IADC/SPE 77213

Cement Matrix Evaluation

Mansour A. Al Hammad, Saudi Arabian Oil Company, Yehya M. Altameimi, SPE, BJ Services Co.

Copyright 2002, IADC/SPE Asia Pacific Drilling Technology

cement job during the design phase and prior to the physical
This paper was prepared for presentation at the IADC/SPE Asia Pacific Drilling Technology placement. The slurry design and placement procedure are the
held in Jakarta, Indonesia, 9–11 September 2002.
key factors to long term cement job success which can be
This paper was selected for presentation by an IADC/SPE Program Committee following
review of information contained in an abstract submitted by the author(s). Contents of the
measured by the zonal isolation, Lack of micro-annulus and
paper, as presented, have not been reviewed by the International Association of Drilling the cement sheath integrity during the well life. The cement
Contractors or 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 IADC or sheath defects and failure to seal a pressurized fluid containg
SPE, their officers, or members. Papers presented at the IADC/SPE meetings are subject to
publication review by Editorial Committees of the IADC and SPE. Electronic reproduction,
zone is usually refered to inadequate mud displacement,
distribution, or storage of any part of this paper for commercial purposes without the written annular fluid migration and possibly mechanical factors. Few
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 cement specialists consider the failure of cement sheath
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
integrity as primary reason for CCA to exist.
01-972-952-9435.
Unique cement sheath failure was observed during the early
Abstract stage of gas wells in Saudi Fields. This was indicated by the
In the Saudi Aramco gas campaign, during the last few years, presence of high CCA pressures with effluent presenting deep
a high casing-casing annulus phenomenon was observed formations. Such incidents drove to a study to evaluate the
clearly days or months after rig released. Some of these casing possible source and path of this effluent to prevent the re-
or liner set and cemented across abnormally pressured occurrence if possible. It was found that for CCA pressure to
formations, which contains salt water, gas or combinations of appear at surface the effluent should pass several barriers over
both. These zones were cemented using various cement thousands of feet such as cemented casing / open hole annulus,
density ranging from normal to high-density slurries. cemented casing /casing annulus and external casing packers.
The integrity of cement sheath is commonly measured by These lead to possible causes such cement design, placement
its ability to provide long term zonal isolation, at the time of procedure, and liquid to solid transition control and post
completion and for the life of the well. It also has to physically placement well environment.
support the casing, withstand pressure cycles, temperature Among all of the above the cement sheath failure "cracking"
cycles and protect the casing against corrosive fluids. Not all and cement design was most logical issue to re-consider due to
cement applications perform these functions with equal the similarity feature of expansive cement which focused the
success, and many of those that do, they do not stand up study on the expanding property of cement during the post
against the forces that exert upon them during time. placement and side effects if any.
This paper is evaluating the expansion and the shrinkage
property using class “ G” cement in Saudi Arabia, and its When cement is correctly placed in the wellbore and provides
effect on the cement matrix. One of the potential problems initially a good zonal isolation, the zonal isolation often
causing the failure of cement zonal isolation, can be attribute disappears overtime, which in most cases is caused by micro
to the over use of chemical expansion additives. The cement annulus or reduction in cement permeability.
bond failure can cause migration of fluid from one zone to Expansive cement systems have been advocated for use across
another, losses of reservoir fluid, well control issues and poor formations, where micro-annulus is the primary cause of
stimulation operations. formation’s fluid migration. Controlled expansion, through the
modification of the additive concentration, will assist in
Introduction sealing small gaps between the cement and the casing or
High casing-casing Annulus (CCA) pressure is an alarming formation, even though expansive cement undergoes a bulk
sign during Gas well production life, from safety point of view dimensional expansion, they still exhibit a net chemical
especially if effluent is hydrocarbon base fluid. Its presence shrinkage and the same hydrostatic pressure decrease as
could also impose operational limitations, which could easily normal cement. This is possibly ineffective in controlling or
hinder the well productivity or even force the operator to sealing large channels and washouts.
consider an expensive work over on emergency basis. The micro-annulus that could lead to sustained casing
Eliminating the CCA pressure is one of the objectives of every pressure caused by the casing test, reduction of the fluid
2 M. AL HAMMAD AND Y. AL TAMEIMI IADC/SPE 77213

weight form the drilling phase to the completion phase, and 3. System 3 Class “G” Saudi cement, 35% silica flour
poor hole cleaning techniques. bwoc, 1% magnesium oxide an expanding additive
The cement sheath under testing, it is subjected to high- bwoc, 0.8% proprietary fluid loss agent bwoc, 0.6%
pressure differential during the fracturing treatments ranging dispersant agent bwoc and 0.5% lignosulfonate
from 6,000 psi to 10,000 psi and pressure drop as high retrder bwoc.
as 100oF. 4. System 4 Class “G” Saudi cement, 35% silica flour
bwoc, 0.8% proprietary fluid loss agent bwoc, 0.6%
Well Design dispersant agent bwoc and 0.5% lignosulfonate
There are two type of casing design which had a common retrder bwoc.
factor of which annuli suffer the cement failure consequences 5. System 5 Class “ G” Saudi cement, 35% silica flour
or communicating to surface. Although these designs are bwoc, 0.75% magnesium oxide an expanding
called big hole and slim hole, yet the intermediate casing additive bwoc, 1.5 gps latics, .15 gps stabilizer, 0.2%
string which supposedly case an abnormally pressure zone is dispersant agent bwoc and 0.5% lignosulfonate
the inner string of an active annuli. The casing size is either retrder bwoc.
13-3/8” or 9-5/8” that is run inside 18-5/8” or 13-3/8” 6. System 6 Class “ G” Saudi cement, 35% silica flour,
respectively (figure-1). The cement placement is performed in 90% bwoc hematite, 0.75% magnesium oxide an
three or two stages depending on wellbore conditions. One expanding additive bwoc, 1.5 gps latics, .15 gps
multi-stage packer is placed across a firm formation interval in stablizer, 0.75% dispersant agent bwoc and 0.85%
the open hole and the second one is placed inside the CCA for lignosulfonate retrder bwoc.
additional insurance and avoiding any cement loose into
weak formations. In system 1 the cement started to expand significantly at 18
hours and continued to increases until the monitoring device
bottomed-in at around 26 hours, as shown in figure 2. The
Cement Slurry Expansion and Shrinkage Properties linear expansion was at 10.7 % and major cracks were evident
Saudi class “G” cement is manufactured to certain chemical in the set cement. A reduction in the expansion additive was
and physical standards to meet API specifications as shown in applied to system 2, the linear expansion rate has slowed down
Table1. An independent party performed the test. and linear expansion values has decreased to 6.5%, as shown
An API rheology test carried out to determine the ease of in figure 3, in picture 1 & 2 cracks can be noticed by the naked
flow of the slurries, and to avoid shut downs during the eyes. On system 3 the expansion additive concentration was
operation. A fluid loss test determined the slurries has less reduced to 1%, the final linear expansion value reached 0.1%
than 100cc in 30 minutes, this was accomplished when tested the cracking has disappeared as shown in figure 4. Shrinkage
with a differential pressure of 1000 psi forming a filter cake to test was performed to determine the shrinkage significant on
prevent further fluid loss. An API free water test conducted on the system using the base additives using the Saudi class”G”
the slurries to confirm that water will not separate from the cement, however this test was carried out prior to system 3.
slurry and migrate upward, which can result in an incomplete The shrinkage process started 12 hrs into the test to reach to
zonal isolation. 0.08%, refer to figure 5.
An expansion test was performed using a new expanding
and shrinkage apparatus. The slurry was tested at BHST of Magnesium oxide provides an expansive force within the
298 deg F and pressure of 3000 psi. All expansion or cement matrix, as a result of the hydration to magnesium
shrinkage measurements were based on the data points taken hydroxide. The hydrated material occupies more space than
after the temperature has stabilized at 298 deg F or at around 4 the original ingredients 1.
hours. It is imperative to determine how much expansion is
enough per cement type, and if the expansion will adversely MgO + H2O→Mg (OH)2
have a negative effect on the cement set properties. The
cement systems used in the expansion and shrinkage test are as We have known that higher compressive strength will
follow: lead to durable cement sheath; this would be true if all loads
1. System 1 Class “G” Saudi cement, 35% silica flour placed on the cement column were compressive. However
bwoc, 5% magnesium oxide an expanding additive cement can be subjected to radial and tangential stresses. More
bwoc, 0.8% proprietary fluid loss agent bwoc, 0.6% tests performed to determine the flexural and tensile strength
dispersant agent bwoc and 0.5% lignosulfonate of the modified slurries, to better understand the cement
retrder bwoc. mechanical properties.
2. System 2 Class “G” Saudi cement, 35% silica flour Cement tensile strength is the force required to pull apart,
bwoc, 2% magnesium oxide an expanding additive or mechanically fail specially formed cement sample. Cement
bwoc, 0.8% proprietary fluid loss agent bwoc, 0.6% flexural strength in the other hand, is the bending force
dispersant agent bwoc and 0.5% lignosulfonate required to break, or mechanically fail a bar of set cement.
retrder bwoc. Since API has no procedure to test the flexural or tensile
IADC/SPE 77213 CEMENT MATRIX EVALUATION 3

strength of oilfield cement, most lab currently use ASTM test Acknowledgements
equipment to run these tests, these test are conducted under
atmospheric and unconfined conditions. Prior to the start of The authors wish to thanks The Management of Saudi Aramco
the test, samples were cured at BHST and 3,000 psi pressure. for permission to publish this paper. Special thanks to Zuhair
The radial stress is compressive and the tangential stress is AL- Hussain, Omar Husaini, Wafik Turki and Scot Jennings
tensile, however the highest value of the tangential stress is at for their support and cooperation.
the steel/cement interface where failure should be
occurring first2. References
Figure 6 shows system 1 without the fluid loss agent and
reduction of the expansion additive to 0.75%, indicates that 1- R.B.Carpenter,et al. 1991 “The effect of Temperature and
the tensile strength is as low as 225 psi, another test was ran Cement Admix on Bond Strength”, SPE 22063.
using 0.8% fluid loss agent and better flexural strength was
achieved as indicated in figure 7. X will be assigned to a new 2- K.J, Goodwin and R.J. Crook, 1991, ” Cement Sheath
material that is under testing, figure 8 shows that system 5 Stress Failure”, SPE 20453.
produced some tensile strength at 366 psi. High-density slurry
that is commonly used in Saudi was also subjected to the same 3- M. J. Thiercelin, et al, 1997 “Cement Design Based on
testing methodology, see figures 9 & 10. Cement Mechanical Response”, SPE 38598.

4- G. Di lullo and P Rae, 2000 “Cement for long Term

Field results Isolation- Design Optimization by Computer Modeling and
Prediction” SPE 62745.
CCA pressure disappeared once the low expansion slurry was
placed in three different wells although these wells have been 5- Nelson, E.B; Well Cementing section 7, pp3-5 (1990).
fractured, showed no CCA pressure after rig moved and
after fracturing.

Conclusions

This project lead to uncovering some technical and field

realities regarding the limit of expansion agent, and it’s side-
effect on the cement matrix in a confined annulus. It is time to
readdress the expansion theory and how much expansion is
suitable for the cement.
It was clearly indicated that the tensile strength of the cement
is the most critical feature. Flexural strength can be achieved
using chemicals that are readily available in the market,
without the need to need to use expensive additives.
4 M. AL HAMMAD AND Y. AL TAMEIMI IADC/SPE 77213

36" @ 120' 36” @ 120'

34 34"
L L

30" @ 700'
” 30” @ 700'
C C

Z Z

? ?

28"
28"
24" @ 2450' 24” @ 2450
’
L
L
C
C
Z
Z

?
?

22"

CASING -
18 5/8" @ 3800
'
22"

17"
DESIGN
13 - 3/8” @ 6300’ - @ 6300
18 5/8”
(TOP Arab - D) '

Fig - 1 12"
17"

9 5/8” @ 9500’
-
13 -3/8" @ 9500
(BASE Jilh)
12"
8 - 3/8"
-
9 5/8” into K-D
7" into K - D

5 - 7/8" 8 - 3/8"

. 4 -1/2" @ TD 7" @ TD

K1 DESIGN K2 DESIGN
IADC/SPE 77213 CEMENT MATRIX EVALUATION 5

Run B. J. Services class G Operator: Kevin Koerner

Date Silo Mills # Silos used %Ret -20 Physical
2/12/2002 1 0 ok
Blaine fCaO Loss Ins Res. S/R
2920 0.39 0.56 2.78
SI02 Al203 Fe203 CA0 MgO SO3 Alk.Equiv.
22.23 3.13 4.87 63.44 2.37 1.76 0.36
C3S C3A C2 S C4AF+2C3A CaS04 C4AF
56.31 0.05 21.26 14.93 2.99 14.82

Table 1: Chemical composition of Class”G” Saudi Cement

Figure 2: The expansion property using 5% Expansive Agent.

6 M. AL HAMMAD AND Y. AL TAMEIMI IADC/SPE 77213

Picture 1:cracks associated with 2% Picture 2: cracks with 5% Expansion agent

Expansion agent

Figure 3: Expansion property using 2%

IADC/SPE 77213 CEMENT MATRIX EVALUATION 7

Figure 4: Expansion property using 1%

Figure 5: Cement Shrinkage

8 M. AL HAMMAD AND Y. AL TAMEIMI IADC/SPE 77213

Saudi G + 35% Silica Flour + .75% Expansive Agent + additives

@ 16.7 ppg

6000
5094 5131
5000

4000
Strength (psi)

3000

2000

1000
225
0
0
48 96
Time (Hrs)

Compressive Strength Flexural Strength

Figure 6: Modified system 1 with no Fluid Loss Agent

Saudi G + 35% Silica Flour + .8% Fluid Loss+ .75% Expansive Agent
+ .6% Dispersnt +.6% Retarder @ 16.7 ppg

7000
6038
6000

5000
Strength (psi)

4000

3000
1984
2000
1158
1000
287 317
0 0 0
0
0 5 10 15
X Concentration (% bwoc)

Compressive Strength Flexural Strength

Figure 7: Modified system 1 with Fluid Loss Agent

IADC/SPE 77213 CEMENT MATRIX EVALUATION 9

Saudi G + 35% Silica Flour + .8% Fluid Loss + .75% Expansive

Agent + .6% Dispersant + .5% Retarder @ 16.7 ppg

9000
7976
8000

7000
6250
6038
6000
Strength (psi)

5000

4000

3000
1984 1780
2000 1380
1000 534 704 670
395
0
48 Hrs. 48 Hrs 96 Hrs. 96 Hrs
20 % X 20 % X
Time (hrs)

Compressive Strength Flexural Strength Tensile Strength 190 F

Figure 8: Modified System 1 with tensile strength

System 5 @ 16.7 ppg

6000

5000 4774
4620
4492

4000
Strength (psi)

3460

3000

2000

1000
591
465
318 297366
0 0 0
0
48 Hrs. 48 Hrs. 96 Hrs. 96 Hrs.
20% X Time (hrs) 20% X

Compressive Strength Flexural Strength Tensile Strength 190 F

Figure 9: System 5 mechanical property

10 M. AL HAMMAD AND Y. AL TAMEIMI IADC/SPE 77213

System @ 20.7 ppg

4000
3727

3500

3000

2500
Strength (psi)

2000

1500
1248

1000

460
500 358

0
.75% Expansive Agent 5% Expansive agent
Time (Hrs)

Compressive Strength Flexural Strength Tensile Strength 190 F

Figure 10: High density compression relative to Expansion agent loading