Copyright 2002, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the SPE Asia Pacific Oil and Gas Conference and Exhibition held in Melbourne, Australia, 8–10 October 2002. This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435. Abstract Recent advances in oil and gas cementing technology allow for the modeling and prediction of both compressive and tensile stresses upon an annular cement sheath, throughout the life of a well. Given the knowledge of the type and magnitude of stresses likely to be encountered in a specific location in a wells annulus gives designers target parameters for designing the mechanical properties necessary in the set cement to be able to sustain those stresses without failing. Such a mechanical failure in a cement sheath can cause a loss of annular isolation. However, the authors feel the ability to model these stresses is only one-half of the information necessary to design cement systems for long-term zonal isolation. While some good work has been done on certain lower density cement systems in an attempt to develop fit-for- purpose designs with improved tensile and flexural strengths, the authors have found that some wells requiring higher density cement systems, also need cements with “enhanced” mechanical properties. Towards this end, the authors have conducted mechanical properties research of several relatively common cement additives. These included organic materials as well as non-organic materials. For this study, these materials were added to oilfield cements with water contents averaging from 50 to 66 % by weight of cement (bwoc). Besides the more common unconfined compressive strength tests, the samples are also subjected to tensile and/or flexural strength testing. While the API has long ago established procedures for running unconfined compressive strength tests, there are currently no API standards in place covering the testing methodology for tensile and/or flexural strengths of oilfield cements. Accordingly, the authors present not only the mechanical properties achieved with the use of the various materials tested, but also the methodology used to achieve their data. In an effort to more closely scrutinize the effect each individual material has on the mechanical properties of the set cement, each additive is examined independently. Armed with this information, design engineers should be equipped to propose cement systems that produce effective long-term zonal isolation at the induced annular stresses of their own wells. Introduction In the process of oil and gas well drilling various types of cement systems are being placed into the annular space between the casing and the formation. The purpose of this cement is to structurally support the casing string and prevent casing corrosion, as well as to create a competent hydraulic seal for long-term zonal isolation during the entire operational life of the well. As mentioned by Ravi 1 , the cement should meet a wide range of short-term criteria such as free water, thickening time, filtrate loss, gelling, strength development, shrinkage, etc., as well as certain long-term requirements like resistance to chemical attack, thermal stability and mechanical integrity of the cement sheath. In today’s oil and gas fields, it is common to find design engineers who understand that changes throughout the life of a well can significantly impact induced stresses on the annular cement sheath responsible for maintaining annular isolation. Changes in wellbore stresses can affect the mechanical integrity of the cement sheath and can be caused by a variety of different factors such as: - production rate changes - depleting reservoirs - formation compaction - workovers - stimulation treatments - pressure and temperature changes - secondary and tertiary recovery methods Models have been developed to quantify induced stresses on cement sheaths, and in most instances, if a failure occurs, the models tend to predict that failure will usually not occur under compressional stresses, but rather under tensile stresses 2 . This realization has led many engineers to conclude that the ultimate compressive strength of a cement system may not be the best indication of the durability of an annular cement seal SPE 77867 The Effect of Key Cement Additives on the Mechanical Properties of Normal Density Oil and Gas Well Cement Systems Thomas Heinold, SPE, Robert L. Dillenbeck, SPE, Murray J. Rogers SPE, BJ Services Company
Transcript
Copyright 2002, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the SPE Asia Pacific Oil and Gas Conference and Exhibition held in Melbourne, Australia, 8–10 October 2002. This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Models have been developed to quantify induced stresses on
cement sheaths, and in most instances, if a failure occurs, the
models tend to predict that failure will usually not occur under
compressional stresses, but rather under tensile stresses2. This
realization has led many engineers to conclude that the
ultimate compressive strength of a cement system may not be
the best indication of the durability of an annular cement seal
SPE 77867
The Effect of Key Cement Additives on the Mechanical Properties of Normal Density Oil and Gas Well Cement Systems Thomas Heinold, SPE, Robert L. Dillenbeck, SPE, Murray J. Rogers SPE, BJ Services Company
2 T. HEINOLD, R. DILLENBECK, M. ROGERS SPE 77867
over the life of a well. Instead, many are now attempting to
develop more elastic cement systems that are specifically
designed to withstand the induced stresses anticipated for a
specific well. However, as most researchers skilled in the art
of oilfield cementing can attest, the task of trying to design a
complex cement system for fit-for-purpose applications can be
a real challenge. Many, if not most, of the more common
cementing admixtures used in today’s cementing systems can
have a multitude of primary and secondary effects on the
performance of both the slurry, as well as the final, hydrated
cement. Given the realization that as an industry, much more
attention must now be placed on the mechanical properties of
the set cement than has been given in the past, most
researchers are looking beyond compressive strengths to
cement attributes such as the flexural and tensile strength.
In our industry, new materials are being developed and
evaluated in an attempt to expand the elasticity and tensile
failure resistance of specialized cementing systems to allow
for better long-term zonal isolation at the anticipated induced
stresses throughout the life of a well. In many instances these
new systems are of a medium or low density in order to try to
reduce the hydrated cement’s Young’s Modulus, and hence to
reduce induced stress for a given set of wellbore changes.
While this strategy can be successful, the authors recognize
that it may not be applicable in all instances. Many fields
throughout the world still require cementing systems of higher
density. These higher cement densities may be required for
reasons such as pressure maintenance/well control or for
maintaining hole stability. Given this need for more “normal”
density cement systems with improved resistance to induced
annular stresses, the authors have become aware through
laboratory testing and field applications, that some very
common cement additives can impact the flexural and tensile
strengths of these systems in a multitude of ways. Therefore,
the decision was reached that in order to develop cement
systems with enhanced flexural and tensile strength properties,
it would behoove them to first understand the impact of
commonly used cementing additives on these set cement
properties. It was felt that such an understanding would then
allow for making better choices of basic cementing additives,
such that these additives could act more synergistically with
other purpose designed materials to yield a system with better
elastic properties.
A review of the available literature led the authors to the
conclusion that much of the cement tensile strength testing
done to date had been performed at relatively low
temperatures or were derived from assumptions about certain
relationships between flexural and tensile strength3,4. Due to
the size and shape of the molds used to cure cement specimens
for tensile strength tests, they typically will not fit in high
pressure, high temperature (HTHP) curing chambers.
Therefore, the authors discovered that even though most
induced stress models calculate induced stress values in
compressional and tensile components, much of the higher
temperature testing of cement mechanical properties involved
testing the cement flexural strengths only. The authors believe
this was the case because the molds to cure cement test
specimens for flexural strength would typically fit in the
HTHP curing chamber. In order to explore any possible
correlation between flexural and tensile strengths in normal
density cements at multiple temperatures, the authors made the
decision to use specially made molds that would allow for the
curing of tensile strength cement test samples inside HTHP
curing chambers.
Materials The cement used for this investigation was a monogrammed
API Class G cement. Cements marked by the API monogram
are manufactured within certain parameters and have specified
chemical and physical characteristics. To investigate the
effect various cement additives have on the mechanical
properties of set cement, seven cementing additives commonly
used in oil and gas well cementing were chosen for evaluation
at 100ºF and 200ºF. Following is a short description of these
materials, their applications and typical additive
concentrations.
a) Polyvinyl Alcohol (PVA)5 is a white, powdered, synthetic
polymer, manufactured by the hydrolysis of polyvinyl
acetate. It is readily cold water-soluble. This additive,
dependant on its hydrolysis, is typically used at
temperatures up to 300ºF with concentrations commonly
ranging from 0.2 to 2.0 % bwoc. The film forming
properties of this material can contribute to a reduction in
permeability by limiting the inner particle flow within the
cement matrix. Because of its adhesive properties, it is
also considered to enhance bonding between cement and
casing/wellbore.
b) Silica Fume (SF)6 is a by-product of silicon metal, or
ferrosilicon alloy production and consists primarily of
amorphous silicon dioxide. When added to Portland
cement, the silicon dioxide in the SF reacts with calcium
hydroxide to form calcium silicate hydrate gel. This
reaction will contribute to compressive strength
improvement, as well as a reduction in the permeability of
the cement matrix. The additive is typically used in
temperatures of up to 400ºF with concentrations
commonly ranging from 1.5 to 15.0 % bwoc. When used
in conjunction with suitable fluid loss control additives, it
can improve antigas migration properties. It is also used
to enhance compressive strength, improve free fluid
control as well as CO2 resistance of oil and gas well
cement systems.
c) Metakaolin (HRM) is a white, amorphous, alumino-
silicate and is produced through the calcination of the clay
mineral kaolin. Metakaolin is a highly reactive pozzolan
with a high specific surface area. When added to Portland
cement, it reacts aggressively with calcium hydroxide, a
normal cement hydration by-product, which makes it very
suitable as a cementing material. This additive is typically
used over a temperature range from 32ºF to 180ºF in
concentrations ranging usually from 1.0 to 25.0 % bwoc.
It contributes to the improvement of various properties of
oil and gas well cements, such as permeability,
THE EFFECT OF KEY CEMENT ADDITIVES ON THE MECHANICAL SPE 77867 PROPERTIES OF NORMAL DENSITY OIL AND GAS WELL CEMENT SYSTEMS 3
compressive strength development, flexural and tensile
strength, gas control and sulfate resistance.
d) Wollastonite is a white calcium-silicate powder and is
sometimes referred to as a naturally occurring mineral
fiber. It is less reactive than HRM but will produce a
range of benefits when added to Portland cement. This
additive can typically be used over a wide temperature
range from 32ºF to 400ºF in concentrations ranging
usually from 10.0 to 50.0 % bwoc. Similar to HRM it can
contribute to the improvement of various properties of oil
and gas well cements, such as permeability, compressive
strength development, flexural and tensile strength, gas
control and sulfate resistance.
e) Styrene Butadiene Latex6 is a milky suspension of small
spherical particles and is normally stabilized with a
surfactant package to improve freeze/thaw resistance and
prevent coagulation when added to Portland cement. This
additive can typically be used in temperatures up to 400ºF
with concentrations ranging usually from 0.5 to 4 gal/sk.
It is considered that the film forming properties of a latex
modified cement system can lead to a reduction in
permeability, increased tensile strength, better fluid loss
control, enhanced bonding between cement and
casing/wellbore and improved acid resistance.
f) Hydroxyethyl cellulose (HEC)5 is a white, powdered,
water-soluble polymer. Manufactured in wide range of
grades, these principally differ in molecular weight. This
additive can typically be used at temperatures in excess of
300ºF with concentrations commonly ranging from 0.1 to
1.0- % bwoc. HEC-based materials are used to enhance
the fluid loss control properties of various cement systems
and are effective over a broad range of well conditions.
g) Sodium Metasilicate (SMS)6 is a white, water-soluble
powder, which is produced through the fusing of silica
(sand) with sodium carbonate at 1,400°C. When added to
Portland cement, silicates will react with lime to form
calcium silicate gel. The resulting gel structure can
provide enough viscosity to allow for the usage of large
quantities of mixwater without compromising slurry
stability due to the separation of excessive freewater. The
additive is typically used in temperatures of up to 200ºF
with concentrations normally ranging from 0.1 to 4.0 %
bwoc. This additive is commonly used to provide an
economical low-density slurry system with good free
water control and enhanced compressive strength
development.
Testing Methodology All slurries presented in this paper were prepared and cured
according to the API RP 10B7 document, using the previously
mentioned Class G cement. All systems were mixed at 15.0
ppg containing only one individual additive per test. The
authors felt that by omitting other cementing additives, such as
retarders, dispersants and fluid loss control, which are used to
optimise slurry systems, they could better gauge the effect
each individual additive has on the mechanical properties of
the set cement. The exemption to the above were slurries that
contained latex. All latex slurries were mixed at 16.0 ppg in
order to generate a stable cement slurry. Initial testing had
shown that by using a 15.0-ppg system, considerable settling
occurred. No stabilizer additive was used. This was done to
eliminate any positive or negative impact an additional
chemical may have on the overall strength for each individual
specimen. All slurries were mixed at room temperature in
either one of the following orders.
1) fresh water + defoamer +dry blended cement
2) freshwater + defoamer + latex + cement
The different mechanisms of cement sheath failure due to the
impact of various mechanical stresses in a downhole
environment were until recently not fully understood. As
already mentioned in the introduction, in the past, very little
attention had been paid to the importance of these properties
for the mechanical integrity of various cement systems to
ensure good long-term zonal isolation. The American
Petroleum Institute (API) currently has no testing procedures
in place to determine flexural or tensile strength of oilfield
cement systems under simulated downhole conditions.
Historically, most cement testing was governed by procedures
in API RP 10B and concentrated on the properties of cement
slurries prior to setting. Such tests included thickening time,
fluid loss, rheology and free water. Until a few years ago, the
only test carried out on the set cement was a compressive
strength test under unconfined conditions. Typically, “rules -
of – thumb” were applied for required minimum values and
compressive strength as high as possible was usually
considered better. However, for many years now, our
colleagues in the construction industry have carried out
destructive strength tests to determine the flexural and tensile
properties of concrete. Utilising this expertise, many
laboratories are currently using equipment and procedures as
specified by the American Society for Testing and
Materials (ASTM)8,9.
In order to measure the mechanical properties of the set
cement, all slurries were cured under simulated downhole
conditions in a standard HPHT curing chamber. For several
days (72 hrs), each specimen was exposed to a constant
pressure of 3,000 psi and temperatures of 100ºF or 200ºF,
respectively. After the curing period had expired, samples
were cooled down over a period of 1 hour and 30 minutes and
then slowly de-pressurized. During this period, all samples
remained under water. All specimens for flexural, tensile and
compressive strength tests were cured simultaneously, with
molds in a vertical orientation, in the same curing chamber
throughout the testing process.
Flexural strength testing was performed on specimens
measuring approximately 1.6 x 1.6 x 6.3 in. utilising a three-
point bend fixture in a Gilson Co. HM 138 strength tester, as
illustrated in Figure 1. It should be noted that the rectangular
shaped cement sample is placed in the fixture, with the two
end supports pulling downwards, while the middle support
pulls upwards. The exerted force is gradually increased until
mechanical failure of the specimen occurs.
4 T. HEINOLD, R. DILLENBECK, M. ROGERS SPE 77867
Tensile strength testing was performed using briquette mold
specimens that are commonly referred to as “Dog Bones”.
Using fundamentally the same testing machine as for flexural
strength testing, the differently shaped sample fixture should
be noted in Figure 2. With a gradual increase in exerted force,
the upper half of the fixture is pulling the cement sample
slowly upwards, until mechanical failure of the specimen
occurs. As mentioned in the introduction, these curing molds
were specifically manufactured to enable the authors to cure
cement specimens under temperature and pressure.
Compressive strength of the specimens was obtained
through destructive crush tests on specimens measuring 2 x 2
x 2 in.
Results The mechanical strength data of the various slurry systems are
graphically presented in Figures 3 through 9. In order to allow
for easier comparison of the gathered information for each
individual additive, data for both test temperatures is plotted in
one single graph. The plots are representations of flexural,
tensile and compressive strength for various additive
concentrations at 100ºF and 200ºF respectively. Finally Tables
1 through 4 are tabular summaries of the data used to produce
the above mentioned graphs.
Discussion Before launching into a more detailed discussion of the
recorded results, the authors would like to point out a few
issues related to the chosen testing methodology.
As mentioned previously, the majority of slurries tested
were mixed at 15.0 ppg. The authors acknowledge the fact that
these systems were not optimised and might not be utilised in
this form to cement a wellbore. However, since the goal was
to investigate the effectiveness of individual additives, the use
of other common cementing admixtures such as dispersants,
retarders or fluid loss control may have masked the effects of
the materials under investigation. Because of the substantial
water requirements of such materials like silica fume and
metakaolin, and a desire to investigate the effects of the
chosen additives over a relatively wide range of
concentrations, the decision was made to lower the density
slightly to compensate for these higher water requirements. In
the course of researching this paper, the authors also decided
to maintain a constant cement density rather than a constant
cement to water ratio. Through an optimisation of the water to
