113 (2023) 204974

Contents lists available at ScienceDirect

Gas Science and Engineering

journal homepage: www.journals.elsevier.com/gas-science-and-engineering

Utilization of fly ash-based geopolymer for well cement during CO2

sequestration: A comprehensive review and a meta-analysis
H.B.S. Sathsarani a, K.H.S.M Sampath a, *, A.S. Ranathunga b
a
Department of Civil Engineering, University of Moratuwa, Sri Lanka
b
Department of Civil Engineering, School of Civil Engineering, University of Leeds, Leeds, UK

A R T I C L E I N F O A B S T R A C T

Keywords: Global warming is a pressing issue caused by the increase in greenhouse gas emissions, with CO2 contributing to
CO2 sequestration 64% of total emissions. To reduce anthropogenic CO2 emissions, several options have been proposed, including
Well cement CO2 sequestration. A key requirement for a successful and sustainable geo-sequestration process is the use of
Ordinary portland cement
appropriate zonal isolation provided by the cementing material used between the annular surface and injection
Fly ash
Hydro-mechanical properties
well. Although, Ordinary Portland Cement (OPC) is typically used as the well cement, it has shown failures
during the process, including degradation issues, carbonation, shrinkage and microcracking, increased perme­
ability in CO2-rich environments, and loss of sealing properties in a short period of time. To address these
problems, fly ash (FA)-based geopolymers have been introduced as a better well cement replacement. This study
provides a comparative review between OPC and FA-based geopolymers in the context of CO2 sequestration. The
review comprehensively analyses the behaviour of FA-based geopolymer cement with its chemical composition,
the impact of preparation conditions on the mechanical behaviour of geopolymers, and CO2 permeability
through FA-based geopolymer. Furthermore, a meta-analysis was conducted to develop statistical models for
predicting the pertinent hydro-mechanical properties of FA-based geopolymer, including dry density,
compressive strength, autogenous shrinkage strain, and CO2 permeability during the geo-sequestration process.
The outcomes of the meta-analysis can aid decision-making regarding the appropriateness of applying FA-based
geopolymer as a replacement for OPC to conduct a sustainable and safe CO2 geo-sequestration process under
proper isolation conditions.

production to low-carbon or non-carbon fuel sources, and CO2 seques­

tration (Ji and Zhu, 2015). Among them, “CO2 sequestration” can be
1. Introduction
considered one of the promising methods to address the issue. In the CO2
sequestration process, the atmospheric CO2 level is reduced by capturing
In the 20th and 21st centuries, fossil fuels are mostly used as power
and injecting the anthropogenic CO2 into the underground deep
generation sources due to their vast availability. It fulfils about 82% of
geological settings including exhausted oil and gas pools, saline aquifers,
the world’s energy requirement which is rapidly increasing due to the
sedimentary basins, and deep coal layers (Ji and Zhu, 2015; Sampath
growth of the population (Newell et al., 2019). CO2 is the main output of
et al., 2020). In fact, anthropogenic CO2 can be captured from mainly
the burning of fossil fuels, which causes the acceleration of the global
three types of activities: 1) industrial process, 2) electricity generation,
warming process by contributing 64% of total emissions of greenhouse
and 3) hydrogen production (Kaldi et al., 2009). Fig. 1 shows the
gases (Bertier et al., 2006). China, the United States, the European
capturing process of natural gas with high CO2 content and preparing it
Union, India, Russia, and Japan are the main CO2 emitters in the world
for geo-sequestration. During the sequestration process, various gasses
and at the end of 2020, the world’s total fossil fuel-related CO2 emissions
and oil trapped in deep aquifers can be extracted from the geological
(FFCO2) were estimated about 34.8 billion tons of CO2 (Karakurt and
media and are produced from production wells. The injection of CO2
Aydin, 2023). As a result of finding a solution to reduce substantial
into geo-reservoirs and the subsequent production of trapped oil/gas
anthropogenic CO2 emissions, many options have been proposed. Some
from aquifers without any leakages is a key requirement for this tech­
of them are; upgrading fossil fuels’ efficiency and energy conversion,
nique. The cementing material used in between the annular surface and
increasing interest in terrestrial and marine biomass, shifting energy

* Corresponding author. University of Moratuwa, Sri Lanka.

E-mail address: [email protected] (K.H.S.M Sampath).

https://doi.org/10.1016/j.jgsce.2023.204974
Received 27 October 2022; Received in revised form 8 March 2023; Accepted 6 April 2023
Available online 11 April 2023
2949-9089/© 2023 Elsevier B.V. All rights reserved.
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

Abbreviations FAG Fly Ash Geopolymer

API American Petroleum Industry
AA/FA Alkaline activator to Fly ash ratio GGBFS Ground Granulated Blast Furnace Slag
B Binder (mass of fly ash and alkaline activator that is XRF X-Ray Fluorescence
included in m3 of FAG mixture) XRD X-Ray Diffraction
CT Curing Temperature(0C) SEM Scanning Electron Microscopy
L-MVR Linear Multivariable Regression EDX Energy Dispersive X-ray
M NaOH concentration ASTM American Society for Testing and Material
NL-MVR Non-Linear Multivariable Regression UCS Uniaxial Compressive Strength
Pc Confining Pressure (Pa) UTS Uniaxial Tensile Strength
Pi Injection Pressure (Pa) MVR Multivariable Regression
SS/SH NaOH to Na2SiO3 ratio L-MVR Linear Multivariable Regression
T Curing time NL-MVR Non-Linear Multivariable Regression
OPC Ordinary Portland Cement ANN Artificial Neural Network
FA Fly Ash

injection well plays a major role in conducting a successful and sus­

tainable geo-sequestration process by providing appropriate zonal
isolation (Nasvi et al., 2013).
In general, Ordinary Portland Cement (OPC) is commonly used as
well cement; however, it shows some failures in a CO2-rich environment
and loses its isolation properties within a short period of time (Nasvi
et al., 2012a). Due to these reasons, geopolymer is introduced to replace
Portland cement; among which, fly ash geopolymer (FAG) has a higher
prominence due to its ability to reduce the gigantic amounts of fly ash
(FA) piled up during coal-fired power plant operations (Hardjito et al.,
2004). Fig. 2 illustrates the production and the re-used amount of
thermal power plant ash in different countries. The re-used percentages
of FA in India, China, the UK, the USA, and Germany are quite low
compared to the produced quantity, which signifies the importance of
the utilization of this waste product in a productive manner (Ionescu &
Lӑzӑrescu, 2020). Therefore, the total dumps of FA that would lead to
environmental pollution can be significantly minimized by encouraging
the use of FA-based geopolymers as well cement during the CO2 Fig. 2. Production and usage of thermal power plant ash in different countries
sequestration process. Thus, more focused research on the applicability (million tons per year) (Developed after Ionescu & Lӑzӑrescu, 2020).
of FA-based geopolymers, as a substitute to OPC for gas or well
cementing during CO2 sequestration is more essential. for well cementing during CO2 sequestration.: 1) Role of well cement in
a successful CO2 geo-sequestration process, 2) OPC as a well cement and
its problems, 3) FA-based geopolymers as an alternative well cement,
2. Review methodology
highlighting the roles of alkaline activator, curing condition and tem­
perature, 4) Mechanical behaviour of FA-based well cement including
This paper is mainly focused on comprehensively reviewing the
the variation in compressive strength, tensile strength, and
literature on the following subcategories to identify the existing research
permeability.
gaps on the suitability of FA-based geopolymers as a replacement to OPC

Fig. 1. Overview of carbon dioxide (CO2) sequestration processes (Developed after Kaldi et al., 2009).

2
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

Fig. 3 summarises the total number of publications reviewed in this

study, including peer-reviewed journal publications, refereed confer­
ence publications, and research articles from different countries that are
published over 36 years.

3. Role of well cement in a successful CO2 geo-sequestration

process

Wellbore integrity is more important to conduct a successful CO2

geo-sequestration process. For that, the leakage rate should be limited to
0.01%–1.0% per year, and if this rate is exceeded, it will be harmful to
the groundwater table and the ecosystem in the surrounding (Kutchko
et al., 2007). The cement that is used for wellbore integrity can be
Fig. 4. A typical wellbore with primary and secondary cementing zones
mainly divided into primary cement and secondary cement. The primary
(Developed after Nasvi et al., 2013).
cement or annular cement fills the annular surface, connecting the
casing and forming region, and the secondary cement material (i.e.,
High Pressure and High Temperature (HPHT) vessels to maintain or
cement plug) is poured inside the casing (refer to Fig. 4) (Nasvi et al.,
simulate downhole conditions by controlling temperature and pressure
2013). This cementing material requires significant isolation charac­
gradients. These vessels usually consist of brine-saturated CO2 and su­
teristics to maintain a leakage-free wellbore, which include imperme­
percritical CO2. These apparatuses were used to examine the carbon­
ability, non-shrinking ability, resistance to different chemicals, and
ation of cement, the effect of the curing period, durability changes, and
ability to withstand mechanical loads (Khalifeh et al., 2014). The
identify the mechanical behaviour of well cement in different states of
American Petroleum Industry (API) has classified different types of well
CO2 environment (Barlet-Gouédard et al., 2009). Some of the most
cement, based on different temperature and pressure ranges. The most
pertinent literature on different types of OPC in various conditions and
commonly available API class well cement types are listed in Table 1
the key findings are summarised in Table 2.
(Robins and Milodowski, 1986). Based on the percentage content of
The cement sheath in a well bore is the first material exposed to the
tricalcium aluminate (C3A), oil well cements are classified as 1) ordinary
injected CO2 in the sub-surface during the sequestration process. Long-
sulphate resistant, 2) moderate sulphate resistant, and 3) high sulphate
term CO2 sequestration includes contact with supercritical CO2 and
resistant well cements. Effective zonal isolation can be obtained by using
brine solutions at increased pressure and temperature and decreased pH
the correct cement for specific well environments. In general, API classes
levels. When OPC-based well cement is exposed to CO2-saturated brine
of G and H, which are based on OPC have been used as well cement
under down-hole pressure and temperature conditions, a series of rings
(Shahriar, 2011). However, with the variation of temperature and
are observed, as shown in Fig. 6 (Brandvoll et al., 2009; Duguid and
pressure conditions through the large depths, these well cements un­
Scherer, 2010). In fact, the different zones observed are: 1) an
dergo radial cracking, durability issues, strength reductions, and
un-reacted dark grey core in the middle, 2) a grey layer with depleted Ca
permeability issues. These failure conditions may create leakage path­
(OH)2 (zone 1), 3) a white layer precipitated with CaCO3 due to
ways in different ways through the wellbore and reduce the sealing
carbonation (zone 2), and 4) an orange layer with highly depleted Ca
capability with time (refer to Fig. 5) (Nasvi et al., 2012a). In fact, tem­
(OH)2 and C–S–H paste (zone 3).
perature differences in the period of curing of cement, weak bonding
Generally, the carbonation of cement minimizes porosity and
between the cement and casing, and swelling and shrinkage of the
permeability under atmospheric conditions. However, under downhole
cement during the curing period are the main reasons for generating the
conditions, OPC-based well cement reacts with dissolved CO2 and is
leakage pathways shown in Fig. 5. Accordingly, the cement material that
prominent to produce weak porous gel owing to the absence of a C–S–H
is used between the annular surface and wellbore plays a key role in
layer, which causes an increase in the porosity and permeability of well
achieving the appropriate isolation properties to carry out a sustainable
cement (Kutchko et al., 2007). In addition, they are influenced by the
CO2 sequestration process.
degradation rate and degradation method of well cement in environ­
ments with high concentrations of CO2. As listed in Table 3, pressure and
4. Ordinary Portland Cement (OPC) as a well cement and its
temperature variations, state of the fluid, type of fluid, cement compo­
problems
sition, water-cement ratio, pH gradient, and mechanical loading are
some of the influential factors that affect the integrity of the OPC-based
Researchers have done many experimental studies for identifying the
well cement. As summarised in Table 3, the effects of the influential
behaviour of OPC-based well cement in an environment with a high
factors were studied by many researchers, and those observations are
concentration of CO2. These experiments were mainly conducted using
important to assess the performance of OPC-based well cement under
downhole conditions. The problems about generally used OPC as a well
cement that has been recognized in laboratory tests are experienced in
practical situations as well. For instance, among the research done with
316,500 abandoned wells in Canada, 4.6% showed leakages, and of
those leakages, 81% reportedly specified that OPC-based well cement is
the main factor causing the leakages (Nasvi et al., 2013). Therefore,
there is a requirement for finding a new well cement material to replace
the generally used OPC under partial or full conditions, to carry out safe
CO2 sequestration processes.

5. Fly ash (FA)-based geopolymers as an alternative well cement

The manufacturing process of OPC produces approximately 1.35

billion tons of greenhouse gases per year and it is estimated as 7% of the
Fig. 3. Overview of the publication years of the reviewed studies. entire greenhouse gases which are emitted to the atmosphere annually

3
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

Table 1
The most commonly used API class well cement types (Developed after Robins and Milodowski, 1986).
API Class Depth (m) Temperature (◦ C) Sulphate resistance Comments

A 0–1830 77 Ordinary Regular cement

B 0–1830 77 Moderate Regular cement
C 0–1830 77 Ordinary-High High early strength
D 1830–3050 110 Moderate-High For high pressure and temperature
E 1830–4270 143 Moderate-High
F 3050–4880 160 Moderate-High
G 0–2440 93 Moderate-High OPC
H 0–2440 93 Moderate OPC

(Hardjito et al., 2004). By considering this issue, geopolymers were

introduced as a construction material to build high-quality in­
frastructures and used as an environmentally friendly material in the
construction industry with the requirement of 50% less energy con­
sumption in the manufacturing method due to the low process temper­
ature procedure, as shown in Fig. 7 (Hewayde et al., 2006). Not only
environmental friendliness but also geopolymer has a cost-effective
production method compared to OPC, as the cost of geopolymer pro­
duction is normally 10%–30% less than that of OPC (Rangan, 2008). The
raw materials of geopolymer include metakaolin, ground granulated
blast furnace slag (GGBFS), fly ash, sedimentary and rock powder, silica
fume, and alkali feldspars (Hardjito et al., 2004). Among them, world­
wide fly ash production was estimated at around 780 million tons per
annum. Generally, most of the ashes produced by burning coal in power
plants are disposed of in landfills. These improper disposal methods
create significant environmental and social problems due to the accu­
mulation of heavy metal substances in the ground, causing the
contamination of the groundwater table and damaging the surrounding
flora and fauna. Therefore, the use of FA-based geopolymer for well
cement has a higher prominence due to its ability to reduce the gigantic
amounts of fly ash piled up due to coal-fired power plant operations
(Ridha et al., 2018). On the other hand, in the production of 1 tonne of
geopolymeric cement, only 0.184 tonnes of CO2 is emitted, which is
approximately one-sixth of the CO2 amount produced during the
Fig. 5. Possible leakage pathways through the wellbore: (a) in the middle of manufacturing process of 1 tonne of OPC (Yang et al., 2009). Therefore,
cement and casing, (b) between cement plug and casing, (c) through the frac­ the above facts confirm the appropriateness of FA-based geopolymer for
ture of cement, (d) through the corroded casing, (e) between the formation and the replacement of OPC as an alternative well cement material for the
cement, and (f) through the pores of cement. CO2 sequestration process, in the context of environmental and technical
feasibility.

Table 2
Previously reported key findings about different classes of OPC-based well cement.
Reference Tested cement Experimental conditions Findings
type
State of CO2 Pressure Temperature Exposure
environment Time

Duguid and Scherer (2010) Class H cement CO2-saturated brine – 50 ◦ C 1, 2, 3, 6, and Degradation of the grout was shown as 0.58 mm after
(pH varied from 3 to 12 months 6 months.
7) The rate of degradation of the cement was controlled
by the rate of dissolution of the calcium carbonate-
rich layer.
Condor and Asghari (2009) Class A cement Wet supercritical CO2 15 MPa 55 ◦ C 3 months Permeability was reduced initially but increased after
Class G cement CO2-saturated brine a few months.
Compressive Strength was increased initially but
reduced with the time.
Zhang and Talman (2014) Class G cement CO2-saturated brine 10 MPa 53 ◦ C 3, 7, 14, 28, The durability of the cement was decreased, and
and 84 days permeability was changed from 0.16 mD to 1.1 mD,
at the final stage of the observation period.
Pratt et al. (2009) Portland Wet supercritical CO2 10 MPa 50 ◦ C 84 days A 200 μm carbonation depth was observed.
cement
Santra et al. (2009) Cement + silica CO2-saturated water 14 MPa 93 ◦ C 15 and 90 A 7 mm penetration depth was observed, and no
fumes + fly ash days improvement was observed when increasing the
silica fumes.

4
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

Fig. 6. Schematic view of different zones formed in OPC-based well cement in CO2-rich environments (Developed after Duguid and Scherer, 2010).

Table 3
The effect of different factors on the behaviour of OPC-based well cement during the CO2 sequestration process.
References Influential Cement Experimental conditions Key findings Conclusion
factors class

Laudet et al. Pressure and Class G Pure class G cement was tested at 90 ◦ C, After 3 months, a 5 mm carbonation front A higher rate of carbonation occurs at
(2011) temperature and 35% of silica mixed class G cement was developed in the outer region at 90 ◦ C elevated temperatures.
was tested at 140 ◦ C, where the samples and the propagation front reached the
were exposed to CO2-saturated water at 8 entire cross-section of 20 mm at 140 ◦ C.
MPa.
Sauki and Pressure and Class G Tested under 10.5 MPa and 14 MPa 50% Ca depletion was observed at 10.5 The highest depth of penetration
Irawan temperature pressures and 40 ◦ C and 120 ◦ C MPa, whereas the reduction was only 2% occurs at lower pressures and higher
(2010) temperatures. at 14 MPa. 0.9 mm and 1.5 mm penetration temperatures.
depths were observed at 40 ◦ C and 120 ◦ C
of temperatures, respectively.
Duguid and State of fluid Class H Tested the effect of moving brine at room Severe cement degradation (0.244 mm/ Under moving brine conditions, the
Scherer temperature and 50 ◦ C of temperature. day) was observed after 31 days, and total transport of reactants and reaction
(2010) loss of portlandite [Ca(OH)2] and calcium products by the forced advection
silicate hydrate (C–S–H) was observed in enhances the chemical reaction,
the outer regions. leading to severe degradation.
Brandvoll State of fluid Class G Tested under both static CO2-saturated The propagation of carbonates was limited High degradation rate yields at the
et al. water and moving fluid, and CO2-brine to the surface during the static fluid moving fluid condition.
(2009) system at 50 ◦ C temperature and 10 MPa condition (i.e., approx. 200 μm after 30
pressure. days), whereas extensive cement
degradation (producing porous amorphous
silica gel) was observed under the moving
brine condition.
Huerta et al. Mechanical Class H Subjected to the coupled effect of An increase in confining pressure reduced Leakage paths through the cement can
(2009) loading confining stress and acid treatment. the effective fracture aperture size and acid be self-healing under down-hole stress
The effect of the following on aperture size treatment altered the mechanical and acid exposure conditions.
distribution was studied: cyclic loading properties of the cement, and as a result,
and unloading; re-assembling of cores to the fracture aperture was very narrow, (i.
study the new alignment of fractures; and e., 10 μm at 3.5 MPa confinement)
acid treatment (HCl exposure of cement compared to that of un-reacted samples.
cores for 7–12 days).
Huerta et al. Cement Class H Two cement mixtures were tested with 65:35 pozzolan blends had severe Higher penetration rates occur at high
(2009) composition pozzolan: cement ratio of 65:35 and penetration rates (i.e., complete pozzolan: cement ratios.
35:65, using fly ash as pozzolan and class degradation within 2 days) compared to
H cement under sequestration 35:65 blends (i.e., penetration depth is
environments (i.e., 50 ◦ C temperature and approximately 5 mm after 9 days)
15 MPa pressure).

With the introduction of FA-based geopolymer as a replacement for cium; n refers to the amount of polycondensation or polymerization and
OPC as well cementing material, it was first researched by Davidovits z is the ratio of Si/Al, that is 1, 2, 3, or higher (Nasvi et al., 2014).
(1994). Geopolymer is an alumino-silicate cementitious substance, The alumino-silicate source substances can be synthetic pozzolanas
which is activated by an alkaline solution with the combination of NaOH or Alumina Silicate-based industrial by-products, or a mixture of these.
and Na2SiO3 (refer to Fig. 8). The chemical formula of geopolymers can From previous studies, it is found that FA-based geopolymer dem­
be generally expressed as in Eq. (1). onstrates better performance under wellbore conditions compared to the
OPC in terms of several properties including high strength, low perme­
Mn[ − (SiO2 )z − AlO2 ]n.wH2 O (1)
ability, high pump-ability, long-term durability, high sulphate resis­
Where, M is the alkaline element such as Potassium, Sodium, or Cal­ tance, high volume stability, high thermal stability, less creeping action,
less drying shrinkage, high surface smoothness and high resistance to

5
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

in its microstructure, and appeared to be showing high compressive

strength values, compared to that of OPC (Uehara, 2010).
From the preliminary review, it was found that the FA-based geo­
polymer demonstrates a better performance over OPC as a well cement
for the CO2 sequestration process. The following sub-sections compre­
hensively discuss the characteristics of FA-based geopolymer and its
superior performance in different aspects as a well cement over con­
ventional OPC-based well cement.

5.1. The role of chemical characteristics of FA in FA-based geopolymer

The chemical characteristics of the FA-based geopolymer used as the

Fig. 7. Comparison between OPC and geopolymer productions (Developed well cement significantly depend on the chemical composition of the
after Hewayde et al., 2006). particular FA used for the production – as FA is the major component of
the geopolymer. Therefore, it is essential to thoroughly study the
chemical composition of the FA used for the geopolymer, before using it
on an industrial scale.
The chemical composition of FA is controlled by the source of the
coal, the type of furnace, the operating conditions, and post-combustion
Fig. 8. The schematic view of Geopolymer preparation. parameters (Kutchko and Kim, 2006). The chemical composition of FA
was analysed either by X-Ray Fluorescence (XRF) or X-Ray Diffraction
(XRD) analyses. In general, the chemical composition of FA is very
Table 4
similar to that of bottom ash. FA is primarily composed of Silica (SiO2),
Chemical analysis of OPC and class F fly ash (Developed after Sagoe–Crentsil
Ferric Oxide (Fe2O3), and Alumina (Al2O3), with smaller quantities of
et al., 2010).
Calcium Oxide (CaO), Sodium Oxide (Na2O), and Sulphur Trioxide
Chemical element OPC (wt %) Fly Ash (wt %) (SO3). In fact, the Calcium Oxide (CaO) percentage decides the fly ash
SiO2 20.20 47.19 category, where the fly ash with higher than 20% of CaO is referred to as
Al2O3 4.16 29.79 Class C, whereas lower-calcium fly ash is classified as Class F (Nasvi
Fe2O3 5.30 13.93
et al., 2013). The chemical configuration of the fly ash class is highly
CaO 64.80 3.29
MgO 1.29 1.38 important for the geo-polymerization process, as it decides the charac­
Na2O 0.22 0.24 teristics and microstructure of the final product (refer to Table 5). By
K2O 0.42 0.49 considering these effects, class F FA is mostly used for FA-based geo­
TiO2 – 1.77 polymer cement (Nasvi et al., 2013).
SO3 2.67 0.13
LOI 1.34 1.3
The chemical composition of FA in different countries that were
collected from the literature is graphically represented in Fig. 9. A
detailed summary including the country, generated power plant, and the
acidity (Khalifeh et al., 2014). As listed in Table 4, the chemical com­ chemical composition of the FA are given in Annex A. In fact, the
positions of OPC and FA mainly affect their respective properties and the chemical properties of FA play an important role in the utilization of FA,
consequent mechanical behaviour. as they may cause some adverse environmental impacts, such as trace
Experimental studies were conducted to identify some specific metal contamination in the groundwater, health hazards to living be­
properties of the FA-based geopolymer and OPC, which give a ings, and loss of fertility in the soil (Fulekar and Dave, 1986). Therefore,
comparative idea between OPC and FA-based geopolymer. In fact, FA it is essential to conduct a chemical analysis to assess whether it contains
class C and OPC class H were examined under different test procedures any constituents that are harmful to the environment and its living be­
by Ahdaya and Imqam (2019). Their stability test results proved the ings, before utilizing them for producing FA-based geopolymers.
inability of OPC to hold the water, as it showed a loss of water per­
centage of 2.28%. However, the tested FA-based geopolymer did not 5.2. The role of alkaline activator and other elements in FA-based
give any loss of water during the observed period. The sedimentation geopolymer
test, which was one of the stability tests showed the difference in density
between FA-based geopolymer and OPC with time, which were 0.008 The source materials are activated by alkali to take part in the geo-
g/cm3 and 0.028 g/cm3, respectively. This specifies that there is no polymerization process and reacted with Si and Al to make a geo­
considerable particle sedimentation occurred in the FA-based geo­ polymer binder (Ridha et al., 2018). The alkali activator is generally an
polymer during the observed time period. However, the selected class C alkali hydroxide or alkali silicate solution such as Sodium Silicate
geopolymer and class H cement exhibited the same rheological behav­ (Na2SiO3), Sodium Hydroxide (NaOH), Sodium Carbonate (Na2CO3),
iour (Ahdaya and Imqam, 2019). Especially in downhole conditions, Potassium Carbonate (K2CO3), Potassium Hydroxide (KOH), water, and
OPC generally forms CaCO3 under CO2 rich environment and causes a some mixtures of these (Nasvi et al., 2013). The most common activators
pH drop from 12–13 to 7–8. This in turn causes the degradation of the are NaOH and Na2SiO3 due to their availability, economic feasibility,
OPC applied between casing and caprock interfaces (Nasvi et al., and efficiency. As a result of a combination of Na2SiO3 and NaOH, it
2012b). In the case of FA-based geopolymer, it was found that Na2CO3 or creates a solid substance with fewer apertures and provides a sturdy
K2CO3 with a pH of approximately 10 delivers high resistance to connection between the geopolymer matrix and the aggregate. A high
degradation under an acid-rich environment (Davidovits, 2005). percentage of NaOH concentration involves leaching the Si and Al ions
Further, OPC and geopolymer were tested by Uehara (2010) in different in NaOH solution, which is directly influenced to generate higher
acidic environmental conditions, where OPC failures were observed compressive strength (Ahdaya and Imqam, 2019). Therefore, the
after 4 months in a 10% H2SO4 experimental condition. Since OPC chemical concentration and the ratios, which are mixed in geopolymer,
minerals reacted with an acidic solution, a high degradation of the OPC are directly involved in deciding the properties of the final FA-based
sample occurred because of its inability to withstand acidic surround­ geopolymer cement. Due to that effect, researchers have done experi­
ings. In contrast, FA-based geopolymer did not show significant changes mental work on identifying the variation of geopolymer characteristics

6
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

Table 5
Chemical composition (%) of Class F fly ash and Class C fly ash (Developed after Guo et al., 2010).
Element SiO2 Al2O3 Fe2O3 MgO CaO SO3 K2O Na2O LOI

Composition Class F 50.0 28.0 12.0 0.6 6.5 – 1.5 0.2 –

Class C 38.0 19.0 9.0 5.0 20.0 3.0 0.4 1.0 3.5

Fig. 9. Variation in the chemical composition of FA in different countries.

Fig. 12. Variation of Compressive strength with the different alkali contents
(Developed after Thakur and Ghosh, 2009).

with the changes in content ratios of chemical elements. In fact, many

research works have shown the effect of changing several physical
properties with the variation of the 1) alkaline activator to fly ash ratio,
2) Sodium Silicate to Sodium Hydroxide ratio, and 3) Sodium Hydroxide
concentration.
Fig. 10, developed after Ahdaya and Imqam (2019) shows the vari­
ation of physical properties of class C FA-based geopolymer paste
including density, compressive strength, and fluid loss under different
chemical ratios. Further, Hardjito et al. (2008) tested both class C and
class F fly ash by increasing the alkaline activator to fly ash ratio and
observed an increment in the compressive strength, in which 0.40 of
optimum activator to fly ash ratio yielded the highest compressive
strength (refer to Fig. 11). In addition, the influence of mixed compo­
sition on compressive strength was deeply examined by Thakur and
Fig. 10. Variation of physical properties with the changes in chemical con­
Ghosh (2009) using different percentages of alkali content and silica
centration of fly ash class C (Developed after Ahdaya and Imqam, 2019).
content. In this case, compressive strength of class F FA-based geo­
polymer with different alkali percentages of 0.46%, 0.50%, 0.54%,
0.58%, 0.62% was tested in 3 days, 7 days, and 28 days, where the test
results showed a linear increment with increasing alkali content (refer to
Fig. 12). The highest compressive strength of 48.20 MPa was observed
for maximum alkali content (Na2O/Al2O3) of 0.62. Thakur and Ghosh

Fig. 11. Variation of Compressive strength with the different alkaline activator
to fly ash ratio (Developed after Hardjito et al., 2008).

Fig. 13. Variation of Compressive strength with the different silica contents
(Developed after Thakur and Ghosh, 2009).

7
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

(2009) also examined the silica content (SiO2/Al2O3) vs. compressive the cement slurry. Therefore, investigations have been conducted to
strength using different silica contents, such as 3.70%, 3.85%, 4.00%, observe the suitability of retarders for the geopolymers to improve the
4.15%, 4.30% (refer to Fig. 13). Noticeably, it has shown a similar setting time (Nasvi et al., 2013).
behaviour as in Fig. 11, where the compressive strength variation Curing time is also an important aspect that affects the characteris­
showed a slight increment, yielding an optimum compressive strength at tics of FA-based geopolymer well cement. Many works done in this
a silica content of approximately 4.15%, beyond which the compressive aspect have shown that the increase in curing time tends to increase
strength reduces with the increasing silica percentage. In addition to the compressive strength. However, there is no substantial increase
above characteristics, the physical appearance at the elevated temper­ observed after 48 h (Thakur and Ghosh, 2009). In fact, class F FA-based
ature also depends on the Si/Al ratio of the FA-based geopolymer geopolymer samples have been tested with alkali and silicate contents,
mixture (Thokchom et al., 2012). and water-to-geopolymer solid ratios of 0.62, 4.0, and 0.228, respec­
Further, the cracks that appeared in the microstructure of FA-based tively. These specimens were placed in an oven under air pressure and
geopolymer well cement was investigated by researchers under temperature at 85 ◦ C for a variable duration of 4–72 h. The highest
different elevated temperatures, where the microstructure effects have compressive strength of 40.8 MPa was obtained with 48 h of curing time,
been scanned by using Scanning Electron Microscopy (SEM) and Energy but the rise of curing time did not affect the increase of compressive
Dispersive X-ray (EDX) tests. Geopolymer cement with a higher Si/Al strength further (Thakur and Ghosh, 2009). This confirms that geo­
ratio shows lesser crack formation, comparatively to the mixtures with polymer attains its ultimate compressive strength in a short period of
lesser Si/Al ratios (Thokchom et al., 2012). These results were obtained curing time with a significantly high geo-ploymerization process
by using the geopolymer samples with Si/Al ratios of 1.9, 2.2, and 1.7, (Hardjito et al., 2004). However, it is found that quick curing at higher
under three main elevated temperatures of 300 ◦ C, 600 ◦ C, and 900 ◦ C. temperatures would produce cracks, which makes a negative impact on
According to the X-ray diffractogram, geopolymers with 1.7 and 2.2 of the characteristics of geopolymer cement. On the other hand, the
Si/Al ratios showed different behaviour under exposure to 900 ◦ C. As a compressive strength has shown a significant reduction when cured at a
result, the geopolymer with a low Si/Al ratio of 1.7 has given a highly higher temperature for a longer period, as this breakdowns the granular
amorphous nature and showed less amorphous phases in the geo­ structure of the geopolymer mixture (Hardjito et al., 2004).
polymers with a high Si/Al ratio of 2.2. According to these experimental In fact, the impact of curing temperature was studied by Hardjito
results, it was confirmed that FA-based geopolymers with high Si/Al et al. (2004) using specimens with 0.62 of alkali content and 4.0 of silica
ratios have shown high stability under elevated temperatures (Thok­ content. These samples were cured for 48 h changing the curing tem­
chom et al., 2012). perature from 45 ◦ C to 120 ◦ C under atmospheric pressure. The
Especially, at downhole conditions, the temperature gradient is maximum compressive strength of 48.2 MPa was obtained at the opti­
assumed to be 30 ◦ C/km, which can be varied with the location (Nasvi mum temperature of 85 ◦ C. Nasvi et al. (2015) investigated the me­
et al., 2012a). The well cement which is used in the CO2 sequestration chanical behaviour of geopolymers at various curing temperatures from
process should have a greater stability and a crack-free surface to carry 20 ◦ C to 80 ◦ C, under triaxial experimental conditions. Their experi­
out a successful CO2 sequestration process under elevated temperatures mental result also confirmed the above observations by getting the
and high pressures. Therefore, it is highly important to carefully deter­ deviatoric strength increment from 20 MPa to 105 MPa, when increasing
mine the correct elemental compositions of FA-based geopolymer the curing temperature from 23 ◦ C to 60 ◦ C, and showed a 15% of
mixture to produce a well cement that would be stable under downhole deviatoric reduction with the rapid change of curing temperatures from
conditions. 60 ◦ C to 80 ◦ C (Nasvi et al., 2015). As illustrated in Fig. 14, beyond
60 ◦ C, the strength level is closer to the stabilized level, which does not
5.3. The role of curing conditions in the mechanical aspects of FA-based give further strength increment with the curing temperature.
geopolymer Not only compressive strength but also stress-strain behaviour and
crack occurrence also depend on curing temperature. In fact, the tested
The depth of the injection wells which are used in the CO2 seques­ samples have shown higher strains of approximately 6–8% at failures, at
tration process can vary from 0.8 to 2 km and the temperature can vary lower temperatures such as 23 ◦ C and 30 ◦ C, whereas high-temperature
from 30 ◦ C to 80 ◦ C at the deepest location of the well with a pressure cured samples (i.e., cured at 40 ◦ C–80 ◦ C temperature) have experienced
variation up to 50 MPa (Nasvi et al., 2012b). In most cases, curing low strains of approximately 0.8–3.5% at failure. These results conveyed
conditions such as curing temperature, curing time, and categories of that the FA-based geopolymer well cement may have behaved as a
curing support to predict the performance of geopolymer at downhole brittle material at high temperatures (Nasvi et al., 2012b).
conditions (Nasvi et al., 2013). Numerous experimental studies were The curing environment was examined by Sagoe–Crentsil et al.
done to investigate the compressive strength, Poison’s ratio, stress-strain (2010) to identify the effect on compressive strength by using OPC and
changes, and crack initiation at altered curing temperatures. Heat is a
major accelerator for the geo-polymerization process, due to which, the
curing was normally carried out at elevated temperatures. According to
past studies, it was revealed that by increasing the curing temperature
up to 60 ◦ C, geopolymer cement achieves up to 70% of its strength in the
first 4 h of setting time, where the setting time and curing temperature
show an inverse relationship (Nasvi et al., 2013). When the temperature
was raised from 30 ◦ C to 50 ◦ C to 75 ◦ C, the Al and Si precursors were
highly dissolved from the source material and the related setting time
was reduced (Tempest et al., 2009). Since high pressure and temperature
are used in general CO2 sequestration wells, reduced setting time is
possible under wellbore conditions, and this has to be adjusted to ach­
ieve a workable mix to fill the annular surface and allow sufficient time
before hardening. According to the practical circumstances, FA-based
geopolymer cement can be handled for a period of up to 2 h without
any setting, in the cured temperature range of 65 ◦ C–80 ◦ C (Hardjito
et al., 2008). In general, the nature of the downhole conditions requires Fig. 14. Variation of compressive strength of geopolymer well cement with
1–4 h setting time, according to the temperature and the pumping rate of curing temperature (Developed after Nasvi et al., 2015).

8
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

Table 6
Compressive strength development for ambient and steam curing conditions
(Developed after Sagoe–Crentsil et al., 2010).
Compressive OPC Class F FA-based geopolymer
strength
Steam curing Ambient Steam curing Ambient
environment curing environment curing
(MPa) environment (MPa) environment
(MPa) (MPa)

1 day 22.50 9.10 42.30 –

3 days – – – 2.00
7 days 28.00 35.40 42.30 7.20
28 days 38.10 43.70 44.00 35.50

class F FA-based geopolymer samples, during which two curing envi­

ronments were used, i.e., 1) heat curing and 2) ambient curing (refer to
Table 6). According to the results, A relatively speedy early strength gain
was shown by OPC samples compared to geopolymer samples at Fig. 15. Compressive strength variation of class G (G) and Class A (T10) well
ambient curing environment. These results showed that ambient tem­ cements in CO2 environment (Developed after Condor and Asghari, 2009).
perature can also make an influence on the rate of early strength
increment (Sagoe–Crentsil et al., 2010). 5.5. Hydro-mechanical behaviour of FA-based geopolymer well cement
It is evident from the above studies that curing condition has a sig­
nificant effect on the performance of FA-based geopolymer; thus, needs 5.5.1. Compressive strength and tensile strength of FA-based geopolymer
to be controlled cautiously to achieve the required level of performance. Compressive strength is one of the key characteristics that have a
great influence on wellbore integrity at higher pressure conditions to
5.4. The role of elevated temperature on the behaviour of FA-based ensure the structural support for the casing and hydraulic/mechanical
geopolymer cement isolation in the downhole environment (Di Lullo and Rae, 2000). In­
vestigations on class G cement have concluded that conventional OPC is
Different studies were conducted to identify the surface evaporation not capable of bearing the stress created by frequent pressure and
effect using sealed and unsealed conditions during the curing process temperature changes in wellbores (Pedersen et al., 2006). Researchers
and the strength variation of different FA-based geopolymer pastes in have experienced that the performance of OPC-based wells dropped
different elevated temperature exposure conditions (Kannangara et al., after 15 years when the wells were exposed to 180 ◦ C and 22% CO2, and
2021). These experimental results concluded that less excess initial it was noticed that CO2 is the main reason for degradation. In addition, it
surface evaporation can occur with high strength values for sealed was concluded that higher temperature and CO2 are the causes of the
geopolymer specimens under elevated temperatures. Unsealed geo­ loss of compressive strength and cement integrity in the wells (Krilov
polymer samples have shown approximately 35% and 25% lesser values et al., 2000). Fig. 15 shows the compressive strength reduction of Class G
for the residual strength results at 400 ◦ C and 800 ◦ C, respectively, (G) and Class A (Type T10) well cements in a CO2 environment under
where they have caused a higher degree of thermal cracking and split­ 2200 psi and 55 ◦ C pressure and temperature conditions. It is noticeable
ting compared to the sealed geopolymer samples (Kannangara et al., that the compressive strength of both class G and class A well cement
2021). Experimental works have been conducted to study the compar­ reduces gradually with time, confirming the adverse impact imposed by
ative behaviour of geopolymers made with metakaolin and FA after CO2 on conventional well cement (Condor and Asghari, 2009).
exposure to elevated temperatures. The FA-based geopolymer has The durability of class G well cement was investigated by Lécolier
shown an increment in strength after exposure to elevated temperatures et al. (2007), where class G cement samples were cured at 80 ◦ C for one
(i.e., 800 ◦ C), while metakaolin-based geopolymer has shown a reduc­ month, and samples were aged in water, brine, and crude oil at 80 ◦ C. It
tion in strength after similar exposure. The availability of a large number was observed that samples aged in water and crude oil did not show
of small pores on FA-based geopolymer cement is the reason for the considerable strength reduction after one year, whereas the uniaxial
observed result, where these pores facilitate the escape of moisture when compressive strength (UCS) of the samples aged in brine fell to 50% of
heated, triggering minimal damage to the geopolymer matrix (Kong the initial strength. These results concluded that the mechanical
et al., 2007). Although FA-based geopolymer did not show any notice­ strength of OPC-based cement reduces in CO2- rich downhole conditions
able cracks on the surface of the specimens after exposure to elevated due to cement degradation.
temperature, metakaolin geopolymer has shown macro-cracks with
widths of 0.1–0.2 mm due to the absence of pore distribution structures Table 7
similar to the FA-based geopolymer (Kong et al., 2007). Ridha et al. A summary of previous studies related to compressive strength and tensile
(2020) showed that the crystal-like shape identified as calcium car­ strength of geopolymers.
bonate was formed at the surface of spherical fly ash particles after CO2
Reference Source material, Test UCS UTS UCS/UTS
interaction-induced carbonation. Although, the carbonation process Variables (MPa) (MPa) (%)
occurred, no strength reduction was observed during the study. Further,
Hardjito and FA, different NaOH 44–89 4.4–7.4 8.3–10
microstructure analysis has discovered that zeolite has been formed Rangan molarity and curing
during CO2 acid exposure for geopolymer cement, which causes a slight (2005) temperatures
development of strength. According to the above findings, FA-based Mishra et al. FA, different NaOH 13–27 0.9–3.1 5.6–9
geopolymers have shown to be suitable for well cement in CO2 injec­ (2008) molarity values
Sofi et al. different FA types, H2O/FA 47–57 2.8–4.1 5.2–8.3
tion wells under elevated temperature conditions.
(2007) ratio and slag content
Olivia and different activator/FA ratio 56–60 4–4.3 7–7.3
Nikraz and superplasticizer
(2011) contents
Raijiwala et al. FA, different curing 39–55 4.1–6.5 10.5–11.7
(2012) temperatures

9
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

cement matrix, leading to higher tensile strength values (Sarker, 2011).

5.5.2. Permeability of FA-based geopolymer

The permeability of the well cement is the major problem compared
to strength-related problems. The low permeability is the main indicator
that promotes the leakage-free nature of the sequestration process
(Nasvi et al., 2014). Especially at downhole conditions, the generally
used OPC was degraded due to the CO2-rich environment, causing the
increment in permeability and porosity (Nasvi et al., 2014). The
permeability of FA-based geopolymer well cement was tested under
different mix compositions to recognize the effect of the chemical
mixture on the permeability variation. An experiment was done by Van
den Heede et al. (2010) to check the Oxygen permeability using four
different FA-mixed samples, which included 0%, 35%, 50%, and 67% of
FA amounts. The test results showed low permeability values of 0.5–2.0
× 10− 16 m2 for most of the FA-containing samples, compared to
generally used OPC samples with a 3.0 × 10− 16 m2 permeability. These
findings indicate that the pores in FA-containing samples are less con­
Fig. 16. Comparison of splitting tensile strength and the compressive strength
nected due to the additional hydration products created by the pozzo­
of OPC and FA-based geopolymer well cements (GPC) (Developed after
Sarker, 2011). lanic FA reaction, causing the low permeability. Not only the FA
composition but also researchers have investigated the permeability of
FA-based geopolymers by adding different slag amounts and compared
Not only compressive strength, but also tensile strength of well
the permeability values with class G cement, under CO2 environments
cement plays a vital role in cement integrity due to the following rea­
(Nasvi et al., 2014). The test samples were made by partially replacing
sons: 1) if there are sufficient induced stresses to cause a mechanical
FA with alkali-activated slag in 0%, 8%, and 15% by mass. The CO2
failure of a set cement sheath, failure is probably of tensile nature
permeability of these samples has been varied from 0.002–0.019 μD,
(Goodwin, 1997), and 2) if the formation gas enters the cement pores
0.002–0.021 μD, and 0.0005–0.002 μD, respectively. The outcomes of
and the gas pressure exceeds the tensile strength value of cement, a
this study concluded that the permeability of geopolymer without
fracture is created leading to gas migration problems (Backe et al.,
alkali-activated slag and the geopolymer with 8% slag is 100 times lower
1999). Therefore, cement with higher tensile strength and higher tensile
than that of class G cement, where the slag with 15% mixture has given
strength to Young’s modulus ratio is preferred for well cementing
the lowest permeability which is nearly 1000 times lower than the
(Mavroudis, 2001). In fact, the tensile strength of carefully mixed
permeability of class G cement in general. This is mainly attributed to
OPC-based well cement under API-recommended formulations varies
the improvement of the microstructure of geopolymer with the addition
from 1 to 2 MPa (Labibzadeh, 2010).
of slag (Nasvi et al., 2014).
The tensile strength of geopolymer needs to be studied to predict its
In typical downhole conditions, the well cement is exposed to a
suitability as well sealant material. Many researchers have studied the
temperature range of 20–80 ◦ C with a depth variation of 1.5–2 km.
tensile strength of alkali-activated geopolymer materials (Hardjito and
Therefore, the permeability variation under different temperatures was
Rangan, 2005; Raijiwala et al., 2012; Sakulich et al., 2009; Sarker, 2011;
also examined by many researchers. An experimental study conducted
Sofi et al., 2007). A summary of their test methods and their salient
by Nasvi et al. (2014) reported that the permeability values of FA-based
findings are shown in Table 7, including the source material and test
geopolymer well cement increase with the increment in temperature for
variables employed, the UCS, uniaxial tensile strength (UTS), and the
a given inlet and confining pressure. The most important finding is that
ratio of UTS/UCS.
the highest permeability value of geopolymer is nearly 5000 times lower
Table 7 indicates that the tensile strength values range from 5.2% to
than the recommended limit of 200 μD for well cement by the American
11.7% of the compressive strength, and the values vary significantly
Petroleum Industry (API). In addition to the above findings, a summary
with the geopolymer mix compositions and test conditions used. The
of the permeability values between OPC and FA-based geopolymer is
relationship between tensile and compressive strength depends on many
shown in Table 8. In the case of conducting a successful CO2 seques­
factors such as age, strength of concrete, liquid/solid ratio, curing
tration project, the idea about the permeability variation of well cement
temperature, and type of curing (Zain et al., 2002).
should be focused on; however, only limited studies have been con­
Sarker (2011) compared the splitting tensile strength of
ducted on testing the CO2 permeability of fly ash-based geopolymers up
FA-based-geopolymers and OPC based on their results and the findings
to date.
from the literature. It was noticed that, for a given compressive strength,
the splitting tensile strength of FA-based geopolymer is higher than that
5.5.2.1. Available models for predicting the CO2 permeability of FA-based
of OPC (refer to Fig. 16). The higher tensile strength is attributed to the
geopolymer. Some different empirical equations and models were
use of soluble silicates in geopolymers, where the soluble silicates pro­
developed by researchers to estimate the permeability of different
duce denser interfacial transition zones between the aggregate and the

Table 8
Previously reported permeability values for FA-based geopolymer and OPC-based well cement.
Reference Permeating fluid Permeability

Geopolymer OPC

Van den Heede et al. (2010) Oxygen 3.0 × 10− 16 m2 0.5–2.0 × 10− 16 m2
Zhang and Talman (2014) Water 0.5–1.5 × 10− 12 m2 1.0 × 10− 10 m2
Sagoe–Crentsil et al. (2010) Oxygen 6.19 × 10− 17 m2 6.32 × 10− 17 m2
Water 1.52 × 10− 10 m/s 1.73 × 10− 11 m/s

10
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

liquids, materials, and gases. Some of these models have been tested to scenarios, Nasvi et al. (2014) performed an undrained triaxial experi­
identify their applicability in the prediction of CO2 permeability in FA- ment to improve Eq. (5). Through that, they developed a formula for
based geopolymer. Studies have been performed using the pressure downstream pressure using regression techniques as shown in Eq. (6).
transient approach to measure the permeability of brittle materials. In ( ) ( )
T − T0 Pin
the pressure transient method, the boundary condition shown in Eq. (2) Pout = A +B − C (6)
T + T0 Pc
has been used to calculate the flow rate through the sample (Siriwardane
et al., 2009).
Where T0 is the reference temperature (K), Pin is the inlet pressure
ʚp(L, t) Q (MPa), Pc is the confining pressure (MPa), A and B are material constants
= (2)
ʚt βV depending on the temperature and mechanical loading, respectively,
and C is a constant. The values of A, B, and C for geopolymer are 49.16,
Where p is the pressure, L is the length of the sample, t is the time, Q is 4.47, and 1.32, respectively. It is stated that the proposed formula is
the flow rate through the sample, V is the downstream volume, and β is valid only up to 80 ◦ C, beyond which the estimation may not accurate
the adiabatic compressibility of the gas. enough.
The Hagen–Poiseuille expression (refer to Eq. (3)) can be used to Eq. (5) was then adjusted to calculate the CO2 permeability, and the
calculate the permeability of a porous solid body for a compressible values of the material constants AT and AP of FA-based geopolymer were
fluid. This equation assumes a linear variation between the volumetric evaluated based on the experimental results. The modified equation to
flow rate and pressure gradient (Ranjith and Perera, 2011). predict the permeability of geopolymer at different temperatures is
given in Eq. (7).
2QμLpout
kA = (3) ( )0.65
A(pi 2 − pout 2 ) Pin
kT = k0 100.016(T− T0 ) (7)
Pout
Where kA is the apparent gas permeability, μ is the fluid viscosity, L is the
length of the sample, A is the area, Q is the flow rate, pi and pout are Where Pin is the inlet pressure, Pout is the downstream pressure given by
upstream and downstream pressures, respectively. Eq. (6), and AT and AP are 0.016 and 0.65, respectively for fly ash-based
According to the experimental work that was conducted under geopolymer.
drained conditions, a linear relationship has been derived to estimate There are a number of theoretical and empirical models developed
the variation of CO2 flow rate with injection pressures under different and progressed to predict the CO2 permeability in FA-based geopolymer.
confining pressures (refer to Eq. (4)). Therefore, Eq. (4) can be used to However, still there are various parameters to be incorporated in the
find the CO2 permeability through FA-based geopolymer (Nasvi et al., models to accurately predict the permeability values.
2013).
Q = − 0.000417 pc + 0.00165 pin + 0.00484 (4) 6. Meta-analysis with current experimental results on FA-based
geopolymer well cement
Where Q is the CO2 flow rate (l/h), pc is the confining pressure (MPa),
and pin is the injection pressure (MPa). Through the thorough literature review, the weaknesses of OPC as a
REFPROP database was used by Huber et al. (2008) to derive the well cement were identified, and the suitability of FA-based geo­
viscosity (μ) and adiabatic compressibility (β) of CO2 used in Eqs. (2) and polymers to replace the OPC in geo-sequestration projects is investi­
(3) to develop a model regarding CO2 permeability in a solid body. The gated. In this context, the well designers have a responsibility to provide
REFPROP program, developed by the National Institute of Standards accurate details about mix composition and other preparation condi­
and Technology (NIST), provides information about the thermodynamic tions for achieving the required properties of FA-based geopolymer to
and transport properties of industrially important fluids and mixtures. maintain the wellbore integrity at different temperature and pressure
The viscosity and adiabatic compressibility values of CO2 taken from the variations under downhole conditions.
REFPROP database for different mean pressure (Pm) and temperature For encouraging the utilization of FA-based geopolymer as a well
(T) conditions are utilized in the test to identify the behaviour. Results cement, apart from comprehensively reviewing pertinent studies, an
showed a closer relationship between viscosity and adiabatic analytical study was carried out to develop predictive models for dry
compressibility with pressure than temperature. In the subcritical density, compressive strength, autogenous shrinkage strain, and
pressure conditions (P < 7.2 MPa) the viscosity and adiabatic permeability of FA-based geopolymer using different influential pa­
compressibility has not shown a significant variation with the temper­ rameters. For this purpose, databases were developed by collecting data
ature, whereas in the supercritical CO2 conditions (P > 7.2 MPa and T > from many experimental studies available in the literature and using
31.8 ◦ C), the variation of viscosity and compressibility has shown a
considerable variation with both pressure and temperature.
Table 9
In the case of developing the equations to predict the permeability of Coefficients of the nonlinear equation developed
FA-based geopolymers at different temperatures under different for dry density of FA-based geopolymer (refer to
confining pressures, the formula that was developed by Gawin et al. Eq. (8)).
(1999) is widely used. This formula predicts the permeability of concrete
Coefficient Value
at different temperatures and different gas pressure conditions using a
a1 31202.079
mechanistic approach (refer to Eq. (5)).
a2 513.493
( )Ap a3 5302.295
Pg
kA = k0 10AT (T− T0 ) (5) a4 − 55729.
P0 a5 3664.797
a6 90213.612
Where k0 is the intrinsic permeability at the reference temperature a7 43.769
(T0 = 293 K), Pg is the gas pressure, P0 is the atmospheric pressure, and b 0.328
c − 0.793
AT and Ap are material-dependant constants.
d 0.005
The above formula (refer to Eq. (5)) has been used to predict e 0.005
permeability under drained conditions, where the downstream pressure f − 0.968
is always 0.1 MPa. Since the downhole pressure is not constant in actual g 0.01

11
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

different analytical techniques including Multivariable Regression ( )

S
(MVR) and Artificial Neural Network (ANN) analyses. The mainly used CS = − 1.015 + 0.08615(CT) + 18.4 + 0.0002946 (CT)2
input variables are alkaline activator to FA ratio (AA/FA), Na2SiO3 to A
( ) ( )2 (9)
NaOH ratio (SS/SH), NaOH concentration (M), water to solid ratio (W/ − 0.01148(CT)
S
− 4.449
S
S), curing temperature (CT), and curing time (T). A A

6.1. Predictive model for the dry density of FA-based geopolymer Where CS is the compressive strength in MPa, CT is the curing tem­
perature in 0C and S/A is the Si/Al ratio in the geopolymer mixture.
The developed model predicts the dry density of the FA-based geo­ From the review, it was identified that developing a model for pre­
polymer mix, considering Na2SiO3 to NaOH ratio (SS /SH), alkaline dicting the compressive strength of FA-based geopolymer with different
activator to fly ash ratio (AA /FA), and NaOH concentration (M) as Water to geopolymer solid ratios (W /S), Curing temperature (CT), and
influential parameters. 45 data sets were collected from the experi­ curing time (T) is important to prepare an FA-based geopolymer mixture
mental investigations carried out by Ahdaya and Imqam (2019) and with required compressive strength characteristics. Hence, model 3 was
Mermerdaş et al. (2020), in which 34 and 11 data sets were used for developed considering these parameters, where 32 samples were
model development and model validation, respectively. The developed collected from the experimental investigation done by Nikolić et al.
non-linear multivariable regression model (NL-MVR) is given in Eq. (8), (2015). Out of the collected 32 data sets, 24 samples were used for the
which has a strong coefficient of determination value (R2) of 0.964, development of the model and 8 samples were used for the validation.
confirming the higher accuracy of the model. The relevant coefficients of Linear multivariable regression (L-MVR) analysis was performed, and a
the non-linear equation are listed in Table 9. model was developed with an R2 value of 0.932, which indicates a strong
( )b ( )c ( )e ( )f relationship between independent variables and the dependent variable
SH AA SH AA (refer to Eq. (10)).
Dry Density = − a1 + a2 − a3 − a4 M d + a5
SS FA SS FA ( )
( ) W
gAA AA CS = − 122.2217499 + 0.163917701(CT) + 0.358590226(T)
+ a6 M FA − a7(M) S
FA
+ 69.86642578 (10)
(8)

6.2. Predictive model for compressive strength of FA-based geopolymer 6.3. Predictive model for autogenous shrinkage strain of FA-based
geopolymer
Mainly three different models were developed for predicting the
compressive strength of FA-based geopolymer using various input var­ Since prediction of shrinkage of FA geopolymer is important at the
iables, as compressive strength is highly important in preserving the well primary stage of well cement preparation, an NL-MVR model was
integrity during the CO2 sequestration process. Among them, model 1 developed by using 160 data sets, where 132 and 30 data sets were used
was developed using ANN due to the complex relationship between the to develop and test the model, respectively. This model predicts the
selected input variables and the output variable. The model parameters autogenous shrinkage strain (ASS) of the FA-based geopolymer mix after
are given in Annex B. This statistical model can be used to predict the the curing period, considering Na2SiO3 to NaOH ratio (SS /SH), Age of
compressive strength of FA-based geopolymer when the alkaline acti­ cement paste (T), NaOH concentration (M) and binder (B) as indepen­
vator to FA ratio (AA/FA), Na2SiO3 to NaOH ratio (SS/SH), NaOH dent variables. Binder means the mass of fly ash and alkaline activator
concentration (M), curing temperature (CT), and curing time (T) are that is included in m3 of FA-based geopolymer mixture. The developed
known. The developed model has an R2 value of 0.916 which confirms NL-MVR model is given in Eq. (11), which has an R2 value of 0.862.
the reasonable accuracy of predicting the compressive strength of FA- Hence, the developed reasonably accurate model can be used to
based geopolymer. approximately estimate the autogenous shrinkage of FA-based geo­
Model 2 for compressive strength estimation was developed using Si/ polymer during the pre-feasibility stage of the CO2 sequestration project.
Al ratio and curing temperature (CT) as independent variables. The data [
set of FA-based geopolymers with different Si/Al ratios and curing ASS = 153.606(M)0.862 − 153.094(B)0.536 + 30021.428(T)0.002
temperatures for the analysis was gathered from a research study that ( ) ( )
SS SS
was conducted by Zhou et al. (2016). The collected data was divided by + 0.588(M) + 0.025 (B) + 4.258(B) + 77.402(M) (11)
SH SH
a 3:1 ratio for model development and validation. The resulting NL-MVR ]
model is a second-degree polynomial equation (refer to Eq. (9)) with an + 13.007(SS / SH) − 28208.902 x10− 6
R2 value of 0.976, and it coincides with the findings shown in Fig. 13 in
section 5.2. In fact, according to the graphical representation presented
in Fig. 13, the compressive strength rises near-linearly with silica con­ 6.4. Predictive model for CO2 permeability of FA-based geopolymer
tent up to a certain extent, beyond which the compressive strength was
reduced with the increment of silica percentage. This behaviour is To predict the CO2 permeability (k) in FA-based geopolymer, five
confirmed by the developed Eq. (9). linear multivariable regression (L-MVR) equations were developed
using the independent variables as injection pressure (pi ) and curing
temperature (T) under different confining pressures including 12, 16,
20, 25, 35 MPa. The developed equations for the selected parameters
under each divided group are shown in Eq. (12).

⎧ ( 2 )
⎪
⎪ − 0.003103851pi + 0.000459223 T + 0.017956233; Pc = 12 ( R2 = 0.880)
⎪
⎪
⎨− 0.001069696pi + 0.000254524T + 0.007122723; Pc = 16 ( R2 = 0.995)
k (Pi , T) = − 0.000245848pi + 0.000124239T + 0.001017792; Pc = 20 ( R2 = 0.959) (12)
⎪
⎪
⎪−
⎪ 0.003950112pi + 0.001158095T + 0.050207161; Pc = 25 ( R2 = 0.964)
⎩
− 0.000391676pi + 0.000175025T + 0.006027776; Pc = 35 R = 0.980

12
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

parameters and the resultant well behaviour, which confirms the

applicability of FA-based geopolymers as an effective well cement.
These results further concluded that the linear relationship between Hence, a meta-analysis was carried out to develop accurate correlations
injection pressure and the temperature with CO2 permeability has been between such parameters. The outcomes of the meta-analysis will be
strengthened with the increment of confining pressure values. Accord­ helpful for decision-making regarding the appropriateness of applying
ing to the developed equations, CO2 permeability increases with FA-based geopolymer as a replacement to OPC in order to conduct a
increasing temperature and decreases with increasing injection pressure sustainable geo-sequestration process under proper isolation conditions.
under specified confining pressures. The resulting R2 values for these
developed models conclude that the predicted results from these 8. Recommendations for future work
regression models exhibit a reasonable correlation with the experi­
mental values for the selected input parameters. The main objective of this research study was to investigate FA-based
geopolymer well cement during the CO2 sequestration process, through
7. Summary an analytical and comparative study. A systematic review on the usage
of FA as geopolymer material during CO2 sequestration was conducted
Well cement plays a major role in the CO2 sequestration process, and the findings were discussed comprehensively in the 1st phase of the
where the sealing property of the well cement is highly important to study. In the 2nd phase, statistical models were introduced using MVR
confirm the leakage-free process. With these requirements, the existing analysis and ANN technique. Four important properties of FAG were
research work was generally carried out in different paths, where they targeted for the model development including FAG dry density,
have mainly focused on generally used API-class well cement, associated compressive strength, autogenous shrinkage strain, and CO2 perme­
issues with the traditionally used well cement at CO2 rich environment, ability in FAG. Since there is numerous research being carried out
leakage pathways of CO2, the flow and mechanical behaviour changes considering different other influential parameters, it is suggested to
under different temperature and pressure variations and the quality of collect the experimental data and create a database, which can be used
the cement slurry. According to past-observed results, the generally used to develop more accurate correlations to predict the mechanical
OPC has shown considerable degradation at downhole conditions, behaviour of FA-based geopolymer as well cement for the CO2 seques­
reducing the sealant properties of the well cement. Based on these tration process. In fact, the density of the FA-based geopolymers varies
experimental works, most of the researchers confirmed the inability of with the curing type and the conditions; as such, further studies are
OPC to perform in acidic downhole conditions as a stable well cement. In recommended to improve the developed models by using the densities of
addition to the degradation, the OPC has shown some other weaknesses the FAG samples that are prepared at different curing conditions. Due to
such as higher shrinkage properties, high permeability and porosity the limited availability of data, three models were developed to predict
values, and strength reductions. the compressive strength of FAG using different input variables in each
To address these issues of OPC well cement, inorganic geopolymer model. It is recommended to develop a single model that represents all
was introduced as an alternate solution, where different alumino-silicate the above-mentioned input parameters, which will be more effective in
source materials were proposed as geopolymers. Due to the ability to taking direct decisions on the compressive strength of FAG.
reduce the gigantic amounts of fly ash piled up during coal-fired power
plant operations, FA-based geopolymer as an alternative well cement 9. Statements and declarations
has gained vast popularity in the industry. Therefore, the properties of
FA-based geopolymer and its applicability as a more effective well The authors have no competing interests to declare that are relevant
cement in the CO2 sequestration process were examined widely by many to the content of this article.
researchers.
Through this comprehensive review, major influential parameters of
FA-based geopolymer well cement were analysed, mainly focusing on Declaration of competing interest
the compressive strength changes under different curing temperatures
and various chemical compositions. Further, the other main factor that The authors declare that they have no known competing financial
has influenced the maintenance of required wellbore integrity and zonal interests or personal relationships that could have appeared to influence
isolation is the permeability of bonding material. Therefore, the relevant the work reported in this paper.
literature on permeability changes under different mix compositions and
temperatures has also been investigated. Data availability
From the thorough literature review, it was found that there was no
proper representation or relationship developed between the influential Data will be made available on request.

Annex A.
Table A.1
A summary of literature on chemical composition of FA in different countries

Reference Country Power plant Fly ash SiO2 Al2O3 Fe2O3 CaO MgO K2O Na2O SO3 LOI
type

Nasvi et al. (2012b) Australia Gladstone (Queensland, Australia) Class F 48.3 30.5 12.1 2.8 1.2 0.4 0.2 0.3 1.7
Gunasekara et al. Gladstone (GFA) Class F 50.82 29.89 10.26 3.24 0.8 0.58 0 0.28
(2014) Port Augusta (PAFA Class F 49.97 31.45 3.22 5.03 1.54 1.87 1.85 0.33
Collie (CFA) Class F 52.67 29.6 11.27 0.94 0.72 0.65 0 0.48
Mount Piper (MPFA) Class F 65.18 25.3 1.9 0.63 0 3.65 0 0.23
Tarong (TFA) Class F 73.12 21.5 1.36 0.29 0 0.63 0 0
Provis et al. (2009) Gladstone Power Station in Class F 46.4 28.3 11.7 5.1 1.4 0.6 0.3 0.3 3.3
Queensland
(continued on next page)

13
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

Table A.1 (continued )

Reference Country Power plant Fly ash SiO2 Al2O3 Fe2O3 CaO MgO K2O Na2O SO3 LOI
type

Rickard et al. (2011) Collie power station in Western Class F 51.38 26.9 13.2 1.74 1.41 0.9 0.41 1.15
Australia,
Eraring power station in New Class F 65.47 23 4.03 1.59 0.51 1.68 0.56 1.37
South Wales
Tarong power station in Class F 73.68 22.4 0.64 0.08 0.17 0.53 0.09 0.79
Queensland
Shill et al. (2020) Eraring thermal power plant Class F 62.19 27.15 3.23 1.97 0.4 0.89 0.3 0.07 1.75
Xie and Kayali (2014) Eraring thermal power station Class F 59.6 29.1 3.3 0.4 0.48 0.28 0.2

Nasvi et al. (2014) India Kolaghat Thermal Power Station Class F 56.01 29.8 3.58 2.36 0.3 0.73 0.61 0.4
Nath et al. (2014) Tata Power, Jojobera plant, Class F 52.6 26.55 5.29 5.1 1.76 1.12 0.61 3.1
Jamshedpur

Hardjito et al. (2008) Malaysia Sejingkat Power Plant in Kuching, Class F 59.9 24.7 6.3 2 1.9 2.9 0.3 0.1 0.3
Sarawak
Al Bakri Abdullah et al. Sultan Abdul Aziz power station Class F 52.11 23.59 7.39 2.61 0.78 0.8 0.42 0.49
(2012)
Abdulkareem and Ramli Manjung Power Station Class C 26.4 9.25 30.13 2.16 0.27 2.58 1.3 3.02
(2015)
Abdulkareem et al. Sultan Abdul Aziz Power Station Class F 52.11 23.59 7.39 2.61 0.78 0.8 0.42 0.49 5.59
(2012)
Al Bakria et al. (2011) Manjung power station Class F 52.11 23.59 7.39 2.61 0.78 0.8 0.42 0.49

Nikolić et al. (2015) Sebia FA Morava, TPP Morava, Svilajnac. Class F 55.23 21.43 7.42 7.94 2.61 1.35 0.64 0.81 1.66
FA Kolubara, TPP Kolubara Class F 62.13 17.2 5.95 5.67 2 1.04 0.58 0.67 2.88
FA Kostolac B1, TPP Kostolac Class F 46.85 23.2 12.14 8.26 2.77 0.81 0.4 1.48 3.44

Álvarez-Ayuso et al. Spain Spanish power plants Class F 54.1 23.3 8.5 3.5 2 3.2 0.9 0.4 2
(2008) Spanish power plants Class F 43.4 25.9 19 5.3 1.2 1.2 0.1 0.9 1.5
Spanish power plants Class F 58.1 22.7 6.1 3.5 1.8 1.6 0.6 0.2 3.5
Spanish power plants Class F 51.3 25.5 6.9 2.9 1.8 3.6 0.7 0.5 5
Kovalchuk et al. (2007) Spanish power plant Class F 54.42 26.42 7.01 3.21 1.79 3.02 0.59 0.01 2.19

Palomo et al. (1999) United Pennsylvania Power and Light Class F 53.2 26 7.95 3.57 0.97 2.59 0.29 2.22
States Co.‘s Montour County power plant

Panias et al. (2007) Greece Greek Public Power Corporation S. Class F 48.95 18.61 7.99 10.91 2.76 1.73 0.8 4.11
A

Lăzărescu et al. (2017) Romania Mintia power plant Class F 53.61 26.16 7.58 2.42 1.49 2.6 0.59 0.26 3.57

Zhou et al. (2016) China Shenhua Junggar Energy Class F 52.4 18.09 0.42 0.33 0.02 0.19 0.03 20.59
Corporation in Junggar
Thokchom et al. (2012) plant in Shanxi Province Class F 52.79 20.95 7.76 6.95 3.42 0.51 0.09

Helmy (2016) Egypt Geos, Cairo Class F 55.819 28.112 7.488 2.71 0.846 1.515 0.215 0.344

Cho et al. (2019) South Class F 55.4 22.2 6.84 5.12 1.84 1.55 1.26 0.71 3.7
Korea Class F 59.1 20 6.22 3.65 1.71 1.62 0.99 0.36 4.43
Class F 62.6 20 7.13 2.83 1.2 1.2 0.65 0.32 2.62
Class F 54 22 6.43 4.76 1.48 1.21 1.34 0.5 6.7
Class F 62.4 17.7 6.89 4.15 1.55 0.97 1.24 0.34 2.53
Class F 62.3 19 6.3 3.42 1.49 1.62 0.75 0.37 3.55
Class F 57.7 21.1 6.39 4.26 1.8 1.67 1.06 0.52 3.91
Class F 53 20.7 6.94 6.17 2.31 1.21 2.3 0.51 4.93
Class F 56.6 20.9 8.09 4.66 1.82 1.2 1.27 0.72 2.61
Class F 58.3 20.8 6.83 3.44 1.39 1.15 0.94 0.35 5.16
Class F 60 19.8 6.41 3.14 1.32 1.18 0.9 0.49 4.76
Class F 61.9 18.7 6.15 3.28 1.33 1.19 0.81 0.48 4.43
Class F 62.3 20.2 6.66 2.54 1.15 1.18 0.64 0.42 3.28
Class F 52.2 22.4 7.57 5.22 1.93 1.12 1.46 0.82 5.16
Class F 57.5 20.5 7.16 5.07 1.72 1.43 0.78 0.71 2.75
Class F 52.4 23 8.85 5.51 2.06 0.79 1.26 0.47 2.94
Moon et al. (2016) Class F 53.04 21.38 5.77 3.1 1.41 1.45 0.61 0.2
Class F 39.87 18.2 8.07 6.49 1.67 1.17 1.56 0.4
Class F 39.62 14.08 6.07 4.94 1.6 1.17 1.12 0.37
Class F 41.53 15.16 6.87 5.8 2.13 1.08 0.65 0.43
Class F 39.1 16.06 6.82 5.45 1.8 1.73 1.06 0.43
Class F 49.91 16.73 6.01 3.9 1.46 1.14 0.04 0.24

Kiattikomol et al. Thailand Class F 46.25 26.43 10.71 7.61 2.21 3.07 1.11 1.85 0.23
(2001) Class F 45.02 36.21 4.09 3.64 0.54 0.31 0.44 0.48 5.32
Class F 43.92 36.62 3.97 3.05 0.55 0.44 0.38 0.64 7.52
Class F 47.39 22.73 6.29 8.36 2.64 2.95 0.63 3.38 3.12
Class F 49.04 37.91 2.75 1.03 0.39 0.52 0.38 0.18 4.7
Somna et al. (2011) Mae Moh power plant Class C 31.2 18.9 16.5 20.8 1.86 2.8 1.53 4.1 1.8
Rattanasak and Class C 39.5 19.5 14.1 17.3 1.3 2.9 1.3 2.6 0.8
Chindaprasirt (2009)
Chindaprasirt et al. Class C 38.7 20.8 15.3 16.6 1.3 2.1 1.3 2.6 0.8
(2009)
Chindaprasirt et al. Class C 38.7 20.8 15.3 16.6 1.5 2.7 1.2 2.6 0.1
(2007)

14
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

Annex B. Table B.2

Model parameters used for the ANN model.

Parameter Value

Number of input neurons 5

Number of output neurons 1
Number of hidden layers 1
Number of hidden neurons 8
Number of training epochs 602
Number of total datasets 86
Number of training datasets 71
Number of validation datasets 15
Network type Feed-forward back-propagation
Transfer function TANSIG
Error (Performance) function MSE
Training function TRAINLM
Adaptation learning function LEARNGDM

Fig. B.1. Network architecture of the ANN model.

References Backe, K., Lile, O., Lyomov, S., Elvebakk, H., Skalle, P., 1999. Characterizing curing-
cement slurries by permeability, tensile strength, and shrinkage. SPE Drill. Complet.
14 (3), 162–167.
Abdulkareem, O.A., Al Bakri Abdullah, M.M., Kamarudin, H., Khairul Nizar, I., 2012. The
Barlet-Gouédard, V., Rimmelé, G., Porcherie, O., Quisel, N., Desroches, J., 2009.
influence of curing periods on the compressive strength of fly ash-based geopolymer
A solution against well cement degradation under CO2 geological storage
at different aging times 479, 512–516.
environment. Int. J. Greenh. Gas Control 3 (2), 206–216.
Abdulkareem, O.A., Ramli, M., 2015. Optimization of alkaline activator mixing and
Bertier, P., Swennen, R., Laenen, B., Lagrou, D., Dreesen, R., 2006. Experimental
curing conditions for a fly ash-based geopolymer paste system. Mod. Appl. Sci. 9
identification of CO2–water–rock interactions caused by sequestration of CO2 in
(12), 61.
Westphalian and Buntsandstein sandstones of the Campine Basin (NE-Belgium).
Ahdaya, M., Imqam, A., 2019. Fly ash Class C based geopolymer for oil well cementing.
J. Geochem. Explor. 89 (1–3), 10–14.
J. Petrol. Sci. Eng. 179, 750–757.
Brandvoll, Ø., Regnault, O., Munz, I., Iden, I., Johansen, H., 2009. Fluid–solid
Al Bakri Abdullah, M.M., Kamarudin, H., Abdulkareem, O.A., Ghazali, C.M.R., Rafiza, A.,
interactions related to subsurface storage of CO2 Experimental tests of well cement.
Norazian, M., 2012. Optimization of alkaline activator/fly ash ratio on the
Energy Proc. 1 (1), 3367–3374.
compressive strength of manufacturing fly ash-based geopolymer 110, 734–739.
Chindaprasirt, P., Chareerat, T., Sirivivatnanon, V., 2007. Workability and strength of
Al Bakria, A.M., Kamarudin, H., BinHussain, M., Nizar, I.K., Zarina, Y., Rafiza, A., 2011.
coarse high calcium fly ash geopolymer. Cement Concr. Compos. 29 (3), 224–229.
The effect of curing temperature on physical and chemical properties of
Chindaprasirt, P., Jaturapitakkul, C., Chalee, W., Rattanasak, U., 2009. Comparative
geopolymers. Phys. Procedia 22, 286–291.
study on the characteristics of fly ash and bottom ash geopolymers. Waste Manag. 29
Álvarez-Ayuso, E., Querol, X., Plana, F., Alastuey, A., Moreno, N., Izquierdo, M., Font, O.,
(2), 539–543.
Moreno, T., Diez, S., Vázquez, E., 2008. Environmental, physical and structural
Cho, Y.K., Jung, S.H., Choi, Y.C., 2019. Effects of chemical composition of fly ash on
characterisation of geopolymer matrixes synthesised from coal (co-) combustion fly
compressive strength of fly ash cement mortar. Construct. Build. Mater. 204,
ashes. J. Hazard Mater. 154 (1–3), 175–183.
255–264.
Condor, J., Asghari, K., 2009. Experimental study of stability and integrity of cement in
wellbores used for CO2 storage. Energy Proc. 1 (1), 3633–3640.

15
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

Davidovits, J., 1994. Properties of Geopolymer Cements, vol. 1, pp. 131–149. Mishra, A., Choudhary, D., Jain, N., Kumar, M., Sharda, N., Dutt, D., 2008. Effect of
Davidovits, J., 2005. Geopolymer, Green Chemistry and Sustainable Development concentration of alkaline liquid and curing time on strength and water absorption of
Solutions: Proceedings of the World Congress Geopolymer 2005. Geopolymer geopolymer concrete. ARPN J. Eng. Appl. Sci. 3 (1), 14–18.
Institute. Moon, G.D., Oh, S., Choi, Y.C., 2016. Effects of the physicochemical properties of fly ash
Di Lullo, G., Rae, P., 2000. Cements for Long Term Isolation-Design Optimization by on the compressive strength of high-volume fly ash mortar. Construct. Build. Mater.
Computer Modelling and Prediction. IADC/SPE Asia Pacific Drilling Technology. 124, 1072–1080.
Duguid, A., Scherer, G.W., 2010. Degradation of oilwell cement due to exposure to Nasvi, Gamage, R.P., Jay, S., 2012a. Geopolymer as well cement and the variation of its
carbonated brine. Int. J. Greenh. Gas Control 4 (3), 546–560. mechanical behavior with curing temperature. Greenhouse Gases: Sci. Technol. 2
Fulekar, M., Dave, J., 1986. Disposal of fly ash—an environmental problem. Int. J. (1), 46–58.
Environ. Stud. 26 (3), 191–215. Nasvi, M.M.C., Gamage, R.P., Jay, S., 2012b. Geopolymer as well cement and the
Gawin, D., Majorana, C., Schrefler, B., 1999. Numerical analysis of hygro-thermal variation of its mechanical behavior with curing temperature. Greenhouse Gases:
behaviour and damage of concrete at high temperature. Mech. Cohesive-Frict. Sci. Technol. 2 (1), 46–58. https://doi.org/10.1002/ghg.39.
Mater.: An International Journal on Experiments, Modelling and Computation of Nasvi, Ranjith, P., Sanjayan, J., 2013. The permeability of geopolymer at down-hole
Materials and Structures 4 (1), 37–74. stress conditions: application for carbon dioxide sequestration wells. Appl. Energy
Goodwin, K., 1997. Oilwell/gaswell cement-sheath evaluation. J. Petrol. Technol. 49 102, 1391–1398.
(12), 1339–1343. Nasvi, Ranjith, P., Sanjayan, J., 2014. Effect of different mix compositions on apparent
Gunasekara, M.C., Law, D., Setunge, S., 2014. Effect of Composition of Fly Ash on carbon dioxide (CO2) permeability of geopolymer: suitability as well cement for CO2
Compressive Strength of Fly Ash Based Geopolymer Mortar, pp. 113–118. sequestration wells. Appl. Energy 114, 939–948.
Guo, X., Shi, H., Dick, W.A., 2010. Compressive strength and microstructural Nasvi, Ranjith, P., Sanjayan, J., 2015. A numerical study of triaxial mechanical
characteristics of class C fly ash geopolymer. Cement Concr. Compos. 32 (2), behaviour of geopolymer at different curing temperatures: an application for
142–147. geological sequestration wells. J. Nat. Gas Sci. Eng. 26, 1148–1160.
Hardjito, D., Cheak, C.C., Ing, C.L., 2008. Strength and setting times of low calcium fly Nath, S., Mukherjee, S., Maitra, S., Kumar, S., 2014. Ambient and elevated temperature
ash-based geopolymer mortar. Mod. Appl. Sci. 2 (4), 3–11. geopolymerization behaviour of class F fly ash. Trans. Indian Ceram. Soc. 73 (2),
Hardjito, D., Rangan, B.V., 2005. Development and Properties of Low-Calcium Fly Ash- 126–132.
Based Geopolymer Concrete. Newell, R., Raimi, D., Aldana, G., 2019. Global energy outlook 2019: the next generation
Hardjito, D., Wallah, S. áE., Sumajouw, D. áM.J., Vijaya Rangan, B., 2004. On the of energy. Resources for the Future 1, 8–19.
development of Fly ash-based Geopolymer concrete. ACI Mater. J. 101 (6), 467–472. Nikolić, V., Komljenović, M., Baščarević, Z., Marjanović, N., Miladinović, Z., Petrović, R.,
Helmy, A.I.I., 2016. Intermittent curing of fly ash geopolymer mortar. Construct. Build. 2015. The influence of fly ash characteristics and reaction conditions on strength and
Mater. 110, 54–64. structure of geopolymers. Construct. Build. Mater. 94, 361–370.
Hewayde, E., Nehdi, M., Allouche, E., Nakhla, G., 2006. Effect of geopolymer cement on Olivia, M., Nikraz, H., 2011. Durability of Fly Ash Geopolymer Concrete in a Seawater
microstructure, compressive strength and sulphuric acid resistance of concrete. Mag. Environment. Proceedings of the CONCRETE 2011 Conference.
Concr. Res. 58 (5), 321–331. Palomo, A., Grutzeck, M., Blanco, M., 1999. Alkali-activated fly ashes: a cement for the
Huber, M.L., Lemmon, E.W., Diky, V., Smith, B.L., Bruno, T.J., 2008. Chemically future. Cement Concr. Res. 29 (8), 1323–1329.
authentic surrogate mixture model for the thermophysical properties of a coal- Panias, D., Giannopoulou, I.P., Perraki, T., 2007. Effect of synthesis parameters on the
derived liquid fuel. Energy Fuel. 22 (5), 3249–3257. mechanical properties of fly ash-based geopolymers. Colloids Surf. A Physicochem.
Huerta, N.J., Bryant, S.L., Strazisar, B.R., Kutchko, B.G., Conrad, L.C., 2009. The Eng. Asp. 301 (1–3), 246–254.
influence of confining stress and chemical alteration on conductive pathways within Pedersen, R., Scheie, A., Johnson, C.R., Hoyos, J.C., Therond, E., Khatri, D.K., 2006.
wellbore cement. Energy Proc. 1 (1), 3571–3578. Cementing of an Offshore Disposal Well Using a Novel Sealant that Withstands
Ionescu, B.A., Lӑzӑrescu, A., 2020. A Review Regarding the Use of Natural and Industrial Pressure and Temperature Cycles. IADC/SPE Drilling Conference.
by-products in the Production of Geopolymer Binders 877 (1), 012033. Pratt, A., Talman, S., Zhang, M., Thibeau, Y., 2009. Characterization of Portland cement
Ji, X., Zhu, C., 2015. CO2 storage in deep saline aquifers. In: Novel Materials for Carbon reacted with supercritical CO2 4–7.
Dioxide Mitigation Technology. Elsevier, pp. 299–332. Provis, J.L., Yong, C.Z., Duxson, P., van Deventer, J.S., 2009. Correlating mechanical and
Kaldi, J.G., Gibson-Poole, C.M., Payenberg, T.H., 2009. Geological Input to Selection and thermal properties of sodium silicate-fly ash geopolymers. Colloids Surf. A
Evaluation of CO2 Geosequestration Sites. Physicochem. Eng. Asp. 336 (1–3), 57–63.
Kannangara, T., Guerrieri, M., Fragomeni, S., Joseph, P., 2021. Effects of initial surface Raijiwala, D., Patil, H., Kundan, I., 2012. Effect of alkaline activator on the strength and
evaporation on the performance of fly ash-based geopolymer paste at elevated durability of geopolymer concrete. Journal of Engineering Research and Studies 3
temperatures. Appl. Sci. 12 (1), 364. (1), 18–21.
Karakurt, I., Aydin, G., 2023. Development of regression models to forecast the CO2 Rangan, B.V., 2008. Fly ash-based geopolymer concrete.Research Report GC 4.
emissions from fossil fuels in the BRICS and MINT countries. Energy 263, 125650. Ranjith, P., Perera, M., 2011. A new triaxial apparatus to study the mechanical and fluid
Khalifeh, M., Saasen, A., Vralstad, T., Hodne, H., 2014. Potential utilization of class C fly flow aspects of carbon dioxide sequestration in geological formations. Fuel 90 (8),
ash-based geopolymer in oil well cementing operations. Cement Concr. Compos. 53, 2751–2759.
10–17. Rattanasak, U., Chindaprasirt, P., 2009. Influence of NaOH solution on the synthesis of
Kiattikomol, K., Jaturapitakkul, C., Songpiriyakij, S., Chutubtim, S., 2001. A study of fly ash geopolymer. Miner. Eng. 22 (12), 1073–1078.
ground coarse fly ashes with different finenesses from various sources as pozzolanic Rickard, W.D., Williams, R., Temuujin, J., Van Riessen, A., 2011. Assessing the suitability
materials. Cement Concr. Compos. 23 (4–5), 335–343. of three Australian fly ashes as an aluminosilicate source for geopolymers in high
Kong, D.L., Sanjayan, J.G., Sagoe-Crentsil, K., 2007. Comparative performance of temperature applications. Mater. Sci. Eng., A 528 (9), 3390–3397.
geopolymers made with metakaolin and fly ash after exposure to elevated Ridha, S., Hamid, A.I.A., Setiawan, R.A., Ibrahim, M.A., Shahari, A.R., 2018.
temperatures. Cement Concr. Res. 37 (12), 1583–1589. Microstructure behavior of fly ash-based geopolymer cement exposed to acidic
Kovalchuk, G., Fernández-Jiménez, A., Palomo, A., 2007. Alkali-activated fly ash: effect environment for oil well cementing. Arabian J. Sci. Eng. 43 (11), 6413–6428.
of thermal curing conditions on mechanical and microstructural development–Part Ridha, S., Setiawan, R.A., Pramana, A.A., Abdurrahman, M., 2020. Impact of wet
II. Fuel 86 (3), 315–322. supercritical CO2 injection on fly ash geopolymer cement under elevated
Krilov, Z., Loncaric, B., Miksa, Z., 2000. Investigation of a Long-Term Cement temperatures for well cement applications. J. Pet. Explor. Prod. Technol. 10 (2),
Deterioration under a High-Temperature, Sour Gas Downhole Environment. SPE 243–247.
International Symposium on Formation Damage Control. Robins, N., Milodowski, A., 1986. Borehole cements and the downhole environment—a
Kutchko, B.G., Kim, A.G., 2006. Fly ash characterization by SEM–EDS. Fuel 85 (17–18), review. Q. J. Eng. Geol. Hydrogeol. 19 (2), 175–181.
2537–2544. Sagoe–Crentsil, K., Brown, T., Yan, S.Q., 2010. Medium to long term engineering
Kutchko, B.G., Strazisar, B.R., Dzombak, D.A., Lowry, G.V., Thaulow, N., 2007. properties and performance of high-strength geopolymers for structural applications
Degradation of well cement by CO2 under geologic sequestration conditions. 69, 135–142.
Environ. Sci. Technol. 41 (13), 4787–4792. Sakulich, A.R., Anderson, E., Schauer, C., Barsoum, M.W., 2009. Mechanical and
Labibzadeh, M., 2010. Assessment of the early age tensile strength of the oilfield class g microstructural characterization of an alkali-activated slag/limestone fine aggregate
cement under effects of the changes in down-hole pressure and temperature. Trends concrete. Construct. Build. Mater. 23 (8), 2951–2957.
Appl. Sci. Res. 5 (3), 165–176. Sampath, K.H.S.M., Ranjith, P.G., Perera, M.S.A., 2020. A Comprehensive Review of
Laudet, J.-B., Garnier, A., Neuville, N., Le Guen, Y., Fourmaintraux, D., Rafai, N., Structural Alterations in CO2-Interacted Coal: Insights into CO2 Sequestration in
Burlion, N., Shao, J.-F., 2011. The behavior of oil well cement at downhole CO2 Coal. Energy Fuel. 34 (11), 13369–13383.
storage conditions: static and dynamic laboratory experiments. Energy Proc. 4, Santra, A.K., Reddy, B.R., Liang, F., Fitzgerald, R., 2009. Reaction of CO2 With Portland
5251–5258. Cement at Downhole Conditions and the Role of Pozzolanic Supplements. In: SPE
Lăzărescu, A., Szilagyi, H., Baeră, C., Ioani, A., 2017. The effect of alkaline activator ratio International Symposium on Oilfield Chemistry, Woodlands, TX, Apr., pp. 20–22.
on the compressive strength of fly ash-based geopolymer paste 209 (1), 012064. SPE Paper No. SPE-121103-MS.
Lécolier, E., Rivereau, A., Le Saoût, G., Audibert-Hayet, A., 2007. Durability of hardened Sarker, P.K., 2011. Bond strength of reinforcing steel embedded in fly ash-based
portland cement paste used for oilwell cementing. Oil & Gas Science and geopolymer concrete. Mater. Struct. 44 (5), 1021–1030.
Technology-Revue de l’IFP 62 (3), 335–345. Sauki, A., Irawan, S., 2010. Effects of pressure and temperature on well cement
Mavroudis, D., 2001. Downhole Environmental Risks Associated with Drilling and Well degradation by supercritical CO2. International Journal of Engineering &
Completion Practices in the Cooper/eromanga Basins, vol. 9. Report Book. Technology IJET-IJENS 1 (4), 53–61.
Mermerdaş, K., Algın, Z., Ekmen, Ş., 2020. Experimental assessment and optimization of Shahriar, A., 2011. Investigation on Rheology of Oil Well Cement Slurries. Electronic
mix parameters of fly ash-based lightweight geopolymer mortar with respect to Thesis and Dissertation Repository, 113.
shrinkage and strength. J. Build. Eng. 31, 101351.

16
H.B.S. Sathsarani et al. Gas Science and Engineering 113 (2023) 204974

Shill, S.K., Al-Deen, S., Ashraf, M., Hutchison, W., 2020. Resistance of fly ash based Uehara, M., 2010. New concrete with low environmental load using the geopolymer
geopolymer mortar to both chemicals and high thermal cycles simultaneously. method. Quarterly Report of RTRI 51 (1), 1–7.
Construct. Build. Mater. 239, 117886. Van den Heede, P., Gruyaert, E., De Belie, N., 2010. Transport properties of high-volume
Siriwardane, H., Haljasmaa, I., McLendon, R., Irdi, G., Soong, Y., Bromhal, G., 2009. fly ash concrete: capillary water sorption, water sorption under vacuum and gas
Influence of carbon dioxide on coal permeability determined by pressure transient permeability. Cement Concr. Compos. 32 (10), 749–756.
methods. Int. J. Coal Geol. 77 (1–2), 109–118. Xie, J., Kayali, O., 2014. Effect of initial water content and curing moisture conditions on
Sofi, M., Van Deventer, J., Mendis, P., Lukey, G., 2007. Engineering properties of the development of fly ash-based geopolymers in heat and ambient temperature.
inorganic polymer concretes (IPCs). Cement Concr. Res. 37 (2), 251–257. Construct. Build. Mater. 67, 20–28.
Somna, K., Jaturapitakkul, C., Kajitvichyanukul, P., Chindaprasirt, P., 2011. NaOH- Yang, Z., Ha, N., Jang, M., Hwang, K.H., 2009. Geopolymer concrete fabricated by waste
activated ground fly ash geopolymer cured at ambient temperature. Fuel 90 (6), concrete sludge with silica fume 620, 791–794.
2118–2124. Zain, M.F.M., Mahmud, H., Ilham, A., Faizal, M., 2002. Prediction of splitting tensile
Tempest, B., Sanusi, O., Gergely, J., Ogunro, V., Weggel, D., 2009. Compressive strength strength of high-performance concrete. Cement Concr. Res. 32 (8), 1251–1258.
and embodied energy optimization of fly ash based geopolymer concrete 1–17. Zhang, M., Talman, S., 2014. Experimental study of well cement carbonation under
Thakur, R.N., Ghosh, S., 2009. Effect of mix composition on compressive strength and geological storage conditions. Energy Proc. 63, 5813–5821.
microstructure of fly ash based geopolymer composites. ARPN J. Eng. Appl. Sci. 4 Zhou, W., Yan, C., Duan, P., Liu, Y., Zhang, Z., Qiu, X., Li, D., 2016. A comparative study
(4), 68–74. of high- and low-Al 2 O 3 fly ash based-geopolymers: the role of mix proportion
Thokchom, S., Mandal, K.K., Ghosh, S., 2012. Effect of Si/Al ratio on performance of fly factors and curing temperature. Mater. Des. 95, 63–74. https://doi.org/10.1016/j.
ash geopolymers at elevated temperature. Arabian J. Sci. Eng. 37 (4), 977–989. matdes.2016.01.084.

17