Influence of Supercritical CO 2 on the Mobility and Desorption of Trace Elements from CO 2 Storage Rock Sandstone and Caprock Shale in a Potential CO 2 Sequestration Site in Taiwan

Excess carbon dioxide emission was considered as the most important cause of increased trend of global warming. Significant amounts of research were devoted to the reduction of CO2 emission and CO2 sequestration. Sequestration of CO2 in empty oil reservoirs was considered as one of the most promising options. However, the effects of metal release and mobility as a result of CO2 injection were not studied in detail, particularly under super-critical CO2 (scCO2) conditions. In this study, the release of selected metals immersed in distilled water for varying amounts of time in the absence and presence of scCO2 was assessed in simulated conditions at 90°C and 24 MPa. Significant increases in dissolution of Fe, Sr, and Ba by 3, 8, and 24 times were found when the storage rock sandstone or caprock shale was immersed in DI water for different time period. However, in the presence of scCO2, the dissolution of these metals was reduced by 80% for Fe, suggesting permanent sequestration of scCO2 into carbonate minerals. The trend in changes of pore water chemistry in the sandstone and shale after being immersed in DI water showed dissolution of Sr-bearing mineral and precipitation of Ba-bearing mineral.


INTRODUCTION
Carbon dioxide contents in the atmosphere had a strong connection with the global temperature as shown by the Antarctic ice core data covering the past 400,000 years (Oelkers and Cole, 2008).In Taiwan alone, the annual CO 2 emissions are about 200 million tons, of which about 90 million tons will be geo-sequestrated in order to meet the government's green initiatives (Sung et al., 2014).As such, reducing the impact of CO 2 emissions to the atmosphere was one of the most urgent needs and challenges (Oelkers and Schott, 2005).For Taiwan, its carbon emission reduction targets for 2050 were set to 50-100% of its 2000 levels, or 79-89% below the business as usual levels (Tsai and Chang, 2015).
There are several approaches to reduce CO 2 input to the atmosphere, among them carbon capture and storage (CCS) is the most promising.CCS refers to a number of technologies risks from many aspects such as technology, health, safety and environment, market, energy and resources, as well as policy and regulations (Xuan, 2014).In Taiwan, the Changhua Coastal Industrial Park site, located in the center of the west coast was viewed as a good potential storage site because of its proximity to massive stationary sources of CO 2 emissions from nearby petrochemical and coalfired power plants (Yu et al., 2011).However, leakage from subsurface storage sites is one of the main concerns of the CCS technology (Peter et al., 2012).
In addition to EOR, there are growing interests in CCS in saline aquifers due to their large potential storage capacities (Soong et al., 2014).As such, several projects were in the pipeline demonstrating its viability (Leung et al., 2014).Another area of interests in CCS studies was via permanent CO 2 sequestration, in which CO 2 is chemically stored in solid carbonates by carbonation of minerals (Huijgen and Comans, 2003;Chang et al., 2012;Pan et al., 2015).Direct mineral carbonation was investigated as a process to convert gaseous CO 2 into a geologically stable, solid final form utilizing a solution of sodium bicarbonate, sodium chloride, and water, mixed with a mineral reactant, such as olivine or serpentine (O'Connor et al., 2001).
With regard to the release and mobility of trace elements associated with CCS, most tests were conducted at the batch and column scales (Little and Jackson, 2010;Lu et al., 2010;Soong et al., 2014).Leaching tests were conducted to study and assess the mechanisms and efficiency of metal release processes from solids in contact with liquids to simulate the consequences of CO 2 leakages from CCS (Ardelan and Steinnes, 2010;de Orte et al., 2014).CO 2 seepage caused a significant increase in the total and dissolved Fe, Mn, and Co concentrations (30, 40, 50 times for the total and 20, 30, 40 times for the dissolved fractions, respectively) during the early phase (Ardelan and Steinnes, 2010).Major increases in the concentrations of Ca, Mg, Fe, and Mn, following injection of food-grade CO 2 at the Zero Emission Research and Technology field site, Bozeman, Montana were also observed in laboratory studies (Kharaka et al., 2010).However, more studies on CO 2 -water-rock interactions induced by CO 2 injection were needed, particularly under the elevated temperature and pressure conditions (Bertier et al., 2006).
Although the technology of CCS is currently considered to be technically feasible, information on the environmental impacts of accidental leakage is still lacking (de Orte et al., 2014).Moreover, information on caprock reactivity with CO 2 is limited compared with those on the reservoir rocks and long term containment of stored CO 2 in deep geological reservoirs will also depend on the performance of the caprock to prevent the buoyant CO 2 from escaping to shallow drinking water aquifers or the ground surface (Liu et al., 2012).As one of the important clay minerals for caprocks, kaolinite could adsorb up to 3.4 mg CO 2 g -1 after acid treatment (Chen and Lu, 2015).Thus, the goals of this study were two folds: (1) to contrast the differences in metal dissolution in the CCS sandstone formation and in the CCS caprock shale in the absence and presence of supercritical fluid CO 2 (scCO 2 ); and (2) to illustrate possible mechanisms associated with the changes in metal compositions in the formation rocks and formation water in the presence of scCO 2 .We hope that the results will add more valuable data to the assessment of the feasibility of CCS in sandstone formations after EOR.

GEOLOGIC SETTING
The potential CO 2 storage site is located on the Yunghoshan-Chinshui structure, which is in the foothills belt in northwestern Taiwan.On the surface, the foothills belt is characterized by several anticlines that are separated by thrust faults (Fig. 1(a)).In the eastern part of the belt, the trend of the anticlines with the associated thrust faults, including the Paoshan (Hsincheng thrust, HCNF), Yunghoshan-Chinshui (Luchukeng thrust, LCKF) anticlinal structures, are mostly NNE-SSW, parallel to the general trend of the mountain range to the east (Fig. 1(a)).In the western part, where it is adjacent to the coastal line, the trends of the anticlines, including Chingtsaohu, Chiting, Chunan, Paishatun, and Sanhu anticlines, are NEE-SWW, parallel to the faults trending at high angle to the mountain belt.
The Chinshui structure can be viewed as the southern extending part of the Yunghoshan structure, but is cut off by a NEE-SWW trending high angle faults to its north (Fig. 1(a)).On the surface, the outcropped formation along its fold axes is the Chinshui Shale (Pch).In the subsurface, the structure is characterized by an asymmetric anticline, with western limb steeper than the eastern one (Fig. 1(b)).The western limb of the anticline is also cut off by a smallscale normal fault (Fig. 1(b)).
The target reservoirs considered for possible future CO 2 storage are the sandstone layers (named as Talu Sandstone, Mtlss) in the lower part of the Talu Formation (Mtl) and its underlying 363 meter-thick Peiliao Sandstone (Mpl) at depth of 2279-2642 meters (Figs. 1(b) meters (Figs. 1(b) and 1(c)), which belongs to the lower Middle Miocene (~14 m.y.bp).The Talu Sandstone is about 50 meters thick at the depth of 2300 m, which is similar in depth at 2450 m for the first China EOR demonstration project (Su et al., 2013).It contains one thin layer of shale dividing the sandstone into two separated layers, and is composed of mid-to-coarse grained wellcompacted sandstones.The Talu Sandstone is capped by the Talu Shale of more than 250 meter thick (Figs.1(b) and 1(c)), which is mainly composed by dark-grey to black-grey shales and sandy shales, with thin interlayers of muddy sandstones.

Sample Collection
Sandstone (depth = 2292-2333.3m) and shale (depth = 2279-2280 m) samples were collected from the gas well CS59 (Fig. 1) on May 1, 2014 from the Geology Core Repository of China Petroleum Corporation, Miaoli, Taiwan.Both the formation water and rock samples were shipped to the Hydrogeology Laboratory, National Cheng Kung University, within five hours after collection.
The formation water was sampled at depth of 2292- 2333.3 m on June 13, 2014 from an oil and water separation tank at the gas well CS59 (Fig. 1) and stored in a 7 L vacuum tank under anaerobic conditions.The physicochemical parameters of the formation water were measured on-site.The salinity, electrical conductivity (EC), and total dissolved solids (TDS) were measured using a portable conductivity meter (Suntex: 162 WTW, LF30, Kaohsiung, Taiwan).The pH and redox potential (ORP) were measured using a redox meter (Suntex: SP701, Kaohsiung, Taiwan).

Experiments
For the metal releasing experiments in the absence of scCO 2 , 5 g of rock was immersed in 50 mL of deionized (DI) water for 10, 20, 30, and 40 d.To further investigate the influence of scCO 2 in metal mobility and desorption, 5 g of rock was immersed in 50 mL DI water for 40, 60, 80, and 100 d after which, the mixtures were placed into a highpressure autoclave (filled with 12 L DI water to maintain the desired pressure) (Fig. 2) for 20 d in the presence of scCO 2 .The temperature and pressure setting was 90°C and 24 MPa (240 bars) to simulate the field condition corresponding to the injection depth.
After each treatment, 1 g of rock fragment (without grinding) was placed into 10 mL of 0.1 M HNO 3 and immersed for 24 h without agitation to monitor the release of trace elements under static conditions.

Chemical and Instrumental Analyses
The trace element concentrations of the formation water after in contact with the rocks and the extractants after the rock being in contact with HNO 3 were analyzed using an ICP-MS (Aligent 7500cs).Proper dilutions were made to bring the final concentrations into the working concentration range.The detection limit for all trace elements was < 20 ng L -1 , except for Se and Zn (50-100 ng L -1 ).A 10 mg L -1 multi-element standard solution (High-Purity, USA) was diluted to proper element concentrations to establish calibration curves.The accuracy and precision were determined to be within ± 5% for most elements by running a secondary standard (J T Baker, USA).
For powder X-ray diffraction analyses bulk sandstone and shale specimens were grounded to powder that was finer than sieve #375, from which 0.5 grams were sampled and uniformly placed on the sample holder.The measurements were performed on a Dandong Fangyuan DX-2700 X-ray diffractometer (Liaoning, China) with a Sol-X solid state detector and CuKα (λ = 1.5406Å) radiation at 35 kV and 30 mA.Samples were scanned from 5-80° 2θ with a scanning step of 0.04° per step.After the powder specimens were added with 1 wt% polyvinyl alcohol, 12 g of the specimens were pressed under 80 MPa to a tablet (12 mm in diameter), which was placed in a sample holder and then measured for oxides using X-ray fluorescence (XRF) (ZSX100e, Rigaku, USA).The bulk sandstone/shale slices were placed on the Campton tape and observed under a scanning electron microscopy (SU1510, Hitachi, Japan) and analyzed for trace elements on the surface of sandstone and shale using EDS (XFLASH-Detector 6/10, Bruker, Germany).

Physicochemical Parameters of the Formation Water
The pH of the formation water was 6.3, while the ORP and dissolved oxygen (DO) were -5 mV and 0.6 mg L -1 , respectively, reflecting a moderate reducing environment.The salinity, TDS, EC, and water temperature were 26.3‰, 24,400 mg L -1 , 42.7 mS cm -1 , and 21.1°C, respectively.

Leaching and Release of Metals from Storage Rock Sandstone
The release of minor elements and heavy metals after the sandstone samples were immersed in DI water for 10, 20, 30, and 40 days is presented in Fig. 3(a).Iron is the most released element with concentrations in the range of 245 to 433 mg L -1 in comparison to 274 mg L -1 for the sandstone without soaking in formation water.The high Fe concentrations suggested the presence of Fe(II) in the sandstone formation.The Mn concentrations were in the range of 4.3 to 6.6 mg L -1 after the sandstone was immersed in DI water, in comparison to 5.5 mg L -1 for the raw sandstone.For Zn, its concentrations varied from 1.4 to 1.8 mg L -1 , while it was 1.9 mg L -1 for the original sandstone.Other elements were in the microgram per liter ranges and did not show significant changes after being immersed in DI water except Sr and Ba.The Sr and Ba concentrations were 5.1 and 2.5 mg L -1 , respectively for the original sandstone.Their concentrations increased up to 50 and 17.6 mg L -1 , respectively, after in contact with DI water for 30 d, (Fig. 3(a)).The 10-and 6-fold increase in Sr and Ba concentrations suggests the presence of water-soluble mineral species associated with these elements in the sandstone storage rock.
In comparison to the rock samples immersed in DI water, the extended immersion of samples in DI water at 40, 60, 80, and 100 d followed by equilibration with scCO 2 for 20 d showed drastic different dissolution behaviors for many elements (Fig. 3(b)).The Fe concentrations deceased to 50 to 161 mg L -1 , in comparison to 274 mg L -1 for the sandstone without DI water immersion.Similarly, Mn concentrations decreased to 0.4 to 1.6 mg L -1 , in contrast to 5.5 mg L -1 for the raw sandstone.And the Zn concentrations decreased to 0.3-0.9mg L -1 compared to 1.9 mg L -1 for the raw sandstone.Most importantly, significant decreases of Sr and Ba concentrations (in the ranges of 6.4-19.5 mg L -1 for Sr and 4.9-12.4mg L -1 for Ba) were found in the presence of scCO 2 in comparison to 17.0-50.2mg L -1 for Sr and 12.1-17.6mg L -1 for Ba for samples in the absence of scCO 2 , although their concentrations are still higher than those of 5.1 and 2.5 mg L -1 for raw sandstone.These results indicated that the presence of scCO 2 would decrease the concentrations of Fe, Mn, and Zn significantly by forming mineral precipitates such as siderite, rhodochrosite, and smithonite, thus, reducing the desorption and mobility of these metals on one hand.Meanwhile, the formation of these carbonate minerals will be able to permanently sequestrate the injected CO 2 on the other hand.These dual features should add valued assets to the scCO 2 sequestration in formations where significant amounts of Fe(II), Mn(II), and Zn were present under a reducing environment.Similarly an initial increase in Fe concentration was observed following the CO 2 breakthrough in a pilot study of CO 2 injection into the Frio Brine Pilot field (Xu et al., 2010).Subsequently, observed aqueous Fe concentrations decreased due to carbonate precipitation and were confirmed by a reactive transport modeling simulation (Xu et al., 2010).

Leaching and Release of Metals from Caprock Shale
In comparison to sandstone, the leaching and release of metals from caprock shale are somehow slightly different (Fig. 4(a)).For Fe, its concentrations varied in the range of 231 to 881 mg L -1 after 10-40 d of immersion in DI water, in comparison to 292 mg L -1 for the shale without DI immersion.The extracted Mn concentrations were much higher, in the ranges of 124 to 281 mg L -1 after the shale was immersed in DI water, in comparison to 120 mg L -1 for the raw shale.The Zn concentrations released varied from 2.2 to 3.0 mg L -1 after DI water soaking, in comparison to 2.9 mg L -1 for the raw shale.The significant increase in Mn could be attributed to desorption of Mn from clay minerals of the shale.Again, the large amounts of Fe and Mn release may signal a reducing environment under which Fe(II) and Mn(II) would dominate.Similar to Mn, the release of Sr was much higher, varying from 52 to 93 mg L -1 after the shale being immersed in DI water in comparison to 12.7 mg L -1 for the raw shale.The Ba concentration was only 1.1 mg L -1 for the raw shale and it increased to 16-27 mg L -1 after being immersed in DI water for up to 40 d (Fig. 4(a)).These results suggested the presence of more water-soluble mineral species in the caprock shale in comparison to the sandstone or more adsorbed Sr and Ba on the surface of clay minerals that desorbed into the solution after being immersed in DI water.
Similar to the case of sandstone, in the presence of scCO 2 , the Sr release increased from 12.7 mg L -1 for the raw shale to 26-41 mg L -1 after being immersed for 40-100 d and then equilibrated with scCO 2 for 20 more d (Fig. 4(b)).These values are two to four folds smaller than the values of 52-93 mg L -1 in the absence of scCO 2 .Similarly, the Ba concentration was 1.1 mg L -1 in the absence of water immersion, and it increased to 6.5-14 mg L -1 after being immersed in DI water for 40-100 d and then equilibrated with scCO 2 for 20 more d (Fig. 4(b)).These values again showed two to four folds reduction in Ba release in comparison to 16-27 mg L -1 when the samples were immersed in DI water for 10-40 d.A drastic reduction in Mn concentrations (10-138 mg L -1 for the samples with scCO 2 vs. 124 to 281 mg L -1 for the samples immersed in DI water only) also indicated reduced mobility for Mn in the presence of scCO 2 .Similar to the sandstone, the introduction of scCO 2 to the caprock shale could also permanently sequestrate CO 2 into mineral phases such as rhodochrosite.In contrast, when CO 2 seepage was in effect, the distribution of Fe, Mn, and Co in seawater was changed and these metals were mobilized and transported from the sediment to the overlying water both as dissolved and suspended particulate forms of the metals (Ardelan and Steinnes, 2010).

Changes in Pore Water chemistry of Storage Rock Sandstone
The trend in changes of pore water chemistry in the sandstone is different from that in sandstone immersing solution.The trace elements in the pore water were Sr followed by Ba with concentrations of 1.6 and 7.4 mg L -1 for the original formation water, respectively.After the sandstone was immersed in DI water for 10-40 d in the absence of scCO 2 , the Sr concentrations in formation water increased to 16.3-17.8mg L -1 , while the Ba concentrations decreased to 1.8-2.0mg L -1 (Fig. 5(a)).After the sandstone was immersed in DI water for 40-100 d and then in scCO 2 for 20 d, the Sr concentrations varied between 10.5 and 29.2 mg L -1 , while the Ba concentrations were in the range of 1.6-2.9mg L -1 (Fig. 5(b)).These results are in agreement with those in the absence of scCO 2 , suggesting dissolution of Sr-bearing mineral and precipitation of Ba-bearing mineral.In a different study, the Ba concentration increased up to 2 orders of magnitude after in contact with gaseous CO 2 at atmospheric pressure over 100 d (Little and Jackson, 2010).

Changes in Pore Water Chemistry of Caprock Shale
The trend in changes of pore water chemistry of the shale was similar to that of sandstone.The Sr concentrations increased to 13.0-13.7 mg L -1 while Ba concentrations decreased to 0.2-0.8mg L -1 after the shale was immersed in DI water for 10-40 d in the absence of scCO 2 (Fig. 6(a)).After the shale was immersed in DI water for 40-100 d and then in scCO 2 for 20 d, the Sr concentrations varied between 12.3 and 26.5 mg L -1 , while the Ba concentrations were in the range of 0.1-0.6 mg L -1 (Fig. 6(b)).Again, the results suggested that dissolution of Sr-bearing mineral and precipitation of Ba-bearing mineral also occurred in shale.The Rb concentrations varied from 0.1 to 0.5 mg L -1 for shale and 0.2 to 0.6 mg L -1 for sandstone after being immersed in DI water in the absence or presence of scCO 2 for 10-100 d, in comparison to 0.5 mg L -1 for the Rb concentration in the formation water.These results suggest the alkali and alkaline earth elements contributed significantly to the trace element chemistry of the formation water.
To confirm the contribution of alkali and alkaline earth elements to the formation water chemistry, major elements were analyzed by ICP-MS.For the formation water, the Na, Mg, K, and Ca concentrations were 11000, 46, 155, and 47 mg L -1 , respectively.And the changes in their concentrations were in the ranges of 10600-14700, 45-66, 157-207, and 13-64 mg L -1 for the sandstone and 10000-13500, 66-136, 144-193, and 8-86 mg L -1 for the shale.These results suggested that the water is in a sodium type.Similarly, a slight increase in release of major cations accompanied with decrease in major anions was also observed after 10 d injection of CO 2 into a shallow aquifer mainly made of medium sand (Peter et al., 2012).However, laboratory incubations of CO 2 infiltration under oxidizing conditions for > 300 d on samples collected from four freshwater aquifers showed 2 orders of magnitude increase in concentrations of the alkali and alkaline earths, manganese, cobalt, nickel, and iron (Little and Jackson, 2010).

Changes in Mineral Phases
The XRD analyses showed quartz as the major mineral phase in the sandstone with trace amount of microcline (Fig. 7).In contrast, significant amounts of illite and chlorite were present in the shale in addition to quartz (Fig. 8).The SEM and EDS analyses also confirmed the presence of chlorite in the shale formation (Fig. 9).Chlorite is an important primary mineral in caprock shale for CO 2 sequestration.It can be readily dissolved in formation water after reacting with scCO 2 (Xu et al., 2001).The main source of Mn and Fe besides the carbonates is chlorite.Chlorite is also among the most dissolving minerals present in the caprock shale.As the dissolution was minute in comparison to the sensitivity of XRD analyses, the changes in mineral phases before and after the rock samples were in contact with scCO 2 were hardly seen in the XRD analyses (Liu et al., 2012).Amorphous silica was the most abundant secondary mineral in the dissolution and precipitation of CO 2 -brinepholopite system (Shao et al., 2010).However, in order to confirm the change of the chlorite content after caprocks reacting with scCO 2 , the ratios of the integrated XRD peak intensity of chlorite (2θ = 12.4°) to that of quartz (2θ = 26.5°)for the caprock shales are compared (Table 1).It indicated that the amount of chlorite in the caprocks after reacting with scCO 2 decreased in comparison to the amount  of chlorite in the samples immersed in DI water only.These results suggest that the significant increase in Fe and Mn after immersing in DI water and reacting with scCO 2 could be attributed to the dissolution of chlorite.

PHREEQC Simulation on Rock Water Interactions
Simulation using the chemistry data from formation water showed precipitation of aragonite, calcite, dolomite, goethite, hematite, hausmannite, pyrolusite, stronitainite, and witherite.Meanwhile, rhodochrosite, siderite, smithsonite showed a dissolution feature (Table 2).For the samples that were immersed in DI water, the witherite is not stable and goes into dissolution.However, precipitate of witherite would occur in the presence of scCO 2 (Table 2).The trend in witherite dissolution and precipitation agreed well with the experimental data plotted in Figs.3-6.

ENVIRONMENTAL IMPLICATIONS
The results from this study showed that the injection of scCO 2 could provide an environment for permanent scCO 2 sequestration as revealed by drastic decrease in concentrations of metals that could form carbonate minerals in the presence of scCO 2 .
In a previous study, two types of cations were recognized according to their trends in changes of concentrations with respect to CO 2 flux (Lu et al., 2010).The first type includes Ca, Mg, Si, K, Sr, Mn, Ba, Co, B, Zn, and their concentrations increased rapidly after initial CO 2 flux and reached stable concentrations before the end of the experiment, which was attributed to dissolution of dolomite and calcite (Lu et al., 2010).The second type includes Fe, Al, Mo, U, V, As, Cr, Cs, Rb, Ni and Cu, with their concentrations increased at the beginning of CO 2 flux, and then declined to the levels lower than pre-CO 2 flux concentrations, in most cases (Lu et al., 2010).Reaction path and kinetic models indicate that geochemical shifts caused by CO 2 leakage are closely linked to mineralogical properties of the receiving aquifer (Wilkin and Digiulio, 2010).In this study, the opposite trend in cation concentrations after in contact with scCO 2 could be due to the absence of the carbonates in the formation rock sandstone and caprock shale, As the concentrations of the elements Ca, Mg, Sr, Ba, Mn, and U are controlled by carbonate dissolution (Mickler et al., 2013).Therefore, the release and mobility of metals after CO 2 injection is site-specific and use of universal rules to judge the release of metals accompanying CCS should be with caution.
As studied before, most potable groundwater samples in the United States were reducing, under which minerals containing many hazardous trace elements were stable (Apps et al., 2010).As their finding was controversial, while the authors were confident that the supporting evidence was strong, thus, further studies to test their claims were needed (Apps et al., 2010).This study provided part of the evidences to support their simulation results.

Fig. 1 .
Fig. 1.(a) Regional geological map of northwestern Taiwan (from Chinese Petroleum Corporation, 1994).The highlighted rectangle is located on the southwestern part of the Yunghoshan-Chinshui structure.The black dot in (a) shows the location of well CS-59 from which rock and groundwater samples were collected.(b) Subsurface geological profile of the Chinshui structure.(c) Lithofacies of the Talu Formation and Peiliao Sandstone shown by spontaneous potential and resistivity electric logs of well C-59.The 363-meter thick Peiliao Sandstone, the potential reservoir for CO 2 storage, is capped by the 220-meter thick Talu Shale.

Fig. 2 .
Fig. 2. Experimental setup for dissolution of trace elements from sandstone and shale.

Fig. 3 .
Fig. 3. Changes in minor element contents of sandstone before (a) and after (b) in contact with CO 2 .The insert is for Fe.The y-axis is in µg L -1 .

Fig. 4 .
Fig. 4. Changes in minor element contents of shale before (a) and after (b) in contact with CO 2 .The insert is for Fe.The yaxis is in µg L -1 .

Fig. 5 .Fig. 6 .
Fig. 5. Changes in minor element contents of pore water of sandstone before (a) and after (b) in contact with CO 2 .The yaxis is in µg L -1 .
Fig. 7. X-Ray diffraction patterns of sandstone before and after in contact with CO 2 .

Fig. 9 .
Fig. 9.The presence of chlorite in shale as revealed by SEM observation and EDS elemental analyses.

Table 1 .
Ratios of the integrated XRD peak intensity of chlorite (2θ = 12.4°) to that of quartz (2θ = 26.5°)for the caprock shales after different days of immersion in DI water.

Table 2 .
Results from PHREEQC simulation based on the water chemistry from samples of the original formation water, samples immersed in DI water, and samples immersed in DI water followed by scCO 2 .