Accepted Manuscript

A novel method to investigate cement-casing bonding using digital image correlation

F. Nath, R.J. Kimanzi, M. Mokhtari, S. Salehi

PII: S0920-4105(18)30254-7
DOI: 10.1016/j.petrol.2018.03.068
Reference: PETROL 4809

To appear in: Journal of Petroleum Science and Engineering

Received Date: 16 February 2018

Revised Date: 12 March 2018
Accepted Date: 15 March 2018

Please cite this article as: Nath, F., Kimanzi, R.J., Mokhtari, M., Salehi, S., A novel method to investigate
cement-casing bonding using digital image correlation, Journal of Petroleum Science and Engineering
(2018), doi: 10.1016/j.petrol.2018.03.068.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT
1 A Novel Method to Investigate Cement-Casing Bonding using Digital Image Correlation
2

3 Nath, F., University of Louisiana at Lafayette, E-mail: fxn1459@louisiana.edu

4 Kimanzi, R.J., University of Oklahoma, E-mail: raymos.j.kimanzi-1@ou.edu
5 Mokhtari, M., University of Louisiana at Lafayette, E-mail: mxm4487@louisiana.edu

PT
6 Salehi, S., University of Oklahoma, E-mail: saeads@gmail.com
7

RI
8

9 Abstract

SC
10 The main focus of current industry practices on the cementing of oil and gas wells is short-term cement-
11 design goals such as high strength, adequate pumpability and proper placement of cement in the wellbore.
12 However, the bond between cement and casing is a major factor for long-term integrity of wellbore which

U
13 requires more attention with the current increasing trend in horizontal drilling and hydraulic fracturing
14 operations in unconventional resources. These wells experience various kinds of forces throughout their
AN
15 life in which the casing and cement bond can play a critical role in the integrity or failure of these wells. In
16 this paper, we apply digital image correlation (DIC) technique to monitor strain development over the
17 surface of a casing filled with cement while it is diametrically compressed. In this technique, the
M

18 casing/cement sample is painted with random white and black speckles and the surface of sample is
19 recorded by camera. Finally, strain development with time is calculated using image processing technique.
20 This visualization helps to find out the points of weakness for strain accumulation which eventually end
D

21 up for fracture initiation and propagation. In this laboratory setup, the tensile strain accumulation is the
22 highest on the casing/cement interface 90-degree away from casing/loading point and the strain gradually
TE

23 increases at the casing/cement interface toward the casing/loading point. Then there is fracture generation
24 at the center of cement sample propagating toward casing/loading point. We also review the impact of
25 cement contamination, barite and water-cement ratios on cement/casing bonding and their associated DIC
EP

26 results.

27
C

28 Keywords: Casing/cement bonding, water-cement ratio, deformation, strain development, digital image
29 correlation, fracture
AC

30

31 1. Introduction
32 The main function of cement in oil and gas wells is to support casing and to prevent unintentional flow of
33 fluids into annulus, aquifers and surface. Cement failure can trigger wellbore integrity. Well integrity
34 issues could result in a number of researched environmental issues including contamination of
35 groundwater, fluid spills and escape of gas to the surface environment. These issues are most commonly
36 associated with the cement as a seal or barrier for subsurface fluids (Kiran et al. 2017). Since the drilling
 ACCEPTED MANUSCRIPT
1 engineer desires a cement with adequate strength, pumpability, fast-setting capability and a strong cement-
2 casing bond, cement composition selection is an important aspect of the drilling program.

3 Poor cement quality, casing corrosion, change in the temperature and pressure applied to the casing and
4 cement are among the main factors impacting well integrity throughout the life of a well (Nygaard et al.
5 2014). It is important to consider not only short-term well integrity factors such as cement quality and
6 pumpability at the time of cementing operation but also the long-term integrity of cement sheath such as
7 cement/casing bonding throughout the well-life.

PT
8 Cement-casing bonding issues usually arise from a number of factors which include improper casing
9 centralization, improper mud removal and improper cement mixing at the surface. Evans and Carter
10 (1962) tested the relationship between compressive strength and shear bond strength and proposed a

RI
11 correlation to obtain shear bond strength from compressive strength. The correlation determines an
12 increase in shear bond strength with the increase in compressive strength of the cement samples. Carter
and Evans (1964) applied experimental technique to study the cement bonding issues with casing. They

SC
13
14 defined the shear bond strength as the force required for casing to move in a cement sheath. Casings with
15 rough outer surfaces exhibited stronger shear bond with cement. Scott and Brace (1966) also examined the
16 effect of casing roughness on cement/casing bonding. They observed improvements in cement-casing

U
17 bonding with resin-sand coating on the outer surface of the casing before cementing. Lancaster and Gray
AN
18 (1989) studied subsurface gas migration between formations through the annular space due to poor zonal
19 isolation increased the probability of cement bond failure. They also concluded that bond strengths
20 between cement and casing increases with pipe surface roughness and the change in the internal pressure
21 of the casing will decrease the bond stability. Typical cement properties, such as tensile and compressive
M

22 strength, porosity, and permeability can be altered significantly due to cement compositional variation
23 such as water quality used in the cement slurry (Saleh et al. 2018). Salehi et al. (2016) studied the
D

24 development of geopolymer based cement slurries with regards to thickening time, compressive strength,
25 shear bond strength and durability. There was increased compressive strength (with increased
TE

26 temperature) as curing took place, and higher average shear strengths as compared to neat Portland
27 cement.

28 The effect of water-cement ratio on cement quality has been one of the earliest major factors studied to
EP

29 improve the quality of cementing such as slurry placement, thickening time and strength. In most cement
30 applications, the water-cement ratio is in the range of 40% - 50% by weight of cement (BWOC). For
31 instance, Baret (1988) studied the impact of water loss to the formation during placement of cement on
C

32 cement thickening time and yield strength. The yield strength of the cement slurry reduced and thickening
33 time increased with increase in the water content. This would affect the bonding between the casing and
AC

34 formation with the cement during hardening. Excess water that is not hydrated reduces cement strength by
35 making the cement more porous and permeable, thus reduces the strength of cement (Bourgoyne et al.,
36 1991). Shear bond strength was redefined as the shearing force between the casing and surrounding
37 cement binder with high significance placed on negative effect it has on well stability if it is low (Salehi et
38 al. 2017). Investigation done on the field data from geothermal wells in relation to oil and gas wells
39 showed that the inevitable expansion of the casing over time has significant negative effect on the cement-
40 casing bonding (Teodoriu et al. 2018).
41 Saidani et al. (2015) investigated the effects of barite as a sand substitution in cement. The results showed
42 a slight increase in compressive strength initially. This was due to the addition of high-density barite at
 ACCEPTED MANUSCRIPT
1 5% concentration by weight of cement (BWOC) which can affect the long-term durability of the cement
2 bonds in oil wells.

3 The interaction of cement with different cement additives during cement preparation on the wellsite can
4 impact the quality of cement/casing bond. Similarly, subsurface contamination with drilling fluids and
5 mud cake can also have negative impact on cement quality such as lower compressive strength and erratic
6 thickening times. This would increase the uncertainty of cement-casing bonds which can eventually lead
7 to major well integrity issues (Beach and Goins, 1957).

PT
8 Morgan and Dumbauld (1952) experimentally tested the harmful effects of contamination by treated muds
9 on cement bonding. According to his study, one of the major issues of mud contamination was the
10 extreme reduction of tensile strength of the setting cement. Treated muds are comprised of chemicals and

RI
11 polymers to improve various rheological mud properties. To counter this effect, activated charcoal was
12 added to the cement during preparation which significantly reduced the cement strength degradation and
13 improved cement bonding with casing.

SC
14 Both oil and water, as continuous phases in drilling fluids, can dissolve most of the cement additives
15 which leads to the compromise of cement integrity during cementing operations. Li et al. (2016) studied
16 the effects of oil based mud on the compressive strength, thickening time and fluidity of cement slurries.

U
17 This research revealed an increase in porosity and decrease in strength of cement after setting when
AN
18 initially mixed with mud. This affected properties such as pumpability, thickening time and strength of the
19 cement.

20 Youzhi et al. (2015) studied the effects of drilling fluid additives on cement slurry and the research
M

21 showed that particular high molecular components in the drilling fluid formed highly complex network
22 structures with the cement causing a reduction in pumpability of the slurries. With regard to particular
23 additives, Jordan et al. (1965) examined the effects of chrome-ferro lignosulfonate, as an increasingly used
D

24 drilling mud additive at the time, on cement casing bonding. The results showed drastic reductions in
25 cement strength and weakened shear bonds with the casing. However, Harder et al. (1993) studied the
TE

26 optimization of a partially water soluble emulsifier in oil based mud which increased cement performance
27 by reducing the effective mixing of the mud and cement increasing well integrity.
EP

28 The negative effect of mud cake on the cement-casing bonding is a major factor to be considered. Drilling
29 fluid will create a layer of mud cake along the permeable formation walls in the wellbore due to high
30 pressure and high temperature conditions. This layer reduces the efficiency of the bond between the
C

31 cement and the formation. Ma et al. (2007) developed a way to evaluate the effect of mud cake on cement
32 bond quality with the formation and casing. The shear bond strength of the samples created for the
AC

33 experiment increased when there was an increased rate of displacement of drilling fluids. This is because
34 the mud cake formation along the walls was reduced.
35 Other studies on mud cake removal and inhibition have also been done to reduce mud cake effect on
36 cement bonding. Gu et al. (2017) studied the influence of mud cake solidification on the thickening time
37 of cement slurry. The research showed that the addition of mud cake solidification agents (MCSAs), used
38 to counteract the sticky property of mud cake, caused a remarkable increase in hydration of the cement.
39 This resulted in shorter thickening times and increased probability of well cement failure. Hao et al.
40 (2016) studied the chemical improvement of the MCSAs and further research is being done on other
41 cement additives to solve the cement-casing bonding issues other than mud cake. Ladva et al., (2004),
42 Ladva et al., (2005) and Bybee (2005) also worked on cement to formation bonding and the existing
 ACCEPTED MANUSCRIPT
1 interfacial transition zone. They tested shear bond strength between cement and sandstone in the presence
2 of oil and water based muds. They found out that the presence of mud cake drastically reduces the cement-
3 to-sandstone bond strength from 0.8 MPa (no mudcake) to 10-3 MPa in water based muds (WBM) and
4 even more so to 10-5 MPa in oil based muds (OBM). They also documented that since oil based mud does
5 not react with the cement, the shear bond strength for the OBM cake is weaker than for the WBM cake.
6 The objective of this study is to evaluate casing-cement bonding using digital image correlation technique.
7 To this end, casing filled with cement samples were prepared and they were diametrically compressed

PT
8 while they were monitored with camera. Captured images were processed to calculate strain for six types
9 of samples with various amount of solid and liquid additives.
10

RI
11

2. Experimental Methodology

SC
12
13
14 2.1. Cement sample preparation

U
15 Class H cement was mixed with water. The other additives were mixed according to API cement
16 preparation requirements. All liquid components mixed together and all solid components mixed together
AN
17 before adding these to a cement mixer. The mud samples used in this experiment were prepared by
18 adding water, 10% BMOW (by mass of water) bentonite and barite to obtain mud density of 11 lbm/gal.
19 Cement samples were prepared in a cement mixer according to API recommendations: at 4000 RPM for
M

20 15 seconds, during which all solid components were added to the liquid components in the mixer, and then
21 12000 RPM for 35 seconds. The different samples were prepared as presented in Table 1. For 600 cc final
22 mix volume, Table 2 shows the numerical amounts and measured densities for each sample.
D

23 Table 1: Cement sample descriptions

TE

Sample Descriptions
1 Class H cement 44% water
EP

2 Class H cement 60% water

3 Class H cement 44% water with 5% barite
4 Class H cement 44% water with 5% barite and 2% bentonite
5 Class H cement 44% water with 10% mud by volume
C

6 Class H cement 44% water with 20% mud by volume

AC

Class H cement 44% water with 10% mud by volume and mud cake smeared
7 along the inner walls
24 All samples were each mixed in a cement mixer to 600cc final volume.

25 Table 2: Cement sample mix amounts for solid and liquid additives

Sample Component Density

Class H cement Water Barite Bentonite Mud (11ppg) Measured
grams grams grams grams cc ppg
1 792.1 348.5 15.8
2 654.0 392.4 14.6
 ACCEPTED MANUSCRIPT
3 779.9 343.2 39.0 16.1
4 772.2 339.8 38.6 15.4 16.1*
5 712.9 313.7 60.0 15.3
6 633.7 278.8 120.0 14.7
7 712.9 313.7 60.0 15.3
1 *This sample was not included in DIC testing.
2

PT
3 The different sample mixtures were poured into steel cylindrical molds (from the same pipe) of ID 2- inch
4 and height 1-inch, sealed with a gasket underneath and left to set and harden in air for 24 hours as shown
5 in Figure 1(a). The final step in sample preparation was to spray paint on the front face of the sample with

RI
6 a black on white speckled pattern (Figure 1b and c). During the diametrical compressive test, camera
7 records the surface of the sample and those images will be used for image processing to calculate strain.

SC
8

U
AN
M
D

10 a)
TE

11

12
EP

Loading direction
13 Steel mold
14 Cement
C

15
AC

16

17

18

19

20

21

22
 ACCEPTED MANUSCRIPT
1

4 b) c)
5

6 Figure 1: Sample preparation - a) set up showing the steel molds with cement samples, b) before painting,

PT
7 c) sample with spackle pattern after painting
8

RI
9

10 2.2. Testing Procedure: Diametrical Compression associated with optical measurements

SC
11 The cement/casing samples were tested similar to the testing conditions described by Mokhtari et al.
12 (2017) and Nath et al. (2017) for rock samples. Interested readers can explore the literatures (Sutton et al.,

U
13 2009; Pan et al., 2009; Melenka and Carey, 2015; Stirling et al., 2013) to understand the DIC principles in
14 details. The diametrical compression test was conducted using a universal testing machine. The curved
AN
15 test platens were magnetically attached to the load frames following ASTM standards (ASTM, 2008) as
16 shown in Figure 2. Servo-controlled load frame set up recorded the load and the displacement (in the
17 direction of load) over time throughout the test.
M

18 The camera was synchronized with the load frame to get the synchronized load data with camera at similar
19 time steps. The images was later processed to provide strain from this non-contact optical method.
D

20 While diametrically compressing the platens at a rate of 0.05 mm/min, the load frame was programmed to
21 record compression data every tenth of a second. The LED lights were leveled with the sample disc to
TE

22 ensure the speckle paint pattern on the disc was identified by the camera.

23 In this work, camera was operated by photogrammetry technique and recorded photographs at the same
EP

24 rate as the load frame data which were utilized to obtain strain map. Based on the test duration, the
25 captured number of images varies approximately 2,500 to 3,000 for the tested samples to capture fracture
26 growth and propagation. Once the test was completed, the images were analyzed using digital image
C

27 correlation to evaluate tensile, compressive and shear strain.

28
AC
 ACCEPTED MANUSCRIPT

PT
RI
1
2 Figure 2: Schematic diagram for experimental set up with DIC

SC
3 3. Results and Discussions
4 3.1.Force-strain diagram

U
5 The failure behavior of materials is usually characterized by load (or stress) – displacement (or strain)
6 relationship. The load–strain curves for six tested samples are shown Figure 3. Strain is calculated based
AN
7 on the comparison of the change in the length of sample with the initial length of the sample (outer
8 diameter of casing). The slope of load/strain curve gradually increases with time due to the nature of
M

9 casing/cement combination. All samples show a sudden drop in their applied force (e.g. at the strain of
10 0.0112 for sample#3) which is indicative of failure in cement/casing interface or cement itself. However,
D

11 the sample will continue building force since the casing is withholding the applied force.
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Time, sec
0 100 200 300 400 500 600 700
3500

3000 Sample
#1 Cement failure

PT
#2
2500
#3
#5

RI
#6
Load,lbf

Strain development
2000 #7 due to steel casing

SC
1500

U
1000
AN
500
M

0
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
D

Axial Strain (∆/D), inch/inch

1
TE

2 Figure 3: Load-strain diagram for tested samples

3 With DIC visualization, Figure 4 illustrates the load-strain nature of sample#1 over time while
EP

4 diametrically opposed compressive load was applied. The horizontal strain maps (εxx) at corresponding
5 time intervals are also depicted. This figure shows how horizontal strain is developed over time to
6 eventually cause failure at the cement- casing interface and later on at the center of cement sample itself.
C

7
8 3.2.Strain development
AC

9 The strain development map was obtained over time and the snapshot of these continuous measurements
10 in a few critical times are provided here (Figure 5). The terms , , , , , represent respectively the
11 initial time, and time intervals between the initial and final time. The strain maps include the horizontal
12 strain (εxx),vertical strain (εyy) ( in the direction of load) and shear strain (εxy). Analyses of these three
13 different strain maps can reveal the nature of failure in tension, compression or shear. Finally fracture
14 initiation and propagation was observed until the end of the test.
 ACCEPTED MANUSCRIPT

Time, sec
0 100 200 300 400 500 600 700
3000

2500

2000 Sample ID

PT
#1

1500
Load,lbf

RI
1000

SC
500

0
U
AN
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018
Axial Strain (∆/D), inch/inch

1
M

2 Figure 4: Load-strain nature for sample #1 with DIC visualizations

3 In the base case scenario (sample#1), the maximum horizontal strain accumulation occurs at the cement-
D

4 casing interface 90º away from the pointing load. Tensile strain eventually increases at that point to cause
TE

5 tensile failure. Later on, tensile strain builds up inside the cement sample itself at the central point of the
6 disc. Tensile strain gradually increases causing fracture initiation at the center and its propagation toward
7 loading points. On the other hand, the interface between cement and casing in the direction of pointing
8 load is the point of maximum vertical strain which eventually leads to compressive failure in some cases.
EP

9 3.2.1. Effect of water-cement ratio

10 It is expected that the more the W/C is increased, the more the strength of the cement is reduced. Under
C

11 applied testing conditions, increase of W/C ratio from 44% to 60% in neat Class H cement (Sample#2)
AC

12 resulted in a 10% less in peak load (Figure 3) as shown by the less time before initial fracture formation
13 and higher strain in both the horizontal and vertical planes (Figure 6). One should notice that the cement is
14 surrounded by the casing in this experiment, therefore the observed peak load does not represent the
15 tensile strength of the cement sample by Brazilian testing. While pure central crack observed for both
16 samples, the crack propagation was observed more erratic in sample#2. The vertical strain accumulation
17 along the lower side of the steel mold walls was irregular. This was a result of existence of free water
18 during curing that reduced the bonding effect between the pipe and the cement. When there is an excess of
19 water, it bleeds out onto the surface which may be responsible for subsurface gas migration through the
20 micro-crack and passages during cement bond failure. This is in agreement with the proposed reduction in
21 mechanical properties due to increased W/C ratio by Baret 1988.
 ACCEPTED MANUSCRIPT

PT
RI
SC
1

U
2 Figure 5: Strain (εxx, εyy and εxy) development in Class H cement with 44% water (Sample#1)
AN
M
D
TE
C EP
AC

4 Figure 6: Strain (εxx, εyy and εxy) development in Class H cement with 60% water (Sample#2)

5 3.2.2. Effect of barite

6 The use of barite in cement typically increased the slurry density and caused significant reduction in
7 pumpability. In this study, addition of barite in Class H cement resulted slightly slower in fracture
8 initiation with higher peak load. As investigated by Sadani et al. (2015), addition of 5% or less of barite
9 would cause an initial increase in the compressive strength of the sample. The fracture propagation was
 ACCEPTED MANUSCRIPT
1 not along the central axis but it was slightly off-center as shown in Figure 7. This could be result of
2 inhomogeneous distribution of barite in the sample.

PT
RI
U SC
AN
3
4 Figure 7: Strain (εxx, εyy and εxy) development in Class H cement with 44% water
M

5 and 5% barite (Sample#3)

6
D

7 3.2.3. Effect of mud contamination

Generally, water-based systems are the most susceptible to contamination. Failure of the cement/casing
TE

8
9 bond occurs almost everywhere in the cement/casing interface in samples #5 (Figure 8) and sample#6
10 (Figure 9) With a 10% mud contamination, the initial fracture was formed slower in sample #5 than in
11 neat Class H cement (Sample#1). Sample #6 had a much faster strain deterioration than sample #5 because
EP

12 of the higher mud contamination (20%).

13
C

14 3.2.4. Effect of mud contamination with mud cake bond hindrance

AC

15 In sample#7, there was complete failure of cement along the inner walls as the mud cake reduced the
16 bonding strength between the cement and steel mold. Figure 10 shows evidence of this failure at the top
17 and bottom with the significant irregularities on the left and right of the mold. There was also a much
18 faster deformation reported before initial fracture development. The failure at the cement-casing interface
19 was a mixture of tensile failure on left and right sides of sample and compressive failure on the top and
20 bottom sides.
21
 ACCEPTED MANUSCRIPT

PT
RI
U SC
1
AN
2 Figure 8: Strain (εxx, εyy and εxy) development in Class H cement with 44% water and 10%
3 mud by volume (Sample#5)
M
D
TE
C EP
AC

4
5 Figure 9: Strain (εxx, εyy and εxy) development in Class H cement with 44% water and 20% mud
6 by volume (Sample#6)
 ACCEPTED MANUSCRIPT

PT
RI
U SC
1
AN
2 Figure 10: Strain (εxx, εyy and εxy) development in Class H cement 44% water with
3 10% mud by volume and mud cake smeared along the inner walls (Sample#7)
4
5 4. Conclusions
M

6 Digital image correlation is an effective non-contact full-field technique to investigate strain pattern in
7 complex systems such as the interface of casing with contaminated cements. In this research, we applied
D

8 DIC to study systematically different causes of cement–casing bonding issues including cement
9 contamination, addition of barite and change in water-cement ratios during cementing operations.
TE

10 Following conclusions were made by comparing neat Class H cement, with other cement slurry recipes
11 used in this study:

•
EP

12 The highest point of tensile strain is at the cement/casing interface 90-degree away from the
13 loading points and the tensile fracture initially starts there and propagates toward the loading point
14 along the cement/casing interface.
•
C

15 The highest point of compressive strain is at the cement/casing interface at the loading point and
16 the compressive strain decreases gradually away from the loading point. In some cases such as
AC

17 mud contamination and the presence of mudcake, there is compressive failure at the loading point.
18 • After the initial crack generated at the cement/casing interface 90-degree from the pointing load,
19 crack initiates at the center of cement sample and propagates toward the loading point.
20 • In some cases such as barite or mud cake addition, there is irregular or off-center fractures and
21 strain accumulation due to the inhomogeneity in the samples.
22

23 Acknowledgements
24 The authors are thankful to Correlated Solutions for providing DIC software.
25
 ACCEPTED MANUSCRIPT
1 Abbreviations
2
3 BMOW = By mass of water
4 BWOC = By weight of cement
5 MCSA =Mud cake solidification agents
6 DIC = Digital image correlation
7 = Strain in the x-direction
8 = Strain in the y-direction
9 = Strain in the xy-plane

PT
10

11 References

RI
12 ASTM, 2008. Standard test method for splitting tensile strength of intact rock core specimens.
13 Baret, J.F. 1988. Why Cement Fluid Loss Additives are necessary. Paper SPE 17630 presented at the SPE

SC
14 International Meeting on Petroleum Engineering, Tianjin, China, 1 – 4 November.
15 http://dx.doi.org/10.2118/17630-MS

16 Bybee, K. (2005, August 1). The Cement-to-Formation Interface in Zonal Isolation. Society of Petroleum

U
17 Engineers. http://dx.doi.org/10.1016/10.2118/0805-0041-JPT

18 Beach, H.J. and Goins, Jr., W.C. 1956. A Method of Protecting Cements against the Harmful Effects of Mud
AN
19 Contamination. Paper SPE 679 presented at the Petroleum Branch Fall Meeting, Los Angeles, California,
20 U.S.A., 14 – 17 October.

21 Bourgoyne Jr., A.T., Chenevert, M.E., Millheim, K.K. and Young Jr., F.S. 1991. Applied Drilling Engineering, Vol.
M

22 2. Richardson, Texas: Textbook Series, SPE.

23 Carter, L.G. and Evans, G.W. 1964. A Study of Cement-Pipe Bonding. Journal of Petroleum Technology 16(2):
D

24 157-161. https://dx.doi.org/10.2118/764-PA
TE

25 Evans, G.W. and Carter, L.G. 1962. Paper API 67-072 presented at the API Division of production spring meeting
26 of the South-western district, March.

27 Gu, J., Huang, J. and Hao, H. 2017. Influence of mud cake solidification agents on thickening time of oil well
EP

28 cement and its solution. Construction and Building Materials 153: 327-336.
29 https://doi.org/10.1016/j.conbuildmat.2017.07.128

30 Harder, C.A., Carpenter, R.B., Freeman, E.R., Brookey, T.E. and Gandy, R.G. 1993. Optimization of Oil-Base Mud
C

31 Chemistry for Cementing. Paper SPE 25183 presented at the SPE International Symposium on Oilfield
32 Chemistry, New Orleans, Louisiana, U.S.A., 2 – 5 March. https://dx.doi.org/10.2118/25183-MS
AC

33 Hao, H., Gu, J., Huang, J., et al. 2016. Comparative study on cementation of cement-mudcake interface with and
34 without mud-cake-solidification-agents application in oil and gas wells. Journal of Petroleum Science and
35 Engineering 147: 143-153. https://doi.org/10.1016/j.petrol.2016.05.014

36 Jordan, A.L., Overton, H.L., and Weaver, W.B. 1965. The Effect of Chrome-ferro Lignosulfonate on cement
37 bonding. SPWLA The Log Analyst 5(4): 13-25.

38 Kiran, R., Teordoriu, C., Dadmohammadi, Y. et al. 2017. Identification and evaluation of well integrity and causes
39 of failure of well integrity barriers (A review). Journal of Natural Gas Science and Engineering 45: 511-526.
40 https://doi.org/10.1016/j.jngse.2017.05.009
 ACCEPTED MANUSCRIPT
1 Lancaster, G. and Gray, S. 1989. Advances in Cement Slurry Application. Paper CIM 558909 presented at the
2 Petroleum Society of CIM technical meeting of the South Saskatchewan Section, Regina, Sasketchewan,
3 Canada, 25 – 27 September. http://dx.doi.org/10.2118/SS-89-09

4 Ladva, H. K. J., Craster, B., Jones, T. G. J., Goldsmith, G., & Scott, D. (2005, September 1). The Cement-to-
5 Formation Interface in Zonal Isolation. Society of Petroleum Engineers. http://dx.doi.org/10.2118/88016-PA.

6 Ladva, H. K. J., Craster, B., Jones, T. G. J., Goldsmith, G., & Scott, D. (2004, January 1). The Cement-to-Formation
7 Interface in Zonal Isolation. Society of Petroleum Engineers. http://dx.doi.org/10.2118/88016-MS

PT
8 Li, Z., Liu, H., Guo, X., et al. 2016. Contamination of cement slurries with oil based mud and its components in
9 cementing operations. Journal of Natural Gas Science and Engineering 29: 160-168.
10 https://dx.doi.org/10.1016/j.jngse.2016.01.003

RI
11 Ma, Y., Cui, M., Guo, X., Shi, Q. and Li, L. 2007. How to evaluate the Effect of Mud cake on Cement Bond Quality
12 of Second Interface. Paper SPE/IADC 108240 presented at the SPE/IADC Middle East Drilling Technology
13 Conference and Exhibition, Cairo, Egypt, 22 – 24 October. http://dx.doi.org/10.2118/108240-MS

SC
14 Melenka, G.W. and Carey, J.P. 2015. Evaluation of Fiber Reinforced Cement Using Digital Image Correlation.
15 PLoS ONE 10(6): https://doi.org/0.1371/journal.pone.0128644

U
16 Mokhtari, M., Hayatdavoudi, A., Nizamutdinov, R., Rizvi, H. and Nath, F. 2017. Characterization of Complex
17 fracture propagation in Naturally Fractured Formations Using Digital Image Correlation technique. Paper SPE
AN
18 184826 presented at the SPE Hydraulic Fracturing Conference and Exhibition, Woodlands, Texas, U.S.A., 24 –
19 26 January. http://dx.doi.org/10.2118/184826-MS

20 Morgan, B.E. and Dumbauld, G.K. 1952. Use of Activated Charcoal in cement to combat effects of contamination
M

21 by Drilling Muds. Journal of Petroleum Technology 4(9): 225-232. https://doi.org/10.2118/952225-G

22 Nath, F., Salvati, P. E., Mokhtari, M., Seibi, A., and Hayatdavoudi, A., 2017. Observation of Fracture Growth in
D

23 Laminated Sandstone and Carbonate Rock Samples under Brazilian Testing Conditions Using Digital Image
24 Correlation Technique. Society of Petroleum Engineers. http://dx.doi.org/10.2118/187515-MS
TE

25 Nygaard, R., Salehi, S., Welderman, B. and Lavole, R. 2014. Effect of Dynamic Loading on Wellbore Leakage for
26 the Wabamun Area CO2-Sequestration Project. Journal of Canadian Petroleum Technology 54(1): 69-82.
27 https://doi.org/10.2118/146640-PA
EP

28 Pan, B., Qian, K.M., Xie, H.M., and Asundi, A., 2009. Two-dimensional digital image correlation for in-plane
29 displacement and strain measurement: a review. Measurement Science Technology 20:062001.
C

30 Saleh, F.K., Rivera, R., Salehi, S. et al. 2018. How Does Mixing Water Quality Affect Cement Properties. Paper
31 SPE 189505-MS presented at the SPE International Conference and Exhibition on Formation Damage Control,
AC

32 Lafayette, Louisiana, U.S.A. 7 – 9 February.

33 Salehi, S., Khattak, M.J., Ali, N. and Rizvi, H.R. 2016. Development of Geopolymer-based Cement Slurries with
34 Enhanced thickening time, Compressive and Shear bond strength and durability. Paper IADC/SPE 178793
35 presented at the IADC/SPE Drilling Conference and Exhibition, Fort Worth, Texas, U.S.A., 1 - 3 March.
36 http://dx.doi.org/10.2118/178793-MS

37 Salehi, S., Khattak, M.J., Bwala, A.H. and Karbalaei, S.F. 2017. Characterization, morphology and shear bond
38 strength analysis of geopolymers: Implications of oil and gas well cementing applications. Journal of Natural
39 Gas Science and Engineering 38: 323-332. http://dx.doi.org/10.1016/j.jngse.2016.12.042

40 Scott, J.B., and Brace, R.L. 1966. Coated Casing “A Technique for Improved Cement Bonding.” Paper API 66-043
41 presented at the API Spring meeting of the Mid-Continent District, New York, New York, U.S.A., 1 January.
 ACCEPTED MANUSCRIPT
1 Shook, E.H., Frisch, G.J. and Lewis, T. 2008. Cement Bong Evaluation. Paper SPE 108415 presented at the SPE
2 Western Regional and Pacific Section AAPG Joint Meeting, Bakersfield, California, U.S.A., 31 March – 2
3 April. http://dx.doi.org/10.2118/108415-MS

4 Stirling, R.A., Simpson, D.J., and Davie, C.T., 2013. The application of digital image correlation to Brazilian testing
5 of sandstone. International Journal of Rock Mechanics and Mining Sciences 60:1-11.
6 http://dx.doi.org/10.1016/j.ijrmms.2012.12.026

7 Sutton, M.A., Orteu. J.J., and Schreier, H.W., 2009. Image Correlation for Shape, Motion and Deformation
8 Measurements. Springer. http://dx.doi.org/10.1007/978-0-387-78747-3

PT
9 Teodoriu, C., Yi, M. C., Ichim, A. and Salehi. S. 2018. A Novel View of Cement Failure with Application to
10 Geothermal Well Construction. Paper presented at the 43rd Workshop on Geothermal Reservoir Engineering,

RI
11 Stanford, California, U.S.A. 12 – 14 February.
12 Youzhi, Z., Chaoyi, S., Kunquan, Y. et al. 2015. Contamination effects of drilling fluid additives on cement slurry.
13 Natural Gas Industry B 2: 354-359. https://doi.org/10.1016/j.ngib.2015.09.009

U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

-a review of significant cement-casing bonding issues in past research

-experimental analysis using Digital Image correlation

-investigation of deformation and strain development in cement

PT
RI
U SC
AN
M
D
TE
C EP
AC