Cement and Concrete Composites 74 (2016) 54e70

Contents lists available at ScienceDirect

Cement and Concrete Composites

journal homepage: www.elsevier.com/locate/cemconcomp

Degradation of well cement in HPHT acidic environment: Effects

of CO2 concentration and pressure
Omotayo Omosebi a, Himanshu Maheshwari a, Ramadan Ahmed a, *, Subhash Shah a,
Samuel Osisanya a, 1, Shokrollah Hassani a, 2, Gunnar DeBruijn b, Winton Cornell c,
Dave Simon d
a
The University of Oklahoma, United States
b
Schlumberger Limited, United States
c
The University of Tulsa, United States
d
DES Consulting, United States

a r t i c l e i n f o a b s t r a c t

Article history: Application of cement in high-pressure high-temperature (HPHT) carbonic acid containing environment
Received 24 November 2015 poses serious well integrity issues. HPHT condition facilitates acid attack on well cement resulting in the
Received in revised form loss of structural integrity.
30 July 2016
This paper presents the results of an experimental study conducted to investigate the effects of CO2
Accepted 9 September 2016
Available online 9 September 2016
concentration (mole fraction) and pressure on mechanical integrity of degraded Classes G and H cements
after exposure to carbonic acid environment. Cement cores were prepared and aged in autoclave filled
with 2% NaCl solution saturated with mixture of methane and carbon dioxide gases for durations up to
Keywords:
Well cement
28 days. Experiments were performed by varying test pressure and CO2 concentration. Compressive
Compressive strength strength, porosity and permeability of unaged and aged samples were measured to assess the level of
Carbon dioxide degradation. Besides, analysis of the mineralogy, the microstructure, and the morphology of specimens
Acid attack was conducted.
Degradation After exposure, incomplete hydration of clinker materials, carbonation of cement hydrates, leaching of
Carbonation by-products of carbonation, and structural transformation of amorphous calcium silicate hydrate to its
crystalline form were observed. Generally, compressive strength increased with CO2 concentration due to
carbonation of calcium hydroxide (CH) and calcium silicate hydrate (C-S-H). Correspondingly, porosity
and permeability decreased. Furthermore, degree of degradation is greatly influenced by pressure.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction tetracalcium aluminoferrite (Ca4Al2Fe2O10 or C4AF). The major dif-

Cement is placed around the casing during drilling operation to
protect it from corrosive fluids, to provide mechanical support, and
to isolate productive zones thereby ensuring long-term operational
integrity of the well. The most commonly used API well cements in
the oil and gas industry are Classes G and H cements. These con-
ventional Portland-based cements contain four main compounds
namely, tricalcium silicate, (Ca3SiO5 or C3S), dicalcium silicate
(Ca2SiO4 or C2S), tricalcium aluminate (Ca3Al2O6 or C3A) and
* Corresponding author.
E-mail address: r.ahmed@ou.edu (R. Ahmed).
1
Currently with the Petroleum Training Institute Abu Dhabi.
2
Currently with BP.
ferences between these two classes of cements are their relative
grain sizes, slurry density and quantity of mixing water required
per sack (i.e. 94 pounds in oilfield standard) of cement. Class G
cement is finer than Class H cement [1]. API recommend adding
44% by weight of cement (BWOC) mixing water to Class G cement
and 38% BWOC mixing water to Class H cement to attain the same
level of slurry consistency. In special applications, Class H cement
can be formulated with lower water contents than Class G cement.
Compressive strength of cement decreases with increase in water
content [2e4]. However, the performance of Class G cement rela-
tive to Class H cement depends on other parameters such as cement
composition, particle size and curing time.
C3S, C2S, C3A, and C4AF are hydrated through a curing process.
Hydration of C3S and C2S produces amorphous calcium silicate

http://dx.doi.org/10.1016/j.cemconcomp.2016.09.006
0958-9465/© 2016 Elsevier Ltd. All rights reserved.
 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70
hydrate (CSH) and calcium hydroxide (CH), both of which aid early
strength development. Although compressive strength in the range
of 0.7e5 MPa is generally sufficient to continue drilling after the
casing is cemented, further hydration and structural (i.e., phase)
transformations occur throughout the productive life of the well,
even after plugging and abandonment. These phase trans-
formations, which are mainly dictated by prevailing down-hole
temperature and lime-to-silica ratio (i.e., CaO/SiO2 or C/S ratio) in
the cement, can alter the physical and mechanical properties of the
set slurry. Therefore, normal field practice in slurry design involves
addition of 35e40% silica BWOC to the base slurry to prevent
strength retrogression that occurs at elevated temperature [1,5,6].
In wells where CO2 is injected (e.g. CO2-EOR flooding, geological
carbon storage, etc.) or produced as associated gas (e.g. deep sour
gas condensate and volatile oil fields), high down-hole temperature
and pressure conditions result in further alteration in mechanical
properties of cement, which can cause degradation after prolonged
exposure. This aggressive environment results in severe mechani-
cal damage and ultimate failure of cement sheath, potentially
leading to micro-channelling and formation of micro-annuli.
Comprehensive review of previous studies [2,5,7e37] indicates
that structural transformation, carbonation and leaching are the
primary mechanisms responsible for physical and mechanical
degradation of well cement after exposure to carbonic acid under
HPHT condition.
Cement degradation due to structural transformation is a phe-
nomenon that occurs mainly during cement curing at elevated
temperature [7,27e32]. High temperature favors structural trans-
formation of amorphous CSH to several crystalline polymorphs,
generally identified as C-S-H. In temperature range of 60e150
C
and C/S ratio of approximately 1, tobermorite gel is the predomi-
nant phase formed with good binding properties, high compressive
strength and low permeability. Several forms of tobermorite have
been reported which include 9 Å tobermorite or riversideite
(Ca5Si6O16(OH)2), 10 Å tobermorite or oyelite (Ca10Si8-
B2O29$12H2O), 11 Å tobermorite or normal tobermorite (Ca5-
xSi6O17-2x(OH)2x.5H2O), and 14 Å tobermorite or plombierite
(Ca5Si6O16(OH)2$7H2O) [30,31,38e41]. Above 150
C (302
F), a
moderately permeable but strong calcium silicate hydrate phase,
called xonotlite (Ca6Si6O17(OH)2 or C6S6H), is formed [8e10]. In
addition to these, other phases such as truscottite and pectolite can
be formed at different temperatures and C/S ratios. The micro-
structure, mineralogy and morphology of these minerals are
commonly characterized using techniques such as Fourier trans-
form infrared (FTIR), scanning electron microscopy (SEM), X-ray
diffraction (XRD), energy-dispersive X-ray (EDX) spectroscopy and
nuclear magnetic resonance (NMR) [9e15].
Cement degradation due to CO2 attack involves the following
chemical reaction steps:
Step 1: Formation of carbonic acid
CO2(aq.) þ H2O(l) # H2CO3 (aq.)
Step 2: Carbonation of the cementitious compounds
(i) Carbonation of calcium hydroxide
Ca(OH)2(s) þ H2CO3(aq.) / CaCO3(s)þ2H2O(l)
(ii) Carbonation of calcium silicate hydrate
(a) Complete carbonation
55
3CaO$2SiO2$3H2O(s) þ 3H2CO3(aq.) / 3CaCO3(s)
þ 6H2O þ 2SiO2(s) (3)
(b) Incomplete carbonation
3CaO$2SiO2$3H2O(s) þ H2CO3(aq.) / C-S-C (4)
Step 3: Leaching out and precipitation of calcium carbonates
(i) CaCO3(s) þ H2CO3 (aq.) # Ca(HCO3)2(s) (5)
(ii) Ca(HCO3)2(s) þ Ca(OH)2(s) / 2CaCO3(s) þ 2H2O(l) (6)
Carbonation produces solid carbonates (e.g. calcite, aragonite,
etc.), which reduces transport properties (porosity and perme-
ability) and improves mechanical strength. Conversely, leaching
consumes additional portlandite (the naturally occurring form of
CH) and dissolves calcium carbonate. This increases transport
properties and decreases mechanical strength. Generally, por-
tlandite is attacked at faster rate and is the first to completely leach
out before C-S-H because it is more reactive than C-S-H
[8,11e13,16]. Field results of cement samples retrieved from three
wells confirm this phenomenon [21e23]. In Eq. (4), C-S-C refers to
calcium silicate carbonate group which comprises scawtite
(Ca7(Si6O18)(CO3)$2H2O), fukalite (Ca4Si2O6(OH)2CO3), spurrite
(Ca5(SiO4)2CO3), tilleyite (Ca5(Si2O7)(CO3)2) and galuskinite
(Ca7(SiO4)3CO3) [42]. Formation of these carbonates limits the re-
action in step 3. Previous studies [28,42] conclude that scawtite is a
good mineral for strength reinforcement if it is present in trace
amount. However, strength degradation occurs if it becomes the
dominant phase in cement.
The rates of carbonation and leaching of the hydrated products
(i.e. CH and C-S-H) are controlled by several parameters such as
temperature, pressure, cement formulation, CO2 concentration, and
salt concentration in the surrounding liquid. At elevated tempera-
ture and pressure, carbonation and leaching fronts progress faster
in Class G than Class H cement [13,15e18]; hence, strength is lost
more rapidly in Class G than Class H cement, producing a porous,
permeable and weak cement. This trend continues as long as the
supply of carbonic acid is uninterrupted until the cement is
completely leached out resulting in weak amorphous silica.
Although carbonated cement is susceptible to leaching, disso-
lution induced leaching is possible even in the absence of carbonic
acid [24,34,35]. However, this is comparatively a very slow process.
This type of leaching occurs due to the concentration gradient
developed when cement encounters deionized water. This results
in diffusion of calcium ions out of the cement, which triggers the
(1) dissolution of calcium hydroxide that often leads to reduction in the
calcium content of the solid matrix.
Limited studies have been reported on cement degradation due
to carbonic acid attack under HPHT condition. To provide better
understanding on the structural integrity of cements used to
complete wells that are exposed to CO2-containing fluids, addi-
tional studies are still needed. A scientific understanding of how
(2) cement performance is altered by environmental variables to
which it is exposed is an important precursor for engineering
cement that would resist such extreme downhole conditions. This
study investigates the effects of CO2 concentration and pressure on
cement integrity. New experiments were conducted and com-
plemented with data from previous study [43]. Throughout the
56 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70
tests, temperature was held constant at 177
C (350
F) while
pressure and CO2 gas composition were varied from 21 MPa (3000
psi) to 62 MPa (9000 psi) and 10e100%, respectively.
2. Materials and methods
2.1. Materials
2.1.1. Slurry preparation
Table 1 presents the composition of baseline cement, additives
and water in each slurry formulation. API RP 10B [44] procedure
was used to prepare the slurry. Density, filtration loss and thick-
ening time were measured after preparation of cement slurries.
Table 2 summarizes slurry properties in each batch of curing and
CO2 experiment. Slurry density was measured using oilfield stan-
dard atmospheric OFITE mud balance. A good match was obtained
when measurement was compared to theoretical computation and
these were within practical range used in the field. CTE Model 200
atmospheric consist-o-meter was used to measure consistency and
thickening time of the slurries at 135
F. Omosebi et al. [43] pre-
sents consist-o-meter readings for both classes of cement. Low-
pressure filter press (OFITE Model 142-53) was used to measure
filtration loss at 100 psi and ambient temperature. The press has a
cell body to hold slurry sample, a pressure inlet for pressurization, a
base cap with screen and filter paper to simulate permeable for-
mation, CO2 charger to pressurize the sample and pressure regu-
lator to accurately control the cell pressure.
2.1.2. Curing and specimen preparation
The slurry was poured into rubber moulds and cured in 2% NaCl
solution for 5 days at 93
C using a pressure vessel placed in an
oven. Cylindrical cores were cut from the cured moulds. Dimension
of cores is approximately 38.10 mm in length and 25.4 mm in
diameter. The cores were divided into two parts. The first part,
which comprises 10 Class G and 10 Class H samples, were aged in
HPHT autoclave. The second part (unaged specimens), which con-
tains 5 Class G and 5 Class H samples, were kept in 2% NaCl solution
for 14 days at ambient condition.
2.2. Methods
2.2.1. Specimen aging experiment
Cement cores were aged for 14 days in HPHT autoclave (Fig. 1)
containing static 2% NaCl solution. Some specimens were aged for 6
and 28 days to study the effect of exposure time. Oil is circulated
through the heating jacket to increase temperature. Methane (CH4)
and carbon dioxide (CO2) were used to pressurize the system. The
compositions of these gases were varied using an injection cylinder.
In each batch of experiment, no measurable changes were observed
in the compositions of the injected CH4 and CO2 gases because the
Table 1
Composition of cement, water and additives.
Materials Composition
Class G
Cement 100% BWOCa
Water for Cement 44% BWOCa
Water for Silica Flour 38.5% BWOSFb
Silica Flour 35% BWOCa
HEC 0.1% of water for Silica
Anti-Foaming Agent 0.1 gal/sack (94 lb.) of cement
a
BWOC ¼ by weight of cement.
b
BWOSF ¼ by weight of silica flour.
Table 2
Properties of cement slurries.
Property Equipment Class H Class G
Theoretical Density, kg/m3 1912 1859
Measured Density, kg/m3 OFITE mud balance 1941 1851
API fluid loss (mL) OFITE Model MB Filter Press 200 240
Thickening Time, min CTE Model 200 consist-o-meter 170 135
amount of these gases dissolving in 2% NaCl solution is small
compared to the high pressures at which they were injected into
the HPHT autoclave. Throughout the experiment, temperature was
kept constant at 177
C while CO2 concentration and total test
pressure were varied. After each test, the system is gradually dep-
ressurized. System pressurization and depressurized rates were
maintained approximately at 0.12 MPa/min (1000 psi/hr). Table 3
shows the test matrix for 11 batches of experiments. Some data
from previous study [43] are combined with new data to complete
the dataset required for investigating the effects of variation in CO2
concentration and pressure on cement integrity.
2.2.2. Measurement of mechanical and transport properties
A number of techniques [24,45,46] have been proposed for
measuring and reporting compressive strength of cements. The
ASTM C39/C39M [46] standard procedure is adopted in this study.
Mechanical testing apparatus [43] was used to measure the
compressive strength at 6.9 MPa confining pressure. ASTM stan-
dard stress loading rate of 0.24 ± 0.05 MPa/s was maintained
throughout the test.
Previous similar studies [14,17,47,48] on compressive strength
showed significant scattering in measurements. In order to mini-
mize scattering and establish a reasonable trend, the values of
compressive strength reported for each class of cement were
averaged (three for unaged samples and six for aged samples). In
addition, the measured values of compressive strength were cor-
rected according to ASTM C39/C39M procedure since length to
diameter ratio (L/D) of the cores was slightly less than the standard
(2.00).
Porosity and permeability were obtained to analyze and inter-
pret results obtained from compressive strength measurements.
Exposed and unexposed cores were dried in an oven at 43
C for
24 h, then porosity and gas permeability were measured using
Automated Permeameter (Coretest AP-608). The values of porosity
and gas permeability reported for each class of cement were
averaged (two for unaged samples and four for aged samples).
2.2.3. Material characterization
Dried cores were cut transversely and information about
mineralogy were collected from three unique zones (inner, middle
and outer zones). Using FTIR equipment (Nicolet 6700), FTIR
mineralogy analysis was conducted on the specimens to measure
Class H
100% BWOCa
38% BWOCa
38.5% BWOSFb
35% BWOCa
0.1% of water for Silica
0.1 gal/sack (94 lb.) of cement
 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70 57

P3
Injection Cylinder
DAQ System

Oil P1
Reservoir

Oil Pump

Gas
P2
Heating Jacket
CH4 CH4-H2S CO2
T Cylinder Cylinder Cylinder
Heating Oil Specimens

Aging Cell

Fig. 1. Schematic of the aging cell setup (adopted from Omosebi et al. [43]).

Table 3
Test matrix during CO2 aging experiment.

Batch number Temperature (
C) Total pressure (MPa) CO2 (%) CH4 (%) Exposure time (days)

01 177 21 10 90 14
02 177 21 40 60 14
03a 177 21 100 0 14
04 177 42 10 90 6
05 177 42 10 90 28
06a 177 42 10 90 14
07a 177 42 40 60 14
08a 177 42 100 0 14
09 177 62 10 90 14
10 177 62 40 60 14
11a 177 62 100 0 14
a
Data source: Omosebi et al. [43].

the degree of carbonation and leaching in each of these regions. In
addition, XRD analysis was performed on some of the cores to
quantify the composition of minerals present. XRD runs were car-
ried out on a RIGAKU MiniFlex 600 outfitted with a Cu X-ray tube, a
six-position sample changer, a NaI scintillation counter, and a
diffracted-beam monochromator. Powdered samples were pressed
into standard 1-inch (25-mm) round holders and run from 4 to 65

two-theta in step-mode, with two seconds/step counting time. A
Class H reference cement was also run. Sample mineralogy was
quantified using Rietveld refinement; for this, RIQAS software from
MDI Inc. was used. The amorphous component was quantified by
adding an internal standard (alpha-alumina) at a loading of 15 wt
percent and rerunning the powders. Modeling was then carried out
with no amorphous component, thus yielding an over-
determination of the alumina content (e.g. 25%). Normalization to
the known content then allowed for estimation of the general
amorphous content. Besides, SEM imaging and EDX spectroscopy
were obtained using FEI equipment equipped with back-scattered
electron (BSE) imaging to ascertain and examine different
degraded regions, identify the interface between them, and to es-
timate elemental composition of minerals in these zones.
3. Results and discussion
Effects of CO2 concentration and pressure on cement degrada-
tion are presented in this section. Evidence of alteration in miner-
alogy, microstructure and morphology are first presented using
visual inspection, FTIR, XRD, SEM, and EDX measurements. Subse-
quently, measurement of compressive strength is presented and
supported with porosity and permeability.
3.1. Visual inspection and FTIR mineralogy
Photos of the degraded specimens are shown in Figs. 2e4. The
cracked pieces in Class H are not due to aging but the specimens
were deliberately broken in order to conduct FTIR studies. In order
to study the effect of exposure time on degradation, first three
batches of experiments were conducted varying aging time from 6
to 28 days (Fig. 2). Then, the remaining tests were performed aging
the specimens for 14 days while varying CO2 concentration and
total pressure (Figs. 3 and 4). Class G cement shows uneven visual
degradation (discoloration) than Class H cement although the later
suggests deeper CO2 penetration than the former. Evidently, in-
crease in time, CO2 concentration and total test pressure leads to
58 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70

Fig. 2. Photos of degraded specimens after exposure to carbonated brine at 177
C, 42 MPa and 10% CO2 concentration: (a) Class G; (b) Class H (* from Omosebi et al. [43]).

Fig. 3. Photos of degraded Class G specimens after exposure to carbonated brine at 177
C and: (a) 21 MPa; (b) 42 MPa; (c) 62 MPa (* from Omosebi et al. [43]).

further chemical attack. However, variation in CO2 partial pressure addition, increased exposure time allows more reaction to take
and time has more impact than variation in total test pressure. This place. As discussed previously, this dissolution leads to carbonation
is expected because methane has low solubility in brine, which of cement hydrates (CH and CSH). Visual inspection of the degraded
means that CO2 solubility governs the degradation process. In cores confirms that the degree of cement degradation increases
 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70
Fig. 4. Photos of degraded Class H specimens after exposure to carbonated brine at 177
C and: (a) 21 MPa; (b) 42 MPa; (c) 62 MPa.
with CO2 concentration.
FTIR-based mineralogy analysis was conducted to characterize
the minerals present in the degraded specimens and to assess the
degree of degradation and alteration due to chemical attack. In this
analysis, three points were selected to collect FTIR samples, one
Fig. 5. Three selected zones in degraded specimens used for FTIR analysis: inner (red),
middle (blue) and outer (yellow) zones. (For interpretation of the references to colour
in this figure legend, the reader is referred to the web version of this article.)
59
from each of the outer, middle and inner zones of Class H speci-
mens (Fig. 5). Absorbance peaks of FTIR spectrum at wavenumbers
specific to a mineral can be used to identify the presence of the
mineral. The intensity (i.e. height) of this peak is directly related to
the amount of the mineral present. For instance, qualitative
assessment of carbonation and leaching can be conducted using
FTIR absorbance peaks. Ylme
n et al. [49] applied this technique
successfully to monitor the formation of carbonates in cement.
Each peak of absorbance at a certain wavenumber in the FTIR
spectra denotes a specific mineral. The distinct absorbance peaks of
binding components of cement (CH and CSH) and expected
degradation products were used to identify the minerals present.
Calcium silicate hydrate, calcium carbonate and calcium hydroxide
are the major minerals expected after exposure, each exhibiting
approximate peaks of 950, 1480 and 3645 cm1, respectively
[49e58].
Fig. 6 compares FTIR spectra of unaged and aged specimens. At
wave number 3645 cm1, the unaged cores and inner/middle zones
of the aged cores show CH absorbance peak. This implies that CH
has been consumed in the outer zone through carbonation and/or
leaching reactions. At 950 cm1, similar observation is made in the
inner and middle zones of the aged specimen as well as the unaged
specimen. This indicates that C-S-H is preserved in these zones. In
the outer zone, absorbance peak of CaCO3 occurs at 1480 cm1 due
to the consumption of CH and C-S-H in this zone. In the inner and
middle zones, weak absorbance peaks of CaCO3 suggest limited
penetration of carbonic acid. This mineralogical proof substantiates
the existence of maximum carbonation in the outer zone of the
aged sample, which arises from direct exposure of this zone to
60 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70
Fig. 6. FTIR spectra of unaged and aged specimens of Class H cement.
carbonated brine.
To investigate the effect of CO2 concentration and pressure on
mineralogy, FTIR absorbance peaks of calcium hydroxide (CH),
calcium silicate hydrate (C-S-H) and carbonates for the three
degraded zones were obtained and presented in Fig. 7 in normal-
ized form. CO2 concentration effect at 21 MPa (Fig. 7a) confirms
formation and precipitation of carbonates, which results from the
reaction of cement hydrates, especially portlandite. Higher absor-
bance peak in the outerzone indicates that the carbonates are
derived from the carbonation of both CH and C-S-H. In addition to
carbonation across all zones, the retention of C-S-H in the inner and
middle zones preserves mechanical strength. At 42 MPa (Fig. 7b),
both calcium-bearing hydrated products (i.e. portlandite and C-S-
H) are intact in the inner and middle zones, especially with 10% and
40% CO2 concentrations. At 100% CO2 concentration, calcium sili-
cate hydrate was preserved in the inner region; however, calcium
hydroxide was completely consumed across all zones to form
CaCO3. In the outer zone, regardless of CO2 concentration, calcium-
bearing hydrates were completely converted to carbonates (i.e.
CaCO3 polymorphs and C-S-C).
At 62 MPa (Fig. 7c), FTIR results suggest total consumption of CH
across all zones; nonetheless, C-S-H was retained in the inner and
middle regions. Therefore, deeper penetration of reaction fronts
(i.e. carbonation and leaching fronts) can be inferred.
3.2. XRD analysis
The FTIR data presented in the previous section identified cal-
cium hydroxide, calcium silicate hydrate and carbonates in the
exposed and unexposed specimens. However, this data does not
fully characterize the minerals present for the following reasons.
Firstly, calcium silicate hydrate exhibits structural polymorphs
depending on temperature and C/S ratio. Characterization of these
polymorphs requires a more sophisticated technique. Secondly,
FTIR uses carbonate bond (CO2 3 ) to identify the presence of car-
bonates. When cement is exposed to carbonated brine, several
types of carbonates could be formed, depending on the cement
hydrates being carbonated and environmental conditions of tem-
perature and pressure. These include calcite, aragonite, vaterite,
ikaite, scawtite, fukalite, spurrite, tilleyite and galuskinite. FTIR is
not able to detect the specific carbonate formed under the aging
conditions. Thirdly, cement clinker compounds hydrate at different
rates. Unhydrated clinker materials are not easily detected and/or
quantified by FTIR. For these reasons, additional technique is
required to ascertain and quantify minerals that are not easily
characterized by FTIR. XRD analysis is a useful means of obtaining
this information. XRD data were obtained in previous study [43].
Relevant data for unaged specimen (cured for 5 days in 2% NaCl
solution at 93
C) and specimen aged for 14 days in 2% NaCl solution
at 177
C, 21 MPa, and 40% CO2 are extracted and re-presented in
this study to explain observations in compressive strength, porosity
and permeability. Table 4 summarizes the results.
Fig. 8 shows the XRD spectra of an unaged specimen. The data in
Table 4 suggests complete hydration of C3S and C3A to produce
hydrates (portlandite and calcium silicate hydrate) and katoite (i.e.,
tricalcium aluminate hexahydrate, C3AH6) respectively. However,
C2S and C4AF are incompletely hydrated, which indicates slower
rate of hydration than C3S and C3A. Since the slurry was cured at
93
C, it is likely that pozzolanic reaction occurs between silica flour
and portlandite. Occurrence of this reaction will reduce the amount
of portlandite but increase the amount of calcium silicate hydrate.
In the presence of atmospheric CO2, which dissolves in water dur-
ing slurry mixing to form low concentration of carbonic acid, por-
tlandite reacts to form calcite while reaction of calcium silicate
hydrate account for the scawtite detected. Calcite-aragonite phase
diagram [59] show that aragonite is typically formed at high
pressure. So, the absence of aragonite could be due to low curing
pressure. Quartz is the stable crystalline form of amorphous silica
which is produced as a by-product of leaching. Also, it may have
been formed at 93
C from the added silica that has not undergone
pozzolanic reaction with calcium hydroxide. Clinotobermorite is
the only crystalline C-S-H that was detected by XRD, which in-
dicates that CSH is partly transformed at elevated temperature. This
crystalline phase of CSH is structurally related to 11 Å tobermorite
[40], which is typically formed at temperatures near and above
200
F [43].
Fig. 9 shows the XRD spectra of the three degraded zones of
aged specimen. For this specimen, similar observations are made,
especially in the inner zone. Comparison of the mineral composi-
tion in this zone with the unexposed specimen (Table 4) proves that
calcium hydroxide is completely reacted while only a proportion of
calcium silicate hydrate is carbonated. As stated above, pozzolanic
 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70 61

Fig. 7. Normalized absorbance peaks extracted from FTIR spectra of degraded zones of Class H cement aged at 177
C and: (a) 21 MPa; (b) 42 MPa; and (c) 62 MPa.

reaction between silica flour and portlandite might have reduced
the amount of portlandite being carbonated. The uncarbonated
portion of amorphous CSH is transformed to clinotobermorite. Like
CSH, clinotobermorite can also be carbonated. Complete carbon-
ation of these calcium silicate hydrates forms calcium carbonate
(calcite and aragonite) while incomplete carbonation forms cal-
cium silicate carbonate. These carbonates, in addition to the cal-
cium silicate hydrates (amorphous CSH and clinotobermorite) aid
the development of compressive strength. Amorphous silica is the
by-product of bi-carbonation and leaching of calcium carbonate
due to incessant exposure to carbonated brine (Eqs. (5) and (6)).
The composition of these amorphous silica indicates a severely
leached specimen. In the exposed specimen, the compositions of
calcite, aragonite and calcium silicate carbonate are highest in the
outer zone. Evidently, this zone is the most degraded followed by
the middle zone because it is exposed for the longest duration as
dissolved CO2 diffuses radially through the pore solution.
Compared to the unexposed specimen, the composition of katoite
across all zones is roughly the same after exposure, which suggest
that this mineral has low reactivity with carbonic acid. Similarly,
62 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70

Table 4 (C-S-H) is darker. Therefore, the white spots in the unpenetrated

XRD data of unexposed and exposed Class H cement specimens (Data from Omosebi region refer to the unhydrated clinker materials (i.e. C3S, C2S, C3A
et al. [43]).
and C4AF). The light grey corresponds to CH while the dark grey
Mineral Composition (%) signifies C-S-H. Fig. 12a shows the SEM image of a location in the
Ageda Unaged uncarbonated zone. In this zone, CH and CSH are the primary
compounds present, both being cement hydrates that are produced
Inner zone Middle zone Outer zone
after hydration of clinker compounds. To confirm the elemental
Calcium hydroxide 0.0 0.0 0.0 7.0
composition of the hydrate present, EDX analysis (Fig. 12b) was
Clinotobermorite 5.6 6.2 6.3 9.0
Calcite 2.3 2.5 9.8 1.0 conducted at the location of the image. The elements detected by
Quartz 22.9 24.4 13.8 25.0 EDX are typical elements in calcium hydroxide and calcium silicate
Katoite 3.8 4.2 3.5 4.0 hydrate. Using this information together with the FTIR and XRD
Dicalcium silicate 1.0 1.0 1.6 5.0 mineralogy, the presence of CH and CSH is inferred. Although dis-
Tetra calcium aluminoferrite 5.6 7.2 6.3 6.0
Aragonite 0.0 0.0 2.5 0.0
solved CO2 has not penetrated this zone, calcium leaching can occur
Calcium silicate carbonate 10.4 8.6 33.5 0.0 due to the difference between calcium ion concentration in solid
Amorphous phases 48.4 45.9 22.8 0.0 CH and calcium ion concentration in the pore solution. This con-
Calcium silicate hydrate 0.0 0.0 0.0 7.0 centration difference establishes a gradient that leads to the
Scawtite 0.0 0.0 0.0 36.0
diffusion of calcium ions into the pore fluid, which results in
a
Aging condition: 177
C, 21 MPa, 40% CO2, 2% NaCl, 14 days.

Fig. 8. XRD spectra of unaged Class H specimen cured at 200
F.

further hydration of tetra calcium aluminoferrite is insignificant. In
addition, the fraction of dicalcium silicate decreases in the exposed
specimen, which indicates further hydration.
Fig. 10 shows the overall contribution of the major minerals post
degradation. Structurally weak amorphous silica (a by-product of
leaching) is the dominant mineral followed by strength-
compensating carbonates.
3.3. SEM and EDX
Fig. 11 presents SEM images obtained for aged Classes G and H
specimens. These pictures show the interface between the CO2
penetrated zone (usually carbonated) and the unpenetrated zone
(uncarbonated). The back-scattered electron (BSE) images show
zones, which consist of three color-coded sections: white, light grey
and dark grey. Kutchko et al. [2] deduced from BSE image that
unhydrated cement compounds are usually the brightest while
calcium hydroxide (CH) is less bright and calcium silicate hydrate
dissolution of solid CH. Fig. 13a shows the SEM image at a location
in the penetrated zone of a specimen that was aged at 21 MPa
(3000 psi). The pore at this location is packed with CaCO3, which
arises from carbonation of cement hydrates. Fig. 13b displays the
SEM image at a location in the penetrated zone of a specimen that
was prepared, cured and aged under the same conditions of tem-
perature and CO2 concentration with exception of pressure
(62 MPa). This image shows additional grain packing in the pores,
which suggests high degree of carbonation at this pressure. EDX
data (Fig. 13c), in combination with the FTIR and XRD mineralogy,
shows that the penetrated zone is composed of carbonates, espe-
cially calcium carbonate (CaCO3). Subsequently, CaCO3 is leached
after exposure to additional carbonated brine. This observation is
consistent with the mechanical and transport properties presented
in Section 3.4 of this article. The leaching front (i.e. dissolution of
CaCO3) occurs behind the carbonation front. In the leached zone,
structurally weak amorphous silica is the by-product of cement
degradation.
 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70 63

Fig. 9. XRD spectra of the degraded zones of Class H specimen aged for 14 days in 2% NaCl solution at 177
C, 21 MPa, and 40% CO2: (a) inner zone; (b) middle zone; (c) outer zone (*
from Omosebi et al. [43]).
64 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70
Fig. 10. Contribution of chemical phases in Class H cement aged for 14 days at 177
C,
21 MPa, 40% CO2 in 2% NaCl solution.
3.4. Compressive strength, porosity and permeability
3.4.1. Effect of exposure time
Effect of variation in exposure time is investigated at 177
C,
42 MPa and 10% CO2 concentration. The result is presented in
Fig. 14. After the initial 6 days of exposure, changes in mechanical
and transport properties of Class G cement is insignificant, indi-
cating less reactivity with CO2. In Class H cement, strength in-
creases considerably while porosity and permeability decreases.
This indicates higher reactivity leading to carbonation reaction. As
time increases, compressive strength decreases in Class G cement
while transport parameters generally increase. Similar trend is
observed in Class H cement, although loss in strength is more sig-
nificant after 28 days. This is due to leaching which arises from the
increased time of exposure permitting more time for the carbon-
ated brine to leach the specimens after initial carbonation.
3.4.2. Effect of CO2 concentration
This section discusses the effect of variation in CO2 gas
composition on cement degradation at 177
C and total test pres-
sures of 21 MPa, 42 MPa, and 62 MPa. In Fig. 15, porosity and
permeability data are presented along with compressive strength
to explain observations. Since the unaged specimen was first cured
Fig. 11. SEM images showing penetrated and unpenetrated zones in (a) Class G; (b) Class H cements (Omosebi et al. [43]).
for 5 days at 93
C and then kept at ambient conditions for 14 days,
it is presented along with the exposed specimens (aged for 14 days
at 177
C), not as a control but to observe the initial alteration in
cement parameters after exposure to carbonated-brine. Strength of
Class H cement noticeably increased while porosity and perme-
ability generally decreased as CO2 gas composition was increased at
3000 psi (Fig. 15b). This can be attributed to chemical degradation
dominated by carbonation reaction and transformation of crystal-
line cement hydrates at elevated temperature. Both XRD and FTIR
data confirms the occurrence of carbonation. At 100% CO2 gas
composition, the increase in porosity could be due to leaching of
specimen. Similarly, mechanical strength increased slightly in Class
G cement as CO2 gas composition is increased (Fig. 15a). Corre-
sponding reduction in porosity and permeability implies that the
effect of carbonation overrides leaching of carbonates. Further-
more, addition of 35% silica flour minimizes strength retrogression
at elevated temperature (i.e. 350
F) but does not hinder structural
transformation of amorphous CSH to its crystalline phases (i.e. C-S-
H). Addition of silica to the neat cement reduces the C/S (or CaO/
SiO2) ratio from 3 to 1, approximately [43,60]. According to C-S-H
phase diagram (Fig. 8b), calcium silicate hydrate produced during
curing transforms to xonotlite or clinotobermorite (11 Å tober-
morite) at aging temperature of 350
F and C/S ¼ 1, approximately.
As presented in Section 3.2, XRD analysis of specimen that was aged
at 350
F, 3000 psi and 40% CO2 confirms the transformation of
amorphous CSH to crystalline clinotobermorite. This crystalline
phase of CSH is believed to have acceptable physical and mechan-
ical properties required for good zonal isolation.
At 6000 psi, measurements show slightly different trend. At low
CO2 concentration (10%), aging improved the strength of Class H
cement while strength of Class G cement was essentially constant.
This suggests that mechanical performance was dominated by
carbonation since availability of CO2 is limited; hence preventing
further leaching. This observation is verified by FTIR mineralogy,
especially in the outer zone of the specimen. Correspondingly,
permeability decreased in both cements while porosity increased
in Class H suggesting the occurrence of diffusion-controlled
leaching of portlandite in the pores while induced fractures were
healed by reaction-controlled precipitation of carbonates. This
phenomenon was observed in cement samples retrieved after 30
years of CO2 exposure in the SACROC unit, West Texas [21]. At
higher CO2 concentration, rate of precipitation of carbonates is
faster than the rate at which water-soluble calcium bicarbonate is
 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70 65

Fig. 12. (a) SEM image showing formation of hydrates in unpenetrated zone of Class H cement; (b) composition of elements obtained from EDX analysis.

Fig. 13. (a) SEM image showing formation of calcium carbonate in penetrated zone of Class H cement at 21 MPa; (b) SEM image showing formation of calcium carbonate in
penetrated zone of Class H cement at 62 MPa; (c) composition of elements obtained from EDX analysis for calcium carbonate.
66 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70

Fig. 14. Effect of aging time on cement degradation at 177
C, 42 MPa and 10% CO2.

Fig. 15. Effects of CO2 concentration on compressive strength, porosity and permeability of: (a) Class G cement; (b) Class H cement.
 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70
Fig. 16. Effects of total test pressure on compressive strength, porosity and permeability of: (a) Class G cement; (b) Class H cement.
leached out. This leads to further strength development due to the
reduction in solubility of portlandite [61] at high temperature,
which preserves original matrix strength in addition to the for-
mation of carbonates in the pores. Additionally, severe strength
retrogression at elevated temperature is not expected since suffi-
cient amount of silica flour was added to the neat cement. Using the
C-S-H phase diagram, it is deduced that clinotobermorite is the
most likely C-S-H phase present in the aged specimens.
At 9000 psi, the overall mechanical behavior is similar to the
trends observed at lower pressures. That is, the specimens of both
cements are progressively carbonated as CO2 gas composition in-
creases. However, strength improvement in Class H cement is more
significant than strength development in Class G cement. FTIR
mineralogy shows that Class H cement is increasingly carbonated
as CO2 gas composition increases. In this cement, further hydration
and structural transformation at elevated temperature may have
contributed to additional compressive strength development. The
degree of carbonation in both API cements is lesser at low CO2
partial pressure than at high partial pressure. Therefore, evolution
of mechanical and physical properties of well cement in the event
of exposure to varying composition of CO2 gas under high-
temperature and high-pressure condition is direct consequence of
interrelated chemical processes. The dominant chemical process
will dictate the mechanical behavior of cement.
3.4.3. Effect of pressure
Theoretically, pressure variation alters CO2 phase behavior and
its solubility in NaCl solution. The effects of total test pressure at
67
10%, 40% and 100% CO2 concentration are presented in Fig. 16. At
10% CO2 composition, increase in compressive strength is observed
in Class G cement as pressure increased to 42 MPa after which
strength decreased (Fig. 16a). This results from complete carbon-
ation of a portion of the portlandite, which precipitates calcite and
its possible polymorphs (i.e. aragonite, vaterite and/or ikaite),
thereby reducing porosity and permeability. However, strength
decreases at extremely high pressure (62 MPa) while porosity and
permeability increases. This is likely due to leaching of the fully-
carbonated portion of the cement. Although carbonation and
leaching may have occurred, the residual compressive strength is
still sufficient for casing support. Furthermore, experimental data
suggests that Class G cement is more susceptible to carbonation
than leaching at low aging pressures (less than 42 MPa). However,
the degree of calcium bi-carbonation and subsequent leaching is
higher at high pressures (greater than 42 MPa). At moderately high
CO2 concentration (i.e. 40%), the trend differs from result at low CO2
concentration. Strength development takes place as pressure in-
creases due to increased rate of carbonation. This result suggests
that Class G cement is carbonated faster than it is leached.
Furthermore, effect of pressure on compressive strength was
investigated at 100% CO2 concentration. Result shows that Class G
cement was subjected to further carbonation at high pressure,
although the onset of leaching starts at extremely high pressure (i.e.
62 MPa). This implies that carbonation dominates leaching in Class
G at low-pressure and extremely high CO2 concentration.
At 10% CO2 composition, increase in pressure (up to 42 MPa)
leads to the development of mechanical strength in Class H cement
68 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70
due to the overriding effect of carbonation over leaching (Fig. 16b).
Both CH and C-S-H are typically carbonated when cement is
exposed to carbonated brine. Moreover, C-S-H is less reactive to
carbonic acid than portlandite and complete carbonation requires
longer time. With short duration (i.e. 14 days), incomplete
carbonation of C-S-H to C-S-C may have occurred. Fig. 7 of FTIR
measurements (presented in Section 3.1) show that C-S-H is mostly
preserved in the inner and middle zones of the degraded cement.
This implies that the degree of carbonation of C-S-H is limited to
the outer zone. Furthermore, FTIR shows that portlandite and C-S-H
were preserved in the inner and middle zones at 42 MPa; hence, in
addition to complete carbonation, the formation of C-S-C leads to
strength development. However, leaching of completely carbon-
ated cement hydrates offsets carbonation at higher pressures. This
is more pronounced at low CO2 concentration in which limited
carbonation occurs. At high CO2 concentration, increased carbon-
ation plays a dominant role up to 62 MPa. In addition, since water
requirement for Class H cement is less than Class G, the degree of
hydration of Class H is expected to be slightly lower than that of
Class G. Therefore, further hydration and/or CSH phase trans-
formation occurs at elevated temperature, which contributes to
strength development. Trends in porosity and permeability are
mixed and cannot be directly correlated to strength evolution
possibly due to calcium leaching in unreacted portion of the
cement, structural transformation and self-healing (i.e. uneven
precipitation, dissolution and further precipitation of calcium car-
bonate). At 40% CO2 composition, strength development in Class H
cement is insignificant due to carbonation and leaching, which take
place at a comparable rate. Similarly, high CO2 concentration (40%)
favored additional strength development as total pressure in-
creases because of increased rate of carbonation of cement hydrates
over subsequent leaching of calcium bicarbonate. Moreover, the
contribution of C-S-H phase transformation cannot be under-
estimated as the phase transformation diagram (Fig. 8b) suggests
the likelihood of xonotlite as C-S-H phase present which is a strong
but 10 to 100 times more permeable than tobermorite gel [28,29].
Assessment of the overall trend in compressive strength shows
relatively small variations at 21 MPa, though it increases slightly
with CO2 concentration. However, the deviation increases with
total test pressure. This indicates more interaction between cement
materials and solute species in the carbonated brine. Sudden
reduction in mechanical strength at 62 MPa (for 10% CO2) could be
due to the presence of microcracks which may have developed
either due to this high pressure or during depressurization of the
HPHT autoclave. Interestingly, the porosity and permeability of
Class G cement also increases. Pang et al. [62] observed the
development of cracks in well cement after curing for 3 days and
depressurizing at the rate of 0.345 MPa/min. Their study shows that
Class H cement is more resistant to depressurization damage than
Class G cement. Class G cement specimens that were damaged by
depressurization tends to have uneven fracture surfaces during
splitting tension tests. This corroborates the uneven pattern of
degradation fronts observed from visual inspection (Section 3.1).
Since CO2 concentration is low, it is possible that the carbonation
that takes place is not sufficient to heal the cracks. On the contrary,
high CO2 concentration enhances carbonation thereby healing the
cracks and significantly improving mechanical strength.
Overall, Class H cement demonstrates better capability to sus-
tain mechanical strength than Class G cement because it is stronger
after chemical degradation.
4. Conclusions
The principal objective of this study is to expand the existing
knowledge base in degradation of well cements by carbonated
brine under HPHT conditions. To achieve this goal, effects of CO2
concentration and pressure on alterations in physical, chemical,
mechanical, and transport properties of oilwell cements were
investigated. The following conclusions can be drawn from this
study:
 Mechanical behavior of well cement exposed to CO2-rich brine
is governed by the rate of reaction of two competing mecha-
nisms; carbonation and leaching.
 Strengths of Classes G and H cements increase while porosity
and permeability decrease with increase in CO2 concentration;
the governing mechanisms are carbonation of cement hydrates
and subsequent leaching, which can occur simultaneously.
Carbonation leads the reaction front while leaching concur-
rently takes place behind this front. In most cases, cement
carbonation was dominant even though some level of leaching
was observed.
 Carbonation process dominates at low pressures (less than
42 MPa). However, at high pressures, leaching offsets the effect
of carbonation. Porosity, permeability, FTIR and XRD measure-
ments are consistent with this observation.
 FTIR, XRD and SEM mineralogy analyses confirm that por-
tlandite is the most consumed as concentration of CO2 gas in-
creases. On the contrary, calcium silicate hydrate is mostly
preserved, which helps to retain mechanical strength. At high
temperature, Classes G and H cements initially form poly-
morphs of calcium silicate hydrate and calcium carbonate,
which contribute to strength development.
 Class H cement demonstrates better structural integrity for
casing support and zonal isolation than Class G cement after
exposure to carbonic acid attack. Predominantly, it exhibited
increase in compressive strength and reduction in porosity and
permeability after the exposure.
Acknowledgement
The authors acknowledge Bureau of Safety and Environmental
Enforcement (BSEE) for sponsoring this project (E12PC00035).
Technical contributions from industry advisory board members are
also appreciated. Many thanks to Jeffery McCaskill and Joe Flen-
niken for assisting to setup the equipment used in this study.
Special thanks to Halliburton Energy Services, Mountain Cement
Company, Lone Star Portland Cement, Fritz Industries and
Schlumberger Limited for donating materials for this research.
Nomenclature
API American Petroleum Industry
ASTM American Society for Testing and Materials
BWOC By Weight of Cement
CaCO3 Calcium Carbonate
CH Calcium Hydroxide
C-S-C Calcium Silicate Carbonate
CSH Calcium Silicate Hydrate
C-S-H Calcium Silicate Hydrate polymorphs
CO2 Carbon Dioxide
EOR Enhanced Oil Recovery
FTIR Fourier Transform Infra-Red
HEC Hydroxyl Ethyl Cellulose
NaCl Sodium Chloride
SEM Scanning Electron Microscopy
XRD X-ray Diffraction
NMR Nuclear Magnetic Resonance spectroscopy
 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70 69

References [28] L.H. Eilers, E.B. Nelson, L.K. Moran, High temperature cement compositions e
pectolite, scawtite, truscottite or Xonotlite: which do you want? SPE J. Pet.
Tech. 35 (7) (1983) 1373e1377.
[1] J. Bensted, P. Barnes, Structure and Performance of Cements, second ed., Spon
[29] E.B. Nelson, L.H. Eilers, G.L. Kalousek, formation and behavior of calcium sil-
Press, London and New York, 2008.
icate hydrates in a geothermal environment, Cem. Concr. Res. 11 (1981)
[2] B.G. Kutchko, B.R. Strazisar, D.A. Dzombak, G.V. Lowry, N. Thaulow, Degra-
371e381.
dation of well cement by CO2 under geologic sequestration conditions, Envi-
[30] F. Matsushita, Y. Aono, S. Shibata, Calcium silicate structure and carbonation
ron. Sci. Technol. 41 (2007) 4787e4792.
shrinkage of a tobermorite-based material, Cem. Concr. Res. 34 (2004)
[3] K. Rahmani, A. Shamsai, B. Saghafian, S. Peroti, Effect of water and cement
1251e1257.
ratio on compressive strength and abrasion of microsilica concrete, Middle
[31] I.G. Richardson, The calcium silicate hydrates, Cem. Concr. Res. 38 (2008)
East J. Sci. Res. 12 (8) (2012) 1056e1061.
137e158.
[4] O. Alawode, O.I. Idowu, Effects of water-cement ratios on the compressive
[32] A.J. Allen, J.J. Thomas, H.M. Jennings, Composition and density of nanoscale
strength and workability of concrete and lateritic concrete mixes, Pac. J. Sci.
calcium-silicate-hydrate in cement, Nat. Mater. 6 (2007) 311e316.
Technol. 12 (2) (2011) 99e105.
[33] V. Barlet-Goue
dard, G. Rimmele
, B. Goffe

, O. Porcherie, Mitigation strategies
[5] R.A. Bruckdorfer, Carbon dioxide corrosion in oil well cements, in: Pro-
for the risk of CO2 migration through wellbores, in: Proceedings of the IADC/
ceedings of the Rocky Mountain Regional Meeting of the Society of Petroleum
SPE Drilling Conference, 2006, pp. 1e17. Miami, FL, February 21-23, 2006,
Engineers, Billings, MT, May 19 e 21, 1986, 1986, pp. 531e539. SPE Paper No.
IADC/SPE Paper No. 98924.
15176.
[34] M. Mainguy, C. Tognazzi, J.M. Torrenti, F. Adenot, Modelling of leaching in
[6] A.T. Bourgoyne Jr., K.K. Millheim, M.E. Chenevert, F.S. Young Jr., Applied
pure cement paste and mortar, Cem. Concr. Res. 30 (1) (2000) 83e90.
Drilling Engineering, 2, Society of Petroleum Engineers, Richardson, 1986.
[35] M. Mainguy, O. Coussy, Propagation fronts during calcium leaching and
[7] C.D. Saunders, W.A. Walker, Strength of oil well cements and additives under
Chloride penetration, J. Eng. Mech 126 (2000) 250e257.
high temperature well conditions, in: Proceedings of the Fall Meeting of
[36] N. Agbasimalo, M. Radonjic, Experimental study of the impact of drilling fluid
the Petroleum Branch, American Institute of Mining and Metallurgical Engi-
contamination on the integrity of cementeformation interface, ASME J. En-
neers, San Antonio, TX, 1954, pp. 1e8. October 17-20, 1954, AIME Paper No.
ergy Resour. 136 (2014) 1e5.
390-G.
[37] M.C.M. Nasvi, P.G. Ranjith, J. Sanjayan, A. Haque, Sub- and super-critical car-
[8] A. Brandl, J. Cutler, A. Seholm, M. Sansil, G. Braun, Cementing solutions for
bon dioxide permeability of wellbore materials under geological sequestra-
corrosive well environments, SPE Drill. Complet. J (2011) 208e219.
tion conditions: an experimental study, Energy J. 54 (2013) 231e239.
[9] Ch. Noik, A. Rivereau, Oilwell cement durability, in: Proceedings of the SPE
[38] S. Merlino, E. Bonaccorsi, T. Armbruster, The real structure of tobermorite
Annual Technical Conference and Exhibition, 1999, pp. 1e6. Houston, TX,
11Å: normal and anomalous forms, OD character and polytypic modifications,
October 3-6, 1999, SPE Paper No. 56538.
Eur. J. Mineral. 13 (2001) 577e590.
[10] M. Fabienne, Ch. Noik, R. Alain, Z. Helene, Complementary analyses of a tri-
[39] S.V. Churakov, Structural position of H2O molecules and hydrogen bonding in
calcium silicate sample hydrated at high pressure and temperature, Cem.
anomalous 11Å tobermorite, Am. Mineral. 94 (2009) 156e165.
Concr. Res. 32 (2002) 65e70.
[40] C. Henmi, I. Kusachi, Clinotobermorite, Ca5Si6(O,OH)18.5H2O, a new mineral
[11] G. Le Saout, E. Lecolier, A. Rivereau, H. Zanni, Study of oilwell cements by
from Fuka, Okayama prefecture, Japan, Mineral. Mag. 56 (1992) 353e358.
solid-state NMR, C. R. Chim. 7 (2004) 383e388.
[41] N. Hara, C.F. Chan, T. Mitsuda, formation of 14 Å tobermorite, Cem. Concr. Res.
[12] G. Le Saout, E. Lecolier, A. Rivereau, H. Zanni, Chemical structure of cement
8 (1978) 113e116.
aged at normal and elevated temperatures and pressures, Part I: class G oil-
[42] Y.Q. Zhang, A.V. Radha, A. Navrotsky, Thermochemistry of two calcium silicate
well cement, Cem. Concr. Res. 36 (1) (2006) 71e78.
carbonate minerals: scawtite, Ca7(Si6O18)(CO3).2H20, and spurrite,
[13] G. Le Saout, E. Lecolier, A. Rivereau, H. Zanni, Chemical structure of cement
Ca5(SiO4)2(CO3), Geochim. Cosmochim. Aeta 115 (2013) 92e99.
aged at normal and elevated temperatures and pressures, Part II: low
[43] O. Omosebi, H. Maheshwari, R. Ahmed, S. Shah, S. Osisanya, A. Santra,
permeability class G oilwell cement, Cem. Concr. Res. 36 (3) (2006) 428e433.
A. Saasen, Investigating temperature effect on degradation of well cement in
[14] M. Lesti, C. Tiemeyer, J. Plank, CO2 stability of Portland cement based well
HPHT carbonic acid environment, J. Nat. Gas Sci. Eng. 26 (2015) 1344e1362.
cementing systems for use on carbon capture & storage (CCS) wells, Cem.
[44] API RP, Recommended Practice for Testing Well Cements 10B, Standard by
Concr. Res. 45 (2013) 45e54.
American Petroleum Institute, 1997.
[15] N. Neuville, G. Aouad, E. Lecolier, D. Damidot, Innovative leaching tests of an
[45] B.R. Reddy, A. Sandra, D. McMechan, D. Gray, C. Brenneis, R. Dunn, Cement
oilwell cement paste for CO2 storage: effect of the pressure at 80oC, Energy
mechanical-property measurements under wellbore conditions, SPE Drill.
Proced. 23 (2012) 472e479.
Complet. J (2007) 33e38.
[16] Santra, A., Reddy, B. R., Liang, F., and Fitzgerald, R., 2009, reaction of CO2 with
[46] ASTM C39/C39M-12, Standard Test Method for Compressive Strength of Cy-
Portland cement at downhole conditions and the role of pozzolanic supple-
lindrical Concrete Specimens, ASTM International, 2012.
ments, Proceedings of the SPE International Symposium on Oilfield Chemistry,
[47] D. Stiles, Effects of long-term exposure to ultrahigh temperature on the me-
Woodlands, TX, April 20-22, 2009, SPE Paper No. 121103, pp. 1e9.
chanical parameters of cement, in: Proceedings of the IADC/SPE Drilling
[17] A.B. Sauki, S. Irawan, Effects of pressure and temperature on well cement
Conference, 2006, pp. 1e11. Miami, FL, February 21-23, 2006, IADC/SPE Paper
degradation by supercritical CO2, Int. J. Eng. Technol. 10 (2010) 47e56.
t-Goue
dard, G. Rimmele
, O. Porcherie, N. Quisel, J. Desroches, No. 98896.
[18] V. Barle
[48] M.C.M. Nasvi, P.G. Ranjith, J. Sanjayan, A. Haque, Xiao Li, Mechanical behav-
A solution against well cement degradation under CO2 geological storage
iour of wellbore materials saturated in brine water with different salinity
environment, Int’l J. Greenh. Gas Control 3 (2009) 206e216.
levels, Energy J. 66 (2014) 239e249.
[19] J.B. Laudet, A. Garnier, N. Neuville, Y. Le Guen, D. Fourmaintraux, N. Rafai,
[49] R. Ylme
n, U. J€ aglid, Carbonation of Portland cement studied by diffuse
N. Burlion, J.F. Shao, The behavior of oil well cement at downhole CO2 storage
reflection Fourier transform infrared spectroscopy, Int. J. Concr. Struct. Mater.
conditions: static and dynamic laboratory experiments, Energy Proced. 4
7 (2) (2013) 119e125.
(2011) 5251e5258.
[50] Y. Matsuda, M. Tsukada, Identification of Calcium Carbonate Contained as
[20] J. Condor, K. Asghari, Experimental study of stability and integrity of cement
Body in Modern Paints by FTIR Spectroscopy, IRUG 2 Postprints, London, 1995,
in wellbores used for CO2 storage, Energy Proced.. 1 (2009) 3633e3640.
pp. 25e34.
[21] J.W. Carey, M. Wigand, S.J. Chipera, G. WoldeGabriel, R. Pawar, P.C. Lichtner,
[51] M.I. Zaki, H. Kno € zinger, B. Tesche, G.A.H. Mekhemer, Influence of phospho-
S.C. Wehner, M.A. Raines, G.D. Guthrie Jr., Analysis and performance of oil well
nation and phosphation on surface acidebase and morphological properties
cement with 30 years of CO2 exposure from the SACROC unit, west Texas,
of CaO as investigated by in situ FTIR spectroscopy and electron microscopy,
USA, Int’l J. Green House Gas Control 1 (1) (2007) 75e85.
J. Colloid Interface Sci. 303 (1) (2006) 9e17.
[22] G.W. Scherer, B. Kutchko, N. Thaulow, Characterization of cement from a well
[52] D. Baciu, J. Simitzis, Synthesis and characterization of a calcium silicate
at teapot dome oil field: implications for geological sequestration, Int’l J.
bioactive glass, J. Optoelectron. Adv. Mater. 9 (2007) 3320e3324.
Greenh. Gas Control 5 (2011) 115e124.
[53] H.N. Rutt, J.H. Nicola, Raman spectra of carbonates of calcite structure, J. Phys.
[23] Z. Krilov, B. Loncaric, Z. Miksa, Investigation of a long-term cement deteri-
C Solid State Phys. 7 (1974) 4522e4528.
oration under a high-temperature, sour-gas downhole environment, in:
[54] R. Frech, E.C. Wang, J.B. Bates, The I.R. and Raman spectra of CaCO3 (aragonite),
Proceedings of the SPE International Symposium on Formation Damage
Spectrochim. Acta 36A (1980) 915e919.
Control, 2000, pp. 1e9. Lafayette, LA, February 23-24, 2000, SPE Paper No.
[55] A.G. Xyla, P.G. Koutsoukos, Quantitative analysis of calcium carbonate poly-
58771.
morphs by infrared spectroscopy, J. Chem. Soc. Faraday Trans. 185 (10) (1989)
[24] C. Carde, R. Francois, Effect of the leaching of calcium hydroxide from cement
3165e3172.
paste on mechanical and physical properties, Cem. Concr. Res. 27 (1997)
[56] N.V. Vagenas, A. Gatsouli, C.G. Kontoyannis, Quantitative analysis of synthetic
539e550.
calcium carbonate polymorphs using FT-IR spectroscopy, Talanta 59 (2003)
[25] K. Takase, Y. Barhate, H. Hashimoto, S.F. Lunkad, Cement-sheath wellbore
831e836.
integrity for CO2 injection and storage wells, in: Proceedings of the SPE Oil
[57] S. Gunasekaran, G. Anbalagan, S. Pandi, Raman and infrared spectra of car-
and Gas India Conference and Exhibition, 2010, pp. 1e11. Mumbai, India,
bonates of calcite structure, J. Raman Spectrosc. 37 (2006) 892e899.
January 20e22, 2010, SPE Paper No. 127422.
[58] R. Ylme
n, U. Ja €glid, B.-M. Steenari, I. Panas, Early hydration and setting of
[26] B.G. Kutchko, B.R. Strazisar, G.V. Lowry, D.A. Dzombak, N. Thaulow, Rate of
Portland cement monitored by IR, SEM and vicat techniques, Cem. Concr. Res.
CO2 attack on hydrated class H well cement under geologic sequestration
39 (2009) 433e439.
conditions, Env. Sci. Tech. 42 (2008) 6237e6242.
[59] W.H. Mulder, I. Hassan, The effect of orientational disorder of the anion in
[27] H.F.W. Taylor, The Chemistry of Cements, Academic Press, London, 1964.
calcite on the aragonite (Pmcn) 4 calcite (R3c) equilibrium, Can. Mineral. 49
70 O. Omosebi et al. / Cement and Concrete Composites 74 (2016) 54e70

(2011) 1035e1043. hydrates coupled with micro-pore formation, J. Adv. Concr. 4 (2006) 395e407.
[60] E.B. Nelson, Well Cementing, Schlumberger Educational Services, Sugar Land, [62] X. Pang, C. Meyer, G.P. Funkhouser, R. Darbe, Depressurization damage of oil
1990. well cement cured for 3 days at various pressures, Constr. Build. Mater. 74
[61] K. Nakarai, T. Ishida, K. Maekawa, Modeling of calcium leaching from cement (2015) 268e277.