close

Вход

Забыли?

вход по аккаунту

?

j.ijggc.2018.07.013

код для вставкиСкачать
International Journal of Greenhouse Gas Control 78 (2018) 27–36
Contents lists available at ScienceDirect
International Journal of Greenhouse Gas Control
journal homepage: www.elsevier.com/locate/ijggc
Formation of an amorphous silica gel barrier under CO2 storage conditions
C.A. Castañeda-Herrera
a
b
a,b
, J.R. Black
a,b,⁎
, E.M. Llanos
a,b
, G.W. Stevens
a,b
, R.R. Haese
T
a,b
Peter Cook Centre for CCS Research, School of Earth Sciences, The University of Melbourne, Parkville, VIC 3010, Australia
CO2CRC Ltd., 11-15 Argyle Pl, Carlton, VIC 3053, Australia
A R T I C LE I N FO
A B S T R A C T
Keywords:
CO2 leakage
Sodium silicate
Mitigation and remediation
Core-flood experiments
Risk assessments of Carbon Capture and Storage (CCS) have identified that in some instances carbon dioxide
(CO2) leakage through the caprock cannot be entirely precluded, e.g. through high permeability undetected
zones. This study recommends the use of a geochemical barrier that forms upon reaction with CO2, as a means to
mitigate and remediate CO2 leakage. The proposed technology is based on the injection of an alkaline sodium
silicate solution that reacts with the leaking CO2, leading to silica gel formation. Laboratory studies undertaken
to evaluate this technology, included flow-through and core-flood experiments at ambient and reservoir conditions, respectively. The tests aim was to assess the barrier formation performance by changes in pressure as an
indicator of permeability reduction. Results show that the formation of the silica barrier was controlled by the
mixing gradient of the two reactants, where the reaction resulted in a permeability reduction between one and
three orders of magnitude under reservoir conditions. Thus, using sodium silicate as a reagent for forming a
barrier is a promising technology to abate CO2 leakage for CCS purposes. Further research using reactive
transport modelling to investigate barrier formation in a reservoir model is needed before applying this technology at the field scale.
1. Introduction
CCS has been proposed as a technology to mitigate CO2 emissions,
however the possibility of CO2 migration through different pathways is
a main concern for CO2 storage (IPCC, 2005). Risk assessments
(Manceau et al., 2014; Pawar et al., 2015) have identified and grouped
CO2 leakage pathways into two different categories: i) engineered,
which is related to the impairment of wells, especially through the
degradation of well materials (Carey, 2013); ii) natural, which refers to
the migration of CO2 through permeable zones, joints or fractures
within the caprock (Tongwa et al., 2013).
A number of technologies and materials derived from industrial
processes have been proposed for mitigating the risk of, and/or remediating a CO2 leak from a CO2 storage site (Castañeda-Herrera et al.,
2018; Manceau et al., 2014). Cements (Kutchko et al., 2007) and geopolymers (Provis and Van Deventer, 2009) are useful for forming barriers in engineered pathways. However, chemical degradation of cements under CO2 storage conditions and changes to injectivity have to
be considered when using these materials. Other technologies have
focused on mitigation of the risk of CO2 leakage via natural pathways,
such as biomineralization (Cunningham et al., 2009). Microorganisms
in combination with certain reactants can be used for
⁎
biomineralization, where calcite precipitates effectively sealing leaks
(Mitchell et al., 2009). The application of microbes requires certain
conditions for rapid microbial growth and the availability of calcium
for the abundant precipitation of calcite (Cunningham et al., 2013).
Another technology is the hydraulic barrier. It can be used to change
the hydrodynamics in a reservoir through pressure management of
aquifers overlying the caprock of the CO2 storage site, redirecting the
CO2 flow path direction (Réveillère et al., 2012). The implementation of
the hydraulic barrier formation technique is site-specific and requires
considerable operational efforts. Finally, the injection of dry CO2 has
also been studied as a way to induce mineral precipitation which results
in self-sealing leakage pathways (Vialle et al., 2016).
Most of these technologies are suitable for high pressures and
temperatures but are not resistant to the low pH of CO2 storage systems.
For example, barriers formed from cements or calcite are reactive under
the mildly acidic conditions (pH ∼ 4) (IPCC, 2005). Ito et al. (2014)
proposed the use of a reactive barrier that is chemically stable in an
acidic environment to overcome this problem. The idea was to induce
the formation of a barrier through the reaction of a sodium silicate
solution with leaking CO2 either in form of dissolved CO2 or as supercritical CO2. For example, the mixing of acidic CO2-enriched water (pH
∼ 4) with alkaline sodium silicate solution (pH ∼ 11) results in a pH
Corresponding author at: Peter Cook Centre for CCS Research, School of Earth Sciences, The University of Melbourne, Parkville, VIC 3010, Australia.
E-mail address: jay.black@unimelb.edu.au (J.R. Black).
https://doi.org/10.1016/j.ijggc.2018.07.013
Received 9 April 2018; Received in revised form 22 June 2018; Accepted 7 July 2018
1750-5836/ © 2018 Elsevier Ltd. All rights reserved.
International Journal of Greenhouse Gas Control 78 (2018) 27–36
C.A. Castañeda-Herrera et al.
neutral to mildly alkaline triggering the precipitation of amorphous
silica (Eq. (1)).
However, other silica solids might precipitate due to the formation
of monomeric and polynuclear species of silica (SixOy(4x-2y)). At low
silica concentrations and high (> 10) and low pH (< 3) silica predominantly forms a monomeric complex, silicic acid (H4SiO4). The
degree of silica polymerisation is a function of the solution composition,
pH, temperature and time (Andersson et al., 1982; Crerar et al., 1981;
Greenberg and Sinclair, 1955; Rothbaum and Rohde, 1979). The
polymerisation of silica at high concentrations eventually leads to the
formation of a gel (Eq. (2)). It is expected that precipitation of either
amorphous silica or silica gel, which are stable under acidic conditions,
will plug the leakage by filling the pores of highly permeable zones.
Amorphous silica precipitation upon mixing with CO2:
formation in porous media, 2) measure the impact this barrier formation has on the permeability of a typical Sandstone reservoir rock
(Berea Sandstone) under CO2 storage conditions, 3) monitor the stability of this barrier over time. Firstly, a series of tests were run using a
flow-through column operated under ambient pressure conditions,
where parameters such as curing periods and temperatures were tested.
And secondly, a multiphase core-flood instrument was used to induce
silica gel barrier formation in a core of Berea Sandstone under CO2
storage conditions.
H3SiO4− + H+ + 2CO2(aq) = SiO2(am) + 2H2CO3
Unconsolidated acid purified sand (purum p.a.) was purchased from
Sigma-Aldrich. A 7.15 wt% silicate solution was diluted from a commercial sodium silicate solution with a molar ratio of SiO2:Na2O = 3.16
(Chemsupply™). This concentration of silica was chosen as the optimal
solution based on previous physical characterization experiments of the
density and viscosity of the solution (Castañeda-Herrera et al., 2017). A
solution of 0.1 M potassium phthalate monobasic (KHP) with a pH of 4
was used in ambient pressure experiments as a proxy for CO2-enriched
waters found under CO2 storage conditions. The KHP buffer capacity
was adjusted by adding HCl 0.1 M. A cylinder of compressed food grade
CO2 with a dip tube (Coregas) was used for experiments involving CO2
injection.
2. Materials and methods
2.1. Materials
(1)
Silica gelation upon mixing with CO2:
SixOy(4x−2y) + (4x-2y)H+ + CO2(aq) = Silicagel + HCO3−
(2)
A silicate solution could be injected on top of the caprock of the
storage reservoir, where it spreads laterally and would lead to a precipitation reaction once it is in contact with leaking supercritical CO2
(scCO2) or CO2-enriched water. Ito et al. (2014) suggested two different
cases where this mechanism can be used. In the first case, a remediation
technique is considered; here, the solution is injected on top of the
caprock after CO2 leakage is detected. In contrast, the second scenario
proposed a mitigation technique, where the solution is emplaced on top
of the caprock before the CO2 is injected into the storage reservoir. The
latter case is a precautionary measure in situations where the seal integrity is uncertain. Here, the precipitation occurs in case of an eventual
leak.
Only a few studies have investigated the injection of a silicate solution into the zone of a CO2 leakage using laboratory core-flooding
techniques and through reactive transport modelling of the process
(Brydie et al., 2014; Druhan et al., 2014; Fleury et al., 2017; Ito et al.,
2014). Simulations using TOUGHREACT by Ito et al. (2014), Druhan
et al. (2014) and Druhan et al. (2015) showed that precipitation of
amorphous silica led to a permeability reduction and CO2 leakage mitigation was possible. Druhan et al. (2014) also found that a well-defined mixing zone between the reagents was necessary in order to have
an effective seal. Brydie et al. (2014) tested the silicate reaction in a
porous medium (Berea Sandstone), controlling flow with a core flood
apparatus. The experiment by Brydie et al. (2014) found that silica
precipitation successfully sealed the flow of fluid through the porous
medium. However, the experiments were run at a pressure of 58 bar
and temperature of 21 °C, which are not geological CO2 storage reservoir conditions. Fleury et al. (2017) also tested the silicate-based
process in a core flood instrument at 40 °C. The experiment did not
account for reaction with CO2 but the injection of a mixture of silicate
and acetic acid that reacts over time. Results showed that a gel with a
strength of 600 bar/m was formed within the core. Additionally, there
is one study using the silicate technology at field scale by Lakatos et al.
(2009), which was not conducted for CO2 storage purposes. The test
used a silicate-polymer-urea method to successfully stop a CO2 leakage
from a gas field in Becej, Serbia. While these initial results are promising, none of the current studies have conducted laboratory experiments under CO2 storage conditions with scCO2 present. Therefore,
further experimental research is required in order to optimise this
technology.
The aim of this research is to assist developing technologies to mitigate and/or remediate a CO2 leakage through natural pathways,
specifically to develop a procedure for producing a silica gel barrier in
situ by utilising the geochemistry of CO2 storage reservoirs. For this
purpose, several laboratory experiments were conducted in order to 1)
improve our understanding of the conditions of in situ silica gel
2.2. Flow-through column
2.2.1. Ambient conditions
The experimental set-up consisted of a flow-through column, a
peristaltic pump, pressure sensors and a flow meter using a similar
design to that used by Castañeda-Herrera et al. (2017). The setup was
modified by adding a second pressure sensor at the outlet of the column
in order to improve the measurement of the pressure differential (ΔP)
across the column during flow. A column 20-cm in length and with an
internal diameter of 2.54 cm was used. The column was packed with the
unconsolidated sand and the pore volume (PV) was determined at the
beginning of each experiment. A PV value of 40 cm3 was calculated
using the weight and density of the sand to estimate the volume fraction
it occupied in the fixed column geometry, which corresponded to a
porosity of ca. 50%. The sand packed column was connected to the
flow-through system and pre-saturated with MilliQ water to check for
any leaks. Subsequently, the solution was injected at 1 mL/min. Two
experiments were conducted at ambient pressure and a temperature of
20 °C:
2.2.1.1. Mitigation scenario. The column was saturated with the 7.15 wt
% silicate solution followed by the injection of 0.1 M KHP solution. This
injection sequence is similar to the case where CO2-enriched water
encounters a silicate solution that was emplaced on top of the caprock
before any leakage occurs.
2.2.1.2. Remediation scenario. The column was saturated with 0.1 M
KHP solution followed by the injection of the 7.15 wt% silicate solution.
This injection sequence refers to the injection of silicate solution into a
porous rock layer partially filled with CO2-enriched water.
In both scenarios, up to two pore volumes of the first solution was
slowly injected to assure displacement of the pre-saturated water and
complete saturation along the column with the solution. Subsequently,
the second solution was injected until half the PV was displaced. At that
point, the interface between both solutions was approximately at the
mid-point of the column. The injection of the second solution was
stopped and an incubation period of 24 h commenced. The incubation
period allowed the two fluids inside the column to react by mixing over
time. This was followed by a sequence of injection of the second
28
International Journal of Greenhouse Gas Control 78 (2018) 27–36
C.A. Castañeda-Herrera et al.
Fig. 1. Simplified schematic diagram of the multiphase core-flood instrument at The University of Melbourne (Oven assembly within dashed-lined rectangle).
injected into the column at 1.25 mL/min for 5 min to saturate the
column with the silicate reagent. The differential pressure across the
column was monitored during injection in order to monitor for changes
in the pressure differential across the column. The permeability was retested after aging the gel in situ for 24 days by attempting to flow water
through the column.
Micro-CT scans of the column were taken before and after injection
of the reagent, when the silica-gel barrier was formed and after 15 days
of aging the gel in situ. Micro-CT scans were conducted using a GE
Phoenix Nanotom m (GE Sensing & Inspection Technologies GmbH)
with an X-ray energy of 60 kV and 400 μA. A diamond target with a
tungsten coating was used with a 0.5 mm aluminium filter to preharden the X-ray energy and reduce beam hardening by the sample. A
resolution of 6 μm was achieved taking 1200 X-ray projections averaging 3 and skipping 1 per projection to improve contrast for a scan
time of 41 min. 3D volumes were reconstructed from 2D projections
using the dataosx reconstruction software (GE Sensing & Inspection
Technologies GmbH). Analysis of the volume fraction of different
phases was conducted using the Avizo software package (ThermoFischer, FEI). Phases were segmented and the volumes occupied by
each segmented phase compared to a mask of the entire sample volume.
solution and periodic incubations until the end of the experiment.
Incubation periods and injected volumes varied depending on the
monitored pressure gradient in the column from one injection period to
another. The experiments were stopped when no further changes in the
pressure gradient was observed or when 1 PV of the second solution
was injected into the column. For each injection period the ΔP was
recorded and fluid samples were collected for chemical analysis (see
Section 2.4).
2.2.2. Elevated temperatures
In order to study the effect of temperature, both the mitigation and
remediation scenarios were tested by running experiments at 30 °C,
45 °C, and 60 °C. The procedure was slightly modified as follows: the
column was detached from the injection caps after 0.5 PV of the second
fluid was injected. Then, the column was plugged with rubber stoppers
and placed horizontally inside a water bath (Thermoline scientific
TWBC-24) at a given water temperature. The first incubation period
started at that point with the temperature held constant throughout.
The limited height of the water bath prevented vertical orientation of
the column. The column was taken from the water bath after the incubation period and placed back into the flow-through setup in order to
continue with the next injection period. A series of incubations in the
water bath followed by injection periods was repeated until 1 PV was
reached or no further changes were observed in ΔP. Lastly, pH and
cation concentrations were measured, as in previous experiments.
2.3. Core-flood experiments at reservoir conditions
2.3.1. Core sample
Berea Sandstone cores were used to conduct core-flood experiments.
The Berea Sandstone is a well characterized siliclastic rock that has
been used in various core-flood studies (Krevor et al., 2012; Pentland
et al., 2011). A core with a length of 30 cm and a diameter of 3.81 cm
was cut into thirds (Cores 1, 2 and 3). A different experiment was
carried out on each core. Prior to commencing an experiment, porosity
was estimated by the water saturation method (Luhmann et al., 2017).
Likewise, permeability was measured using the core-flood instrument
by recording ΔP values at different flow rates when injecting MilliQ
water through the core. ΔP records were used to calculate permeability
using Darcy’s law (Whitaker, 1986).
2.2.3. Micro-CT analysis
A separate flow through column experiment was conducted using a
smaller geometry cylindrical column with a 6.35 mm internal diameter
and packed sand column length of ca. 25 mm (pictured in Fig. 4A, see
Results Section 3.3). A dilution of the sodium silicate reagent to 6 wt%
SiO2 was made using doubled deionised water and 1 M HCl so that the
final pH was 10.76 to slowly induce silica-gel formation (solutions set
as a gel within 12 min of mixing). This ‘pre-mixing’ procedure ensures
gel formation for imaging purposes and may serve as another method of
deploying the reagent in order to modify reservoir properties such as
the sweep efficiency. After mixing the reagent it was immediately
29
International Journal of Greenhouse Gas Control 78 (2018) 27–36
C.A. Castañeda-Herrera et al.
the stability of the barrier with time.
B.) Pure scCO2 was injected at a flow rate of 0.058 mL/min for 2 h. An
incubation period was not applied since a significant decrease in
permeability was immediately observed. The experiment was
stopped, permeability was measured immediately and retested after
3.5 weeks of storage under ambient conditions.
2.3.2. Multiphase core-flood instrument
The core-flood system at the University of Melbourne was custombuilt by DCI Test Systems Corporation. A simplified schematic diagram
of the system is depicted in Fig. 1. Three piston-controlled VPA (Volume-Pressure Actuator) pumps operate three main processes (DCI,
2015):
• VPA 1: used to inject CO
•
•
Permeability for these experiments was measured by re-injecting
MilliQ water for several PV after the injection of CO2 and a respective
incubation period. This step was carried out to displace all possible
residual CO2 and aqueous silica from the system. The core was considered to be completely CO2 depleted when no more bubbles were
observed and neutral pH was obtained in the outlet. At that point, ΔP
values were recorded and used to calculate permeability using Darcy’s
law.
2 into the core holder, where scCO2 pressure
is achieved by using a gas booster pump, and pore pressure is set
using a back-pressure regulator.
VPA 2: controls the flow rate of the injection fluid from the accumulators to the core.
VPA 3: pumps confining oil into the core holder to achieve the desired experimental confining pressure.
The oven assembly is shown as part of Fig. 1, which has four main
components: 1) Two accumulators of 1 L capacity each that are used to
store the injection fluid. The accumulators are filled from a temperature-controlled fluid reservoir so that the accumulators can be refilled
with water/brine at the temperature of experiments. 2) A mixer unit
that is used for simultaneous injection of gas and liquid phases. 3) A
dome-loaded back-pressure regulator that sets the pore pressure at the
outlet of the core by applying the set pressure to a diaphragm using
nitrogen (N2). 4) A Hassler sleeve-type core holder, which contains a
Viton rubber sleeve that holds the core to which confining pressure is
applied.
The Hassler sleeve-type core holder ensures equilibrated pressure
surrounding the core (DCI, 2015). It consists of a rubber sleeve or jacket
and end plugs with a distribution pattern. The sleeve is surrounded by
the confining oil chamber, where the confining pressure is applied. The
core holder can accommodate cores of 3.81 cm diameter and up to
30.48 cm long. Pressure along the core holder is measured by six
pressure transducers: one at the inlet, one at the outlet and four
transducers are placed at equal distances along the column. The series
of pressure transducers are placed outside the oven assembly and can be
used as a bypass when fluid is injected without flowing through the
core.
2.4. Analytical techniques
Fluid samples obtained from the outflow of flow-through columns
and the core-flooding instrument were analysed for pH and cation
concentrations. pH was determined using a 916 Ti-Touch titrator
Methrom pH-meter using a glass electrode immediately after collection
of the sample. pH was only determined in samples collected from experiments at ambient conditions. Depressurization of fluid samples
collected from experiments involving reservoir conditions prohibited
the measurement of a correct pH. Total concentrations of silica, sodium
and potassium were also measured using an Inductively Coupled
Plasma Optical Emission Spectrometer (Agilent 5100 ICP-OES). The
instrument was calibrated using a standard calibration solution for
these cations. The samples were diluted using 2% nitric acid (HNO3) in
order to be in the calibration range of 0.5–50 ppm.
Formation of the silica barrier was also assessed by analysing the
mineral composition of the solid phase. This analysis was not carried
out for all experiments since it was only used to confirm the composition of precipitated barrier as amorphous silica. An X-ray powder diffraction (XRD) technique was used to characterise the precipitate,
which was performed using a Bruker D8 Advance Diffractometer at the
Materials Characterisation and Fabrication Platform (MCFP) at the
University of Melbourne.
2.3.3. Experimental procedure
Barrier formation was studied under mitigation conditions where
the core was pre-saturated with silicate solution. Remediation conditions were avoided to preserve the integrity of the instrument, since the
injection of silicate solution may lead to an unwanted blockage of the
flow lines by silica precipitation.
Two different CO2 sources were applied: CO2-enriched water was
injected into Core 1 while scCO2 was injected into Cores 2 and 3. Each
core was pre-saturated with 7.15 wt% SiO2 solution, wrapped in aluminium foil and placed in the core holder. The core was wrapped in
aluminium foil to avoid damage of the sleeve by contact with CO2.
Next, the accumulators were filled with pre-heated MilliQ water at
60 °C and a confining pressure of 206.8 bar was applied, followed by a
back pressure of 103.4 bar. A temperature of 60 °C and a pressure of
103.4 bar were chosen as conditions of a normal CO2 storage reservoir.
The CO2 source was then injected into the core. Flow rates and pressure
data were recorded during the experiment to calculate permeability
values. Depressurized fluid samples were collected at the outlet and
used for silica analysis. The procedures for the injection of CO2-enriched water (Procedure A) and scCO2 (Procedure B) were different as
explained here:
3. Results
3.1. Mitigation and remediation scenarios
The results of the flow-through column experiments are presented
in Fig. 2 with their corresponding incubation periods marked as vertical
lines. Fig. 2A depicts the behaviour of ΔP for the mitigation scenario
showing that ΔP increased with time, inferring a permeability reduction. ΔP values were averaged from the data recorded for per minute to
avoid the background noise cause by the use of a peristaltic pump. ΔP
increased from around 2 mbar up to 150 mbar, which indicates a barrier
formation. No data are shown in this figure for the first 0.5 PV because
of a recording fault by the software. Fig. 2A also shows an oscillatory
pattern in ΔP that may represent the build-up of the barrier impeding
flow (increasing ΔP), followed by penetration of the barrier (decreasing
ΔP). This behaviour was discussed in preliminary experiments
(Castañeda-Herrera et al., 2017).
Change of water composition was also observed with the mitigation
scenario as shown in Fig. 2B. This figure shows pH and dissolved cation
normalized concentrations as a function of PVs injected. Dissolved silica
and sodium were introduced into the system with the injected silicate
solution, while potassium came from the use of KHP solution. Incubation periods are not shown in Fig. 2B in order to simplify the figure.
However, the incubation periods correspond to the same shown in
Fig. 2A. Silica and sodium concentrations decreased in the outflow,
while the concentration of potassium increased. Silica showed an
A.) CO2-enriched water was injected by simultaneously flowing MilliQ
water at 0.058 mL/min and CO2 at 0.0087 mL/min. This injection
lasted for around 2 h filling ca. 1/3 of the PV. The core was then
incubated at pressure and temperature for 2 days and the permeability was retested. Afterwards, the core was stored at ambient
conditions. Permeability was tested again after 4 weeks to assess
30
International Journal of Greenhouse Gas Control 78 (2018) 27–36
C.A. Castañeda-Herrera et al.
Fig. 2. Effect of ΔP (red line) and fluid chemistry as a function of pore volume for different scenarios. Cation concentration was normalized to the initial concentration of the solutions (C0): A) ΔP change for mitigation, B) Fluid chemistry change for mitigation, C) ΔP change for remediation, D) Fluid chemistry change for
remediation. Vertical lines in ΔP plots are incubation periods (Mitigation: Dashed = 8 h, Solid = 18 h) (Remediation: Dashed = 24 h, Solid = 48 h). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
instantaneous decrease after the first incubation period, decreasing
from a normalized concentration of 1 to less than 0.17. Fig. 2B also
shows the trend in pH, which remains at 10.5 throughout the experiment.
Fig. 2C shows changes for ΔP for the remediation scenario. ΔP
started at 30 mbar for the first 0.5 PV and reached its maximum of
500 mbar after the first 48 h incubation period. The most significant
increase in ΔP occurred exactly after the first 24 h curing period where
it increased from 100 mbar to 500 mbar. After this point ΔP decreased,
returning to values of 100 mbar for the rest of the experiment.
Fig. 2D depicts the results of the fluid chemistry for the remediation
scenario. Trends in pH and cation concentrations were different than
the mitigation scenario, since the sequence of fluid injection was different. The initial pH of the KHP solution on the column was around 4,
and as the sodium silicate reagent was injected the pH of outlet fluids
increased gradually to a pH of 4.5 until flow was ceased for a 48 h
incubation period. After this incubation the pH suddenly increased to
10.5 and remained constant for the rest of the experiment. This change
in pH occurred simultaneously with an increase in silica in the outflow.
Silica and sodium concentrations showed a concurrent increase, where
silica reached a normalized concentration of 0.86 after 1.2 PV. In contrast, potassium decreased with time as expected, given potassium was
derived from KHP.
3.2. Effect of temperature
ΔP values at different temperatures were only obtained for the remediation scenario experiments, since it was observed that for mitigation experiments ΔP reached lower values than the initial values of the
experiment. It was found that the low ΔP values were not related to the
formation of the barrier but to a phenomenon caused by the experimental procedure. This phenomenon is explained in further details in
the discussion (see Section 4.2). However, fluid chemistry results for
mitigation experiments followed the same trend as the previous experiment. Figures of ΔP and fluid chemistry at different temperatures
can be found in the supporting material (Figs. S1–S2).
Values of ΔP for remediation experiments at different temperatures
are compared in Fig. 3, which shows values before and after the initial
24 h incubation period. It is important to point out that the first incubation period gave the most significant change in ΔP. In general, ΔP
started with values between 22 and 45 mbar at all temperatures, and
then increased significantly after the first incubation period reaching
values between 88 and 100 mbar.
3.3. Micro-CT imaging of in situ gel formation
An initial micro-CT scan of the dry column was conducted (Fig. 4A
31
International Journal of Greenhouse Gas Control 78 (2018) 27–36
C.A. Castañeda-Herrera et al.
Table 1
Petrophysical properties of Berea Sandstone cores use in core-flood experiments.
Porosity (%)
Initial
Core 1
Core 2
Core 3
18
17
18
Permeability (mDarcy)a
Initial
after injectionb
after ca. 4 weeksc
105 ± 3.8
114 ± 13.7
133 ± 1.4
11.6 ± 0.6
0.26 ± 0.23
0.11 ± 0.08
57.5 ± 3.04
1.26 ± 0.04
Average of 3 highest flow rate data points, where available, and 2σ error.
Permeability after 2 days of incubating Core 1 in situ, and immediately
after injection for Cores 2 and 3.
c
Permeability after 4 weeks of aging Core 1 and 3.5 weeks of aging Core 2
under ambient conditions.
a
b
space, gel and/or fluid and mineral phases were segmented separately
from the micro-CT histogram and add up to a total volume fraction of 1.
The average porosity of the dry unconsolidated sand in the column was
ca. 40% (segmented from Fig. 4B in region occupied by sand). The blue
line in Fig. 4E shows that the porosity decreased to 8% on average
leaving ca. 32% of the pore space occupied by silica-gel. Porosity decreases to around 3.8% on average in specific sections (10–15 mm,
Fig. 4) of the column. A large dark grey object is seen between the
3–7.5 mm sections of the column (Fig. 4C). This was the filter paper
packed into the base of the column (Fig. 4B) to prevent migration of
unconsolidated sand. Flowing reagent through the column pushed the
filter paper into view. The filter paper is segmented as part of the silicagel phase and included in the volume fraction plotted in Fig. 4E (seen as
a deviation from the trend between ca. 3–7.5 mm).
Fig. 3. Change in ΔP for the remediation scenario experiments at different
temperatures. Permeability values decreased after an initial 24 h incubation
period in all experiments giving an indication of barrier formation.
and B) and showed an average porosity between sand grains of ca. 40%.
A pre-mixed sodium silicate reagent with an adjusted pH of 10.76 took
ca. 12 min to form a stable solid gel from initial mixing of the reagent.
After waiting 12 min for the gel to form the differential pressure across
the column was seen to rise from below the detection limit (ca.
10 mbar) to 5 bar, which may indicate a significant decrease in permeability. However, measuring the permeability in this scenario is
complicated by the changing viscosity of the silica gel reagent. After 24
days of aging the gel in situ an attempt to measure the permeability
through the column was made by injecting water, where again the
differential pressure across the column immediately increased to ca.
6 bar at very low flow rates (0.03 mL/min), which would indicate a
permeability decrease to < 5 mDarcy. However, this is near the upper
limit of the pressure sensors for the flow through system, and no flow
was actually observed at the outlet of the column indicating it may have
been completely clogged making the permeability unmeasurable.
Micro-CT scans of the column were collected after the on-set of gel
formation and 15 days after aging the gel in situ. Fig. 4C shows a microCT cross-section of the saturated column, with the silica-gel phase appearing as a dark grey phase within the pore space (black space) between quartz grains (light grey grains). Some unoccupied pore space
remains, as illustrated in Fig. 4D, which renders the unoccupied pore
space in the column after reagent injection. The volume fraction occupied by different phases is quantified in Fig. 4E, where the pore
3.4. Core-flood experiments under reservoir conditions
Initial porosity and permeability values of the cores used for the
multiphase core-flood experiments are shown in Table 1. Table 1 also
reports values of permeability after experiments, which are more
clearly presented in Fig. 5. Results for Core 1 indicate that the final
permeability was one order of magnitude lower than the original value,
dropping from approximately 105 mDarcy to 11.6 mDarcy.
Results for Core 2 show a permeability decrease from ca. 114
mDarcy to 0.26 mDarcy, which is approximately a change of three
orders of magnitude. Lastly, permeability decreased from 133 mDarcy
to ∼0.1 mDarcy for Core 3, which is similar to the replicate experiment
using Core 2, showing effective reproducibility in the result. Core 3
Fig. 4. Micro-CT analysis of silica-gel barrier
formation within a small flow through column
(A). A 2D. cross section through the middle of
the column is shown where B) the column is
packed with dry unconsolidated quartz sand
(light gray grains) with porosity visible (black
space) between grains; C) the column has been
flushed with 6 wt% silicate reagent and silicagel has formed in the pore space (darker gray
pore filling phase). Some porosity is seen to be
unsaturated after flushing the column with silica reagent (C), which is more apparent when
rendered in 3D (D). E) Volume fractions occupied by empty pore space (blue line), silica-gel
and/or aqueous reagent (green line) and unconsolidated sand grains (brown line). (For
interpretation of the references to colour in this
figure legend and text, the reader is referred to
the web version of this article).
32
International Journal of Greenhouse Gas Control 78 (2018) 27–36
C.A. Castañeda-Herrera et al.
Table 3
Content of amorphous phase of untreated and treated unconsolidated sand and
Berea sandstone.
Sample
Unconsolidated
sanda
Berea sandstoneb
a
b
broke into two pieces when removed from the core holder. Therefore,
the permeability of Core 3 could not be retested after 3.5 weeks.
The permeability of Core 1 and Core 2 were measured after 4 and
3.5 weeks of aging at ambient conditions, respectively, in order to
evaluate the stability of the barrier. The permeability increased from
11.6 to 57.5 mDarcy for Core 1. The latter was still about half of the
initial value before the experiment (105 mDarcy). The permeability of
Core 2 increased by one order of magnitude, from 0.26 mDarcy to 1.26
mDarcy after 3.5 weeks. This change in permeability was still two orders of magnitude less than the original rock permeability of 114
mDarcy.
A silica mass balance was calculated for experiments using Core 1
and Core 2 based on the outlet fluid composition (Table 2). A mass
balance for Core 3 was not calculated because the experiment was not
fully completed since the core broke in half. The silica mass balance for
Core 1 showed that 51.5% of the initial pre-saturated silica precipitated. In contrast, 92.2% of silica precipitated in Core 2 where
scCO2 was injected instead of CO2-enriched water.
The difference between the silica precipitation percentages in the
two experiments are reflected in the results of the permeability tests.
Core 2 had a greater drop in permeability than Core 1. This result shows
that the barrier formed using scCO2 was more efficient than the one
formed with CO2-enriched water.
Core 1
Core 2
882
830
428
65
454
765
51.5
92.2
Crystalline
phase
content (%)
Amorphous
phase content
(%)
99
1
82
18
88
12
75
35
Sample taken from the mitigation scenario experiment.
Sample taken from Core 2 used in the scCO2 injection experiment.
Results showed that for all the scenarios the increase in ΔP were
explained by the formation of silica gel in the pores, leading to a likely
drop in porosity and permeability. This in turn led to a greater restriction of the fluid flow. From preliminary experiments, it was found
that the resulting precipitate was not an amorphous solid but a gel. Gel
formation from sodium silicate has been observed in other studies (Iler,
1982; Tognonvi et al., 2011). It was also observed that ΔP dropped at
the late stage in the remediation scenario. This trend suggest that one
part of the barrier was not strong enough and was displaced by the fluid
injection. Nevertheless, the barrier that remain in the column was able
to keep a high ΔP along the column.
The drop in silica concentration in the mitigation experiment at the
early stage of the experiment suggested that some silica was immobilised. This outcome agreed with the drop in permeability. For the
remediation experiment, the change in pH and increase of silica concentration at around 1 PV following the first 48 h incubation period
suggested a successful mixing between the two fluids leading to a
partial precipitation of the aqueous silica.
However, it was also observed that pH did not change for the mitigation scenario despite silica concentration decreasing. This behaviour can be explained by the high buffer capacity of sodium silicate
(Weldes and Lange, 1969). Mechanisms for the polymerization of silica
gel can be explained by the buffer capacity of this solution (An-Pang,
1963). Predominant species at lower pH (< 10.5), such as H3SiO4− and
H4SiO4 (Eq. (3)), undergo a condensation process leading to the formation of siloxane bonds (SieOeSi) and the release of hydroxyl ions
(OH−) (Eq. (4)).
Table 2
Mass balance of silica in Cores 1 and 2 during barrier formation experiments.
% Si precipitated
Amorphous
phase content
(%)
4.1. Changes of permeability and fluid chemistry at ambient conditions
The formation of the barrier was also evaluated using XRD analysis
for this experiment. The main purpose was to make a comparison between the abundance of crystalline and amorphous phases before and
after the experiment since the precipitation of silica would increase the
amorphous phase fraction. This analysis was conducted on samples of
the unconsolidated sand used in the mitigation scenario at ambient
Precipitated mass of
Si (mg)
Crystalline
phase
content (%)
4. Discussion
3.5. Amorphous phase solid content
Outlet mass
of Sib (mg)
After experiment
conditions at 20 °C and samples of Core 2 used in the scCO2 injection
core flood experiment.
A summary of results is found in Table 3. The percentage of amorphous phase for the unconsolidated sand pack column at 20 °C increased from less than 1% to 18%. In contrast, an increase of amorphous phase from 12% to 35% was observed in a core of Berea
Sandstone where scCO2 was injected into a core pre-saturated with silica reagent. Both results showed a significant increase in the amorphous phase after the experiments.
Fig. 5. Summary of the permeability in three Berea Sandstone cores used in
core flooding experiments under reservoir conditions (60 °C and 100 bar) at
different stages of the experiment. Error bars represent the standard deviation
of the calculated permeability values at different flow rates. Cores were presaturated with the barrier forming sodium silicate reagent following the injection of A) CO2-enriched water and B) scCO2.
Initial mass of
Sia(mg)
Before experiment
H2SiO42− + 2H+ → H3SiO4− + H+ → H4SiO4
(3)
H3SiO4− + H4SiO4 → (OH)3Si-O-Si(OH)3 + OH−
(4)
−
The continuous release of OH increases the pH again, explaining
the high buffer capacity of sodium silicates. Further addition of acid is
necessary to form a polymeric gel with concurrent neutralization of
OH− ions.
The different pH trends between scenarios also suggested that the
sequence of the injected fluids affected the formation of the silica
a
Amount of silica in sodium silicate reagent saturating the cores available
pore space.
b
Amount of silica measured in fluids expelled from the core during barrier
formation experiments.
33
International Journal of Greenhouse Gas Control 78 (2018) 27–36
C.A. Castañeda-Herrera et al.
water bath during incubation periods instead of being vertically oriented as they were at 20 °C. The horizontal position of the column may
have led to a density stratification of the fluids during incubation. In
such as scenario, the less dense fluid (KHP) would migrate upwards
within the horizontal column. This stratification created a fluid layer
with less flow resistance. The fluids were already in different places
when the column was placed back vertically to continue with the injection. Migration of KHP in mitigation experiments was also affected
by formation of viscous fingers, since KHP moved from the bottom to
the top of the column. Therefore, it was not possible to confirm the
formation of the silica barrier in the elevated temperature mitigation
experiments. However, remediation experiments at different temperatures showed better results. These experiments might have had a density stratification but were not affected by viscous fingering since the
lower viscosity KHP was the top layer in the columns from the beginning of the experiments. A significant increase in ΔP was found for all
the experiments suggesting that the silica barrier can be formed regardless of the temperature. It was also confirmed that the main increase in ΔP occurred during the first 24 h of reaction.
Fig. 6. Comparison of dispersion of the fluids using dyes. A) Mitigation scenario
with blue dye and B) Remediation scenario with yellow dye. Red lines represent
the observed interface between both solutions during the experiments. (For
interpretation of the references to colour in this figure legend and text, the
reader is referred to the web version of this article).
4.3. Visualisation of gel formation and aging using Micro-CT
Pre-mixing the reagent with an adjusted pH of 10.76 allowed for the
silica to be injected and fill the pore space of the column before it set as
a gel. Further indication of gel formation in the porous media was given
by a large increase in the ΔP across the column to ca. 5 bar during the
5 min injection of reagent.
Significant syneresis (dehydration) was observed in stock solutions,
which had formed stable gels after 12 min of reaction. Syneresis is the
process where the silica gel strengths over time by shrinking and releasing water (Hamouda and Amiri, 2014). After 1 week of aging the
gel in test tubes, the volume occupied by the gel shrunk by 50% exsolving fluid in its place so that the total volume occupied did not
change. To examine the effect this may have on gel and subsequent
barrier stability within the porous media the difference between microCT scans on the first day of gel formation and after 15 days of aging the
gel were compared. Registration of the two micro-CT datasets showed
little difference between the distribution of phases at day 1 and day 15,
with unoccupied porosity decreasing by ca. 1.3% on average across the
column. The permeability of the column was tested after 24 days of
aging the gel in situ and potentially decreased to < 5 mDarcy as indicated by a rise in the ΔP to ca. 6 bar. However this estimate of permeability is convoluted by the viscous gel in the column and a lack of
flow at the outlet indicating the gel may have hardened in the porous
media. Therefore, it appears as though the process of syneresis may
operate differently in the closed system of the porous media, and did
not appear to impact the stability of the gel or barrier that formed.
barrier. For the mitigation scenario, pH remained within a range of 10.5
and 11, because of the high buffer capacity of the silicate solution.
However, in the case of the remediation scenario, pH was initially low
as it was buffered by the KHP solution (pH = 4) and increased with
time when more alkaline silicate solution was introduced into the
system. The gradual increase in pH suggested a broader transition zone
with a pH below approximately 10.5, where precipitation of silica gel
was expected to occur. As a consequence, the presence of this pH
transition zone provides favourable conditions for the formation of the
barrier since the injected silicate solution was expected to form silica
gel efficiently.
The difference in pH trends between remediation and mitigation
scenarios also suggested a difference in dispersion within the column.
Qualitative tests were run to provide a visual comparison of the fluid
dispersion behaviour in both scenarios. The experiments were conducted using the same conditions of the experiments (column length:
20 cm and flow rate: 1 mL/min at 20 °C), but dyes were added to the
injection solution. The colours of the dyes were blue for the mitigation
scenario and yellow for the remediation scenario.
Results of the dispersion experiments are shown in Fig. 6. Fig. 6A
shows a preferential flow of the blue dyed KHP solution up the left-hand
section of the column in a mitigation scenario, while Fig. 6B shows a
uniform mixing front gradually flowing up the column in a remediation
scenario. The difference between the two cases can be explained by the
difference in viscosity of the two fluids (Viscosity KHP(∼H2O): ∼1 cP;
Viscosity silicate (Castañeda-Herrera et al., 2017): 1.63 cP). Previous
fluid displacement studies (Aminzadeh et al., 2013) have shown that
the flow front is unstable when a less viscous fluid is injected into a
more viscous phase. This instability leads to a fingering of the flow front
in consolidated sandstones (Aminzadeh et al., 2013). Thus, the injection
of a less viscous phase (KHP) into the porous media saturated with a
high viscosity fluid caused the disturbance of the flow front and led to a
preferential flow path forming in the mitigation experiment (Fig. 6A).
This could also explain the rapid drop in silica concentration for the
mitigation scenario. In turn, the nature of hydrodynamic mixing of the
two fluids affected the precipitation of the silica barrier, where the pH
gradient was controlled by the dispersion of silicate solution.
4.4. Formation of the silica barrier with different CO2 sources
The two CO2 sources used to test the formation of the barrier
showed a successful drop in permeability at reservoir conditions.
Results showed that the final permeability for the CO2-enriched water
source was one order of magnitude lower than the initial, while permeability dropped by three orders of magnitude when using scCO2 as a
source. The discrepancy between experiments can be explained by the
assumption that the silica barrier formation process was different for
each case. The reaction between the silicate solution and CO2–enriched
water was only driven by the drop in pH caused by the mixing between
the solutions, while the reaction with scCO2 was controlled by the
dissolution of CO2 in the alkaline silicate solution. Dissolution of CO2 is
accelerated by alkaline solutions as the dissolved CO2 transforms rapidly into carbonate ions allowing further uptake of CO2 into solution
(Takemura and Matsumoto, 2000). Therefore, the dissolution of CO2 in
the silicate solution would lead to a more rapid drop in pH, followed by
silica precipitation and permeability reduction.
4.2. Effect of temperature for the different scenarios
Experiments conducted at higher temperatures had a significant
difference in their setup where the column was placed horizontally in a
34
International Journal of Greenhouse Gas Control 78 (2018) 27–36
C.A. Castañeda-Herrera et al.
completely characterized using XRD analysis. Thus, there is still the
need to understand properties of the amorphous phase since different
gel structures with combination of amorphous silica might be forming
under variable conditions. Finding other or new techniques that could
give a proper structural analysis of amorphous phases would be ideal
for this purpose.
5. Conclusions
This paper studied the geochemical formation of an amorphous silica gel barrier in porous media, as a suitable technology for treating
CO2 leakage using mitigation and remediation strategies. The technology is based on the in-situ reaction between the leaking CO2 and a
commercial solution of sodium silicate to produce the amorphous silica
gel precipitate.
It was found that the effectiveness of the reaction in a column experiment will depend on the concentration gradient given by the mixing
front between the solutions. The mixing front can be also affected by
viscosity of the solutions. A disturbance of the mixing front might lead
to difficulties in forming the barrier because a redistribution of silica
might buffer the pH of the system. Under ambient conditions incubation periods were needed for the reaction to happen, where the first 24hour incubation was the most critical. Experiments testing mitigation
and remediation scenarios at different temperatures suggested that
temperature did not affect the reaction. Micro-CT imaging of gel formation in situ in ambient flow through columns showed that if silicate
solutions are acidified before injection a strong permeability barrier can
be formed by occluding the porosity of an unconsolidated sand. Here,
the formation of silica-gel remained stable over the course of aging for
24 days with no syneresis observed in situ, perhaps due to the column
remaining sealed during the aging process. Implementing the geochemical barrier in this fashion may be useful for targeted permeability
modification in a reservoir, for improving the sweep efficiency.
Injection of two CO2 sources demonstrated a successful barrier
formation at reservoir conditions by finding a significant drop in permeability. Reduction in the permeability of a Berea Sandstone core was
higher in experiments where scCO2 was injected. The higher reduction
in permeability when using scCO2 was attributed to the uptake of CO2
by alkaline solutions. It was also found that permeability values increased over the period of a month. The change in permeability may be
related to a syneresis effect. Nonetheless, permeability remained significantly lower than the initial values despite the silica transformation.
The significant increase in ΔP and reduction in permeability, supported by fluid chemistry and solid analysis, suggests that CO2 leakage
pathways could be completely blocked using sodium silicate solutions
as a reagent for barrier formation. However, further research is needed
before implementing this technology at field scale. Stability of the silica
barrier over time, CO2 leakage monitoring techniques and reactivetransport at field scale using a multiphase numerical simulator would
be necessary for the correct deployment of the technology.
Fig. 7. Pictures of Core 1 before experiment (left) and after 4 weeks of experiment (right). A white precipitate was observed on the surface of the stored
core.
The difference between experiments is also shown by the different
silica precipitation percentages, where Core 2 had almost twice the
amount of silica precipitate than Core 1 (Table 2). This result confirms
that the barrier formed using scCO2 is much more efficient than the one
formed with CO2-enriched water. In general, experiments injecting
scCO2 gave the best results because the dissolution of CO2 into the silicate solution was an effective process for forming the silica barrier.
The stability over time was also tested by measuring the permeability of Core 1 and Core 2 after 4 and 3.5 weeks, respectively. An
increase in permeability to approximately half of the initial value was
found for Core 1. This change was related to a white precipitate that
was observed on the surface of the core (Fig. 7). This observation
suggested that when the core was exposed to the atmosphere during
storage, the fluid started to dry out causing migration of the silica gel to
the surface of the core by capillary pressure. Therefore, the silica gel
migration caused the reconnection of the pore network and the increase
in permeability. Core 2 was wrapped with Parafilm during storage to
avoid dry-out as observed in Core 1. Indeed, a white precipitate on the
outside of the core was not observed after 3.5 weeks. Nevertheless,
permeability increased by one order of magnitude. This change in
permeability was still two orders of magnitude less than the original
rock permeability, which confirmed the strong barrier that was formed
when the silicate solution reacted with scCO2.
The increase in permeability in Core 2 after storage could be explained by two possible causes: Firstly, the silica gel could have redissolved in MilliQ water during the permeability test conducted after the
experiment. However, dissolution could be minimal since the pH of the
water was not alkaline. Secondly, the change in permeability could be
related to syneresis of the silica gel. Thus, it was possible that the silica
gel in the core effectively reduced its volume fraction over time and
transformed into a more consolidated form.
Acknowledgements
We acknowledge the Australian National Low Emissions Coal
Research and Development (ANLEC R&D) for providing funding for this
project. ANLEC R&D is supported by the Australian Coal Association
Low Emissions Technology Limited and the Australian Government
through the Clean Energy Initiative. The authors thank the CO2CRC
and the Peter Cook Centre for CCS Research for their support and
technical oversight of the project. Thanks for support from the Trace
Analysis for Chemical, Earth and Environmental Sciences (TrACEES)
platform from the Melbourne Collaborative Research Infrastructure
Program (MCRIP) at the University of Melbourne. Acknowledgements
to The David Lachlan Hay Memorial Fund for the financial support on
the writing of this manuscript. We would also like to thank the reviewers of this paper for their valuable and constructive comments.
4.5. Amorphous solid content analysis
The formation of the silica barrier was also confirmed by XRD
analysis for the experiments using the unconsolidated acid-washed sand
and the Berea Sandstone core. Both cases had a significant increase of
amorphous phase after experiment, indicating that this increment was
due to the silica gel precipitation. The increase of amorphous phase
from 12% to 35% for the Berea core sample is in agreement with reported changes in permeability and the derived silica mass balance.
However, the structure of the formed amorphous phases could not be
35
International Journal of Greenhouse Gas Control 78 (2018) 27–36
C.A. Castañeda-Herrera et al.
Appendix A. Supplementary data
Iler, R.K., 1982. Colloidal Components in Solutions of Sodium Silicate. ACS Publications.
IPCC, 2005. In: Metz, B., Davidson, O.R., Coninck, H.d., Meyer, L.M. (Eds.), Special
Report on CO2 Capture and Storage, Cambridge, UK, p. 431.
Ito, T., Xu, T., Tanaka, H., Taniuchi, Y., Okamoto, A., 2014. Possibility to remedy CO2
leakage from geological reservoir using CO2 reactive grout. Int. J. Greenh. Gas
Control 20, 310–323.
Krevor, S.C.M., Pini, R., Zuo, L., Benson, S.M., 2012. Relative permeability and trapping
of CO2 and water in sandstone rocks at reservoir conditions. Water Resour. Res. 48.
Kutchko, B.G., Strazisar, B.R., Dzombak, D.A., Lowry, G.V., Thauiow, N., 2007.
Degradation of well cement by CO2 under geologic sequestration conditions. Environ.
Sci. Technol. 41, 4787–4792.
Lakatos, I.J., Medic, B., Jovicic, D.V., Basic, I., Lakatos-Szabo, J., 2009. Prevention of
vertical gas flow in a collapsed well using silicate/polymer/urea method, SPE
International Symposium on Oilfield Chemistry. Old Spec. J.
Luhmann, A.J., Tutolo, B.M., Bagley, B.C., Mildner, D.F., Seyfried, W.E., Saar, M.O., 2017.
Permeability, porosity, and mineral surface area changes in basalt cores induced by
reactive transport of CO2‐rich brine. Water Resour. Res. 53, 1908–1927.
Manceau, J.C., Hatzignatiou, D.G., de Lary, L., Jensen, N.B., Réveillère, A., 2014.
Mitigation and remediation technologies and practices in case of undesired migration
of CO2 from a geological storage unit-current status. Int. J. Greenh. Gas Control. 22,
272–290.
Mitchell, A.C., Phillips, A.J., Hiebert, R., Gerlach, R., Spangler, L.H., Cunningham, A.B.,
2009. Biofilm enhanced geologic sequestration of supercritical CO2. Int. J. Greenh.
Gas Control 3, 90–99.
Pawar, R.J., Bromhal, G.S., Carey, J.W., Foxall, W., Korre, A., Ringrose, P.S., Tucker, O.,
Watson, M.N., White, J.A., 2015. Recent advances in risk assessment and risk management of geologic CO2 storage. Int. J. Greenh. Gas Control 40, 292–311.
Pentland, C.H., El-Maghraby, R., Iglauer, S., Blunt, M.J., 2011. Measurements of the capillary trapping of super-critical carbon dioxide in Berea sandstone. Geophys. Res.
Lett. 38.
Provis, J.L., Van Deventer, J.S.J., 2009. Geopolymers: Structures, Processing, Properties
and Industrial Applications. Woodhead Publishing Limited, Cambridge p. 454.
Réveillère, A., Rohmer, J., Manceau, J.C., 2012. Hydraulic barrier design and applicability for managing the risk of CO2 leakage from deep saline aquifers. Int. J. Greenh.
Gas Control. 9, 62–71.
Rothbaum, H.P., Rohde, A.G., 1979. Kinetics of silica polymerization and deposition from
dilute solutions between 5 and 180°C. J. Colloid Interface Sci. 71, 533–559.
Takemura, F., Matsumoto, Y., 2000. Dissolution rate of spherical carbon dioxide bubbles
in strong alkaline solutions. Chem. Eng. Sci. 55, 3907–3917.
Tognonvi, M.T., Rossignol, S., Bonnet, J.-P., 2011. Physical-chemistry of sodium silicate
gelation in an alkaline medium. J. Solgel Sci. Technol. 58, 625–635.
Tongwa, P., Nygaard, R., Blue, A., Bai, B., 2013. Evaluation of potential fracture-sealing
materials for remediating CO2 leakage pathways during CO2 sequestration. Int. J.
Greenh. Gas Control. 18, 128–138.
Vialle, S., Druhan, J.L., Maher, K., 2016. Multi-phase flow simulation of CO2 leakage
through a fractured caprock in response to mitigation strategies. Int. J. Greenh. Gas
Control 44, 11–25.
Weldes, H.H., Lange, K.R., 1969. Properties of soluble silicates. Ind. Eng. Chem. 61,
29–44.
Whitaker, S., 1986. Flow in porous media I: a theoretical derivation of Darcy’s law.
Transp. Porous Med. 1, 3–25.
Supplementary data associated with this article can be found, in the
online version, at https://doi.org/10.1016/j.ijggc.2018.07.013.
References
Aminzadeh, B., Chung, D.H., Bryant, S.L., Huh, C., DiCarlo, D.A., 2013. CO2 leakage
prevention by introducing engineered nanoparticles to the in-situ brine. Energy
Procedia 37, 5290–5297.
Andersson, K.R., Dent Glasser, L.S., Smith, D.N., 1982. Polymerization and colloid formation in silicate solutions, soluble silicates. Am. Chem. Soc. 115–131.
An-Pang, T., 1963. A theory of polymerization of silicic acid. Sci. Sin. 12, 1311–1320.
Brydie, J.R., Perkins, E.H., Fisher, D., Girard, M., Valencia, M., Olson, M., Rattray, T.,
2014. The development of a leak remediation technology for potential non- wellbore
related leaks from CO2 storage sites. Energy Procedia 63, 4601–4611.
Carey, J.W., 2013. Geochemistry of wellbore integrity in CO2 sequestration: Portland
cement-steel-brine-CO2 interactions. In: DePaolo, D.J., Cole, D.R., Navrotsky, A.,
Bourg, I.C. (Eds.), Geochemistry of Geologic CO2 Sequestration, pp. 505–539.
Castañeda-Herrera, C.A., Black, J.R., Stevens, G.W., Haese, R.R., 2017. Preliminary experiments for a chemical reactive barrier as a leakage mitigation technology. Energy
Procedia 114, 4140–4146.
Castañeda-Herrera, C.A., Stevens, G.W., Haese, R.R., 2018. Review of CO2 leakage mitigation and remediation technologies (in press) In: Vialle, S., Ajo-Franklin, J.B., Carey,
J.W. (Eds.), Caprock Integrity in Geological Storage (Tentative Title). Geophysical
Monograph Series. American Geophysical Union and John Wiley & Sons, Inc..
Crerar, D.A., Axtmann, E.V., Axtmann, R.C., 1981. Growth and ripening of silica polymers
in aqueous solutions. Geochim. Cosmochim. Acta 45, 1259–1266.
Cunningham, A.B., Gerlach, R., Spangler, L., Mitchell, A.C., 2009. Microbially enhanced
geologic containment of sequestered supercritical CO2. Energy Procedia 1,
3245–3252.
Cunningham, A.B., Lauchnor, E., Eldring, J., Esposito, R., Mitchell, A.C., Ebigbo, A.,
Spangler, L.H., Gerlach, R., Phillips, A.J., 2013. Abandoned well CO2 leakage mitigation using biologically induced mineralization: current progress and future directions. Greenh. Gases Sci. Technol. 3, 40–49.
DCI (Ed.), 2015. Handbook: Operation and Maintenance University of Melbourne coreFlood System. DCI Salt Lake City, UT, US.
Druhan, J.L., Vialle, S., Maher, K., Benson, S., 2014. A reactive transport model for
geochemical mitigation of CO2 leaking into a confined aquifer. Energy Procedia 63,
4620–4629.
Druhan, J.L., Vialle, S., Maher, K., Benson, S., 2015. Numerical simulation of reactive
barrier emplacement to control CO2 migration. In: Gerdes, K.F. (Ed.), Carbon Dioxide
Capture for Storage in Deep Geologic Formations: Results from the CO2 Capture
Project. CPL Press.
Fleury, M., Sissmann, O., Brosse, E., Chardin, M., 2017. A silicate based process for
plugging the near well bore formation. Energy Procedia 114, 4172–4187.
Greenberg, S.A., Sinclair, D., 1955. The polymerization of silicic acid. J. Phys. Chem. 59,
435–440.
Hamouda, A.A., Amiri, H.A.A., 2014. Factors affecting alkaline sodium silicate gelation
for in-depth reservoir profile modification. Energies 7, 568–590.
36
Документ
Категория
Без категории
Просмотров
0
Размер файла
1 651 Кб
Теги
013, ijggc, 2018
1/--страниц
Пожаловаться на содержимое документа