determined from the fully
characterized vertical pilot well. The process for
stimulation design comprised the following steps:
Identify intervals that have favorable RQ and CQ
values based on log measurements along the lat-
eral, determine sweet spots based on CQI, cross-
reference CQI with formation properties and
information from the vertical pilot well, and
stimulate the zones highlighted from the
Mangrove processing. The intervals and perforation

34 Oilfield Review
clusters were also grouped based on similar Hybrid Treatment Slickwater Treatment
stress profiles (one of the CQ indicators) to help 80-m stage length, 4 clusters per stage 80-m stage length, 8 clusters per stage
ensure that the treatment breakdown pressures
would be similar for individual stages. Otherwise, 14% 20% 19%
the treatment would have a tendency to follow a 38%
50% 100%
path of least resistance and flow through a lim-
36%
ited number of perforations, which would not 42% 81%
stimulate the rock effectively.
The workflow also allows the design team to Slickwater 100 mesh sand Slickwater 100 mesh sand
simulate scenarios to determine an optimal stim- Gel 40/70 mesh sand Gel 40/70 mesh sand
ulation schedule. For example, for another well in Crosslink 30/50 mesh sand Crosslink 30/50 mesh sand
the project, the predicted results of a hybrid treat-
ment were compared to that of a slickwater-only 1.00
treatment (Figure 11). The hybrid treatment—a
Measured rate
combination of slickwater, gel and crosslink gel— Hybrid treatment
Cumulative production
should produce a higher gas rate and maintain 0.75
Dimensionless gas rate
that output compared to the simulated results
obtained using a slickwater-only treatment. Slickwater treatment
The predicted normalized flow rate per clus- 0.50
ter for the fracture treatment was used to group
similar clusters. The treatment consisted of a
combination of slickwater, 20 lbm [9 kg] cross- 0.25
linked gel and 20 lbm crosslinked gel carrying
Predicted rate
100, 40/80, 30/60 and 20/40 ceramic proppant
(Figure 12). Each stage consisted of 500,000 lbm 0
0 5 10 15 20 25
[227,000 kg] of proppant pumped at a rate of
Days
70 bpm [11 m3/min].
Figure 11. Stimulation simulation comparisons. Engineers modeled and compared treatment scenarios
22. Hryb et al, reference 16. to determine which was more effective. Both designs were based on 80-m [260-ft] stage lengths. The
hybrid design (top left) is based on four clusters per stage, and the slickwater design (top right) is
based on eight clusters per stage. The hybrid design included slickwater, gel and crosslink gel stages
along with a mixture of proppant sizes. The slickwater-only treatment was predominately 100 mesh
sand with some 40/70 mesh sand. The cumulative gas production (red line) of the hybrid design clearly
exceeds that of the slickwater-only treatment (blue line). The predicted hybrid flow rates are higher
than those of the skickwater-only treatment flow rates, and the measured data validates the hybrid
system prediction. (Adapted from Hryb et al, reference 16.)

Simulation Design
Simulation Design 4 proppant mesh sizes 100 mesh proppant 40/80 mesh proppant
Hybrid completion design Slickwater Linear gel Crosslink gel 500,000 lbm proppant at 70 bpm 30/60 mesh proppant 20/40 mesh proppant
100 100

90 90

80 80

70 70

60 60

50 50
%

40 40

30 30

20 20

10 10

0 0
1 2 3 4 5 6 7 8 9 10 Average 1 2 3 4 5 6 7 8 9 10 Average
Stage Stage

Figure 12. Simulated designs. The hydraulic simulation was optimized for the mesh sizes (right), pumped as a percentage of the total volume. (Adapted
parameters of the specific well. Each of the ten stages used a specific fluid from Hryb et al, reference 16.)
type and volume (left). Each stage consisted of a mixture of four proppant

January 2016 35
 Stage 1
Stage 2
Top of Vaca Muerta formation Stage 3
Stage 4
Stage 5
Stage 6
Stage 7
Stage 8
Stage 9
Stage 10
Stage 1
Stage 2
Stage 3
Stage 4
Stage 5
Stage 6
Stage 7
Stage 8
Stage 9 Top of Catriel formation
Stage 10

North–South

Figure 13. Microseismic monitoring in the Vaca Muerta play. The microseismic events in the treatment
well (left, lower well) were monitored in the adjacent horizontal well (upper well). Ten stages
(color-coded on the map display) were performed from the toe of the well (yellow) to the heel (gray).
The survey monitoring tool was positioned across seven intervals (color-coded yellow to green) to
maximize event detection. Stages 3 and 5 through 10 have similar stimulated volumes; the later
Stages 6 through 10 exhibit more downward growth than that seen in Stages 1 through 5 (right).
(Adapted from Hryb et al, reference 16.)

The operator used the parallel offset well for
microseismic (MS) monitoring to evaluate the
effectiveness of the treatment program and to
make adjustments to the job during the course of
the treatment (Figure 13). Prejob modeling
determined that the maximum detectable event
location was about 1,300 ft [400 m] away from the
monitoring array. During the treatment, the array
was moved a total of 7 times during the 10 stages,
thus ensuring the MS events were within the
detection range of the sensors. Event location
uncertainty and event location confidence were
enhanced by repositioning the sensor array.
The MS monitoring results indicated domi-
nant fracture orientation of N110°E and little
propagation downward and out of the Vaca
Muerta formation. Lateral migration was con-
fined to the perforated interval of the stage being
treated, which was indicative of good mechanical
isolation between the newly stimulated areas of
the reservoir and the zones stimulated during
subsequent stages.
Postfracture Analysis
Although the YPF engineers want to avoid a trial-
and-error methodology, they still use an iterative
process to improve job design and execution
based on postjob feedback and continuous
improvement of the workflow process. For the
example well, the microseismic geometry data
acquired using MS monitoring were combined
with the pressure data from the stimulation. The
team used these data to calibrate the UFM simu-
lation results.
A complex fracture network was modeled for
each stage of the stimulation program. Minimum
horizontal stresses were calibrated to the
observed initial shut-in pressures. The horizontal
stress anisotropy was then adjusted to match
observed MS events.23 The fracture direction was
predominantly east-west, parallel to the maxi-
mum stress direction. Effectiveness of fluid
movement and proppant placement was also
evaluated during the postfracture analysis.
Creating a complex fracture network is only
one part of the objective of a stimulation treat-
ment. The complex network must also have suffi-
cient conductivity to move fluids from the
reservoir to the wellbore by way of the newly cre-
ated fracture system. For the job in question, the
modeled results indicated that only about 50% of
the fracture network was conductive. With these
data in hand, the engineers could use the MS
measurements and pressure matching to cali-
brate the UFM software and make adjustments to
future stimulations and help increase the con-
tributing portion of the reservoir. This crucial
step of evaluating results and adjusting the mod-
els is a routine part of the workflow.
After the UFM software output is properly
calibrated to match the observed data, the reser-
voir simulation models can be tested and
adjusted as needed. An additional aspect of sim-
ulators that integrates complex fracture model-
ing is the inclusion of transient behavior typical
of shale wells. Transient behavior can be cap-
tured from small changes in pressures and flow
rates. The Mangrove software used to simulate
23. Horizontal stress anisotropy is the ratio of maximum and
minimum horizontal stresses and directly affects the
fracture complexity. The smaller the ratio, the more
complex will be the fracture network.

36 Oilfield Review
 Propped Fracture Permeability Histogram
16 1.0

14
0.8
Normalized fracture distribution

12
Propped

Permeability increase
10
0.6
Unpropped
8 Matrix
6 0.4

4
0.2
2

0 0
0.12 0.40 0.80 1.20 4.00 8.00 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000
Grid cell fracture permeablilty, mD-ft Pressure, psi

Figure 14. Production predictions from stimulation simulations. The increases in permeability, detect unstimulated matrix, determine unpropped
estimated relative permeability (left) is a key component in predicting fractures and quantify the expected production from the stimulated volume
production from fractured reservoirs. The simulator can produce plots for and propped fractures (right).

production uses small grid blocks in the near
wellbore and in the fractures. In this simulation,
engineers used 445,000 cells and 19 layers,
which provided high-resolution results and fine
details. The team used the simulation to deter-
mine that the proppant had a tendency to settle
out upon fracture closure.
To conduct the history matching, engineers
controlled the total liquid flow rates—both oil
and water—to match the observed fluid phases
and flowing pressures from well tests. From these
results, they determined that the most important
history matching component for the matrix (as it
existed prior to fracture stimulation) was pres-
sure-dependent permeability. For the newly
induced fractures, it was both porosity- and pres-
sure-dependent permeability; however, matching
the flowing pressures using fixed permeabilities
and porosities was not possible. For the model-
ing, a pressure-dependent permeability/porosity
ratio was developed (Figure 14).
Workflow Updates
Now that there are more than 400 wells in the
field, the impact of history matching and job mod-
ifications becomes more evident. The UFM soft-
ware outputs, production models and continuous
feedback provide an understanding of important
variables for future wells and treatments. The
data are used to determine optimal spacing
between lateral wells for maximizing production
but minimizing negative crosswell impacts.
Engineers can use the simulation outputs to
determine the optimal fracturing fluids, prop-
pant type and volumes, pumping schedules, per-
foration cluster locations and placement,
longitudinal staging along the wellbore, horizon-
tal staging related to other wells, stress shadow
effects and the impact of changes made from job
to job and from stage to stage. They use these
data along with sensitivity analyses to determine
which parameters have the greatest impact on
well performance (Figure 15). Future hydraulic

Uncertainty Analysis, Hydraulic Fracture Model Uncertainty Analysis, Production Prediction

DFN Fracture extension

Vertical stress contrast Fracture compaction

Horizontal stress anisotropy Propped fracture

Injected fluid properties Matrix permeability

Maximum stress orientation (angle) Fluid properties

Fracture porosity

Matrix porosity

Matrix compressibility

Figure 15. Postfracture calibration of models. Input parameters have
some degree of uncertainty because of the nature of the shale formation’s
heterogeneity and fracture complexity. Uncertainty analyses can be
performed after the stimulation program to quantify these uncertainties and
determine model sensitivity (left). Similarly, evaluation of production results
can provide levels of uncertainty for the inputs used to predict production
(right). The degree to which this uncertainty affects the simulation can be
used to evaluate the quality of the predictions. (Adapted from Hryb et al,
reference 16.)

January 2016 37
 Optimization Strategy
• Horizontal landing location
• Various proppant systems
• Various fluid systems

• Fibers

• Perforation strategy

• Impact of natural fractures

Figure 16. Optimization strategies as part of the standard workflow. After
the calibrated and corrected models and simulations become available,
engineers can develop optimization strategies. Using the simulations, the
team can vary landing location, proppant characteristics, fluid delivery
methods, perforation strategies and contributions from existing conditions
to optimize well locations and spacings and field development. Simulations
can be presented graphically, and then engineers can compare the results,
which helps them choose the best course of action. (Adapted from Hryb et
al, reference 16.)

fracture stimulation designs are modified from
these simulations to improve efficiency and maxi-
mize the stimulated volume.
The workflows used by YPF and Schlumberger
put all the analysis tools in a single platform,
which simplifies the modeling. The results can be
quickly analyzed, and analysts are able to visual-
ize the effects of parameter changes. The results
include the uncertainties associated with each
variable, which further help in determining which
parameters have the greatest impact on the out-
come and the optimal course of action to take.
Engineers compared simulations results that
were based on previous generation fracture mod-
eling software with simulations that used the
UFM software and found the results more closely
matched the empirical data. The engineers
attributed the more accurate results to the UFM
software’s proper treatment of the development
and propagation of complex fractures. The level
of uncertainty from these simulation models can
be high due to nonunique solutions; however, by
using data from individual wells to calibrate the
simulations, the analysts can improve the models
for future drilling and stimulation programs
(Figure 16).
24. Collins PW, Badessich MF and Ilk D: “Addressing
Forecasting Non-Uniqueness and Uncertainty in
Unconventional Reservoir Systems Using Experimental
Design,” paper SPE 175139, presented at the SPE Annual
Technical Conference and Exhibition, Houston,
September 28–30, 2015.

38 Oilfield Review
 Property
Production history Oil, gas and water volumes
Bottomhole pressure
PVT Oil and gas properties
Changes in rock properties Changes in the matrix
Changes in unstimulated rock
Changes in conductivity of
the propped fractures
Relative permeabilities Oil, gas and water permeabilities
Figure 17. Data requirements for generating production predictions.
Source
Production data
Laboratory measurements
Calculated from wellhead pressure
Laboratory measurements on core
Adjustments in parameters using empirical data
Data from service provider
Adjustments in parameters using empirical data
proppant placement and fracture growth.
Microseismic measurements are used on some
wells in real time to adjust programs during job
execution and then are used to update and fine-
tune existing mechanical models.
The Vaca Muerta Shale is massive, and even
though YPF has drilled numerous wells, this
operator and others have only just begun the
development and commercialization of the play.
As more wells come online, and field complexi-
ties become better understood, this giant has the
potential to revolutionize the oil and gas industry
in Argentina as well as encourage operators to
explore the possibilities for shale resources in
other South American basins. —TS

Not only do the workflows help users under-
stand well dynamics and help them focus on the
most critical parameters for adjusting programs,
they also benefit the operations. Field personnel
are aware of the need for quality measurements
such as permeability, fluid saturations, effective
porosities and pressure data that will be used in
the simulations and modeling. When following
these well-defined programs, field personnel can
focus on what is required to ensure successful job
execution and efficient job operations.
Current Status
Compared to its North American counterparts,
the Vaca Muerta Shale development is a relatively
new play. The first shale gas well in the play that
involved hydraulic fracture stimulation was com-
pleted in mid-2010 and was brought online in
December of that year. Ramped-up development
drilling did not start until 2013, when the first
cluster of 100 wells operated by YPF was drilled in
the Loma Campana block. Spaced at one well per
20 to 40 acres [0.08 to 0.16 km2], the majority of
the early wells were drilled vertically. Production
forecasting is included in the normal workflows,
and specific datasets have been defined to ensure
the necessary inputs are available for the predic-
tive software (figure 17). Because the develop-
ment is new, forecasting medium-to-long-term
well performance is still a challenge.24
Today, several wells that were drilled horizon-
tally and completed using multistage stimulation
programs are producing from the Vaca Muerta
formation. These lateral wells were initially com-
pleted with four to six stages per well; however,
more stages, with as many as fifteen stages per
well, are being attempted. Most of the wells were
drilled vertically, which is in direct contrast to
how wells are drilled in the majority of the uncon-
ventional plays in North America, where horizon-
tal well drilling is standard practice. YPF initially
chose vertical well configurations for targeting
the Vaca Muerta play because of the significant
vertical heterogeneity of the formation and sub-
stantial pay thickness.
The vertical well methodology employed by
YPF engineers is beginning to shift. Although
large-scale drilling of horizontal wells that were
completed using multistage fracturing did not
begin until late 2014, by the end of 2015, approxi-
mately 10% of the producing wells were horizon-
tal. This increase in horizontal drilling as a
proportion of the total wells drilled has been
adopted because engineers at YPF now feel confi-
dent that they have identified sweet spots for
landing horizontal targets within the vertical por-
tion of the reservoir.
The LLL-992h horizontal well that YPF
recently drilled and brought online is a good
example of this shift in methodology. The well is
regarded as a major milestone for the project as it
heralds a transition from drilling primarily verti-
cal wells to drilling a high percentage of horizon-
tal wells. This well, drilled to a lateral length of
approximately 6,600 ft [2,000 m], achieved a
peak production of 1,630 bbl/d [260 m3/d], which
far exceeded initial expectations. These results
are encouraging the operator to migrate more
rapidly from a vertical well architecture to a hori-
zontal well development scheme for wells drilled
in both the oil and the gas windows.
In spite of reservoir complexity and hetero-
geneity, the integrated approach YPF has taken
is helping the company develop the play. By
understanding the discrete fracture networks
through use of the UFM software, engineers and
geologists are able to optimize stimulation
designs and develop realistic modeling of

January 2016 39
 Rod Pumps in Unconventional Resource Wells

Like all artificial lift systems, rod pumping is used to restore production when the
natural drive energy of a reservoir is not strong enough to push hydrocarbons to the
surface. Today, operators are applying this mature technology to enhance production
from unconventional resources that were considered, until recently, economically or
technically inaccessible.

LeMoyne Boyer Because of their high hydrocarbon content and Shale well production is characterized by
Kyel Hodenfield impermeability, shale formations have long been high initial flow rates from natural and induced
Lee Norris considered both source rocks and sealing mecha- fractures accompanied by high producing pres-
Houston, Texas, USA nisms in more traditional oil and gas plays. sure. However, the extremely low permeability of
Having learned to use proven technology to shale slows hydrocarbon movement through the
Fred Clarke
exploit tight formations, the industry is now rock to the fracture, which results in decline
Leslie Malone
working to improve ultimate recovery from shale rates that may reach 70% within the first year
Murphy Exploration & Production Company
Houston, Texas wells through optimal completion and produc- (Figure 1).1 Also as a consequence of extremely
tion strategies. low permeability, shale formations have a limited
Dave Jones
Dallas, Texas

Mark L. Mitchell 450

Oklahoma City, Oklahoma, USA
400
Oilfield Review 28, no. 1 (January 2016). First full month of production
Copyright © 2016 Schlumberger. 350
For help in preparation of this article, thanks to Dewayne
Average oil production, bbl/d

Atwell, Houston, Texas, USA, and Jennifer McMurtry,

Calgary, Alberta, Canada. 300
Initial
OLGA and UniStar are marks of Schlumberger. production year
250
2014
200 2013
2012
150 2011
2010
100 2009

50

0
0 6 12 18 24 30 36 42 48
Time, months
Figure 1. Typical shale resources production history. Initial high production rates, followed by rapid
rate decline in the first months of production as illustrated here in the Eagle Ford Shale, typify
production history in shale formations. (Adapted from the US Energy Information Administration.)

40 Oilfield Review
drainage radius, and their declining production
is accompanied by declining sandface pressure.
When the pressure falls below the bubblepoint in
oil producing wells, free gas forms in the affected
area of the horizontal section, and flow quickly
True vertical depth, ft
becomes unstable, causing large variations in
fluid production rates and slugging.2
Because of their characteristic production
and pressure decline rates, most shale wells
require an artificial lift (AL) system within
months of first production. In these wells, opera-
tors may use any AL system that is able to pro-
duce fluids that may contain various levels of
solids and entrained gas. Today, rod pumps, pro-
gressing cavity pumps, electrical submersible
pumps (ESPs) and gas lift systems are all being
used to boost production in unconventional wells.
In the majority of shale wells, the other AL
systems—jet pumps and plunger lifts—are usu-
ally the least practical choice; the former
requires a separate flow path for power fluid and
the latter has a low liquid removal rate.3 Jet
pumps are applicable within only a narrow range
of circumstances in unconventional resource
wells and require significant capital investment.
Because fluids move slowly through shale,
plunger cycle times are longer than in more per-
meable formations.4 Plunger systems also depend
on bottomhole pressure (BHP) to lift the plunger
and the fluid column to the surface, and as BHP
continues to decline in shale wells, operators
may be forced to add a gas injection system to
help lift the plunger to the surface.5 Typically,
plunger systems are labor intensive and require
continuous operator oversight; this vigilance may
be impractical in unconventional resource pro-
duction strategies that include drilling a high
number of wells per field.
Although gas lift systems are used in shale
production, unconventional plays are typically in
areas without a gas source, pipelines or other
infrastructure required to efficiently deliver
high-pressure gas to a large number of wells. In
addition, liquid production rate changes in
numerous wells may require significant field
crew time to adjust gas injection rates.
In addition to exhibiting rapidly declining
flow rates and pressures, shale well production
typically consists of a multiphase fluid in slug
flow that may include substantial amounts of
sand and increasing gas content; production may
also include mineral-laden, often corrosive,
water that can lead to scaling. Because condi-
tions change as the well is produced, operators
must select an AL system knowing that they may
have to switch system types sooner and more fre-
January 2016
7,000
7,200
7,400
7,600
7,800 Hybrid
Toe-down
8,000 Toe-up
8,200
8,400
8,600
8,800
0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000
Lateral length, ft
Figure 2. Common horizontal shale well profiles. Wells that have significant horizontal extensions
typically include undulation and a general depth relationship between the farthest point from kickoff,
or the toe, and the nearest point, or heel, of the well. When the toe is shallower than the heel, the well
is called toe-up. A well is toe-down when the heel is shallower. In hybrid wells, the well path deviates
in both general directions. (Adapted from Lane and Chokshi, reference 5.)
quently than they do in traditional wells that changes in formation dip direction or rock hard-
have more stable flow regimes. ness force drillers to repeatedly build and drop
This article describes how the industry is angle, creating undulations in the wellbore tra-
adapting rod pump systems to shale resource jectory. Undulation changes the formation inflow
wells. Case histories from various shale plays in distribution along the wellbore, and during multi-
the US demonstrate how operators are incorpo- phase fluid flow, gas may become trapped by
rating the venerable rod pump system to the new accumulated liquid at low spots.6 These liquids
world of unconventional resources. A multiphase can create backpressure and become initiation
simulation used to describe the source of insta- points for slug flow.7
bility in fluid flow from this still little understood As a consequence of the bit following the
resource is also presented. updip or downdip of formations, laterals typically
trend in a general elevation change consistent
Trajectory and New Rules with that formation’s rise or drop. Toe-up wells fol-
Hydraulic fracturing and horizontal drilling are low a formation in an updip direction; toe-down
the two technologies that have revolutionized well profiles result from drilling along a downdip
production from unconventional resources. trend. How a toe-up or toe-down profile affects
However, lateral wells drilled through shale sec- well performance is a matter of some debate
tions are rarely truly horizontal. In some wells, (Figure 2). For instance, one study indicated that
1. Baihly JD, Malpani R, Altman R, Lindsay G and Clayton R: 3. Khan N, Ganzer L, Elichev V and Ali N: “An Integrated
“Shale Gas Production Decline Trend Comparison over Life-Time Artificial Lift Selection Approach for Tight/Shale
Time and Basins—Revisited,” paper SPE 178674/ Oil Production,” paper SPE 169833, presented at the
URTeC 2172464, presented at the Unconventional SPE Hydrocarbon Economics and Evaluation Symposium,
Resources Technology Conference, San Antonio, Texas, Houston, May 19–20, 2014.
USA, July 20–22, 2015. 4. Plunger cycle times are dictated by the time required for
Baihly J, Altman R, Malpani R and Luo F: “Shale Gas formation pressure to build in the casing annulus and
Production Decline Trend Comparison over Time and liquids to accumulate in the production tubing.
Basins,” paper SPE 135555, presented at the SPE Annual 5. Lane W and Chokshi R: “Considerations for Optimizing
Technical Conference and Exhibition, Florence, Italy, Artificial Lift in Unconventionals, paper URTeC 1921823,
September 19–22, 2010. presented at the Unconventional Resources Technology
2. Norris HL III: “The Use of a Transient Multiphase Conference, Denver, August 25–27, 2014.
Simulator to Predict and Suppress Flow Instabilities in a 6. Bond A, Zhu D and Kamkom R: “The Effect of Well
Horizontal Shale Oil Well,” paper SPE 158500, presented Trajectory on Horizontal Well Performance,” paper
at the SPE Annual Technical Conference and Exhibition, SPE 104183, presented at the SPE International Oil
San Antonio, Texas, October 8–10, 2012. and Gas Conference and Exhibition, Beijing,
When pressure of the formation gas trapped behind the December 5–7, 2006.
liquid becomes high enough to move the liquid to the 7. Lane and Chokshi, reference 5.
surface, it does so in a series of liquid slugs separated by
gas pockets. The resulting surges can damage the
formation and surface gas handling equipment.
41
 the highest performing Barnett Shale wells in
Texas, USA, were those drilled toe-up, while the
same study found that the opposite configuration
of toe-down performed better in Woodford Shale
wells in Oklahoma, USA.8
The industry is just beginning to understand
how shale well production is impacted by trajec-
tory and other factors such as completion meth-
ods, operating conditions and fluid behavior.
During initial high-rate, high-pressure produc-
tion in shale wells, the horizontal section is filled
by single-phase liquid flow. However, because of
the very small drainage radius in shale wells, res-
ervoir pressure declines rapidly until sandface
pressure falls below the reservoir fluid bubble-
point. This phenomenon creates a liquid vapor
600
500
flow in the horizontal section, which can lead to
significant variations in fluid production rates.
The effects of these rate changes are com-
pounded by hardware used in the completion.
The horizontal producing sections may include
4- to 6-in. diameter liners necessary to accommo-
date completion tools such as those used in
hydraulic stimulations. These relatively large
cross-sectional flow areas result in low fluid
velocities and stratified flow that can result in
gas volumes that cause slug flow.9
This transient process is affected by wellbore
geometry, reservoir fluid properties, reservoir
pressure, completion and perforation details,
production rates and multiphase fluid dynamics.
Schlumberger engineers recently used the OLGA
0.6
0.5
dynamic multiphase flow simulator to predict
wellbore instability in horizontal shale wells and
to identify possible methods for suppressing pro-
duction fluctuations. The simulation included a
9,500-ft [2,900-m] vertical well with a 10,000-ft
[3,050-m] 4 1/2-in. horizontal section assumed to
have no inclination and 20 perforated sections.
The inputs for the simulation included an initial
reservoir pressure of 6,000 psi [40 MPa], a pro-
ductivity index of 0.005 bbl/d/psi [0.109 m3/d/
MPa], a gas/oil ratio (GOR) of 1 mcf/bbl and
34 degree API oil.10
While reservoir pressure remained high, the
low GOR did not present a substantial threat to
stable fluid flow of 313 bbl/d [50 m3/d]. However,
when the engineers reduced that pressure to
4,000 psi [30 MPa], liquid and gas rates oscil-
lated, and the vertical section loaded up, or filled
with fluid; production was reduced to slugs,
which flowed for about three hours and occurred
every 12 hours (Figure 3).
Analysts then added toe-up and toe-down well
profiles to the simulations and found that both

Gas production, MMcf/d

significantly exacerbated slugging and variations
Liquid production, bbl/d

400 0.4
Total liquid
in liquid production rates, even at 6,000 psi reser-
300 0.3 voir pressure. In these simulations, the problem
Gas was exaggerated in the toe-down well profile
because gas accumulates in the heel; in the toe-
200 0.2
up scenario, liquid would be near the heel of the
well as the gas migrated toward the toe.
100 0.1 In a toe-up scenario, a gas bubble accumu-
Water
lates at the toe and displaces the liquid at the low
0 0 spot—the heel—to the vertical section before
0 4 8 12 16 20 24
liquid can fill the horizontal section. In contrast,
Time, h
in a toe-down geometry, gas has a clear path to
1,000 1.0
the vertical section until liquid completely fills
the horizontal section. As a result, the toe-up
geometry tends to produce small, frequent slugs,
800 0.8 while the toe-down geometry tends to produce
Total liquid
large slugs less frequently.
Gas production, MMcf/d
Liquid production, bbl/d

Simulations established that the prime

600 0.6
source of flow instability in horizontal shale wells
was low flow velocity when sandface pressures
Gas
400 0.4
fall below the bubblepoint. Analysts concluded
therefore that reducing the cross-sectional flow
area of the horizontal section through the use of
200 0.2 slimhole drilling and smaller liners, installation
Water of a velocity string, insertion of a dead string or
injection of fluids into the horizontal flow stream
0 0
0 4 8 12 16 20 24
might remediate the slugging.11
Time, h
Rod Pump Systems
Figure 3. The dynamics of flow instabilities in horizontal hydraulically fractured wells. Simulations of
a 10,000-ft TVD well that has a 10,000-ft horizontal section illustrate that a relatively high reservoir While the industry has learned that shale reservoirs
pressure of 6,000 psi (top) and a low gas/liquid ratio produce stable flow. However, the simulation differ from one another, production from the vast
indicates that when reservoir pressure is 4,000 psi (bottom) and the gas/liquid ratio is 1 Mcf/bbl majority of shale wells occurs in stages. Initially,
[180 m3/m3], the lateral contains a liquid vapor mix that creates highly unstable flow. (Adapted from
high bottomhole pressures provide sufficient
Norris, reference 2.)

42 Oilfield Review
energy for production from the stimulated zones to
lift flow to the surface on its own. Within a short Prime mover Crank arm Beam pumping unit
time, however, liquid flow rates and pressures
decline, and operators must turn to an artificial lift
system. No matter the type of system chosen, instal-
lation of AL systems in shale fields, which are often
located outside traditional oil provinces, usually
demands significant capital investments.
To optimize rate of return on investment from
Casing
shale plays, operators employ strategies rooted in
the certainty that shale well production rates
Tubing
decline during early production. Some operators
select AL systems because they are able to lift the Rods
initially high production volumes that shale wells
can deliver, and as production declines, exchange Tubing
pump
them for lower-rate systems. Insert pump
Other operators choose to sacrifice high ini-
tial rates and instead install a system that will
remain economically efficient even as production
declines. Often, because rod pump systems are
able to efficiently lift fluids from significant
depths and to accommodate a wide range of pro-
duction rates and changing well conditions, oper-
Perforations
ators choose them.12 Operators may install rod
Produced
pumps at first oil or, because they function well at fluids
relatively low rates, install them later when pro-
duction rates slip below those at which higher-
rate systems are efficient.
One North American operator, because of low Plunger
initial production rates, installed electrical sub-
Pump barrel
mersible pumps in its shale wells at first produc-
tion. But as rates fell, the ESPs became inefficient Traveling valve
because of operating costs. The company
switched to rod pump systems because they Standing valve
would ultimately improve the rate of return, par-
ticularly in a low oil price environment. Figure 4. Beam pump system. A traveling ball valve at the end of the rod string
is pushed off seat as it travels downward through the fluid column. When the
Operators turn to rod pumping systems for traveling valve achieves maximum downward reach, and the beam is at its
many reasons. Rod pumps have a long history in lowest point, the beam begins its upward movement, and the rods are pulled
the industry, and engineers are familiar with the upward, which forces the ball of the traveling valve back onto its seat. As
a result, the fluid column (green) is captured above it. As the fluid is pulled
technology. Rod pumps are considered reliable,
toward the surface, the reservoir pressure is greater than the pressure inside
inexpensive and easily installed and maintained. the pump chamber, which forces the standing valve ball off seat. Formation
The most common of all AL systems, rod pumps fluid (green arrows) flows through this lower valve and fills the pump
are composed of a prime mover, a surface pump- chamber. When the traveling valve begins its descent, the pressure of the
fluid column forces the standing valve ball to fall back onto its seat, and the
ing unit, a sucker rod string and downhole pump cycle is repeated.
(Figure 4). The gas engine–driven or electric
engine–driven prime mover transmits power to a
gear reducer, which drives the surface unit. A
  8. Miller C, Waters G and Rylander E: “Evaluation of Installing a velocity string reduces the flow area and
crank arm on the surface unit turns, causing a Production Log Data from Horizontal Wells Drilled in increases the flow velocity to enable liquids to be
beam to reciprocate and lift and lower a rod Organic Shales,” paper SPE 144326, presented at the carried from the wellbore.
SPE North American Unconventional Gas Conference A dead string is a length of tubing closed at the bottom
string attached to one end of the beam. The and Exhibition, The Woodlands, Texas, June 14–16, 2011. and placed inside the production string. Like a velocity
motion of the rod string causes traveling and   9. Norris, reference 2. string, the dead string reduces flow area in the annulus
standing ball valves in the downhole pump to 10. Norris, reference 2. formed by it and the production string and increases
flow velocity.
capture fluid or to allow fluid to flow into the tub- 11. Norris, reference 2.
12. Alvarez G, Ruiz R and Few J: “Shale Oil Production
A velocity string is a small-diameter tubing string run
ing. In some configurations, the valves are part of inside a larger production tubing of a well as a remedial
Challenges in Beam Pumping Being Solved with
Premium Sucker Rod Designs,” paper SPE 171361,
an insert pump—an integrated assembly that treatment to resolve liquid-loading problems. As the
presented at the SPE Artificial Lift Conference and
reservoir pressure in a well depletes, velocity may be
can be retrieved using the rods while leaving the insufficient to transport all liquids from the wellbore. In
Exhibition—North America, Houston, October 6–8, 2014.
production tubing in place. time, these liquids accumulate and impair production.

January 2016 43
 To design rod strings for use in deviated well-
bores, rod pump manufacturers and suppliers
may use deviated-well design software to predict
side loadings and dogleg severity through the use
of 3D images created from operator-supplied
directional surveys (Figure 5). Engineers use
these images to plan the optimal rod string and
pump configuration, including placement of rod
guides and setting depth, type and length of rod
pump to ensure extended operating life and opti-
mal efficiency.
When one operator chose to exchange its
existing AL systems for rod pumps, company
engineers were concerned about the challenges
to installing such systems in wellbores that
included significant deviation and dogleg sever-
ity. Working with Schlumberger engineers, the
operator found that the optimal depth at which
to place the pump, based on pump intake pres-
sure and fluid levels, was in a wellbore section
that included a 70º inclination.
Figure 5. Visualizing tubing wear. Using 3D profile software, operators can visualize tubing wear points
and severity during well planning. In this example of a basic deviated well, severity increases from the The installation design also included molded
least severe (yellow) to moderate (orange) and most severe (red). rod guides placed at strategic locations to coun-
ter the effects of dogleg severity. As a conse-
quence of collaborative designs, the operator’s
wells have been on rod pump systems for more
Because shale reservoirs are marked by such field. Drilling costs are primarily a function of than a year and have had no rod failures while
extremely low permeability and drainage radii, time, which leads operators to drill wells as achieving targeted production rates.
operators not only drill horizontal wells to expose quickly as possible. This can result in unintended
as much of the reservoir to the wellbore as possi- wellbore deviation and dogleg severity as a result Managing Gas
ble, they also drill a high number of wells per of directional changes and corrections. The majority of currently producing shale forma-
tions maintain sufficient energy to push forma-
tion fluids up the horizontal section of the well
for most of the well life. This ability allows rod
100
pumps to be installed in the vertical section of
80 the well but creates challenges associated with
60 the relative positions of the downhole pump and
the perforations.
40 Rod Primary among these concerns are the effects
velocity
20 of gas interference, which results from gas dis-
Velocity, in./s

0 placing fluid entering the pump, which causes

less than optimal production per stroke. If not
–20
managed, gas entering the pump may lead to gas
–40 locking, which is a condition in which the ball of
Stroke the traveling valve cannot be unseated because
–60
reference
gas entering the pump does not create sufficient
–80
Plunger velocity pressure below the traveling valve to overcome
–100 the pressure created by the weight of the fluid
60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360
0
Time, min
above it. Gas interference may also be damaging
to the rod system when the pump travels faster
Figure 7. Evidence of gas interference from remote monitoring. Using remote monitoring, operators
than it would when there is liquid higher in the
are able to detect possible well problems, including gas interference, by measuring surface rod and
downhole plunger velocity. In this case, plunger velocity (green) drops to zero when the plunger pump barrel. This phenomenon causes the pump
strikes fluid about halfway through the downstroke (dashed circle). The Unistar system detects this to land with an accelerated downward force, cre-
condition (circle) and quickly responds by reducing the speed, or strokes per minute, of the pumping ating a condition known as fluid pounding, which
unit (blue) and thus surface rod velocity (red) to minimize fluid pound shock load to the rod string.
(Illustration courtesy of Murphy Exploration & Production.)
can result in shock and vibration large enough to
damage rods and surface gears.

44 Oilfield Review
 In vertical wells, operators overcome gas
interference by placing the pump below perfora-
tion depth. This creates a natural, gravity-based
gas separator in which produced free gas flows
Seal coupling
upward from the perforations while the heavier
fluids fill the casing-tubing annulus around the
pump. However, because the pump is installed
above the perforations in horizontal wells, fluid
Free gas
entering the pump may be foamy when the pump
intake pressure is below the bubblepoint and
Slots on outer tube
gas/liquid ratios are high.
In shale wells, operators attach gas separa-
tors to the bottom of the downhole rod pump to Outer tube
isolate the pump from the direct fluid flow; sepa-
rators direct the liquids into the pump and the Inner tube
gas away from it (Figure 6). When a relatively low
pressure at that higher point in the annulus is Casing
maintained, free gas comes out of solution and
flows upward, and gas-free liquids flow downward
Ported Coupling
to the pump intake.
In many cases, particularly in the presence of
foamy liquids, gas interference cannot be elimi-
nated through separation, and gas consumes
some portion of pump volume with every cycle of
the pump. This condition can be easily discerned
on dynamometer cards, which display rod load
forces versus rod displacement (Figure 7).
In addition to presenting the risk of gas lock or
fluid pound, gas interference reduces pump effi-
ciency. If the pump rate is too rapid, insufficient
volumes of fluid will have time to enter the well-
bore around the pump, which results in pumpoff,
a condition in which no liquid is being lifted.
To allow time for gas to be displaced from the
pump or for formation fluids to reach the well-
bore, operators adjust pump motor speed.
Engineers use pumpoff controllers (POCs) to
turn off prime movers for specific periods. Early Packer
versions consisted of a circular series of pins that Figure 6. Packer-type gas separator. The packer-type gas separator is
revolved on a clock dial to complete or break cir- positioned within the tubing string directly above the packer, which forces
cuits. In the past, the POC timing—when and for all formation fluid to pass through the separator before it enters the pump
intake. Formation fluid (black arrows) travels through the tubing string to and
how long to turn off the unit—was set manually
through the ported coupling of the separator and enters the annulus between
and determined by operators’ experience in the inner flow tube and the outer tube. The fluid then passes through the slots
the field. of the outer tube and falls back into the annulus space between the casing
Today, POCs that have preprogrammed devices and the outer tube, settling on the packer. The flow path creates low-pressure
turbulent flow that helps gas break out of solution. This free gas is diverted
are available to collect, process and analyze data from the pump intake and travels upward through the tubing-casing annulus.
measured by load and position transducers
mounted on the surface pumping unit. From a
load cell mounted on the polished rod, POCs that percentage of pump capacity, the POC sends a above a preset minimum and maximum fill level,
include data about the specifics of the well are command to shut off the motor. After a prescribed VSDs regulate the motor speed to bring fill levels
able to calculate rod load and crank arm position. downtime, the POC restarts the system and 13. A polished rod is a standard alloy steel rod that has
Using those data, POCs determine what portion allows it to continue until it senses the pump is hard-surface spray metal coating; the rod supports the
loads created during the pump cycle and ensures a seal
of the load is being contributed by the pump filling at below targeted levels. through the stuffing box at the top of the well. The
and equate that to the pump fill level, which is Variable speed drives (VSDs) serve the same stuffing box, attached to the wellhead, or pumping tee,
contains packing elements that form a tight seal against
that percentage of the pump volume taken up by purpose as POCs but accomplish the task differ- the polished rod. The resulting seal forms a barrier
fluid.13 When the fill level falls below a preset ently. Instead of stopping the pump when the between the well and the atmosphere and diverts flow
into the flowline via the pumping tee.
level of liquid in the pump falls below or rises

January 2016 45
 Position, %
0 10 20 30 40 50 60 70 80 90 100
0

22,000 100

20,000 B 80
C
60
18,000
16,000 40

14,000 A 20

Load capacity, %
12,000 0
Load, lbf

D –20
10,000
8,000 –40
B C
Pump fill set point –60
6,000
4,000 –80

2,000 –100

0 A –120

–2,000 D –140
0 10 20 30 40 50 60 70 80 90 100 110 120 130
Position, in.

Figure 8. Using dynamometer measurements to avoid pumpoff.
The surface dynamometer card is a plot of rod tension versus
displacement measurements from a load cell during each rod
pump cycle (top). Each color is a single cycle. The upward
stroke of the rod string is recorded as a line from Point A to
Point C through Point B, and the downward stroke is recorded
as a line from Point C to Point A through Point D. At Point A, the
traveling valve closes, and at Point B, the standing valve opens.
At Point C, the standing valve closes, and at Point D the traveling
valve opens. All surface dynamometer cards deviate from the
ideal, and causes for deviation can be interpreted and corrected
automatically or through intervention. Intelligent variable speed
drives (VSDs) interpret the measurements taken at the surface
to create downhole dynamometer cards (bottom). The downhole
card, which, ideally, is an equilateral rectangle, provides more
readily identifiable diagnostic information on current pump
operating condition. When the pump plungers travels through
fluid, gradual tension reduction is visible between Points C
and D. A sudden reduction in load indicates the pump has
traveled quickly through gas and pounded fluid. Intelligent VSD
systems can interpret the shape of the line from Point C to Point D
as a measure of pump fill and adjust pump speed to achieve a
preprogrammed pump fill level (red dot). (Illustration courtesy of
Murphy Exploration & Production.)

within a specified range (Figure 8). Variable
speed drives use the pumping unit geometry and
an inclinometer mounted on the beam of the
pumping unit to determine the position and loads
on the system. These variable speed drives also
measure motor velocity and gearbox torque,
which operators use to monitor unit performance
and power input and output. These measurements
allow operators to recognize incidents and condi-
tions that could result in equipment damage.
Murphy Exploration & Production has more
than 500 wells producing in the Eagle Ford Shale
play of South Texas. Well vertical depths range
from 7,000 to 12,000 ft [2,100 to 3,700 m] and lat-
erals measure 3,000 to 8,000 ft [900 to 2,400 m].
More than 400 of the wells are produced using
beam pumps.
Early in field development, engineers at
Murphy chose to install rod pumps rather than a
higher-rate system that they knew would eventu-
ally be replaced by rod pumping units. To maxi-
mize efficiency, they automated the systems
using the Schlumberger UniStar downhole pro-
tection variable speed drive system. They also
included remote pressure and temperature moni-
toring capabilities, stuffing box leak detection
and safety shutdown capabilities.
In one example of the benefits of automation,
in a well that had a 11/2-in. pump set at 6,450 ft
[1,970 m], average pump speed was maintained
at between 3 and 2.5 strokes per minute, which
rendered a pump efficiency of between 70% and
80% as pump intake pressure fluctuated from an
estimated 800 to 1,200 psi [5.0 to 8.0 MPa]. Over
the course of 90 days, the pump speed, pump
intake pressure and oil production rates all
remained within narrow ranges while producing
only about 0.1 cf/d [0.003 m3/d] of gas.
The operator reported benefits from automa-
tion that included savings on capital expenditures
because the company was able to use smaller
pump units and generators and high-efficiency
motors. Automation and remote monitoring also
provided the operator with equipment protection,
which resulted in reduced operating expenses for
workovers and part replacements. Each unit also
experienced lowered peak load and reduced power
consumption and required fewer routine well site
maintenance visits. In addition, well performance
benefited because pump speed was optimized on
each stroke, the wells produced 24 hours per day
with no down time and gas interference was more
effectively managed.
Rule Changes
After a decade of drilling and producing shale for-
mations at a frantic pace, much remains for opera-
tors and service companies to learn about the
nature of these new plays. Drilling and completion
engineers continue to focus on optimizing well
geometry, well spacing and fracture design. But in
the first flush of shale resources exploitation,
developing methods to optimize production from
these formations has received less attention.
Operator emphasis may change, however, as
circumstances surrounding shale development
evolve. Current oil prices are significantly low,
and today, the most desirable shale leases have
been claimed. As a consequence, funding to con-
tinue the practice of producing from extremely
low-permeability reservoirs by drilling and frac-
turing large numbers of wells in a field may no
longer be viable.
Instead, operators may seek to increase their
return on investment through improved comple-
tion and production practices. That approach is
made all the more practical by the fact that,
according to some experts, 95% of the original
shale oil in place remains in situ and much of it is
accessible through wells that have already been
drilled and fractured. —RvF

46 Oilfield Review
CONTRIBUTORS
Matías Fernández Badessich is a Reservoir Engineer
with YPF, SA. Based in Buenos Aires, he is currently
the Manager of an unconventional resource develop-
ment team, focusing primarily on the Vaca Muerta
Shale play. In his 17-year career with the company, he
has worked in a variety of areas associated with explo-
ration and production of oil and gas. When he began
with YPF, he spent five years in the US working
in offshore fields in the Gulf of Mexico. He has
coauthored several SPE papers on topics ranging
from shale development and the Vaca Muerta Shale
to the integration and characterization of complex
fields. Matías holds a BS degree in chemistry from
Universidad del Salvador, Facultad de Ciencia y
Tecnología, Buenos Aires, a graduate degree in reser-
voir engineering from Instituto Tecnológico de Buenos
Aires and a master’s degree in technology exploration
and production of oil from Instituto Superior de la
Energía, Móstoles, Spain.
LeMoyne Boyer is a Senior Mechanical Engineer for
Schlumberger in Houston, where he is engaged in
perforation research and development of advanced
manufacturing methods and closed-loop control sys-
tems for shaped charges. Before joining the company
in 2008, LeMoyne worked for Lufkin Industries in
Texas, USA, where he developed improved pumping
unit designs, automated controls and the integration
of rod pump controllers with variable frequency drives.
He has also designed gas compressor systems for com-
pressed natural gas refueling applications for Gemini
Engine Company in Texas, performed various fire
tests and simulations to national standards and per-
formed dynamics and thermal simulations for engine
design and test evaluation of a prototype engine for
unmanned aerial vehicles for Southwest Research
Institute, also in Texas. LeMoyne earned BS and
MS degrees in mechanical engineering from Brigham
Young University, Provo, Utah, USA.
Fred Clarke is Senior Artificial Lift Advisor with
Murphy Exploration & Production in Houston. He is
responsible for artificial lift and pumping unit automa-
tion and is Artificial Lift Superintendent and Team
Leader, developing the Murphy remote operating
center for the company’s Eagle Ford tight oil asset
in South Texas. In his 37 years of oil and gas industry
experience, Fred has worked in a variety of technical
and management roles for manufacturers and E&P
companies in the US, Australia and Canada. Prior to
joining Murphy Oil, Fred worked in Australia with the
country’s largest onshore oil producer, focusing on
artificial lift, variable speed drives and remote opera-
tions. He has authored and coauthored numerous SPE
papers on artificial lift.
Gunnar DeBruijn is the Standards and Knowledge
Development Manager for Schlumberger Well Integrity
(cementing) in Sugar Land, Texas. Since joining
Schlumberger in 1991, he has worked in technical
leadership positions across Canada and North America
and in deepwater projects around the world. His work
supports cementing and well integrity initiatives in
environments from coalbed methane to unconven-
tional steam-assisted gravity drainage, extended-reach
drilling and deepwater wells. His focus is training and
competency development for the worldwide cementing
January 2016
population. As a member of API Subcommittee 10, he
contributes regularly to API standards on well cements
and actively contributes to various API standards rel-
evant to well integrity. Gunnar received a BSc degree
in mechanical engineering from the University of
Alberta, Edmonton, Canada. He is a registered SPE
Professional and a Professional Engineer registered in
Texas and in Alberta and Newfoundland, Canada.
Kevin Docherty is the Schlumberger North American
Land (NAL) Domain Manager in Sugar Land, Texas.
He provides support to the NAL market through
technology deployment, technical support and well
integrity solutions. He began his career in 1992 with
Schlumberger as a field engineer in Aberdeen, where
he was responsible for field operations in cementing.
He has since served as a technical engineer, senior
engineer, field service manager, operations manager,
product champion and technology center manager in
Libya, Azerbaijan, Trinidad, the US and France. His
areas of expertise include all aspects of, and an empha-
sis on, deepwater and HPHT cementing operations.
Kevin has a BEng degree in mechanical engineering
from Teesside University, Middlesbrough, England,
and a Scottish Vocational Education Council advanced
diploma in drilling and production engineering.
Anouar Elhancha is the Schlumberger Well Integrity
Technical Head for East and Central-West Africa.
Currently based in Houston, he has 12 years of experi-
ence in the industry and has held several operations,
technical and sales positions for Schlumberger in
North Africa and the Gulf of Mexico. He has most
recently focused on the integrated multisegment
approach to well integrity on both the planning and
evaluation sides. Since 2013, Anouar has been involved
in the development and field testing of next-generation
well integrity workflows and software.
Steve Grayson is a Petrophysics Consultant in
Camarillo, California, USA. He was unconventional
domain champion for Schlumberger in Ventura,
California, until he retired in 2015. He obtained a
BSEE degree from Rensselaer Polytechnic Institute,
Troy, New York, USA, and started as a logging field
engineer in Michigan, USA, in 1977. In his various posi-
tions in the US and Canada, he has had a focus on log-
ging interpretation in thin beds, fractured reservoirs
and dielectric movability. Steve is a recipient of the
American Association of Petroleum Geologists Pacific
Section H. Victor Church Memorial award and served
as SPWLA Distinguished Lecturer in 2013/2014.
James Hemingway is a Nuclear Petrophysics Advisor
for Schlumberger in Sugar Land, Texas. He started his
career as a field engineer in 1980 and has held various
petrophysical and engineering positions since 1984.
Jim was a 2000/2001and 2011/2012 Distinguished
Lecturer for the SPWLA and is a 2014/2015 SPE
Distinguished Lecturer. James holds a BSc degree
in chemical engineering from Texas A&M University,
College Station, and a BSc degree in chemistry from
Emporia State University, Kansas, USA.
Kyel Hodenfield is President of North America
Artificial Lift Schlumberger in Houston. He is respon-
sible for operations, research, engineering and mar-
keting of the company’s artificial lift product lines.
Previously, he was marketing and technology manager
for the production group and was responsible for
research and engineering development of completion
and production technology. He joined Schlumberger in
1985 as a wireline field engineer and later became vice
president of the Drilling & Measurements Segment in
North and South America and vice president of mar-
keting and technology. While in the reservoir sampling
and analysis division of the Testing Services Segment,
he helped integrate and expand the Schlumberger
acquisitions of Oilphase, DBR and TerraTek. Kyel
earned BS degrees in petroleum engineering and
geological engineering from the University of North
Dakota, Grand Forks, USA.
Damián E. Hryb is a Geomechanics Specialist for
YPF, SA in Buenos Aires. He specializes in dynamic
simulation of subsurface geomechanics and fracture
modeling, which are used to optimize drilling and
completion designs. As an Auxiliary Professor at
Universidad de Buenos Aires, he lectures on finite
element methods. He joined YPF in 2009 after
working seven years on numerical simulations of
industrial processes and developing process optimiza-
tion software for Tenaris. Damián has a mechanical
technician degree from Instituto Politécnico Modelo,
Buenos Aires, and a BSME degree from Universidad de
Buenos Aires.
Dave Jones is President of Rod Lift Consulting
Schlumberger Artificial Lift in Dallas. He began his oil
industry career in 1980 with Norris/Dover Artificial Lift
Company, where he was vice president of sales before
joining Schlumberger. Dave received a BA degree in
business from Texas Tech University, Lubbock.
Slaheddine Kefi is a Senior Development Engineer
in well integrity at the Schlumberger Riboud Product
Center in Clamart, France. He is responsible for the
development and commercialization of the UNIFLAC*,
Ultra LiteCRETE*, LOSSEAL* dual-fiber technology
for oil-base mud and CemPRIME* oilfield cement and
OILSEEKER* stimulation formulations. He began his
Schlumberger career in 1998 in Clamart as a develop-
ment engineer for cementing; he next served as a
development engineer for stimulation in Sugar Land,
Texas, and prior to his current position, as a senior
research scientist for drilling fluids in Cambridge,
England. Slaheddine obtained an MS degree in physics
and chemistry from the Ecole Supérieure de Physique
et de Chimie Industrielles in Paris and a PhD degree
in organic chemistry from Université Paris-Sud,
Orsay, France.
Ines Khalfallah was until recently, a Schlumberger
Well Integrity Technology Product Champion in
Clamart, France, where she led the development and
implementation of new technology marketing. She
began her career in the industry in 2006 as a well integ-
rity field engineer in Algeria, where she led cementing
operations for Sonatrach projects. She then served as
a product development engineer in France, where she
led the Cementing for CO2 Wells project, as a technical
and sales engineer in Kuwait and a district technical
engineer and laboratory manager in Brazil. Ines is
the author of several papers on cementing and is the
holder of two patents. She received a master’s degree
in materials science and engineering and a diplôme
grande ecole in chemistry from the Ecole Supérieure
de Physique et de Chimie Industrielles in Paris.
47
CONTRIBUTORS (continued)
Polina Khalilova, based in Houston, is the Invizion*
Product Champion for Schlumberger. When she
began with Schlumberger in 2003, she specialized in
deepwater and HPHT projects, working with offshore
operators in the Gulf of Mexico. In 2008, she moved
to Brunei to serve as cementing manager; her respon-
sibilities included managing cementing services and
tools for deepwater, HPHT and challenging directional
wells. In 2011, Polina returned to the US, where she
became a senior deepwater cementing specialist; in
this position, she reviewed deepwater cement pro-
grams worldwide and coached the Schlumberger deep-
water cementing population. In 2014, she joined the
Invizion project, which integrates data from all aspects
of well construction phases to minimize unknowns
and evaluate cement placement as a well barrier.
She earned a BS degree in mechanical engineering
from Ufa State Petroleum Technological University,
Russia, and an MS degree in mechanical engineering
from Gubkin Russian State University of Oil and Gas
in Moscow. She has an active professional engineer
license in petroleum engineering in Texas.
Leslie Malone is a Senior Staff Engineer in the arti-
ficial lift group of Murphy Exploration & Production
Company in Houston. She began her career as a facili-
ties and operations engineer for ARCO Oil and Gas in
Hobbs, New Mexico, USA, and later served in Midland,
Texas. Leslie was a production and operations engi-
neer for Wentworth Energy and Vernon E. Faulconer,
Inc. in Palestine, Texas. She has a BS degree in
petroleum engineering from Texas Tech University
in Lubbock.
Mark L. Mitchell is Artificial Lift Marketing and
Design Technical Advisor for Rod Lift Consulting
Schlumberger in Oklahoma City, Oklahoma, USA. He
began his career in 1978 in the upstream oil and gas
industry with Noble Drilling Corporation. Before join-
ing Schlumberger in 2010, he worked in the Rocky
Mountain Region of the US for National Oilwell, Spec
Tech and Norris Sucker Rod. Mark is a member of the
API and the SPE and has participated in numerous
artificial lift seminars.
Laurent Mosse is the Schlumberger Wireline
Petrophysics Domain Champion for unconventional
petrophysics in Buenos Aires. He develops interpreta-
tion methods for unconventional play evaluations in
the region. Before moving to the field of petrophys-
ics in 2011, he spent nine years as a physicist in
Schlumberger research centers developing nuclear
and electromagnetic propagation tools. Laurent
received a diplôme grande ecole in physics from the
Ecole Supérieure d’Electricité and DEA and PhD
degrees in physics, both from the Université Paris-Sud,
Orsay, France.
Lee Norris is a Schlumberger Advisor and Flow
Assurance Consultant in Houston. He joined
Schlumberger in 2012 through the Schlumberger
acquisition of SPT Group, which had developed the
OLGA* dynamic multiphase flow simulator. Prior to
joining the SPT Group, he worked at Exxon Production
Research for 20 years in the areas of fluid flow, process
48
design, engineering thermodynamics and hazard risk
analysis; he began his career in the oil and gas indus-
try in 1970. Lee obtained a BS in mechanical engineer-
ing from The University of Texas at Austin and MS and
PhD degrees in mechanical engineering from Stanford
University, California.
Matthew Offenbacher is a Technical Services
Engineering Manager for the Wellbore Productivity
Fluids business line sustaining group with M-I SWACO,
a Schlumberger company, in Houston. He focuses on
the evaluation and commercialization of new technolo-
gies. He began his career with M-I SWACO in 2005
in a three-year program that focused on projects in
the Gulf of Mexico, Canada, Indonesia and Nigeria.
He then served as a project engineer and technical
services engineer, providing support for drilling and
completion fluids operations in Azerbaijan. He next
served as a technical services engineer in Houston,
where he focused on reservoir drill-in fluids and filter-
cake breakers. Prior to his current position, he was a
business development manager. Matthew holds a BS
degree in electrical engineering from The University
of Texas at Austin.
Nuncio Palermo, based in Buenos Aires, is the
Schlumberger South America Unconventional
Resources Group Vice President, a position he has
held since 2015. He leads a team of production, stimu-
lation, geomechanical, petrophysical and geophysical
advisors who provide technical support to uncon-
ventional operations in South America. He began his
career with Schlumberger Wireline in 1978 and has
held many positions in field operations, line manage-
ment and operations support with Wireline, Testing,
Data Services and Oilfield Services in Argentina, Chile,
Bolivia, Venezuela, Mexico, Cuba, Guatemala and
France. Nuncio earned a bachelor’s degree in electron-
ics from ENET in Buenos Aires.
Stéphane Pichon is a Production Stimulation
Engineer for Schlumberger in Buenos Aires, where
he provides technical support to YPF, SA for the
Vaca Muerta Shale unconventional play. Stéphane
began his career in Schlumberger as a fracturing
field engineer in 2008 and has held technical and
support positions in Peru, Colombia and Argentina.
He has an MEng degree in fluid mechanics from
Ecole Nationale Supérieure d’Electronique,
d’Electrotechnique, d’Informatique, d’Hydraulique et
des Télécommunications, Toulouse, France, and an
MSc degree in petroleum engineering from Imperial
College London.
Ray Ravitz is the Senior Business Development
Manager for M-I SWACO Wellbore Productivity Fluids
in Houston. He is responsible for the technical and
commercial aspects of reservoir drill-in fluids and
internal-external filtercake breaker systems. He began
his career in the industry in 1978 as a mud school stu-
dent for IMCO Services (now M-I SWACO). He served in
several sales and service and project engineer positions
until 1995, when he became the West Coast district
manager. He next served as an area manager for the
same region and then as a senior project engineer for
reservoir drill-in fluids. Ray received a BA degree in lib-
eral arts from Villanova University, Pennsylvania, USA.
Laurence Reynolds, based in Buenos Aires, is
the Principal Petrophysicist in the Schlumberger
Unconventional Resources Group for South America,
where he is involved in the development and inte-
gration of petrophysical answers and workflows for
optimizing field and basin development strategies
through close client support and a multidisciplinary
team approach. He joined Schlumberger in 1986 as
a wireline field engineer and has also worked with
Data and Consulting Services and in the Drilling &
Measurements Segment in Scotland, Brazil, Egypt and
Indonesia. He has been involved in the development
and introduction of services that include seismic while
drilling, deep directional imaging, crosswell electro-
magnetic imaging and distributed temperature sensor
optical fiber. Laurence is a member of the SPE and
SPWLA and an author and contributor to 18 techni-
cal papers. He has taught internal and client training
courses and workshops and has been an SPE applied
technology workshop speaker. He obtained a BEng
(Hons) degree in engineering physics from Queen’s
University, Kingston, Ontario, Canada.
Pavel Shaposhnikov is the Schlumberger North
America Offshore Well Integrity Champion in Houston.
He joined Schlumberger Wireline in 1997 in Raduzhny,
Russia, where he worked as a field engineer. He has
since served in various assignments, including as a
land and offshore openhole and cased hole engineer,
deepwater offshore coordinator and field quality
champion. He was involved in development of Lean
Operations and standardized well integrity evalua-
tion operations in the Gulf of Mexico. Pavel holds an
MSc degree in physics from Saint Petersburg State
Technical University, Russia.
Paul Sheperd is a Schlumberger Well Integrity
(cementing) Engineer in Denver. Since joining
Schlumberger in 2004, he has worked in the cementing
discipline in various technical and managerial posi-
tions in the United States, Norway, continental Europe
and Russia. He has expertise supporting cementing
and well integrity initiatives in a wide range of operat-
ing environments including deepwater, arctic, uncon-
ventional and HPHT environments. His current focus is
on improving well integrity in the US Rocky Mountains
by closing the engineering feedback loop between
cement evaluation logs and cement design. Paul
earned a BS degree in chemical engineering from the
University of Notre Dame, South Bend, Indiana, USA.
Mariano Suarez is a Reservoir Engineer with YPF, SA
in Buenos Aires. His primary areas of responsibility are
building simulation models for unconventional reser-
voirs and monitoring operations related to predicted
well dynamics. He joined YPF in 2009 after working as
a process engineer for CH2M Hill. Previously, he was
a process engineer with Fluor Daniel in Buenos Aires.
Mariano has a BS degree in chemical engineering from
Universidad de Buenos Aires.
Oilfield Review
Salim Taoutaou is the Cementing and Well Integrity
Technical Advisor for Schlumberger in Clamart,
France. He manages global development of the well
integrity cementing domain strategy, providing opti-
mal well integrity cementing solutions for clients.
Through his 18 years in the oil and gas industry, he
has held various positions in the North Sea region,
the Middle East and Asia. He has authored more
than 43 international journal and conference papers,
is the holder of two patents and was the recipient of
the SPE 2014 Asia Pacific Regional Technical Award
in Drilling Engineering. He earned a master’s
degree in mechanical engineering from Guelma
University, Algeria.
Coming in Oilfield Review
Log Measurement Integration and Defined
Workflows. In the exploration for hydrocarbons, E&P
companies rely on a range of technologies to evaluate
prospects. Numerous tools and techniques are avail-
able, and workflows used by operators and service
companies alike to analyze well data may not be
created with new measurements in mind. This article
reviews integration and workflow techniques that
are designed from the outset to successfully utilize
new technologies and take advantage of advanced
measurements. These workflows are targeted to those
who can best use the information from new technolo-
gies and are intended to help E&P companies utilize
their assets more effectively.
January 2016
Gioconda Tovar is a Schlumberger People Planning
Manager in Operations Planning in Houston; she has
been in the position since July 2015. Prior to that,
she served as Invizion Product Champion from 2012
to 2015. She began her career with Schlumberger
in 1997 as a field engineer trainee in Venezuela and
consequently served as a field engineer until 2000.
Following a position as a cementing technical leader,
she then served as a Schlumberger recruiter, person-
nel manager, service delivery manager and training
center manager in Venezuela, Ecuador, and the US.
Well cementing, human interface design in automated
control systems for gas plants and chemical and pro-
cess engineering are among her areas of expertise.
Gioconda obtained a bachelor’s degree in chemical
engineering from Universidad Central de Venezuela
in Caracas.
An asterisk (*) denotes a mark of Schlumberger.
Improving Drilling Efficiency While Sliding Improving Exploration Success. Seismic imaging
Operations. Although mud motors have proven their cannot resolve many potential exploration targets
worth in attaining azimuthal control of horizontal drill- lying beneath shallow rock layers. The overburden
ing operations, their use can also result in noticeably behaves as a defective lens, distorting seismic imag-
slower rates of penetration. However, a surface- ing of deeper geologic structures. As a result, targets
mounted torque-rocking system can give operators appear indistinct, distorted, mislocated or, in extreme
significant gains in rate of penetration and azimuthal cases, completely obscured. New developments in
control while decreasing stalls in downhole mud multimeasurement marine seismic acquisition and
motors. This automated surface rotation control sys- imaging are making it possible to compensate for the
tem also helps directional drillers decrease tortuosity distortions, sharpen images of deeper targets and
and extend the reach of horizontal wells. reduce the uncertainty of seismic information about
drilling prospects.
Production Strategies in Unconventional
Resources. Operators have learned that production
from hydraulically stimulated shale wells follows a
reliable pattern of initially high production rates that
decline rapidly soon after going online. Today, some
service companies and operators are questioning the
inevitability of this pattern and are testing the roles
current drilling, completion and production practices
play in shale well deliverability.
49
THE DEFINING SERIES
Geomechanics
John Cook
Scientific Advisor
Geomechanics is the study of how soils and rocks deform, sometimes to fail-
ure, in response to changes of stress, pressure, temperature and other envi-
ronmental parameters. In the petroleum industry, geomechanics tends to
focus on rocks, but the distinction becomes blurred because unconsolidated
rocks can behave like soils.
Geomechanics is relatively young as a science and even younger in its
application to the petroleum industry. However, it applies to nearly all aspects
of petroleum extraction from exploration to production to abandonment and
across all scales, from as small as the action of individual cutters on a poly-
crystalline diamond compact (PDC) bit through drilling wells and perforating
to as large as modeling fields and basins. Over the last 30 years, geomechanics
has come to play an increasingly important role in drilling, completion and
production operations. This trend continues as operators pursue oil and gas
production from shales, in which mechanical anisotropy—the variation of
mechanical properties with orientation—plays a vital role.
At the wellbore scale, geomechanics is central to understanding how
drill bits remove rock, characterizing borehole stability, predicting the sta-
bility of perforation tunnels and designing and monitoring hydraulic frac-
turing stimulation programs. At the reservoir scale, geomechanics helps
model fluid movement and predict how fluid removal or injection leads to
changes in permeability, fluid pressure and in situ rock stresses that can
have significant effects on reservoir performance. Engineers use geome-
chanical modeling to predict and quantify these effects for life-of-reservoir
decisions such as placing and completing new wells, enhancing and sustain-
ing production, minimizing risk and making new investments.
Choosing the correct bit type and design for optimal rate of penetration
and bit life is vital for drilling cost-effective wells. The geomechanics of rock
destruction under the drill bit is complex because of high strain rates and
temperatures, multiple deformation mechanisms and interactions between
the bit, drilling fluid and formation. Many ad hoc approaches to understand-
ing and improving drillbit performance have been taken, and interpretation
methods such as mechanical specific energy—the energy used to remove a
unit volume of rock—have been used since the 1960s to relate drilling per-
formance to rock strength. Recent advances in research methods related to
geomechanics are starting to reveal in more detail the factors, such as the
balance between crushing, disaggregation and brittle cracking or chipping,
that control the loads on the cutter, its wear behavior and the nature of the
rock debris (Figure 1).
Geomechanics also plays a major role in understanding the stability
and integrity of the borehole while drilling. The presence of the wellbore
and the pressure of the drilling fluid induce changes in the stress state or
in the rock. As a result, the rock around the borehole may fail if the redis-
tributed stresses are greater than the rock strength. Tensile cracking
occurs if the mud pressure becomes too high and causes the borehole wall
to go into tension. Breakouts, which occur if the mud pressure becomes too
low, are regions of enlargement on opposing sides of a borehole, where
shear cracking has occurred and the broken rock has been removed by the
Oilfield Review 28, no. 1 (January 2016).
Copyright © 2016 Schlumberger.
50
Approximately 10 mm
Figure 1. Cutting rock. The action of a PDC cutter plowing through
sandstone (orange) is shown in a cross section that was imaged using
microCT scanning. The location and direction of movement of the tungsten
carbide (gray) and diamond (black) cutter are shown schematically. The
section displays the deformation—crushing, disaggregation and brittle
cracking—induced by the cutter.
bit, stabilizers or mud flow. These failures can lead to stuck pipe, lost cir-
culation and other drilling problems but can also be a valuable source of
information about stress magnitudes and orientations.
By computing the stress changes around a planned borehole and com-
paring them to the strength of the rock, engineers can generate a mud
weight program for the well. Typically, the mud weight must be high enough
to suppress shear failure and fluid influx and low enough to avoid tensile
failure and lost circulation. Although the mathematical techniques for cal-
culating the stresses and the failure criteria are established and routine,
the links between rock failure and drilling problems are not so well estab-
lished. Furthermore, some additional failure modes, such as bedding plane
failure, cannot yet be reliably predicted (Figure 2). Consequently, effective
wellbore stability control, especially along challenging well trajectories,
needs real-time monitoring of wellbore conditions as well as predrill predic-
tion of the required mud pressure program.
A challenge for geomechanics modeling and prediction is the availability of
input data—primarily rock strength and in situ stresses. Rock strength is eas-
ily measured on core samples in the laboratory, but the process is time-con-
suming and costly, and the results usually help with future wells rather than
the current one. Consequently, considerable effort is spent deriving rock
strength values from wireline, LWD and sonic data. The trade-off is less accu-
racy but higher spatial coverage along the well than is available from core data.
These data are interpolated or extrapolated to cover sections of interest
in new wells or are used to improve predictions for the current well. The
same data may also be used in geologic burial history models for construct-
ing vertical profiles of the in situ stresses, which then are compared with
and calibrated to discrete stress measurements in a well. More recently,
advanced sonic tools have allowed estimation of rock strength and some
components of the in situ stress for input into geomechanics models.
Events such as induced seismicity in the 1970s at the Rangely field,
Colorado, USA, and compaction and subsidence in the 1980s at the Ekofisk
field, North Sea, offshore Norway, helped engineers realize the role of
geomechanics at the reservoir scale. Examining geomechanical changes on
this scale is routine now, thanks to the development of finite element analysis
Oilfield Review
programs that have been optimized for geologic structures and rock mechani-
cal behavior (Figure 3). Populating these models with rock data can be a chal-
lenge, but since the computation grid is coarse, it can be done using seismic
data. Once the model is populated, the mechanical response of the reservoir
and overburden can be estimated for a variety of operations—including pro-
duction, injection and fracturing. The model can be calibrated or refined with
repeat, time-lapse seismic surveys and the addition of data as new wells are
drilled. Operators can use this type of information to estimate the injection
pressure used in fracture stimulations that risk a breach in a reservoir seal, or
they can predict the fracture gradient after a period of production, allowing
safe and effective drilling of infill wells.
Reservoir stimulation by fracturing, one of the first applications in the
oil field to use geomechanics methods, is still a major development area.
Exploitation of shale reservoirs has caused a surge of interest in the
mechanical anisotropy of rock, which was not widely appreciated until
about 2000. To make improved predictions of fracture geometry and growth,
models for stress and strength and interpretations of sonic and resistivity
measurements must be modified to account for anisotropy. Advances in
sonic logging tools and interpretation have made this possible.
One feature common to all of these areas is the mechanical earth model
Figure 3. Coupled simulation. A geomechanics model of a reservoir
(MEM), which is a collection of the data needed to make quantitative and (background) is shown with the seismic dataset (left) used to build it. Data
qualitative predictions of the subsurface geomechanical environment. for refinement of the model were acquired from several wells (colored
These data include the stresses in the Earth, pore pressure, rock elastic lines), which were used to model changes over time from events such as
depletion. Several faults (colored planes) and the top of the reservoir unit of
properties, strength and fabric and nonnumerical data such as the presence interest (brown surface) are shown. The edges are the model boundaries or
of intense natural fracturing. An MEM can be simple or complex, be large or are reservoir discontinuities caused by faults. Geomechanics models may
be coupled with reservoir simulation models to predict how production will
affect reservoir performance, well integrity or caprock integrity.

small and be 1D, 2D, 3D or 4D—three spatial dimensions plus time—

according to the complexity of the field and phenomena of interest. The
most important defining feature of an MEM is that its data are related to the
rocks that are being drilled, fractured or otherwise affected by field opera-
tions, rather than a particular well or set of wells. A second feature is that it
is designed to be updated as new data become available from ongoing opera-
tions. Data sources for an MEM include any that give information on stress
10 mm and mechanical behavior; such sources include wireline and LWD logs,
cores, cavings and cuttings, regional geology and all types of seismicity.
Ongoing challenges for geomechanics include improvements to:
• sources of data for predicting in situ stress and rock properties
• the use of anisotropic information for predicting deformation during
exploitation of unconventional resources
• the treatment of fault and fracture displacements within numerical models.
In addition, to help improve the application of geomechanics to various sec-
tors in the industry, engineers must have a better understanding of the rela-
tionships between rock failure and operational failure for wellbore
Bed instability and sand production.
ding
plan
e or Geomechanics in the oil field has come a long way from its early days
ient
atio
n as an adjunct to sonic logging. It is recognized as an important part of
nearly all aspects of petroleum extraction and has been crucial in improv-
ing efficiency and driving down costs. The application of geomechanics in
Figure 2. Unpredictable wellbore failure. A laboratory model shows bedding new reservoir types and mature ones and its integration into operators’
plane failure in a hole drilled parallel to bedding in a fissile shale. The stress workflows, along with the introduction of new measurements and tech-
applied to the sample is the same in all directions, in spite of the directionality niques, will ensure its continuing role in the industry. From here, its oper-
of the failure. This geometry is similar to the geometry of failures observed
ational impact will only grow. The application of geomechanics for
in the roofs of mines in fissile rocks, although the driving force in mines is
gravity rather than in situ stress. The borehole pressure needed to prevent revitalizing mature fields is imperative and will affect activities such as
this type of rock failure cannot currently be reliably estimated. infill drilling, compaction mitigation and refracturing.

January 2016 51
THE DEFINING SERIES
Mud Logging
Matt Varhaug
Senior Editor
Since its commercial introduction in 1939, the mud logging unit has become
a hub for monitoring formation responses to the drilling process. Initially,
the mud logger’s mandate was to record the depth and describe the lithol-
ogy of formations encountered by the drill bit then determine whether those
formations contained hydrocarbons. However, the scope of mud logging has
expanded as additional sensors brought more data into the logging unit—
such as gas chromatographs, weight-on-bit and mud pit level indicators.
Basic mud logging services now typically track drilling rates, lithology,
visual hydrocarbon indicators, total combustible gas in mud and individual
hydrocarbon compounds in the gas along with numerous drilling parame-
ters. The mud logger monitors and evaluates a broad range of surface indi-
cators to compile a concise record of subsurface geology, hydrocarbons
encountered and significant drilling events. These days, the term surface
logging is sometimes used to encompass a range of enhanced mud logging
services that incorporate advanced sensor and computing technology to
provide monitoring for wellbore stability and early kick detection .
The practice of mud logging relies heavily on the mud circulation system.
High-pressure mud pumps draw mud, or drilling fluid, from surface tanks and
direct it downhole through the drillpipe (Figure 1). The mud exits the drill-
string through nozzles in the bit. As a bit drills through the subsurface, the
rock it grinds—along with water, oil or gas in the formation—is carried back
up the hole by the drilling fluid. Upon reaching the surface, the fluid exits
through a flowline above the blowout preventer and is deposited over a vibrat-
ing mesh screen at the shale shaker, which separates formation cuttings from
the liquid mud. The liquid portion of the drilling fluid falls through the screens
to the mud pits, ready to be pumped back into the well; the rock cuttings on
the shaker screen provide the basis for determining downhole lithology.
Depth of Samples
The type of material flowing over the shaker and the timing of its arrival are
fundamental to the mud logging process. To characterize the lithology and
fluid content of a particular interval, the mud logger must account for the
transport velocity of the cuttings to determine the time it takes cuttings to
travel from the bit to the shaker. This lag time increases as depth increases,
taking just a few minutes while the upper section of a well is drilled but
extending to several hours in deeper sections. Lag time, a function of depth
and mud pump rate, is usually measured in terms of pump strokes, which
are counted by a pump stroke counter at the mud logger’s console.
The lag time dictates when formation cuttings from a given depth will
arrive at the shaker. Lagged cuttings samples are collected at regular depth
intervals—typically every 3 m [10 ft] or 10 m [30 ft] of drilling—and prior
to tripping out of the hole. Lagged samples are also collected to examine
changes in formation characteristics, as indicated by significant changes in
drill rate or gas curve trends.
Sample Analysis
Inside the logging unit, the mud logger rinses and dries cuttings samples
before examining them under a binocular microscope. The mud logger
describes each sample in terms of lithology, color, grain size, shape, sorting,
Oilfield Review 28, no. 1 (January 2016).
Copyright © 2016 Schlumberger.
52
Standpipe
Kelly Flowline
Gas trap
Mud logging Rotary table
unit Shale
Drill floor Suction line shaker
Bell nipple
Blowout
preventer
Mud Shaker
pump pit
Casing Suction Reserve
pit pit
Drillpipe
Bit
Figure 1. Monitoring the formation. Drilling mud, pumped downhole through
the center of the drillpipe, enters the open borehole and carries away
formation cuttings and fluids as it moves upward to the surface. There, the
mud, formation fluids and cuttings are diverted through a flowline to the
shale shaker. Gas is separated from the mud, and a suction line siphons the
gas to the mud logging unit for analysis. Cuttings are sampled at the shaker
screen and examined inside the unit. When these evaluations are combined
with drilling parameters—drill rate, pump rate, pump pressure, weight on bit,
mud properties and other inputs—the mud logger can identify potential pay
zones and ascertain how the formation is reacting to the drilling process.
porosity, texture and other characteristics relevant to rock type. This infor-
mation is plotted in the lithology column of the mud log, which displays an
estimate of gross lithology as a percentage of cuttings, reported in 10%
increments. Because the presence of hydrocarbons may not be obvious—
even under a microscope—each sample is examined for fluorescence under
ultraviolet (UV) light.
Fluorescence can be an extremely sensitive indicator of the presence of
hydrocarbons in drill cuttings. Sample fluorescence is evaluated in terms of
color (ranging from brown to green, gold, blue, yellow or white), intensity
and distribution. Fluorescence color may indicate oil gravity; dark colors
are suggestive of low API gravity heavy oils, and light colors indicate high
API gravity light oils. Following application of a solvent on the samples,
hydrocarbon fluorescence will appear to flow and diffuse into the solvent as
the oil dissolves. This diffusion is known as cut fluorescence, or more com-
monly just cut. Under UV light, hydrocarbons may be seen to stream from
the rock pores into the surrounding solvent, turning the solvent cloudy.
To measure gas, the mud logger relies on an automated gas detection sys-
tem. Suction lines transport a constant stream of air and gas from the gas
trap, located at the shale shaker, to the logging unit. There, sensitive instru-
ments process the gas samples extracted from the drilling mud. The primary
gas measurement tool is a flame ionization detector (FID), which can sense
hydrocarbon gas concentrations as low as 5 parts per million. From FID mea-
surements, a total gas curve can be plotted on the mud log. Background
gas—a more or less constant, minimum level of gas—establishes a baseline
Oilfield Review
on the total gas plot. A gas show is any
Clay Shale Sandstone Sand Limestone Dolomite Anhydrite Coal
significant increase in detected gas,
FG Formation gas CG Connection gas TG Trip gas Oil and gas Oil Gas Bit change, trip Shoe
which is usually associated with a zone of
Chromatograph, ppm
increased porosity or permeability. ROP
C1 C2 C3
ft/h Depth, Cuttings, Total Gas, units Mud Weight,

Fluorescence
For more detailed hydrocarbon analy- 100 50 0 ft % 0 125 250 375 500
iC4 nC4 ppg
Lithological Description
and Notes
iC5 nC5
sis during shows, the mud logger employs 0.5k 1k 1.5k 2k 2.5k

1,000k
100k
a gas chromatograph. The chromatograph

100

10k
10

1k
7,500 Sandstone: Clr-lt gy-frst, m-f gr,
separates the gas stream into fractions sbelg-elg, sbang-sbrnd, m srt, tr Glau,
calc mtx, p-m cmt, qtzc i/p, m ind,
according to molecular weight. Commonly fri-m hd, p-fr intgran por, no fluor.

detected components fall within the 9 7/8-in. casing set at 7,580 ft MD/
6,691 ft TVD. LOT = 14.8 ppg.
In: 10.8
alkane group: methane [CH4]—denoted Out: 10.8
Clay: Lt brn-tan, arg, calc, plas,
sft, sol, slty, rthy, grty.
7,600
as C1—as well as the following constitu- Shale: Lt gy-lt brn, grnsh gy, arg, calc,
frm-hd, occ sft, p-m cpt, sbblky-blky,
ents: ethane [C2H6] or C2, propane [C3H8] Trip for new bit Increase splty-ppy i/p, rthy, grty.
TG: 1,548 U
MW to 11.3
or C3, the normal and isopolymers of 6.3 bbl gain
Sandstone: Clr-lt gy-frst, m-f gr,
butane [C4H10] or nC4 and iC4 and pentane sbelg-elg, sbang-sbrnd, m srt, tr Glau,
7,700 Increase calc mtx, p-m cmt, qtzc i/p, m ind,
CG: 35 U
[C5H12] or nC5 and iC5. The measurement MW to 11.5 fri-m hd, p-fr intgran por, no fluor.

of these light hydrocarbons helps geolo- CG: 39 U Increase Shale: Lt gy-lt brn, grnsh gy, arg, calc,
MW to 11.7 frm-hd, occ sft, p-m cpt, sbblky-blky,
gists characterize reservoir fluid composi- CG: 45 U splty-ppy i/p, rthy, grty.

tion while drilling. The quantity of gas Clay: Lt brn-tan, arg, calc, plas,
7,800 sft, sol, stky i/p, rthy, grty.
recovered and the ratios of the various
Sand: Clr-frst, trnsp-trnsl, m-c gr,
gases are useful in identifying zones of WOB 38 to 53 klb occ f, sbelg-sbsph, ang-sbang, m srt,
RPM 78 to 84 tr Glauc, uncons-p cmt, p ind, lse, n por,
producible oil or gas. Flow 650 gpm Qtz, no fluor.

FG: 427 U
Clay: Lt brn-tan, arg, calc, plas,
7,900
Coordination with the Drill Floor Increase
sft, sol, stky i/p, rthy, grty.
Sand: Clr-frst, trnsp-trnsl, m-c gr,
Gas monitoring is also important to the FG: 920 U
MW to 12.0 occ f, sbelg-sbsph, ang-sbang, m srt,
tr Glauc, uncons-p cmt, p ind, lse,
driller and company representative. Qtz, no fluor.

Mud gas trends that develop while drill-

8,000
ing are integral to the evaluation of mud
balance and identification of potentially Figure 2. Excerpt from a basic mud log. A mud log typically displays ROP, depth, cuttings lithology, gas measurements
overpressured formations. By carefully and cuttings descriptions along with notes on mud rheology or drilling parameters. This log documents fairly routine
tracking gas and drilling parameters, drilling. Casing was set in a shaly interval at 7,580 ft. After drilling out of casing, the driller ran a leakoff test (LOT).
Drilling ROP was about 25 to 30 ft/h [7.6 to 9 m/h]. A trip for a new bit at 7,650 ft resulted in 1,548 units of trip gas
the mud logger can recognize deviations
(TG). During drilling at near-balanced conditions, small increases in connection gas (CG) were observed
from normal trends and give advanced following each connection, prompting the driller to raise the mud weight. An increase in ROP at 7,890 ft signified a
warning so the driller can mitigate drilling break, which was accompanied by increasing sand content and a gas show, which reached a peak of
impending problems. Thus, the success 920 units of formation gas (FG).
of a well and the safety of the drilling
operation may hinge on how quickly a mud logger can synthesize and inter- logs from offset wells and help the operator track the bit’s position in relation
pret myriad pieces of data. to target formations. Because the mud log is based on physical samples, it can
A sensor mounted on the drawworks tracks the drill rate, or rate of pen- provide a direct, positive identification of lithology and indication of hydrocar-
etration (ROP), to determine the amount of time spent drilling each meter bon content. This information can be especially useful when formation char-
or foot of depth. The mud logger’s role takes on added importance when a acteristics make wireline or LWD log interpretation complicated or ambiguous.
drilling break, or significant increase in ROP, is encountered. Then the mud The mud log provides independent evidence for a more comprehensive under-
logger alerts the company representative to request that drilling be stopped standing of reservoir conditions and geology.
until mud and cuttings from the bit face can be circulated to the surface. If Advances in computing and networking systems, surface sensor design
these cuttings are accompanied by an increase in gas, or if sample analysis and sample analysis are bringing mud logging technology into the 21st cen-
reveals the presence of oil, the mud logger notifies the company representa- tury. Today, even more sensors lead into the logging unit, each acquiring
tive and geologist of a show of gas or oil. The operator then has the option to data at a frequency of several times per second. To handle this increase in
further evaluate the potential pay zone through coring or testing. data rate and volume, a context-aware processing system—based on com-
The mud log serves a variety of functions (Figure 2). The ROP curve is puter-generated trend lines and a library of established models—makes the
plotted as a step chart or a continuous line, increasing from right to left. When data easier for the mud logger and other end users to evaluate. Digital
displayed in this manner, the ROP curve responds to changes in rock type or images of samples viewed under the microscope can be rapidly transmitted
porosity in a manner similar to that of a spontaneous potential or gamma ray from the wellsite to the client office. New approaches to gas sampling and
curve, making for easy correlation between LWD or wireline curves. As a cor- analysis have been developed to extract geochemical properties at the well-
relation tool, the mud log’s ROP and total gas curves often exhibit a remark- site. The mud logging unit has long been a hub for monitoring drilling
able correspondence to gamma ray and resistivity curves, respectively. operations; its role as a source of crucial information for the company rep-
Throughout the drilling process, mud logs provide real-time correlations with resentative, the driller and the geologist continues to evolve.

January 2016 53
THE DEFINING SERIES
Gas Lift
Rick von Flatern
Senior Editor
Artificial lift systems are designed to help natural reservoir energy power
formation fluids to the surface at targeted rates. Gas lift systems aid or
increase production by injecting high-pressure gas from the casing annulus
into fluids that have entered the production tubing from the formation. The
injected gas reduces the fluid density and, thus, the hydrostatic pressure of
the fluid, allowing in situ reservoir pressure to lift the lightened liquids.
The technical applicability and economic viability of gas lift installations
are determined by two factors: gas availability and compression costs. In the
majority of gas lift wells, nearby wells produce enough gas to supply the
system, and after the fluids have been lifted to the surface, the gas can be
separated from the liquids and returned to the casing annulus to maintain
required gas volume and pressure. When circumstances permit, the indus-
try also uses natural, or auto, gas lift systems, which is highly cost effective
because it eliminates the need for compressors, pipelines or a separate
source of natural gas.
System Design
Engineers design gas lift systems by first calculating the production potential
of each well in the network. Then, based on available gas pressure and volume,
they assign each well its optimum production rate and gas lift allocation.
An ideal gas lift system is one in which gas is injected into the fluid col-
umn at a continuous rate and at a constant pressure. This process ensures a
stable liquid flow rate from the reservoir and is possible only in fields in
which sufficient volumes of high-pressure gas are available and liquids can
flow easily through the formation to the wellbore.
Engineers must also construct wells to accommodate the type of injec-
tion system to be used. Gas may be injected into the fluid column through an
open system that has no seal between the tubing and casing annulus or uses
a standing valve in the tubing to isolate the casing annulus from the produc-
tion tubing. However, the most common gas lift well configuration includes
a packer and gas lift valves.
Designing a gas lift system that optimizes production is a complex chal-
lenge. Engineers must account for the interaction of all parts of the produc-
tion system; the potential, constraints and needs of each well must be
considered individually along with those of the network as a whole. Flowline
and downhole tubular sizes and lengths, processing equipment, gas and
compressor availability, fluid composition and other factors impact gas lift
efficiency and production.
Today, designers may address these complex calculations using optimi-
zation software programs that model the production network (Figure 1).
These programs allow the engineer to optimize the system based on a field
economic driver such as total liquid production, total gas produced or total
oil produced and to test the effects of various injection rates.
Because field and individual well conditions are constantly changing,
operators may mate computer optimizing capabilities to real-time data. They
can then adjust injection rates to match changing conditions and quickly
identify and take steps to correct underperforming wells or gas lift systems.
Oilfield Review 28, no. 1 (January 2016).
Copyright © 2016 Schlumberger.
For help in preparation of this article, thanks to Jason Jones, Tyler, Texas, USA, and
Eric Lovie, Singapore.
54
15,000
Well 1
Well 2
Well 3
10,000
Oil production, bbl/d
Well 4
Well 5
5,000 Well 6
Well 7
0
0.5 1.5 2.5 3.5 4.5 5.5 6.5
Injection gas rate, MMcf/d
Figure 1. Gas lift optimization. Every well in a field or network of wells has
an optimal gas lift operating point (black diamonds) at which it will produce
the most fluid. Using gas lift optimization software helps system designers
set an optimal injection gas rate for each well across the network of wells.
Rates are based on gas availability and formation fluid production potential.
Optimal production for each well in a network therefore may be lower than
its potential.
Gas Lift Valves
Gas lift valves are the means by which operators adjust the rate of gas injec-
tion into the liquid column in the production tubing. Check valves within the
gas lift valves allow flow in only one direction—from the casing annulus into
the production tubing. For maximum efficiency, gas lift valves are staged as
deeply in the well as possible; setting depth is limited by available injection
pressure. Injection pressure–operated gas lift valves are designed to open
typically in reaction to a specified gas pressure in the casing annulus.
Gas lift valves may also be placed at various depths between the bottom-
most valve and the surface. These valves are used primarily at the start of
production to reduce the density of the fluid that has risen above the pri-
mary point of gas injection at the deepest set valve. This process is referred
to as well kickoff or well unloading. Typically, the operating valve is the
bottommost valve.
Production pressure–operated gas lift valves, placed at various depths
in the well, may open in response to a preset level of pressure exerted by the
produced fluid column. They remain closed unless the well experiences an
increase in fluid in the tubing, at which point they open to assist in lifting
the excess fluid from the well.
Engineers set conventional gas lift valves in mandrels that are part of
the tubing string; such valves may be retrieved only by pulling the tubing
string. Valves may also be placed in side pocket mandrels, which provide
communication between the annulus and the production tubing and a lock-
ing mechanism to hold the valve in place. These valves may be set in and
retrieved from the mandrels via slickline or coiled tubing tools (Figure 2).
Seals on the valve are matched to a polished bore on the inner wall of
the mandrel; the seal and the polished bore create an isolated pathway for
the gas to flow through the valve and out into the production string. Gas
exits the valve through an orifice that is sized to produce the desired gas
flow rate into the production tubing. To obtain the steady and constant fluid
flow rates that are essential to an optimized gas lift system, the gas must
enter the production string at critical velocity.
Critical velocity is generally defined as the minimum gas velocity in the
production tubing required to move fluid droplets upward. Below the criti-
cal velocity, droplets fall, and liquids accumulate in the wellbore. In addi-
tion, when the rate falls below critical flow, small changes in downstream
Oilfield Review

A B C D
Slickline
kickover tool
Side pocket
mandrel
Gas lift
valve
Figure 2. Setting and retrieving gas lift valves. Gas lift valves are run into the
well and set in side pocket mandrels using a slickline or coiled tubing tool
attached to a kickover tool. The kickover tool contains a spring mechanism,
which is locked into place while the valve is being run into the well. The
slickline or coiled tubing operator runs the tool past the side pocket mandrel (A)
and then reverses direction (B), which unlocks the spring. When the bottom
of the valve is pulled across the mandrel opening, the spring forces it into
the opening. The operator then reverses direction again and the valve is
lowered into the mandrel (C). The operator then pulls upward, which locks
the valve in place and releases the setting tool from the valve (D). The
operator can reverse the process to retrieve the valve using a gas lift valve
pulling tool attached to the kickover tool.
pressures can cause large changes in flow rate that the system may inter-
pret as a need for more gas, which may lead to oscillations in pressure and
production rates. Above critical flow, the system cannot detect changes in
flow downstream of the valve thus will not respond to the situation by
increasing gas flow.
Gas Lift Variations
Gas lift systems are either continuous or intermittent flow. The majority of
wells use continuous flow gas lift because it supplements in situ gas, which
results in a natural, steady flow of formation fluids to the surface. Continuous
gas injection requires a reliable source of high-pressure gas throughout the
life of the well. However, in many cases, in situ gas sources decline before
the zone is fully drained, forcing operators to turn to an outside source to
maintain lift volume.
January 2016
Intermittent flow gas lift periodically injects gas into the production string
to displace slugs of fluid. The high-pressure gas expands rapidly after entering
the tubing, which forces the fluid to the surface. In contrast to the steady flow
created by continuous flow gas lift, intermittent flow results in slug flow,
which may cause gas- and fluid-handling difficulties on the surface. In addi-
tion, the extreme pressure changes caused by the sudden injection of gas into
the production tubing creates pressure surges near the producing zone, which
may cause the onset of sand production in some formations.
Experts recommend the application of continuous flow gas lift for high-
volume wells that have high-static bottomhole pressures (BHPs) and offshore
wells that have strong waterdrive or for formations being produced via water-
flood that have a high productivity index (PI) and high gas/oil ratios (GORs).
Intermittent flow gas lift systems are typically not recommended for wells
producing more than about 30 m3/d [200 bbl/d] or in wells that have high BHP
but low PI or low BHP and high PI. As a consequence, many intermittent flow
systems are deployed in depleted wells that were first using continuous flow
systems and on gas wells that have begun to produce water.
Auto gas lift systems, also referred to as natural or in situ gas lift, have
specific requirements and thus are used less commonly than traditional sys-
tems. Instead of using gas pumped down the casing annulus, these systems
are supplied by reservoir gas caps or noncontiguous gas-bearing formations.
Completion engineers design auto gas lift wells to include a completion design
across the gas cap or gas zone that allows gas to be flowed into the production
tubing. Operators control the gas flow rate with a specifically designed auto
gas lift valve that has a flow area that can be adjusted from the surface.
Operators may use an auto gas lift system in the same way they would a
more traditional continuous or intermittent system. But, because they elim-
inate certain equipment requirements, auto systems provide an advantage
in capital savings and reduce weight on offshore platforms by eliminating
the need for a compressor. Auto gas lift systems also allow the operator to
produce associated gas without recompleting the well. And, because orifice
openings in auto gas lift valves can be adjusted remotely, operators can con-
trol gas injection rates to accommodate changes in formation liquids pro-
duction without an intervention.
The Case for Gas Lift
Gas lift systems offer several benefits not available in other artificial lift
systems. Gas lift valves set in side pocket mandrels may be easily and inex-
pensively removed and be replaced without a workover rig.
Gas lift side pocket mandrels are situated such that the pockets contain-
ing the valves are offset from the centerline of the tubing, which keeps the gas
lift valves from the fluid flow and thus protects them from solids. Consequently,
gas lift technology is a good option for hydraulically fractured wells or for wells
completed in unconsolidated formations that routinely produce sand. Gas lift
systems are applicable in high-angle, low-productivity and high-GOR wells.
Intermittent gas lift systems can also operate efficiently in low-pressure,
low-productivity wells. As a consequence, to capture final production at the
end of a well’s life, operators may replace other artificial lift systems with
gas lift systems until the bottomhole pressure becomes too low to lift the
target fluid volumes, at which point operators typically switch AL systems to
rod pump or electrical submersible pump systems.
Gas lift is an important artificial lift technique. When a reliable source of gas
is available, gas lift systems, whic are highly reliable, flexible and robust, can
operate for the life of the well. The ability to change valves allows operators to
adapt to production changes over the life of the well. As a consequence, gas lift
is often the artificial lift method of choice for use in deepwater subsea wells in
which continuous operation in high-pressure, high–flow rate wells is critical.
55
LOOKING BACK
Looking Back on Artificial Lift
Recent Oilfield Review articles on the use
of artificial lift systems in unconventional
resources inspire a look at the origins of
these methodologies.
The earliest artificial lift systems were sucker rod
pumps that were created in the second half of the
19th century, when prospectors harnessed the drill-
ing rig’s steam-driven oscillating beam—initially
used to lower and raise the cable tool for drilling—
to a long wooden rod that had a plunger attached to
its bottom.1 As the rod was lowered and raised, the
plunger pushed the oil up the tubing to the surface.
The first attempt at using gas lift came in 1864,
when engineers pumped compressed air to the bot-
tom of the well through a tubing string. However,
the mixing of air with the formation fluids was
difficult to control, which made the system highly
inefficient. In 1892, US engineer Julius Pohlé of
New Jersey patented a method of introducing the
air in controlled stages. In 1911, Union Oil began
using produced natural gas rather than air to lift
production. Because the gas was naturally high-
pressured, it was cheaper to compress and would
not absorb the lighter components of oil and, as a
result, become a safety hazard at the surface. The
use of natural gas also eliminated casing and tub-
ing corrosion that resulted from the presence of
oxygen in the compressed air.
In 1932, engineers at Jordan & Taylor Inc. of Los
Angeles had the idea of installing multiple gas lift
valves in the same tubing string at intervals, making
it easier to prime a well. Later that decade, Jeddy
Nixon of the Wilson Supply Company in Houston pio-
neered the actuation of gas lift valves by wireline.
Innovation Brings Electric Submersible Pumps
As wells were drilled to greater depths, sucker rod
pumps became unwieldy and gas was frequently
unavailable for gas lift systems. The electric sub-
mersible pump (ESP) emerged as a viable alter-
native during the 1920s. Such technology can be
traced back to one individual: Armais Arutunoff.
At just 18, he founded the Russia Electrical
Dynamo of Arutunoff Company, producing the first
electric pumps that could be submerged in water.
Arutunoff, with sucker rod salesman Samuel
van Wert, then adapted the pump design for
the oil field. Successful but expensive tests in
56
Baldwin Hills, California, USA, put both partners
in debt. Van Wert attended the 1926 API meeting
in California and encountered Clyde Alexander,
of Phillips Petroleum, who was looking for a high-
In March 1928, the first two ESP systems were
successfully tested in a Phillips well.
Modern sucker rod pumping unit. The sucker rod
pumping unit evolved from the oscillating beam system
originally used to drill wells and has also been known
as a beam pump or nodding donkey. (Illustration courtesy
of and adapted from Mau and Edmundson, reference 1.)
volume lift method. Alexander witnessed the field 1. Mau M and Edmundson H:
test, and on June 15, 1926, he signed a contract for Groundbreakers: The Story
of Oilfield Technology and the
Arutunoff and van Wert to supply an ESP system to People Who Made it Happen.
Phillips Petroleum. Peterborough, England:
Fast-Print Publishing, 2015.
A Company Is Born
In March 1928, the first two ESP systems were suc-
cessfully tested in a Phillips well in the El Dorado
oil field near Burns, Kansas, USA. Later that year,
Phillips Petroleum exercised a contractual option to
create the Bart Manufacturing Company; Phillips
owned 51% and Arutunoff owned 49%. The debt
incurred establishing the manufacturing plant, how-
ever, led Phillips Petroleum to divest Bart soon after.
Once again in need of money to keep the company
going, Arutunoff contacted Frank Phillips’s friend,
Charley Brown, a Bart stockholder and executive
of Marland Oil Company, and received a loan. On
March 15, 1930, they dissolved Bart Manufacturing
and created the REDA Pump Company, which was
eventually sold to Camco in 1987 and acquired by
Schlumberger in 1998.
Today, the majority of oil wells around the world
are equipped with artificial lift systems. Rod pump,
gas lift, ESP and other systems have evolved and
adapted to the modern era through incorporation of
remote monitoring and control and innovative mate-
rials. But in concept and design, these essential sys-
tems have changed little in the decades since their
inception and are certain to be a critical technology
for as long as an oil and gas industry exists.
 For more on shale, see “Shale Gas Revolution”

Knowledge at your fingertips.

from the Autumn 2011 Oilfield Review, available
online or on the app.

http://www.slb.com/oilfieldreview
Since 1989, Oilfield Review has been a reference for E&P professionals
worldwide and encompasses:
· More than 500 in-depth, highly referenced articles across
E&P technology and innovation
· The “E&P Defining Series,” a quick introduction to fundamentals.
The Oilfield Glossary has more than 5,000 oilfield terms, supplementing
and supporting the journal.

http://www.glossary.oilfield.slb.com

Download the app by visiting the Apple or Google Play online store
and searching for Schlumberger Oilfield Review.
Oilfield Review
Authoritative. Relevant. Informative.
Oilfield Review
Wellbore Quality Cement Design and Placement Simulation Cement Evaluation

Well Schematic Gamma Ray Deviation Resistivity Porosity Cement Annular Casing Standoff Cement Simulated Mud Displacement Acoustic Impedance Acoustic Impedance Flexural Attenuation Flexural Attenuation Solids, Liquids and
Density Coverage Simulation Map Map Gas Map

Maximum Maximum

Cement Evaluation Zones

0 Mrayl 10 0 dB/m 200
Bulk Density
Between Centralizers Minimum Minimum
1.95 g/cm3 2.95 Mud
0 % 100 –1 Mrayl 9 0 Mrayl 10 50 dB/m 155 0 dB/m 200
Thermal Neutron Cement Annular Tail slurry Gas
Depth, Well Schematic Gamma Ray Deviation A16H Porosity Density At Centralizers CEM_PERC_CEM_COV Lead slurry Image Orientation, ° Average Image Orientation, ° Average Solid
ft 0 gAPI 150 0 degree 90 0.2 ohm.m 2,000 45 % –15 1 g/cm3 2 0 % 100 0 % 100 Spacer 0 90 180 270 360 0 Mrayl 10 0 90 180 270 360 0 dB/m 200 Liquid

Authoritative. Relevant. Informative. 1,000

Mud
2,000

3,000

Spacer
4,000

5,000

6,000

Lead
7,000

8,000

9,000

Tail
10,000