See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/236330161

Comparison of mechanical behaviors of geopolymer and class G cement as

well cement at different curing temperatures for geological sequestration of
carbon dioxide

Conference Paper · June 2012

CITATIONS READS

25 1,417

3 authors:

Mohamed Nasvi ranjith Pathegama

CINEC maritime campus Monash University (Australia)
1 PUBLICATION 25 CITATIONS 37 PUBLICATIONS 936 CITATIONS

SEE PROFILE SEE PROFILE

Jay G. Sanjayan
Swinburne University of Technology
289 PUBLICATIONS 23,115 CITATIONS

SEE PROFILE

All content following this page was uploaded by Mohamed Nasvi on 27 May 2014.

The user has requested enhancement of the downloaded file.

ARMA 12-232

Comparison of mechanical behaviors of geopolymer and class G

cement as well cement at different curing temperatures for geological
sequestration of carbon dioxide
Nasvi, M.C.M.
Monash University, Melbourne, VIC, Australia
Ranjith, P.G.
Monash Uniersity, Melbourne, VIC, Australia.
Sanjayan, J.
Swinburne University of technology, Melbourne, VIC, Australia.

Copyright 2012 ARMA, American Rock Mechanics Association

This paper was prepared for presentation at the 46th US Rock Mechanics / Geomechanics Symposium held in Chicago, IL, USA, 24-27 June
2012.
This paper was selected for presentation at the symposium by an ARMA Technical Program Committee based on a technical and critical review of
the paper by a minimum of two technical reviewers. The material, as presented, does not necessarily reflect any position of ARMA, its officers, or
members. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of ARMA
is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The
abstract must contain conspicuous acknowledgement of where and by whom the paper was presented.

ABSTRACT: Carbon dioxide (CO2) sequestration into deep underground reservoirs such as saline aquifers, oil reservoirs and coal
seams have found out to be one of the best practical solutions to reduce significant amount of greenhouse gases from the
environment. The success of any large scale CO2 sequestration project depends on many factors, and two of the important factors
are (a) the stability of injection well and (b) well cement. To date, OPC based cement has been used as well cement, and it has
been found that it is unstable in CO2 environments as it undergoes degradation, strength reduction, and shrinkage. Therefore, a
comprehensive experimental study has been undertaken to investigate the suitability of geopolymer as well cement and the
mechanical behavior of geopolymer and class G cement was compared under different down-hole temperatures. When the uni-
axial compressive strength (UCS) of geopolymer and G cement was compared, it was found that geopolymer possess higher UCS
values at elevated temperatures (above 50 ºC) and G cement possesses the highest values at ambient conditions. The peak strength
of both geopolymer and class G cement was observed at curing temperatures of 50-60 ºC. In addition, acoustic emission (AE) test
data revealed that the crack propagation stress thresholds of class G cement are higher at ambient conditions, whereas geopolymer
possesses highest values at elevated temperatures. The photogrammetric results of strain measurement show that geopolymer
undergoes shear failure at lower curing temperatures, whereas the failure was splitting at elevated temperatures. In addition, the
type of failure of class G cement was shear failure for all the curing temperatures and the ultimate failure strain did not vary much
with the curing temperature.
Key words: carbon dioxide, geosequestration, geopolymer, greenhouse gases, well cement
solutions, the best solution to reduce greenhouse gases in
1. INTRODUCTION the near term would be to inject carbon dioxide (CO2)
Greenhouse gas emissions in the globe are rapidly into deep underground reservoirs such as coal seams,
increasing due to the excessive usage of fossil fuel saline aquifers, and depleted oil and gas reservoirs [1, 2]
including coal. Therefore, it is estimated that the global
temperature will increase significantly in the next 50- When CO2 is injected through injection wells it should
100 years. The Intergovernmental Panel on Climate not leach through the surrounding formation as it may
Change (IPCC) estimates that under a “business-as- cause many problems including contamination of ground
usual” energy scenario, global CO2 emissions will reach water resources and have lethal effect on plant and sub
approximately 77Gt/ yr. by 2100, and the average soil animals [3]. Therefore, the integrity of the wells
atmospheric CO2 concentration will reach approximately should be made using a material which is durable, anti-
750 parts per million by volume (ppmv). Greenhouse corrosive, chemically inert, adaptive to pressure
gas emission can be mitigated through improving the variations, and less permeable, in order to maintain well
energy conversion efficiency of fossil fuels, shifting integrity, and to avoid any leakage of CO2. To date, in
energy production to low carbon sources, enhancing almost all sequestration projects, ordinary Portland
uptake by terrestrial and marine biomass, and capturing cement (OPC) based well cement has been used with
and storing CO2 deep underground. Of the above additives and it has been found that Portland cement is

1
unstable in CO2 rich environments as it undergoes in the surface, and the target has to be prepared with
strength reduction, degradation in acidic environment, high contrasting pattern. The Shimadzu 300 kN UCS
shrinkage due to high pressure and temperature and less testing machine was employed for the test, and the
durable in acidic environment [4, 5]. According to an loading rate used was 0.2 mm/min based on strain rate
EOR survey, out of the total of all CO2 EOR wells (Fig. 1b).
(16,348) only 0.15% uses non-Portland cement CO2
zones [6]. On the other hand, it has been found that Table 1: The composition of G cement
geopolymer possesses higher strength, excellent acid API
resistant characteristics, very little shrinkage compared Typical
Composition specification
to OPC, no harm by alkali aggregate reaction (AAR) and values (%)
10A/(max)
has higher pumpability compared to OPC [7, 8]. In our Loss on Ignition(LOI) 3 0.7
previous work [9], we analyzed the mechanical behavior
of geopolymer at different curing temperatures. In soluble residue 0.75 0.4
However, this paper compare the mechanical behavior of Sulphuric Anhydride 3 1.9
geopolymer and Class G cement (OPC) under different Magnesia (MgO) 6 0.69
downhole temperatures as typical well cement is
exposed to a range of temperatures with a geothermal Na2O Equivalent 0.75 0.3
gradient of 30 ºC/km [1]. The curing temperature ranges Tri-calcium Aluminate
of 20-80 ºC were selected for the study, and acoustic (C3A)+Tetra Calcium Alumino 24 20
emission (AE) and ARAMIS image capture testing were Ferrite (C4AF)
employed during the testing to study the macro-micro Tri-calcium Aluminate (C3A) 3 2
behavior of material deformations under various loading
Tri-calcium Silicate (C3S) 65 58
conditions.

Table 2: The composition of fly ash used for geopolymer

2. EXPERIMENTAL METHODOLOGY Composition as Typical values
Oxide (%)
The geopolymer and class G cement samples with a
diameter of 50 mm and 100 mm height were prepared. Al2O3 30.5
The geopolymer paste was prepared according to the
SiO2 48.3
mix design by assuming a density of 1800 kg/m3. The
geopolymer was prepared using ASTM class F (low Fe2O3 2.8
calcium) fly ash, and alkaline liquid used was the
K2O 12.1
combination of 10M NaOH and Na2SiO3. The NaOH
was obtained in pellet form having 68% of pellets and MgO 0.4
32% of water, and was mixed with Na2SiO3 with Na2O 0.2
Na2SiO3/NaOH ratio of 2.0 to produce the alkaline
liquid. The ratio of alkaline liquid (AL) to fly ash (FA) SO3 0.3
used was 0.4. In addition, class G cement paste was Loss on ignition
prepared with water/cement ratio of 0.4. The 1.7
(LOI)
composition of fly ash and G cement used are listed in
Tables 1 and 2 respectively. Once the samples were
prepared, they were oven cured at different curing
temperatures (30-80 ºC) for a period of 24 hours except
samples cured at room temperature. Then all the
samples were kept at ambient condition for another 48
hours prior to testing. The rest time of 48 hours was
selected to facilitate the sample preparation prior to
testing as geopolymer will not develop substantial
strength after 24 hours [10]. Then samples cured at room
temperature (23 ºC) were kept for 72 hours. The end
faces of the samples were machine ground and polished
using mechanical grinder to ensure purely axial loading
during the test. In addition, surface of the samples were
painted with black: white ratio of 1:1 for ARAMIS
image capture (See Fig. 1a) as it captures the black dots
Fig. 1a: The painted sample for ARAMIS image capture

2
 This is due to poor rate of geopolymerization at lower
curing temperatures. It can be seen that both geopolymer
and class G cement gains strength with the curing
temperature up to their optimum curing temperatures.
The optimum curing temperature for both class G
cement and geopolymer for higher strength lies between
55-60 ºC. The strength increment of geopolymer up to
peak strength is 92%, whereas that of class G cement is
46%. Geopolymer gains strength with curing
temperature as Si and Al from the source material
readily dissolves from the source material with the
increase in curing temperature. However, it must be
noted that there is a possibility of breaking up inter
granular structure of geopolymer at very high curing
temperatures (> 100 ºC) and hence it could lead to
strength reduction. In addition, class G cement possesses
higher UCS values at lower curing temperatures (<36
Fig. 1b: The UCS experimental set-up ºC), whereas geopolymer possesses the highest UCS
values at elevated temperatures (> 36 ºC). It is obvious
3. RESULTS AND DISCUSSIONS that geopolymer is unstable at temperatures between 0-
36 ºC compared to class G cement as G cement
3.1 The variation of mechanical strength with possesses the highest strength. Therefore, geopolymer is
curing temperature not a good solution in cold regions where the
temperature is expected to be lower than 30 ºC. As the
The variation of uni-axial compressive strength (UCS)
curing temperature is increased, it can be noticed that
of geopolymer and class G cement with different curing
both class G and geopolymer cement experiences
temperatures is shown in Fig. 2. Even though
strength reductions. However, the reduction rate of
geopolymer was tested at room temperature, it did not
geopolymer is less than that of class G cement.
show considerable strength development and the sample
cured at room temperature could not be tested.

Fig. 2: The variation of UCS of geopolymer and class G cement at different curing temperatures

3
The geopolymer experiences 6% of strength reduction
from the peak strength, whereas class G cement
experiences 48% of strength reduction from the peak
strength. According to basic cement chemistry, OPC
cement experiences strength reduction with the increase
in curing temperature. This is due to the fact that, the
loss of silica at elevated temperature causes strength
reduction for OPC based cement. Therefore, it can be
concluded that class G cement will be a good option at
depth of below 1 km (assuming a geothermal gradient of
30 ºC/km [1]) , while geopolymer is best suited at depth
of above 1 km, where the temperatures is expected to be
above 40 ºC.
The stress-strain behaviors of class G cement
geopolymer and both class G cement and geopolymer
Fig. 3b: The stress-strain curve for geopolymer cement cured
together are shown in Fig. 3a, 3b and 3c respectively. It at different curing temperatures; GP 30-Geopolymer cement
can be noticed that class G cement is unstable at 80 ºC, cured at 30 ºC.
where the stress and strain both reduces significantly
compared to lower curing temperatures. Apart from 80 The overall comparison of stress-strain plots of
ºC cured sample, all the other samples undergo same geopolymer and G cement can be made from Fig 3c. It is
range of failure strain, whereas the failure stresses were very obvious that failure strain and failure stresses are
different depending on the curing temperature. When the generally higher for geopolymer compared to class G
stress-strain plot of geopolymer is considered, it can be cement, especially above 30 ºC. This is due to the fact
seen that 30 ºC cured sample experienced a higher that geopolymer gain strength with the temperature and
failure strain, whereas the peak stress was very low even beyond the optimum temperature the rate of
compared to other samples. This is due the high ductile strength reduction is lower compared to class G cement.
nature of the sample caused by rate of poor On the other hand, class G cement loses the strength
geopolymerisation. As the curing temperature is higher with curing temperature producing a weak matrix.
the final geopolymer matrix will be highly brittle and the Therefore, geopolymer will be a better option than G
peak stress will be higher. Except 30 ºC cured sample, cement at elevated temperatures to adopt the down-hole
all the other samples undergo same range of failure mechanical stresses applied in to the well.
strain with different ultimate strength values.
The variation of Young’s modulus (E) of geopolymer
and G cement with the curing temperature is shown in
Fig. 4. It can be noticed that at lower curing
temperatures, G cement has the highest E value, whereas
at elevated temperatures geopolymer possesses the
highest E values. This is because of higher UCS values
at elevated temperatures, and high UCS values will give
higher E values [11]. In addition, both geopolymer and
G cement shows reduction in E values towards 80 ºC
and it is due to the reduction in UCS values. It can be
concluded that at lower curing temperatures G cement is
stiffer compared to geopolymer, and at elevated
temperatures geopolymer is stiffer than G cement.

Fig. 3a: The stress-strain curve for class G cement cured at

different curing temperatures; GC 23-class G cement cured at
23 ºC.

4
 Fig. 3c: The stress-strain variations for both class G and geopolymer cement cured at different curing temperatures.

brittle material begins with crack closure followed by an

elastic region, and then crack initiation followed by
stable crack growth, and finally crack damage followed
by unstable crack growth. Stress-strain curve, axial
stiffness-stress, and volumetric stiffness-strain curve are
commonly used conventional techniques to study the
stress thresholds [12]. However, this paper presents
crack propagation thresholds obtained from AET. Upon
the application of loading, pre-existing micro and macro
cracks in the specimen tend to close, and this phase is
known as crack closure (σcc). In Acoustic Emission (AE)
very little or no acoustic activity is observed due to the
crack closure of existing macro and micro cracks with no
release of strain energy. Crack initiation (σci) is defined
Fig. 4: The variation of Young’s modulus (E) of geopolymer as the point at which AE event counts initially increase
and G cement with curing temperature above the background level of events. There will be an
increase in acoustic activity in a linear manner due to the
increase in axial loading. When crack damage is found
3.2 Crack propagation stress thresholds using from AET, it is defined as the point, at which curve
acoustic emission testing (AET) depart from linearity to non-linearity (exponential
When a brittle material is subjected to uni-axial variation).
compression or tension, the materials releases its strain The variation of crack propagation stress thresholds of
energy. Sudden release of stored elastic energy generates geopolymer and G cement cured at 40 ºC are shown in
an elastic stress wave which travels from the origin Fig. 5a and 5brespectively. It can be noticed that, at 40
within the material to a boundary where it is observed as ºC, geopolymer possesses highest values of crack
an acoustic emission (AE) event. There are three crack initiation (36.2 MPa) and crack damage (56.9 MPa)
propagation stress thresholds for any brittle materials stress thresholds compared to class G cement (24.5 and
subjected to an axial loading. Fracture phases of any 42.6 MPa respectively). This is attributed to the fact that

5
geopolymer shows higher strength and stiffness (Fig. 3a- propagation stress thresholds depends on the mechanical
3c). However, the crack closure stress was not identified strength of the material considered.
at 40 ºC for both the specimen, and this is perhaps due to
the low sensitivity of the device used in the experimental Table 3: The comparison of crack propagation stress
study. Most of the acoustic events occur prior to the thresholds of geopolymer and class G cement at different
failure of a sample and lack of acoustic activity in the curing temperatures
initial stages is a concern in the AE method [13]. Crack Crack
Crack Damage
Curing closure Initiation
(MPa)
temperature (MPa) (MPa)
GC GP GC GP GC GP
Ambient(23) na* na* 24.4 10.2 27.8 24.2
40 na* na* 24.5 36.2 42.6 56.9
60 8.1 24.2 36.1 51.3 50.7 84.8
80 9.6 na* 23.4 64.6 27.5 71.2
na*-could not be measured

Fig. 5a: The crack propagation stress thresholds of

geopolymer cured at 40 ºC

Fig. 6: The variation of crack propagation stress thresholds of

geopolymer and class G cement at different curing
temperatures

3.3 Deformation characteristics of geopolymer and

class G cement using ARAMIS image
photogrammetry
Fig. 5b: The crack propagation stress thresholds of G cement
cured at 40 ºC
ARMIS is commercial software for three-dimensional
(3-D) image capture. The word ARAMIS is the trade
The over all crack propagation stress thresholds of G name of an optical measurement and strain measuring
cement and geopolymer at different curing temperatures system [14]. This software is an optical 3-D deformation
are shown in Table 03 and Fig. 6. It is obvious that the measuring system that converts the movements of pixels
crack propagation stress thresholds of geopolymer, in a series of stereo photographs into displacement
generally, increases with the curing temperature, and vectors. By combining high stereo photography and
those of G cement shows initial increment and then photogrammetry, the deformation behavior of a target
reduces towards 80 ºC. This is because of larger strength can be studied when it is loaded.
reduction towards 80 ºC. If the crack propagation stress The failure strain and deformation behaviors of
thresholds of G cement and geopolymer are compared, G geopolymer and class G cement are shown in Table 4
cement possesses higher values at ambient conditions, and 5 respectively. For geopolymer, failure strain of the
while geopolymer has higher values at elevated section reduces with the increase in curing temperature,
temperatures. This trend is similar to UCS trend as crack and at lower curing temperatures sample undergoes

6
shear failure whereas at elevated temperatures
geopolymer undergoes splitting failure. This is due to
the fact that as the temperature increases more and more
Si and Al dissolves in alkaline activator increasing the
rate of the reaction, which will produce a brittle matrix.
On the other hand, class G cement undergoes 4-8% of
failure strain at different curing temperatures, and there
is no considerable variation in failure strains with the
temperature. In addition the failure is shear failure
regardless of the curing temperature.
Table 4: The deformation behavior of geopolymer using ARAMIS image capture
Curing Temperature (deg)
23 40 60 80

ARAMIS

Original

Table 5: The deformation behavior of G cement using ARAMIS image capture

Curing Temperature (deg)
23 40 60 80

ARAMIS

Original

7
4. CONCLUSIONS 5. Barlet-gouédard V. 2007. Well technologies for
co geological storage: co-resistant cement. Oil
Based on the finding from this experimental work, the and gas science and technology, 62:325.
following conclusions were drawn; 6. Sweatman RE, Santra A, Kulakofsky DS,
1. The optimum curing temperature for higher Halliburton, Calvert DGJ. 2009. Effective zonal
strength of geopolymer and class G cement lies isolation for CO2 sequestration wells. Society of
in the same range (50-60 ºC). However, G petroleum engineers, SPE International
cement possesses higher strength at ambient Conference on CO2 Capture, Storage,
conditions, whereas geopolymer has higher California, USA: ISSN 978-1-55563-267-0.
strength at elevated temperatures. Generally, 7. Van Jaarsveld J. Van Deventer J, Lorenzen L.
class G cement will be good as well cement up 1997. The potential use of geopolymeric
to a depth of 1 km, whereas geopolymer will be materials to immobilise toxic metals: part i.
suitable at depth of more than 1 km. Theory and applications* 1. Minerals
2. At lower curing temperatures, G cement shows engineering, 10: 659-669.
higher values of Young’s modulus and 8. Diaz E, Allouche E. 2010. Recycling of fly ash
geopolymer has higher values at elevated curing into geopolymer concrete: creation of a
temperatures. This means G cement is stiffer at database. Green technologies conference,
ambient conditions compared to geopolymer, Grapevine, TX: 1-7.
while geopolymer is stiffer at elevated 9. Nasvi MCM, Ranjith PG, Sanjayan J.2011.
temperatures. Geopolymer as well cement and variation of its
3. AET reveals that G cement has highest values of mechanical behaviour with curing temperature,
crack propagation stress thresholds at ambient Greenhouse gases; science and technology, 1: 1-
conditions, and geopolymer has the highest 13.
values at elevated temperatures (above 40 ºC). 10. Mishra A, Choudhary D, Jain N, Kumar M,
However, lack of early counts is a concern in Sharda N, Dutt D. 2008. Effect of concentration
AE method. of alkaline liquid and curing time on strength
4. ARAMIS image capture showed that the failure and water absorption of geopolymer concrete.
strain of geopolymer reduces with the curing Journal of Engineering and Applied Science; 3:
temperature, and the failure is shear at ambient 14-18.
condition, whereas it is splitting failure above 40 11. Hardjito D, Wallah S, Sumajouw D,Rangan B.
ºC. ARAMIS results of class G cement revealed 2005. Introducing fly ash-based geopolymer
that there is no major change in the failure strain concrete: manufacture and engineering
of the sample with the curing temperature, and properties. In: 30th conference on our world in
the type of failure was shear failure for all the concrete and structures, august, Singapore: 271-
curing temperatures. 278.
12. Hoek B, Bieniawski Z. 1965. Brittle fracture
propagation in rock under compression.
5. REFERENCES International journal of fracture, 1:137-155.
13. Eberhardt E, Stead D, Stimpso B, Read R. 1998.
1. Bruant R, Guswa A, Celia M, Peters C. 2002. Identifying crack initiation and propagation
Safe storage of CO2 in deep saline aquifers. thresholds in brittle rock. Canadian geotechnical
Environmental science & technology, 36:240a- journal, 35: 222-233.
245a. 14. Zhang ZYRM. 2003. Structural integrity
2. Perera MSA, Ranjith P, Choi S, Bouazza A, evaluation of polymer composites using optical
Kodikara J, Airey D. 2011. A review of coal strain measurement (aramis). Acmc/sampe
properties pertinent to carbon dioxide conference on marine composites, ISBN 1-
sequestration in coal seams: with special 870918-02-9.
reference to victorian brown coals.
Environmental earth sciences, 64: 223-235.
3. Benso S, Cook P, Anderson J. 2005.
Underground geological storage. Ipcc special
report on carbon dioxide capture and storage:
195-276.
4. Duguid A. 2009. An estimate of the time to
degrade the cement sheath in a well exposed to
carbonated brine. Energy procedia, 1: 3181.

View publication stats