SPE 81001

Design of High Toughness Cement for Effective Long Lasting Well Isolations
Walter Morris, Marcelo A. Criado, SPE, Jorge Robles, SPE and Gustavo Bianchi / San Antonio - Pride Int.
Neuquen, Argentina.

Copyright 2003, Society of Petroleum Engineers Inc.

Introduction
This paper was prepared for presentation at the SPE Latin American and Caribbean Petroleum
Engineering Conference held in Port-of-Spain, Trinidad, West Indies, 27–30 April 2003.
There are many factors that should be considered when
designing slurry for well cementing operations. Some of the
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 most critical ones are temperature, density, rheology, filtration
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
and resistance, among others. A wide variety of additives have
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at been developed to ensure successful cementing operations.
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 API RP-10B presents a number of laboratory procedures to
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
evaluate and predict the slurry performance at specific service
words; illustrations may not be copied. The abstract must contain conspicuous conditions. Several physical properties such as the slurry
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.
rheology, fluid loss control and thickening time are used to
characterize the slurry short-term performance. On the other
hand, less attention is placed on the hydrated cement medium
Abstract
and long-term properties and how they can affect the well
During the well service life the cement isolation is exposed to
isolations mechanical and durable performance.
extreme conditions that can cause its premature failure.
During the well service life the cement isolation is exposed to
Certain well completion operation as perforating and hydraulic
extreme mechanical stress conditions that can cause its
fracturing, the changes in temperature and pressure during
premature failure1. Certain well completion operation as
secondary recovery or the mechanical stress originated by
perforating and hydraulic fracturing or other events as changes
formation displacements may cause severe damage to the
in temperature and pressure during secondary recovery and
cement isolation.
mechanical stress originated by formation displacements may
The mechanical properties (compressive and tensile strength,
cause severe damage to the cement isolation2, 3. Several
toughness, Young modulus and Poisson ratio) of different
mechanical properties of hydrated cement such as compressive
cements were evaluated in order to establish their best service
and tensile strength, toughness, Young modulus and Poisson
performance. Typical slurry designs were tested focusing the
ratio, provide the key for achieving an adequate well isolation
attention on the effect of certain additives such as latex, fibers
service life. During the last years an increasing number of
and other polymers used as fluid loss control or dispersants.
publications have addressed these issues, focusing on the use
The cement mechanical properties were determined according
of additives, like latex, fibers and other polymers that improve
to ASTM and API standard test methods. The cement
the cement elasto-plastic behavior and toughness 3-5. The field
toughness was evaluated following the API RP 43 standard for
performance of theses additives can be evaluated by different
testing well perforators.
logs run after the well has been cemented. Some of the most
The experimental results show that the mechanical properties
widely used are CBL (Cement Bond Log), VCL (Variable
of cement are strongly dependant on the particular additives
Density Log) and SBT (Segmented Bond Tool).
used when preparing the slurry. Even when cement with no
This paper discusses the influence of certain additives as latex
admixtures presents high compressive strength it also shows a
and polymer fibers on the mechanical properties cement. The
fragile behavior with limited strain and low toughness. The
study is based on laboratory and field tests. CBL, VDL and
use of latex improves the cement elastic behavior although it
SBT logs were obtained from two wells, cemented with and
does not better its impact resistance. On the other hand, the
without fibers, in order to evaluate cement damage and de-
addition of polymer fibers improves the cement toughness and
bonding before and after perforating. The paper presents
its elastic behavior. As it was demonstrated by well testing
laboratory and field evidence that show how fibers improve
profiles run before and after perforating, the addition of fibers
the cement toughness behavior and reduces its deterioration
improves the cement performance when exposed to the
under perforation conditions. A discussion about several
different events that may damage the isolation during the well
aspects that influence the cement mechanical properties
service life.
evaluation is presented.
2 SPE 81001
Methodology
The mechanical properties of five hydrated cements prepared
with different slurry designs were determined. The influence
of different additives such as styrene-butadiene latex, a fluid
loss control copolymer and the addition of polymeric fibers
were analyzed. Table 1.presents the slurry designs, its physical
properties and cement specimens identification.
Specimen’s preparation
The slurries were prepared and cured according to API
Recommended Practice for Testing Well Cements, section 5
“Preparation of Slurry”. Cement specimens were cured at
183.2 oF (84 oC) and 3000 psi according to section 7.7. “Well
simulation compressive strength test”, table 2 “Well
simulation test schedules for curing compressive strength
specimens” for a 7900 ft (2400 m) depth and a 1,3 oF/100 ft
(2,4 oC/100 m) temperature gradient.
Four cubes, 2 inch (50 mm) sided, were cast per each slurry
design. Also two cylindrical specimens with a 1.5 inch (38
mm) diameter and a 3 inch (76 mm) length were prepared in
order to evaluate the cement tensile strength according to
ASTM C-496 standard. Theses specimens were cured in a
thermostatic bath at the same temperature than the cube
specimens.
Mechanical tests
The cement compressive strength was evaluated after 48 hours
of curing following the well-simulation compressive strength
test procedure indicated in API RP 10. The tests were
performed using a 15-ton maximum load Carver press. The
vertical (compression) and horizontal (expansion) deformation
of the specimen was evaluated by means of two electronic
displacement transducers. Theses readings allowed the
determination of the material Young modulus (E) and Poisson
ratio (µ). The cement tensile strength was evaluated according
to the ASTM C 496 standard test method for “Splitting tensile
strength of cylindrical test specimens” Two cylindrical
specimens, 1.5 in. (38.1 mm) in diameter and 3.15 in (80 mm)
long were prepared for each slurry design. These specimens
were cured in a thermostatic bath at a temperature of 104 oF
(40 oC). The cement toughness was evaluated by calculating
the area under the tensile stress vs. deformation curve.
Perforator test
In order to determine the performance of standard and fiber
admixed cement under well perforator conditions a testing
procedure based on API RP 43 was followed. The objective of
theses tests was to determine the influence of different
amounts of admixed fibers on the cement integrity. The main
aspects evaluated were, the cement deterioration next to the
perforator hole, the presence of radial and transversal cracks
on the cement disk, and the cement to API concrete bonding.
Four probes were constructed in order to reproduce the
annulus in the well. A 200 mm (8 inch) diameter and 800 mm
(332 inch) long pipe was used as a mold. Figure 1 presents a
scheme showing the mold and probe parts.
The metallic disc represents the casing wall (3/8” thick) to be
perforated; the API concrete represents the formation and the
2 inches (50 mm) thick cement disc, the cemented well
annulus. Four slurry mix designs, identified as NA, F(0.2),
F(0.4) and F(0.6) (see table 1) were 0.2, 0.4 and 0.6 indicates
the percentage of fibers added to the slurry by weight of
cement in each case. The API concrete was prepared using 20-
40 fracturing sand, class A cement and a water to cement ratio,
w/c = 0.45. The cement and API concrete were cured at room
temperature for seven days before conducting the tests. The
probes were tested in a bunker of one of the mayor perforator
charges manufacturers in Argentina.
The molds were placed facing up and a 22 g hollow charge
was positioned on top of the metallic disk at a zero distance
from the disc surface. Figure 2 and 3 present a view of the
probes before conducting the tests, (left) metallic ring and disk
(right) perforator charge placed on the metallic disk.
Field tests
The field performance of fiber admixed cement was evaluated
by running CBL, VDT and SBT logs before and after the
perforating operation. These logs were run on two wells, one
cemented with standard slurry and the other one with a fiber
admixed cement. These wells are located in the Las Heras
oilfield. Table 2 presents some summarized information of
these wells.
Ten wells have been cemented in different oilfields in
Argentina with slurries having admixed fiber contents varying
from 0.3 to 0.5 % BWOC.
Results
Table 3 presents the rheological and fluid loss test results for
each slurry design under study.
Mechanical tests
Figure 4 presents the compressive strength vs. strain
deformation plots for the different cements under study (see
Table 1) tested after 48 hs of curing. In order to better
visualize the different trends, only one curve is plotted per
condition.
As shown in the fig. 4, some trends presented a nonlinear σ vs.
ε behavior during the initial and final stages of the test.
Therefore, the Young’s modulus (E) values were determined
by linear regression, considering the intermediate portion of
the curves. The Poisson’s ratio (ν) were calculated as εh/εv,
were εh and εv are the horizontal and vertical strain
respectively. The average values of compressive strength,
Young’s modulus and Poisson’s ratio are shown in table 4. As
shown in fig. 4, both cements with admixed fibers (F and FL,
see table 1) maintain a certain carrying load capacity after the
maximum compressive strength is reached. At this stage,
cement begins cracking but the specimen parts remain
attached one to the other. A deformation of more than 10 %
was observed on the four specimens that had admixed fibers.
Figure 5 presents the tensile strength vs. strain plots for each
cement type under study. The cement toughness was
determined as the area under the stress strain curve. Table 4
summarizes the mechanical properties of each cement type.
Perforator tests
After the tests were performed, the molds were opened and the
cement disc was removed. In some cases the cement disk was
firmly attached to the API concrete and a cutting tool was used
SPE 81001
to separate both parts. Figure 6 shows the aspect of the
cement disks of each probe after testing.
As shown in fig. 6 a, probe ST presented a radial crack pattern
around the perforator hole with substantial cement loss. The
cement disc was completely de-bonded from the API cement.
As in the previous case, probe F(0.2) (fig. 6 b) presented a
radial crack pattern around the perforator hole. However, a
clean hole with no cement deterioration was observed in this
case. The cement disk was de-bonded from the API concrete.
In the case of probe F(0.4) (fig. 6 c), the cement disk remained
firmly attached to the API concrete. No cement deterioration
was observed around the perforator hole and almost no cracks
were found. Finally, probe F(0.6) (fig. 6 d) presented a similar
behavior than F(O.4), showing good adherence between
cement and API concrete, no cement deterioration nor
cracking. Table 5 summarizes the qualitative evaluation
obtained from the perforator test.
Cement F(0.6) (0.6 % fibers BWOC) presented mixing
inconvenient while the slurry was prepared, as the fibers
would sometimes jam the warring blender blades.
Field tests
Cement with no admixed fibers
Figure 7 presents the CBL. VDT and SBT logs run before and
after perforating well A-487 cemented without fibers.
As shown in fig.7, the pre-perforating logs reveal that cement
presents good quality and 100 % adherence. The CBL shows
values close to the 0 mV, which indicates maximum
attenuation due to optimum adherence between cement and
casing. This is also evident in the VDT log which shows first
wave arrivals disappearances. The SBT log reveals uniform
cement with high resistance within the well perimeter and
along the depth section viewed in the figure.
After perforating the CBL log show significant cement
deterioration. The VDT signal reveals that the casing has lost
adherence compared to the previous case. This is clearly
evident by the appearance of white areas in the SBT map. The
logs shows that the affected area is not only limited to the
cement next to the perforations but also affects the cement
bonding and resistance several feet below it. When comparing
the pre and post perforating logs, it becomes evident that this
operation causes sever damage to the cement annulus due to
the high impact energy developed by the perforator jets.
Cement with admixed fibers
Figure 8 show the CBL, VDT and SBT logs obtained from
well B-614 which was cemented with fibers.
As shown in figure 8, the pre-perforating logs evidence a
sector with fair bonding between casing and cement. This area
corresponds to the production formation zone and the de-
bonding effect is presumably due to micro annulus effects.
Good quality cement is observed in the rest of the well depth
interval analyzed.
After perforating the logs present a similar aspect indicating
that cement was not significantly affected by the perforator
impact energy. The cement above and below the perforated
zone remains bonded to the casing and presents similar
resistance than before the perforating operation.
3
Discussion
Improving the mechanical properties of cements
As could be expected, the mechanical properties of cement are
strongly dependant on the particular additives incorporated to
the slurry. Even when cement prepared with no admixtures or
with a standard slurry design present high compressive
strengths, they show a fragile behavior characterized by a low
tensile strength (σT) and toughness (Τ). On the other hand, its
high Young’s modulus (E) and low Poisson ratio (ν) reduces
its deformation capacity.
The use of latex improves the elastic behavior of cement as it
reduces its Young’s modulus and increases the Poisson’s ratio.
In contrast, latex causes significant compressive strength
reduction and does not improve the materials toughness. As
indicated in table 4, cement with fibers and latex present a
high tensile strength but only fibers enhances the cement
toughness performance. The values of Τ measured on fiber
admixed cements where approximately eight times grater than
those observed on normal, standard and latex cements.
Therefore, cements with admixed fibers provides the best
mechanical properties of this group as it presents high
compressive and tensile strength, very good elastic behavior
and high toughness.
The perforator tests have demonstrated that fibers contribute to
increase the cement to formation bonding. This was later
confirmed by CBL, VDT and SBT logs performed on wells
cemented with and without fibers. The field experience
obtained from 10 cementing operations in Argentina show that
the optimum concentration of polymeric fibers varies between
0.3 and 0.5 % BWOC.
Evaluation of cement mechanical properties
The latest publications on cementing technology agree that not
only compressive strength information should be considered
when designing effective and durable well isolations2-5. The
aim for long lasting isolation has led towards a better cement
mechanical properties characterization. Many parameters such
as tensile strength, Young’s modulus, Poisson’s ratio, and
toughness are being increasingly used to predict the
performance of cement under critical events that may take
place during the well service life5. Certain well completion
operation as perforating and hydraulic fracturing, the changes
in temperature and pressure during secondary recovery or
extreme formation induced stresses may cause severe damage
to the cement isolation2,3. On the other hand, the physical and
chemical nature of hydrated cement introduces certain
difficulties when it comes to establish specific properties and
parameters to predict its service performance. Several factors
such as slurry design, curing time and temperature and cement
age, will strongly affect the final mechanical properties of
cement6. At the same time, the use of additives as latex or
fibers, to improve the material elastic and toughness behavior
will modify the material deformation mechanisms. As
observed on figure 4 and 5 cements F (with fibers), L (with
latex) and FL (with fibers and latex) present a nonlinear stress
– strain relationship. Evermore, the determination of
mechanical properties such as the Poisson’s ratio, the
compressive and tensile strength will depends on other factors
as the specimen geometry and its testing procedure (location
4 SPE 81001
of displacement transductors, load rate application, etc.).
Therefore, in order to standardize the use of other mechanical
parameters further than the compressive strength to better
characterize the cement long-term performance, it will be first
necessary to normalize the different testing procedures to
allow reliable and comparable results.
Conclusions
As could be expected, the mechanical properties of
cement are strongly dependant on the particular additives
incorporated to the slurry. Even when cement prepared with
no admixtures or with standard slurry designs present a high
compressive strengths, they also show a fragile behavior and
limited deformation.
Admixed fibers cements combine high compressive
and tensile strength, very good elastic behavior and toughness
values that are significantly grater than those observed on
normal, standard or latex cements. The use of fibers also
appears to increase the cement to formation bonding.
The well perforation operation causes severe damage
and de-bonding to the cement isolation. The use of polymeric
fibers controls the cement deterioration and cracking around
the perforator hole. This was confirmed by sonic logs
performed before and after perforating wells cemented with
and without fibers.
There are several experimental factors that should be
considered when evaluating the cement mechanical properties.
Standardizing specific testing procedures for determining each
of these parameters (E, σT, ν and Τ) will lead to reliable and
comparable results that may be used for predicting the long
term performance of well isolations.
Acknowledgements
The authors wish to thank San Antonio-Pride for the
encouragement for writing this paper and for the permission to
present it. We also thank San Antonio-Pride field engineers
Raul Gunckel and Juan Carlos Nauman for providing the
operative charts presented in the paper. We would finally like
to acknowledge Mr. Ernesto Ramón for his cooperation and
laboratory work, as well as all the members of the San
Antonio-Pride technology department.
Nomenclature
BOWC = By weight of cement.
CBL = Cement bond log.
E = Elastic or Young´s module
n´ = Behavior index
ν = Poisson´s ratio
K´ = Consistency index
SBT = Segmented bond tool
σc = Cement compressive strength.
σ = tensile strength.
VDT = Variable density log.
w/c = water to cement ratio.
Τ = Cement toughness.
References
1. Goodwin, K. J. and Crook, K. J., “Cement shear
stress failure”, Paper SPE 20453 presented on the
Annual Technical Conference and Exhibition held in
Mew Orleans, September.23-26, 1990
2. Blanco, A., Centeno A., Trujillo, M., Rodríguez, W.
and Rodríguez, S. “Nuevas tendencias en aislamiento
zonal para condiciones de esfuerzos críticos”, XIII
Congreso Latinoamericano de perforación, Colaper,
Venezuela, 2002.
3. Bosma, M., Ravi, K., Van Driel, W. And Schreppers,
G. J., “Design Aproach to Sealant Selection for the
Life of the Well”, Paper SPE 56536 presented on the
Annual Technical Conference and Exhibition held in
Houston, Texas 3-6 October, 1999.
4. Thiercelin, M., J., Baret, J., Dargaud, B. and
Rodriguez, W. J., “ Cement design based on cement
mechanical response”, SPE Drilling and Completion,
December, 1998, (pp. 266-273).
5. Altuna, G., Centurión, S. and Perez Ipiña, J. E.,
“Variation of the mechanical properties for
cementong slurrys with different compositions”,
paper SPE 69616 presented on the Latin American
and Caribean Petroleum Engineering Conference held
in Buenos Aires, Argentina,25-28 March, 2001.
6. Smith, D. K., Cementing, Society of Petroleum
Engineers Inc. ISBN # 1-55568-006, New York,
1990.
SPE 81001 5

Table 1. Slurry design and specimen identification.

Water to cement Fluid loss Dispersant Polymer fibers Styrene-

Slurry Identification ratio w/c additive butadiene
copolymer Latex
NA (no additives) O.44 - - - -
ST (standard slurry) 0.44 1.2 0.6 - -
F (fiber admixed) 0.44 1.2 0.6 0.4 -
L (latex admixed) 0.34 0.3 0.4 - 10
FL (fiber and latex) 0.34 0.3 0.4 0.4 10

Table 2 Well identification and slurry design

Well Ident. Fibers Slurry design

A-487 no Class G cement, w/c = 0.63, 15 % asphaltite, 0.7 %
fluid loss additive+ 0,2% SMSAnh + 0,4% Fibers +
B-614 0.4 % 2% Latex + 0,2% Antifoamer

Table 3 Rheological and fluid loss test results for each slurry design under study.

Rheology API
Slurry Identification Fluid loss test
n´ K´
NA 0.445 0.0512 n/c
ST 0.877 0.0070 20
F 0.792 0.0150 19
L 0.640 0.0205 20
FL 0.786 0.0120 12

Table 4 Mechanical properties of each cement type ( C =compressive strength, T = tensile strength, E = Young’s
module, = Poisson’s ratio and = toughness).

Cement σC σT E ν Τ
type
Kpsi (MPa) psi (MPa) Kpsi (GPa) in/in lb-in
NA 5.24 (36.9) 270.6 (1.90) (5.48) 0.015 75.8
ST 5.05 (35.6) 264.1 (1.86) (6.28) 0.036 100.4
F 5.07 (35.7) 447.3 (31.5) (2.56) 0.08 810.9
L 3.53 (24.9) 258.5 (18.2) (2.62) 0.16 74.5
FL 3.48 (24.5) 487.1 (34.3) (2.54) 0.19 790.1
6 SPE 81001

Table 5. Qualitative evaluation obtained from perforator test.

Aspect evaluated ST F(0.2) F (0.4) F (0.6)

Cement deterioration around hole Yes Some No No
Evidence of significant cracking Yes Some No No
Cement to concrete de-bonding Yes Yes No No
Inconvenient when mixing the slurry No No No Some

Mold

API concrete

Cement

Metallic ring
Metallic disk

Top view

Figure 1. Specimens characteristics used to perform the perforator tests.

Figure 2. Metallic ring and disc of the specimen. Figure 3. 22 g hollow perforator charge placed on
the metallic disk.
SPE 81001 7

6
40
5

30
4 NA
ST

σC (MPa)
σC (kpsi)

3 20 F
L
2 FL
10
1

0 0
0 1 2 3 4
ε (%)

Figure 4. Compressive strength ( C) vs. strain ( )obtained from cement cubes

tested after 48 hs. of curing at 183.2 oF (84 oC).

5 00 3 .5

3
4 00
2 .5
3 00 σT (MPa)
2
σT (psi)

NA
2 00 1 .5
ST
F 1
1 00 L
FL 0 .5

0 0
0 1 2 3 4 5
ε (% )

Figure 5. tensile strength ( T) vs. strain ( ) obtained from diametric compressive test performed
according to ASTMC 496 using cylindrical specimens .
8 SPE 81001

Figures 6. Aspect of the cement disk containing 0 (a), 0.2 % (b), 0.4 % (c) and 0.6 %
(d) of admixed fibers (probes ST, F0.2), F(0.4) and F(0.6) after perforating and removed from the mold.

Figure 7. Pre and post-perforating CBL,VDL and SBT logs run on web A-487.
Cemented without fibers (well depth from 1474 to 1478 m).
SPE 81001 9

Figure 8. CBL, VDT and SBT logs obtained from well B-614 which was cemented with fibers
(well depth from 1623 to 1627 m).