Experimental Investigation and Simplistic Geochemical Modeling of CO2 Mineral Carbonation Using the Mount Tawai Peridotite

In this work, the potential of CO2 mineral carbonation of brucite (Mg(OH)2) derived from the Mount Tawai peridotite (forsterite based (Mg)2SiO4) to produce thermodynamically stable magnesium carbonate (MgCO3) was evaluated. The effect of three main factors (reaction temperature, particle size, and water vapor) were investigated in a sequence of experiments consisting of aqueous acid leaching, evaporation to dryness of the slurry mass, and then gas-solid carbonation under pressurized CO2. The maximum amount of Mg converted to MgCO3 is ~99%, which occurred at temperatures between 150 and 175 °C. It was also found that the reduction of particle size range from >200 to <75 µm enhanced the leaching rate significantly. In addition, the results showed the essential role of water vapor in promoting effective carbonation. By increasing water vapor concentration from 5 to 10 vol %, the mineral carbonation rate increased by 30%. This work has also numerically modeled the process by which CO2 gas may be sequestered, by reaction with forsterite in the presence of moisture. In both experimental analysis and geochemical modeling, the results showed that the reaction is favored and of high yield; going almost to completion (within about one year) with the bulk of the carbon partitioning into magnesite and that very little remains in solution.


Introduction
Carbon dioxide (CO 2 ) is the principal greenhouse gas released into the atmosphere during fuel combustion, particularly due to the extensive use of fossil fuels for energy production from coal, oil, and natural gas since the industrial revolution [1]. Moreover, atmospheric CO 2 has recently surpassed 400 ppm, and is predicted to increase to nearly 1000 ppm by the end of the 21st Century [2,3]. The related global temperature rise (exceeding 2˝C) [3] will almost certainly result in irreversible climate change with potentially disastrous consequences.
Researchers have studied ways to mitigate the amount of greenhouse gases released into the atmosphere by sequestration of CO 2 through different approaches, including aquifer storage, deep sea storage, and mineral carbonation in particular [1,[4][5][6][7][8][9][10][11][12][13][14]. Minerals and rocks rich in magnesium (Mg 2+ )/calcium (Ca 2+ ) are commonly considered as candidates due to their wide availability, low cost, and environmentally benign nature. [8,12,15] During mineral carbonation, CO 2 reacts with Mg 2+ or Ca 2+ -rich minerals (e.g., olivine and gypsum) to form solid carbonates, which are expected to be stable over geologic time periods. The minerals olivine [(Mg,Fe) 2 SiO 4 ] and forsterite [Mg 2 SiO 4 ], containing up to 33.6 wt. % Mg, have the highest capacity to trap CO 2 as magnesium carbonates, and a high rate of dissolution among rock-forming silicate minerals [16]. Formation of magnesite from forsterite, the Mg-end member of the olivine solid solution series, is thermodynamically favorable based on the negative Gibbs free energy of Reaction (1). 1{2Mg 2 SiO 4`C O 2 ÑMgCO 3`1 {2SiO 2`9 5 kJ{mol (Rx. 1) According to Lackner et al. [17] the chemical reactions in mineral carbonation process can be very slow under ambient conditions and, therefore, activation processes such as exposure to acid, heat [8], and water [18][19][20][21] have been used to accelerate the carbonation rate. These processes are tedious and require additional energy input, which has made acid/heating approach less attractive than some others. The rates of carbonation can be raised by increasing surface areas of the mineral or its intermediates and by elevating the temperatures, resulting in lower kinetic constraints [15,16,[22][23][24][25][26][27][28][29]. Moreover, an in situ CO 2 sequestration system with mineral carbonation can be treated as a fully-coupled problem between rock deformation, pore-fluid flow, heat transfer, mass transport, and chemical reaction processes. Three types of models are commonly employed in computational petroleum geoscience and engineering research methodologies. These approaches are geological/geochemical (conceptual), mathematical, and/or numerical simulation.
The purpose of this study was to do ex situ (laboratory) studies of the factors affecting the rates of (forsterite) mineral carbonation (i.e., particle size, water vapor, and reaction temperature) in support of an in situ geochemical conceptual model. A transport reaction modeling software PHREEQC (version 2.18, US Geological Survey (USGS), Reston, WV, USA), was applied for simulating the chemical reactions and transport processes in the forsterite mineral carbonation process. Pre-treatment of forsterite in a quantitative equivalent of mineral acid (HCl), i.e., dissolution to neutrality, was taken as a suitably fast process (in extracting most of the Mg ion) for practical experimentation. One advantage of this process is that the formation of magnesium carbonate (MgCO 3 ) releases heat, which can in principle be cycled back to other endothermic steps (see Reactions (2)-(4) in Section 3.3) through heat integration in a commercial process.

Mineral Characterization
The elemental composition of the peridotite mineral was determined by XRF analysis as MgO (51.9%), SiO 2 (41.1%) as major components with minor levels of FeO, Al 2 O 3 , Na 2 O, K 2 O, and CaO (see details in Table 1).  Morphological and structural changes in the olivine (nominal composition Mg1.84Fe0.16SiO4) at various stages during chemical pretreatment and carbonation, viz., after leaching in HCl, neutralization and precipitation of Mg 2+ [as Mg(OH)2], and exposure to humid CO2 at 4.8 bar and 150 °C, are best seen by SEM micrographs and powder XRD in Figures 2 and 3, respectively.
By SEM, the fresh olivine sample (Figure 2a) consisted typically of polycrystalline grains in the millimeter size range. After acid leaching and neutralization of the sieved fraction <75 µm (Figure 2b), the grains became finer (<5 µm) and the structure more amorphous. In agreement with the findings of Bearat et al. [30] and Kwon et al. [18], this is likely due to an amorphous SiO2 residue after leaching, any (hexagonal) brucite likely being nanocrystalline. Under humid CO2, the development of polyhedral and more sheet-like crystallites (Figure 2c) was evident, probably representing (hydrated) magnesium carbonate, and finally (Figure 2d), the rhombohedral habit typical of magnesite (MgCO3) was evident. Morphological and structural changes in the olivine (nominal composition Mg 1.84 Fe 0. 16 SiO 4 ) at various stages during chemical pretreatment and carbonation, viz., after leaching in HCl, neutralization and precipitation of Mg 2+ [as Mg(OH) 2 ], and exposure to humid CO 2 at 4.8 bar and 150˝C, are best seen by SEM micrographs and powder XRD in Figures 2 and 3 respectively.
By SEM, the fresh olivine sample (Figure 2a) consisted typically of polycrystalline grains in the millimeter size range. After acid leaching and neutralization of the sieved fraction <75 µm (Figure 2b), the grains became finer (<5 µm) and the structure more amorphous. In agreement with the findings of Bearat et al. [30] and Kwon et al. [18], this is likely due to an amorphous SiO 2 residue after leaching, any (hexagonal) brucite likely being nanocrystalline. Under humid CO 2 , the development of polyhedral and more sheet-like crystallites ( Figure 2c) was evident, probably representing (hydrated) magnesium carbonate, and finally (Figure 2d), the rhombohedral habit typical of magnesite (MgCO 3 ) was evident. XRD analysis of the HCl-cleaned starting material ( Figure 1) shows a characteristic pattern of olivine (2θ = 11.8°, 23.6°, 29.3°, 31.1°, 33.6°, 40.8°, and 43.6°), with some contamination by quartz or free silica (2θ = 20.9°, 26.5°, 50.5° and 68.7°). In view of the fairly low level of Fe and its own capacity for carbonation (as FeCO3), modeling studies (vide infra) were based for simplicity on the pure forsterite composition (Mg2SiO4), the Mg-end member of the olivine solid solution series. Morphological and structural changes in the olivine (nominal composition Mg1.84Fe0.16SiO4) at various stages during chemical pretreatment and carbonation, viz., after leaching in HCl, neutralization and precipitation of Mg 2+ [as Mg(OH)2], and exposure to humid CO2 at 4.8 bar and 150 °C, are best seen by SEM micrographs and powder XRD in Figures 2 and 3, respectively.
By SEM, the fresh olivine sample (Figure 2a) consisted typically of polycrystalline grains in the millimeter size range. After acid leaching and neutralization of the sieved fraction <75 µm (Figure 2b), the grains became finer (<5 µm) and the structure more amorphous. In agreement with the findings of Bearat et al. [30] and Kwon et al. [18], this is likely due to an amorphous SiO2 residue after leaching, any (hexagonal) brucite likely being nanocrystalline. Under humid CO2, the development of polyhedral and more sheet-like crystallites ( Figure 2c) was evident, probably representing (hydrated) magnesium carbonate, and finally (Figure 2d), the rhombohedral habit typical of magnesite (MgCO3) was evident.  (chemically un-pretreated) mineral, for which evidence has been reported elsewhere [31]. The abundance of quartz (as co-product) cannot be taken as a reliable indicator of the progress of carbonation because it is also a phase contaminant in the original mineral (see Figure 1). Furthermore, similar chemical treatment (flux extraction of Mg 2+ ) from serpentinites did not produce quartz but instead a silica residue of unusual structure [27]. The unsystematic peak intensity of quartz seen here by XRD may be due merely to local inhomogeneity in the samples.

Effect of Particle size on the Carbonation of Olivine
The leaching rate was found to be much slower than the optimal carbonation rate (achievable at 175 °C vide infra), such that the production of magnesite in these experiments reflects mainly variations in the rate and efficiency of extraction of soluble Mg 2+ ion by acid treatment. Figure 4 shows that decreasing the grain size from >200 µm to <75 µm caused the limiting degree of Mg 2+ leaching to increase from 35% to 99%, respectively. It is well-known and intuitively obvious that the reaction rate and carbonation degree can be raised by increasing the surface area, e.g., by grinding/sieving, as shown for example by Garcia et al. [32]. In this work, the smallest particle size (d < 75 µm) is the only one offering the prospect of full leaching (99%) in a practical time. As a rule of thumb, 1-2 h for a terrestrial (ex-situ) process to achieve >90% conversion is taken as a realistic practical target to limit the scale (and associated costs) of any future installation for CO2 sequestration. This particle cut and leaching procedure (2 h at 60 °C) were therefore applied as standard for all samples in subsequent tests described below.  The ultimate disappearance of olivine is intriguing since it implies direct steam-activated carbonation of the (chemically un-pretreated) mineral, for which evidence has been reported elsewhere [31]. The abundance of quartz (as co-product) cannot be taken as a reliable indicator of the progress of carbonation because it is also a phase contaminant in the original mineral (see Figure 1). Furthermore, similar chemical treatment (flux extraction of Mg 2+ ) from serpentinites did not produce quartz but instead a silica residue of unusual structure [27]. The unsystematic peak intensity of quartz seen here by XRD may be due merely to local inhomogeneity in the samples.

Effect of Particle Size on the Carbonation of Olivine
The leaching rate was found to be much slower than the optimal carbonation rate (achievable at 175˝C vide infra), such that the production of magnesite in these experiments reflects mainly variations in the rate and efficiency of extraction of soluble Mg 2+ ion by acid treatment. Figure 4 shows that decreasing the grain size from >200 µm to <75 µm caused the limiting degree of Mg 2+ leaching to increase from 35% to 99%, respectively. It is well-known and intuitively obvious that the reaction rate and carbonation degree can be raised by increasing the surface area, e.g., by grinding/sieving, as shown for example by Garcia et al. [32]. In this work, the smallest particle size (d < 75 µm) is the only one offering the prospect of full leaching (99%) in a practical time. As a rule of thumb, 1-2 h for a terrestrial (ex-situ) process to achieve >90% conversion is taken as a realistic practical target to limit the scale (and associated costs) of any future installation for CO 2 sequestration. This particle cut and leaching procedure (2 h at 60˝C) were therefore applied as standard for all samples in subsequent tests described below.

Effect of Temperature on the Carbonation of Olivine
The temperature dependence of the carbonation process was studied over the interval from ambient to 175 °C. The effect of temperature on the amount of Mg conversion after standard leaching treatment (99% extraction) is illustrated in Figure 5. As expected, the temperature had an important effect on the mineral carbonation process consisting of Mg extraction and subsequent MgCO3 precipitation. The amount of Mg converted at the desired temperatures was measured continuously several times. The maximum extent of carbonation of (99%) was attained at temperatures in excess of 150 °C. The quantification of the MgCO3 formed in the mineral carbonation process was carried out by titration against HCl at room temperature. Pokrovsky et al. [33] demonstrated that the dissolution rate of magnesite at 150 °C is lower than at 25 °C whereas the rates at 100 and 150 °C in acidic solutions are almost the same [34], suggesting a strong decrease of the apparent activation energy above 100 °C. According to Saldi et al. [34] the tendency for dissolution rates of MgCO3 to decrease with increasing temperature could benefit CO2 sequestration efforts by making magnesite more resistant to dissolution in deeper (hotter) strata, thus preserving the petrophysical integrity of deep carbonate-rich confining reservoirs.

Effect of Temperature on the Carbonation of Olivine
The temperature dependence of the carbonation process was studied over the interval from ambient to 175˝C. The effect of temperature on the amount of Mg conversion after standard leaching treatment (99% extraction) is illustrated in Figure 5. As expected, the temperature had an important effect on the mineral carbonation process consisting of Mg extraction and subsequent MgCO 3 precipitation. The amount of Mg converted at the desired temperatures was measured continuously several times. The maximum extent of carbonation of (99%) was attained at temperatures in excess of 150˝C. The quantification of the MgCO 3 formed in the mineral carbonation process was carried out by titration against HCl at room temperature. Pokrovsky et al. [33] demonstrated that the dissolution rate of magnesite at 150˝C is lower than at 25˝C whereas the rates at 100 and 150˝C in acidic solutions are almost the same [34], suggesting a strong decrease of the apparent activation energy above 100˝C. According to Saldi et al. [34] the tendency for dissolution rates of MgCO 3 to decrease with increasing temperature could benefit CO 2 sequestration efforts by making magnesite more resistant to dissolution in deeper (hotter) strata, thus preserving the petrophysical integrity of deep carbonate-rich confining reservoirs.

Effect of Temperature on the Carbonation of Olivine
The temperature dependence of the carbonation process was studied over the interval from ambient to 175 °C. The effect of temperature on the amount of Mg conversion after standard leaching treatment (99% extraction) is illustrated in Figure 5. As expected, the temperature had an important effect on the mineral carbonation process consisting of Mg extraction and subsequent MgCO3 precipitation. The amount of Mg converted at the desired temperatures was measured continuously several times. The maximum extent of carbonation of (99%) was attained at temperatures in excess of 150 °C. The quantification of the MgCO3 formed in the mineral carbonation process was carried out by titration against HCl at room temperature. Pokrovsky et al. [33] demonstrated that the dissolution rate of magnesite at 150 °C is lower than at 25 °C whereas the rates at 100 and 150 °C in acidic solutions are almost the same [34], suggesting a strong decrease of the apparent activation energy above 100 °C. According to Saldi et al. [34] the tendency for dissolution rates of MgCO3 to decrease with increasing temperature could benefit CO2 sequestration efforts by making magnesite more resistant to dissolution in deeper (hotter) strata, thus preserving the petrophysical integrity of deep carbonate-rich confining reservoirs.  It is clear from Figure 5 that several competing factors in the carbonation process reach equilibrium in the temperature range between 150 and 175˝C. Probably the key factor is the relative humidity (RH). Previous work has shown that carbonation accelerates on cooling below 200˝C in a fixed partial pressure of steam or RH ě 20% [35]. Since the presence of liquid water has been shown to be important, it can also be argued that the effect of temperature on the solubility of CO 2 may influence the carbonation rate. Chen et al. [36] indicated that increasing temperature from 25˝C to 150˝C generally helps the precipitation of magnesite. The carbonate solubility product (K sp ), Henry's constant (K H ), and the first-and second-order dissociation constants of carbonic acid (Ka 1 , Ka 2 ) are all functions of temperature. Increasing the reaction temperature decreases the value of K sp and increases the K H value, which lowers the amount of CO 2 gas in solution at a given pressure. The dissociation constants of carbonic acid (Ka 1 and Ka 2 ) are also increased by increasing temperature and promoting solution speciation and carbonation. Thus, increasing temperature has conflicting effects, lowering the level of dissolved CO 2 gas via its effect on K H but promoting magnesite precipitation via effects on Ka 1 , Ka 2 , and K sp . Furthermore, temperature impacts on the kinetics of magnesite precipitation [37] and also affects the type of Mg-carbonate formed. It is well-known that magnesite precipitation kinetics at ambient temperatures are exceedingly slow, and that metastable hydrated carbonates such as hydromagnesite, dypingite, and nesquehonite almost invariably form instead.

Effect of Water Concentration
Formal carbonation of forsterite [Reaction (1)] does not involve water molecules explicitly, so any beneficial effect of added water is clearly a kinetic effect. As shown in Figure 6, an indication of the importance of water is seen in the modest level of CO 2 removal in its absence. By analogy with recent works on brucite and serpentinites, almost regardless of the CO 2 pressure utilized, the presence of water vapor in high relative humidity appears crucial to obtaining practical carbonation rates [20,31,38], evidently by establishing a highly polar thin-film aqueous overlayer that facilitates CO 2 ingress into the bulk particle. This is supported by independent studies simulating in situ or geochemical carbon sequestration where CO 2 was assumed to be in the supercritical state. Felmy et al. [39] studied high-surface-area forsterite in the presence of water-saturated scCO 2 . They concluded that the nature of the water in contact with the reacting surface is a key factor in the enhanced magnesite formation. When excess water was added to the forsterite particles, a thin water film was formed on the forsterite surface promoting magnesite formation. Loring et al. [40] declared that this water film provides a distinctive situation for the magnesite formation by decreasing the effective Mg 2+ dehydration energy and simplifying the transformation of nesquehonite to magnesite. Otherwise, the presence of liquid water can allow the formation of magnesium bicarbonate in solution that decomposes upon drying to magnesium carbonate. Moreover, Schaef et al. [41,42] revealed that the addition of water to the saturated system noticeably increases the rate of mineral carbonation, facilitating the overall conversion of nesquehonite to magnesite. No evidence of further carbonation was observed under unsaturated conditions below 50˝C. A similar promoting effect of water on brucite carbonation under scCO 2 was reported by Loring et al. [43] using in situ Fourier-Transform Infrared (FTIR) spectroscopic experiments.
To investigate the effect of humidity, the concentration of water vapor was set at various levels, 5, 10, and 20 vol % prior to carbonation of the brucite extract at 175˝C. As illustrated in Figure 6, above 5 vol % steam, the degree of carbonation increases by almost double in the presence of water vapor such that within 5 min, complete removal of CO 2 (15 vol %) was achieved. However, levels of water vapor exceeding 20 vol % had no additional effect on the rate of removal of CO 2 . It can be concluded that water vapor is able to solvate CO 2 , generate carbonate ions and protons [44], and increase the carbonation degree of Mg(OH) 2 as derived from olivine. Moreover, the aforementioned modest uptake of CO 2 (~7 vol %) under "dry" conditions may be due to adventitious water not fully removed from the Mg(OH) 2 -containing residue. Vitillo [45] declared that in the presence of water vapor, MgCO 3 crystalline phase reappeared increasingly, while the magnesium oxide periclase (MgO) phase gradually disappeared. These observations are well in agreement with the thermodynamic reaction kinetics by offering alternative routes to magnesite via hydrocarbonate intermediates such as dypingite. Figure 6. Effect of the presence or absence of water vapor on the rate of CO2 absorption at 175 °C on 10 g Mg2SiO4 (d < 75 µm) activated by chemical pre-treatment and exposed to 0.5 L/min CO2 gas (PCO2 = 4.8 bar, PH2O = 1.6-6.4 bar or 18%-72% RH).
As a comparison to related work on Mg(OH)2 in the literature, Siriwardane and Stevens [46] reported good absorption kinetics and reasonable capacity for CO2 in plug-flow reactor experiments over a promoted brucite (~3 mol CO2/kg or 20 mol %) in "moist" helium at 200 °C. However, the promotional effect of water per se was not explored and the sorbent surface area was low (~2.5 m 2 ·g −1 ). This compares with our thermogravimetric work [38] showing that Mg(OH)2 extracted from the mineral is typically obtained in high-surface-area form (~25 m 2 ·g −1 ). These are evidently important factors in attainment of almost quantitative (~100%) carbonation at lower temperature (150-175 °C) in this work, specifically the higher water levels utilized and the better dispersion of brucite derived from the mineral. Based on the thermodynamic equilibrium of Mg(OH)2 formation from MgO, Siriwardane and Stevens [46] showed that is likely to form Mg(OH)2 under the high steam environment, which accounts for the subsequent CO2 uptake. It is important to note that the Mg(OH)2 system has the far lower heat of sorption. This confirms that the regeneration heat (input) needed to displace CO2 from MgCO3 by water (to form Mg(OH)2) is significantly lower than that required for the decomposition of MgCO3 (to MgO + CO2).

Kinetic Analysis of Mg Extraction by HCl
Different kinetic analyses including expressions for product layer diffusion, film diffusion, chemical reaction control, and a combination of chemical reaction control (Equations (4)- (7), respectively, in Section 3.5), were used to evaluate the integral rate data. The extent of forsterite dissolution, XE, is taken as fitting parameter but this is actually measured from the amount of magnesite, i.e., the extent of carbonation (= RCO2 in Equation (2)), because dissolution is much slower than carbonation (of the Mg(OH)2 extract). Direct carbonation of unreacted forsterite is probably even slower. Thus, dissolution is rate-determining in the overall process. The two best-fit results are illustrated in Figure 7a,b, but the first, a combination of chemical reaction control and product layer diffusion (Equation (7)) provided Figure 6. Effect of the presence or absence of water vapor on the rate of CO 2 absorption at 175˝C on 10 g Mg 2 SiO 4 (d < 75 µm) activated by chemical pre-treatment and exposed to 0.5 L/min CO 2 gas (P CO2 = 4.8 bar, P H2O = 1.6-6.4 bar or 18%-72% RH).
As a comparison to related work on Mg(OH) 2 in the literature, Siriwardane and Stevens [46] reported good absorption kinetics and reasonable capacity for CO 2 in plug-flow reactor experiments over a promoted brucite (~3 mol CO 2 /kg or 20 mol %) in "moist" helium at 200˝C. However, the promotional effect of water per se was not explored and the sorbent surface area was low (~2.5 m 2¨g´1 ). This compares with our thermogravimetric work [38] showing that Mg(OH) 2 extracted from the mineral is typically obtained in high-surface-area form (~25 m 2¨g´1 ). These are evidently important factors in attainment of almost quantitative (~100%) carbonation at lower temperature (150-175˝C) in this work, specifically the higher water levels utilized and the better dispersion of brucite derived from the mineral. Based on the thermodynamic equilibrium of Mg(OH) 2 formation from MgO, Siriwardane and Stevens [46] showed that is likely to form Mg(OH) 2 under the high steam environment, which accounts for the subsequent CO 2 uptake. It is important to note that the Mg(OH) 2 system has the far lower heat of sorption. This confirms that the regeneration heat (input) needed to displace CO 2 from MgCO 3 by water (to form Mg(OH) 2 ) is significantly lower than that required for the decomposition of MgCO 3 (to MgO + CO 2 ).

Kinetic Analysis of Mg Extraction by HCl
Different kinetic analyses including expressions for product layer diffusion, film diffusion, chemical reaction control, and a combination of chemical reaction control (Equations (4)- (7), respectively, in Section 3.5), were used to evaluate the integral rate data. The extent of forsterite dissolution, X E , is taken as fitting parameter but this is actually measured from the amount of magnesite, i.e., the extent of carbonation (= R CO2 in Equation (2)), because dissolution is much slower than carbonation (of the Mg(OH) 2 extract). Direct carbonation of unreacted forsterite is probably even slower. Thus, dissolution is rate-determining in the overall process. The two best-fit results are illustrated in Figure 7a,b, but the first, a combination of chemical reaction control and product layer diffusion (Equation (7)) provided the highest correspondence with the measured data. Thus, it can be concluded that a combination of chemical reaction control and product layer diffusion is rate-limiting for Mg extraction. the highest correspondence with the measured data. Thus, it can be concluded that a combination of chemical reaction control and product layer diffusion is rate-limiting for Mg extraction. Activation energies (Ea) for mineral dissolution were determined from simple log/log plots of the time-independent rate k at various temperatures. These values are presented in Table 2, from which the Arrhenius plots shown in Figure 8 were obtained. Once again, the quality of fit was best for the combination of product layer diffusion and chemical reaction control (R 2 = 0.9917) as compared to product layer diffusion only (R = 0.9764). Considering these models as the controlling mechanisms during the dissolution of forsterite, the Ea value was 15.5 kJ/mol for product layer diffusion control, and 16.0 kJ/mol for the combination of product layer diffusion and chemical reaction control. It is likely that the chemical reaction is initially rate-limiting but product layer diffusion gradually becomes rate limiting as the product layer of silica builds up and the unreacted surface area decreases. According to Gharabaghi et al. [47] a low value of Ea indicates that product layer diffusion is rate-controlling, Therefore, considering the values of multiple regression coefficients for different models and calculated Ea for two selected models, it could be concluded that the dissolution rates of forsterite are kinetically regulated by the combination of chemical reaction control and product layer diffusion. Activation energies (E a ) for mineral dissolution were determined from simple log/log plots of the time-independent rate k at various temperatures. These values are presented in Table 2, from which the Arrhenius plots shown in Figure 8 were obtained. Once again, the quality of fit was best for the combination of product layer diffusion and chemical reaction control (R 2 = 0.9917) as compared to product layer diffusion only (R = 0.9764). Considering these models as the controlling mechanisms during the dissolution of forsterite, the E a value was 15.5 kJ/mol for product layer diffusion control, and 16.0 kJ/mol for the combination of product layer diffusion and chemical reaction control. It is likely that the chemical reaction is initially rate-limiting but product layer diffusion gradually becomes rate limiting as the product layer of silica builds up and the unreacted surface area decreases. According to Gharabaghi et al. [47] a low value of E a indicates that product layer diffusion is rate-controlling, Therefore, considering the values of multiple regression coefficients for different models and calculated E a for two selected models, it could be concluded that the dissolution rates of forsterite are kinetically regulated by the combination of chemical reaction control and product layer diffusion.   The rate constants generated in this study on forsterite are faster than corresponding rate constants reported by Pokrovsky and Schott [48], who only worked at the temperature of 25 °C. Comparison of photomicrographs showing the surface of forsterite presented by Pokrovsky and Schott [48] and the samples from this study (see Figure 2) demonstrates they consist of euhedral and larger crystals due to small adhering particles and their agglomeration in acid solution at the temperature of 25 °C. In the present study, the rate constants are increased because of the far higher density of activated sites per unit surface area at higher temperatures.

Thermodynamic Considerations on the Mineral Carbonation
As regards the effect of temperature on the system, two physical mechanisms interact; the increase of olivine solubility with temperature and, conversely, the reduction of magnesite solubility with increasing temperature. Heat treatment was reported by O'Connor et al. [49] to remove sorbed water and activate the mineral surface. They suggested that the carbonation phase occurs quickly in the olivine powders at high temperatures, from which we expect an increase in the available reactivity of mineral surface with removal of sorbed water and CO2 uptake by the mineral surface and, therefore, more rapid reaction.
Additionally, elevated temperature causes an increase of the olivine dissolution rate, which enhances the overall reaction kinetics. Our results concur with this (see Figure 5) up to the maximum temperature studied (175 °C). The lowest CO2 concentration considered was 0.5% (volume fraction of CO2 unconsumed) indicating olivine samples have efficiently precipitated CO2 as carbonate phase, which is in agreement with the findings of Kwon et al. [18]. Interestingly, the CO2 sequestration capacity of olivine mineral seen here over the temperature range of 150-175 °C was higher than in previous The rate constants generated in this study on forsterite are faster than corresponding rate constants reported by Pokrovsky and Schott [48], who only worked at the temperature of 25˝C. Comparison of photomicrographs showing the surface of forsterite presented by Pokrovsky and Schott [48] and the samples from this study (see Figure 2) demonstrates they consist of euhedral and larger crystals due to small adhering particles and their agglomeration in acid solution at the temperature of 25˝C. In the present study, the rate constants are increased because of the far higher density of activated sites per unit surface area at higher temperatures.

Thermodynamic Considerations on the Mineral Carbonation
As regards the effect of temperature on the system, two physical mechanisms interact; the increase of olivine solubility with temperature and, conversely, the reduction of magnesite solubility with increasing temperature. Heat treatment was reported by O'Connor et al. [49] to remove sorbed water and activate the mineral surface. They suggested that the carbonation phase occurs quickly in the olivine powders at high temperatures, from which we expect an increase in the available reactivity of mineral surface with removal of sorbed water and CO 2 uptake by the mineral surface and, therefore, more rapid reaction.
Additionally, elevated temperature causes an increase of the olivine dissolution rate, which enhances the overall reaction kinetics. Our results concur with this (see Figure 5) up to the maximum temperature studied (175˝C). The lowest CO 2 concentration considered was 0.5% (volume fraction of CO 2 unconsumed) indicating olivine samples have efficiently precipitated CO 2 as carbonate phase, which is in agreement with the findings of Kwon et al. [18]. Interestingly, the CO 2 sequestration capacity of olivine mineral seen here over the temperature range of 150-175˝C was higher than in previous studies [50][51][52][53][54][55]. We tentatively suggest that this effect is due to either an increase of the available surface area or available moisture (or a combination of both factors). At the highest temperatures, (e.g., the range of 150-175˝C), Hänchen et al. [56] suggest the olivine samples released more Mg into the solution, which seems a reasonable assumption to adopt here and is confirmed by the calculation shown below. At lower temperatures (less than 100˝C), King et al. [12] found that MgCO 3 precipitation was kinetically hindered due to the high activation energy for the de-solvation of the strongly hydrated Mg 2+ ions (Reaction (2)). According to Hänchen et al. [57]  The work of Hänchen et al. [57] is intriguing, as we found the evidence of hydromagnesite formation in our samples, suggesting that this phase may be as a metastable "intermediate" phase [38] with respect to magnesite, but kinetically favored as a first reaction product under certain conditions (e.g., at low temperature conditions).

Geochemical Modeling
Equilibrium simulations were performed using PHREEQC (version 2.18-Parkhurst and Appello [58] with data from the Lawrence Livermore National Laboratory (LLNL, Livermore, CA, USA, supplied with the code). The first, very simple calculations examined the saturation state of a solution (initially pure water) when equilibrated with either magnesite, or both magnesite and carbon dioxide gas. Figure 9 shows the saturation index of both solutions with respect to magnesium phases which could be formed from the elements in the simulated system. In each case, the solutions suggest saturation with brucite, Mg(OH) 2 , would be reached above 25˝C, so the simulations were repeated, but allowing brucite to precipitate when it would otherwise be oversaturated at ambient pressure and it is these conditions which are shown in Figure 9. It is interesting to note that the solution remains undersaturated with respect to hydromagnesite at all temperatures and the influence of CO 2 saturation is negligible. This reinforces the view that hydromagnesite is an artifact of rapid precipitation, i.e., a process under kinetic control, in preference to magnesite (in line with experiments by Konigsberger et al. [59] and Power et al. [60]), which, under equilibrium conditions, is the dominant magnesium-bearing phase.
Secondly, attempts were made to simulate the reaction of CO 2 gas with forsterite in the presence of water, with and without varying amounts of HCl. These results also suggest considerable disequilibrium, in that magnesite would not be the dominant "sink" for magnesium in the presence of the silicate ions liberated from forsterite dissolution. Examination of the saturation state of solutions in which CO 2 was sequentially added to forsterite-water mixtures, always suggested a greater tendency for sepiolite precipitation, in preference to magnesite. Over the temperature interval 1 to 175˝C, CO 2 gas was sequentially "reacted" (numerically) with an excess of forsterite in the presence of water and these simulations were repeated in the presence of HCl to represent any carry-over of the acid from the previous processing step. Although we recognize that the presence of hydrochloric acid has little influence on the thermodynamic equilibrium position, it has a major role in promoting a rapid increase in the forsterite dissolution kinetics, so it was considered here. In all cases, thermodynamic equilibrium was only reached when sepiolite was allowed to precipitate, which effectively prevented magnesite formation. No evidence of other phases being important in these reactions was seen and the influence of residual HCl was negligible. Experimentally, no evidence of sepiolite formation was seen and magnesite certainly dominates the solid phase assemblage after carbonation in this work and in the previous studies cited above. Our observations are that forsterite dissolves and magnesite and amorphous silica precipitate as reaction products (see Figures 2 and 3). Experimental studies on the stability of sepiolite have demonstrated that if the silica activity value is high in alkaline solutions, sepiolite precipitates [61][62][63]. From this, we must assume that the kinetics of sepiolite formation are very slow indeed and consequently it was excluded from the mineral assemblage in subsequent calculations. Figure 10 shows the solution chemistry predicted in a system where forsterite is carbonated by gaseous CO2 in the presence of water. One mole of forsterite (in excess) was sequentially reacted with 0.2 mol of CO2 gas, allowing equilibrium to be reached with both amorphous silica and all possible magnesium phases except sepiolite (which was prevented from forming in the simulations). At all temperatures between 25 °C and 175 °C carbonation is predicted to cause dissolution of forsterite and precipitation of magnesite and amorphous silica, with no other magnesium-bearing phases reaching equilibrium, except sepiolite which was allowed to remain oversaturated in the pore solution.
The solid phase chemistry is shown in Figure 11, which plots the ratio of forsterite dissolved against magnesite precipitated (mole ratio) against the number of moles of carbon dioxide consumed by the reaction. This is in effect the "yield" of the reaction, showing the moles of product formed as a function of the moles of reactant consumed. The theoretical stoichiometry (see Equation (1), above) is that two moles of forsterite react with one mole of CO2 to produce one mole of magnesite and one mole of amorphous silica. These simulations suggest that this condition is reached quite rapidly. After reaction of only 0.9 mol of CO2, the reaction is 97% complete in stoichiometric terms; that is to say the bulk of the CO2 reacts to form magnesite and very little partitions into the liquid phase (cf. Figure 10). Figure 11 shows that the temperature does have an effect on the extent of reaction and efficiency of carbonation in terms of the overall yield. The pseudo steady-state conditions established after reaction of 0.2 mol of CO2 are maintained if further reaction is simulated, such that the solution chemistry remains constant, as does the ratio of forsterite and CO2 consumed to the quantities of amorphous silica and magnesite precipitated. This is shown in Figure 12, where an excess of carbon dioxide eventually exhausts the reserve of forsterite, maintaining the stoichiometry of the reaction throughout. It must be stated that since the simplistic geochemical reaction path modeling is used in  Experimentally, no evidence of sepiolite formation was seen and magnesite certainly dominates the solid phase assemblage after carbonation in this work and in the previous studies cited above. Our observations are that forsterite dissolves and magnesite and amorphous silica precipitate as reaction products (see Figures 2 and 3). Experimental studies on the stability of sepiolite have demonstrated that if the silica activity value is high in alkaline solutions, sepiolite precipitates [61][62][63]. From this, we must assume that the kinetics of sepiolite formation are very slow indeed and consequently it was excluded from the mineral assemblage in subsequent calculations. Figure 10 shows the solution chemistry predicted in a system where forsterite is carbonated by gaseous CO 2 in the presence of water. One mole of forsterite (in excess) was sequentially reacted with 0.2 mol of CO 2 gas, allowing equilibrium to be reached with both amorphous silica and all possible magnesium phases except sepiolite (which was prevented from forming in the simulations). At all temperatures between 25˝C and 175˝C carbonation is predicted to cause dissolution of forsterite and precipitation of magnesite and amorphous silica, with no other magnesium-bearing phases reaching equilibrium, except sepiolite which was allowed to remain oversaturated in the pore solution.
The solid phase chemistry is shown in Figure 11, which plots the ratio of forsterite dissolved against magnesite precipitated (mole ratio) against the number of moles of carbon dioxide consumed by the reaction. This is in effect the "yield" of the reaction, showing the moles of product formed as a function of the moles of reactant consumed. The theoretical stoichiometry (see Equation (1), above) is that two moles of forsterite react with one mole of CO 2 to produce one mole of magnesite and one mole of amorphous silica. These simulations suggest that this condition is reached quite rapidly. After reaction of only 0.9 mol of CO 2 , the reaction is 97% complete in stoichiometric terms; that is to say the bulk of the CO 2 reacts to form magnesite and very little partitions into the liquid phase (cf. Figure 10). Figure 11 shows that the temperature does have an effect on the extent of reaction and efficiency of carbonation in terms of the overall yield. The pseudo steady-state conditions established after reaction of 0.2 mol of CO 2 are maintained if further reaction is simulated, such that the solution chemistry remains constant, as does the ratio of forsterite and CO 2 consumed to the quantities of amorphous silica and magnesite precipitated. This is shown in Figure 12, where an excess of carbon dioxide eventually exhausts the reserve of forsterite, maintaining the stoichiometry of the reaction throughout. It must be stated that since the simplistic geochemical reaction path modeling is used in this study, many physical processes, such as material deformation, pore-fluid flow, heat transfer, advection, diffusion/dispersion, are ignored, although they should be considered in future work, as they are expected to be important during industrial-scale-up. In addition, because the chemical dissolution-front instability is also neglected in this study, many important factors, such as mineral reactive surface area, mineral dissolution ratio, solute dispersion, medium anisotropy, medium and fluid compressibility, have also been ignored. To consider these factors appropriately, more comprehensive chemical-transport modeling will be required. this study, many physical processes, such as material deformation, pore-fluid flow, heat transfer, advection, diffusion/dispersion, are ignored, although they should be considered in future work, as they are expected to be important during industrial-scale-up. In addition, because the chemical dissolution-front instability is also neglected in this study, many important factors, such as mineral reactive surface area, mineral dissolution ratio, solute dispersion, medium anisotropy, medium and fluid compressibility, have also been ignored. To consider these factors appropriately, more comprehensive chemical-transport modeling will be required.  this study, many physical processes, such as material deformation, pore-fluid flow, heat transfer, advection, diffusion/dispersion, are ignored, although they should be considered in future work, as they are expected to be important during industrial-scale-up. In addition, because the chemical dissolution-front instability is also neglected in this study, many important factors, such as mineral reactive surface area, mineral dissolution ratio, solute dispersion, medium anisotropy, medium and fluid compressibility, have also been ignored. To consider these factors appropriately, more comprehensive chemical-transport modeling will be required.

Materials
Three kilograms of olivine mineral were excavated and collected from different depths of the Mont Tawai peridotite stratum in Malaysia. Although this source is local, the results can be considered as broadly representative of peridotite as the mineral composition is typical. The samples were crushed by a mechanical grinder and sieved into four different size ranges: <75 µm, 75-125 µm, 125-200 µm and >200 µm. Subsequently, they were dried to a constant weight at 120 °C for 2 h. Ten grams of each size fraction were mixed separately in 500 mL of 0.01 M hydrochloric acid (HCl, QRëC reagent grade, 37%) solution to remove impurities.

Experimental Apparatus
The experimental equipment for CO2 mineralization consisted of a CO2 analyzer system mounted in a flow system connected to a cylindrical (500 mm × 10 mm (d)) autoclave reactor with the means to supply diluted CO2 and flushing with pure N2 (Figure 13). CO2 was introduced into the reactor at different partial pressures (up to 30%). Two flow meters (FM-1050, Matheson Tri-gas, Basking Ridge, NJ, USA) were used to control the flow rate of the inlet gases. The autoclave reactor was loaded with mineral and acid, then placed in a furnace where the temperature was measured and controlled using a thermocouple inserted directly into the reactor.
A water vapor generator was used to humidify the gas stream. CO2 consumption was measured as the difference between supply and vent level at fixed flow (with integration over time) using an optical IR-sensor (GMP221, Vaisala Oyj, Helsinki, Finland), according to Equation (1):

Materials
Three kilograms of olivine mineral were excavated and collected from different depths of the Mont Tawai peridotite stratum in Malaysia. Although this source is local, the results can be considered as broadly representative of peridotite as the mineral composition is typical. The samples were crushed by a mechanical grinder and sieved into four different size ranges: <75 µm, 75-125 µm, 125-200 µm and >200 µm. Subsequently, they were dried to a constant weight at 120˝C for 2 h. Ten grams of each size fraction were mixed separately in 500 mL of 0.01 M hydrochloric acid (HCl, QRëC reagent grade, 37%) solution to remove impurities.

Experimental Apparatus
The experimental equipment for CO 2 mineralization consisted of a CO 2 analyzer system mounted in a flow system connected to a cylindrical (500 mmˆ10 mm (d)) autoclave reactor with the means to supply diluted CO 2 and flushing with pure N 2 ( Figure 13). CO 2 was introduced into the reactor at different partial pressures (up to 30%). Two flow meters (FM-1050, Matheson Tri-gas, Basking Ridge, NJ, USA) were used to control the flow rate of the inlet gases. The autoclave reactor was loaded with mineral and acid, then placed in a furnace where the temperature was measured and controlled using a thermocouple inserted directly into the reactor.
A water vapor generator was used to humidify the gas stream. CO 2 consumption was measured as the difference between supply and vent level at fixed flow (with integration over time) using an optical IR-sensor (GMP221, Vaisala Oyj, Helsinki, Finland), according to Equation (1): where pCO 2 out and pCO 2 in are mean value of pCO 2 (atm) at the inflow and outflow (equilibrium supplyP = 4.8 bar), ∆t and Q are time interval (min) and flow rate (L/min), respectively, R and T are gas constant (0.082057 l.atm/mol.K) and temperature (K), respectively, and M is the mass of forsterite (g). where pCO2 out and pCO2 in are mean value of pCO2 (atm) at the inflow and outflow (equilibrium supplyP = 4.8 bar), Δt and Q are time interval (min) and flow rate (L/min), respectively, R and T are gas constant (0.082057 l.atm/mol.K) and temperature (K), respectively, and M is the mass of forsterite (g).

Experimental Procedure
In this study, 1 M HCl was used for dissolution of Mg from the mineral matrix of peridotite and the dissolution experiment was conducted in a separate vessel. When a stoichiometric amount of HCl solution was added to olivine powder in a reaction vessel and stirred with a magnetic stirrer at 60 °C for two h, the entire mass formed a slurry. This slurry was then transferred into the autoclave reactor and neutralized by adding a base (NaOH: Fisher Chemicals reagent grade, with purity 99.999%) until the final pH was increased to above seven. Before heating the reactor, it was purged with nitrogen in order to replace the air inside the reactor, then the reactor was pre-heated to 175 °C for 1 h to dry the slurry and cooled to ambient. CO2 gas (SFE grade, with a purity of 99.99% contained in a dip-tube cylinder and purchased from MOX Company, KL, Malaysia) was passed through the dried slurry at a typical flue gas level (15 vol %) or 4.8 bar pressure (Ptot = 32 bar) in the absence or presence of water vapor. The temperature of the carbonation process was studied over the interval from ambient to 175 °C.
MgCl2 + 2NaOH → Mg(OH)2 + 2NaCl (Rx. 5) The effect of water vapor level on carbonation was studied in the range 5-20 vol % H2O, corresponding to a range of relative humidity 18%-72% RH. After completion of the experiment, the samples were collected and filtered using <75 µm pore size Whatman filter papers. The MgCO3

Experimental Procedure
In this study, 1 M HCl was used for dissolution of Mg from the mineral matrix of peridotite and the dissolution experiment was conducted in a separate vessel. When a stoichiometric amount of HCl solution was added to olivine powder in a reaction vessel and stirred with a magnetic stirrer at 60˝C for two h, the entire mass formed a slurry. This slurry was then transferred into the autoclave reactor and neutralized by adding a base (NaOH: Fisher Chemicals reagent grade, with purity 99.999%) until the final pH was increased to above seven. Before heating the reactor, it was purged with nitrogen in order to replace the air inside the reactor, then the reactor was pre-heated to 175˝C for 1 h to dry the slurry and cooled to ambient. CO 2 gas (SFE grade, with a purity of 99.99% contained in a dip-tube cylinder and purchased from MOX Company, KL, Malaysia) was passed through the dried slurry at a typical flue gas level (15 vol %) or 4.8 bar pressure (P tot = 32 bar) in the absence or presence of water vapor. The temperature of the carbonation process was studied over the interval from ambient to 175˝C.
MgCl 2`2 NaOH Ñ MgpOHq 2`2 NaCl (Rx. 5) The effect of water vapor level on carbonation was studied in the range 5-20 vol % H 2 O, corresponding to a range of relative humidity 18%-72% RH. After completion of the experiment, the samples were collected and filtered using <75 µm pore size Whatman filter papers. The MgCO 3 formed in the sample was quantified by titration against HCl. For the titration, a certain amount of solid (0.5 g) was weighed in a conical flask. Then, 20 mL of 1 M HCl was added into the flask and was allowed about two hours to react with the MgCO 3 . The excess HCl was then back-titrated using 0.1 M NaOH solution. From the difference in titrant volume, the HCl consumed was calculated from which the content of MgCO 3 was deduced.

Estimation of Carbonation Yield
The extent of CO 2 mineral carbonation Y CO 2 was estimated using the TCA method, which is based on the mineralogy of samples tested and the capacity of carbon sequestratedˆ1 R CO 2˙ [ 64].
R CO 2 is considered as the weight fraction of CO 2 that can be trapped in a specific amount of mineral. According to Gadikota et al. [64] the capacity of CO 2 sequestration in forsterite (Mg 2 SiO 4 ) can be expressed as follows: where W f o and W CO 2 are weights of forsterite before its mineral carbonation and CO 2 sequestrated in the solid phase (i.e., magnesite), respectively. Y Mg is the mass fraction of Mg 2+ in the forsterite (i.e., 34.55%) that can react with carbon dioxide to form stable magnesite. MW mg and MW CO 2 are the formula weights of Mg 2+ (48.61 mol/g in the forsterite) and CO 2 (44.01 mol/g), respectively, in the carbonated forsterite (2MgCO 3 & SiO 2 -see (Reaction (1))). Therefore, Y CO 2 is the amount of carbon dioxide sequestrated (as magnesite) relative to the maximum capacity of CO 2 sequestration in forsterite p 1 R CO 2 " 31.27%q.
Y CO 2 " R CO 2ˆ« 3.67ˆweight fraction of carbon in MgCO 3 1´`3.67ˆweight fraction of carbon in MgCO 3˘ffˆ1 00% where 3.67 is the CO 2 /C mass ratio. The carbonation yields of the forsterite and the effect on these of key empirical variables-reaction temperature, time, and particle size, are compared below.

Modeling System and Kinetic Analysis
Thermodynamic calculations were performed with the PHREEQC program software (version 2.18) with data from the LLNL [58]. Forsterite is the Mg-end member of the forsterite-fayalite solid solution series, and is included in the LLNL database. This program was used to estimate the dissolution and carbonation of forsterite samples in order to predict CO 2 uptake processes and potentials. Moreover, thermodynamic equilibrium constants for the mineral carbonation reactions of forsterite were provided by model databases. In doing so, reaction kinetics were implemented by using a BASIC interpreter. The possibility of implementing reactions kinetic into the code as BASIC statements was also used to predict the reaction progress over time. Consequently, the quality and validity of the model system and the determined rate and equilibrium parameters were verified against the results of carbonation experiments with forsterite samples. The data from a sequence of laboratory efforts were applied for that purpose, which were performed in the aqueous autoclave mini reactor.
The kinetic analysis of the forsterite dissolution rates was determined according to "standard integral analysis Levenspiel's method" [65] in Mg-rich solution using HCl. The results were set into several heterogeneous reaction models represented by integral rate equations and then the multiple regression coefficients (R) were calculated. The shrinking core model described by Dri et al. [66] was applied for the constant size of forsterite particles. Based on this method, reaction rates take place at the outer surface of the unreacted particles, and heterogeneous reactions are controlled by the product layer diffusion (Equation (4)), film diffusion (Equation (5)) and chemical reaction control (Equation (6)). In addition, the possibility of having a compound effect of "chemical reaction control" and "product layer diffusion" was investigated using Equation (7).
kt " X E (6) kt " r1´3p1´X E q 2 3`2 p1´X E q s`r1´3p1´X E q 1 3 s In these equations. "t" is time (s) and "k" (s´1) and "X E " denote the rate constant and the extent of reaction, respectively.

Conclusions
This work has experimentally and numerically modeled the process by which carbon dioxide gas may be sequestered, in situ by reaction with forsterite and/or its extracted intermediate brucite (ex situ) in the presence of moisture. In both cases, we have found that the reaction is favored resulting in a high carbonate yield; going almost to completion with the bulk of the carbon partitioning into magnesite and that very little remaining in solution. In the presence of water vapor, the degree of mineral carbonation was increased, we suggest, due to an alternative carbonation pathway providing for faster reaction kinetics [64]. Despite the observations made in other studies, we suggest that hydromagnesite is an intermediate in these carbonation experiments but is converted into magnesite on the (hours) timescale of reaction studied here, although the mechanism is as yet unclear. Hydromagnesite is less desirable as final product as it corresponds to only 80% sequestration (relative to magnesite) on a molar basis. Moreover, we recognize that this system is itself not at thermodynamic equilibrium, but maybe take a very long period and recrystallization as sepiolite (Mg 4 Si 6 O 15 (OH) 2¨6 H 2 O) is one possible outcome. However, we have found no evidence that this occurs over the period of one year. The clay mineral sepiolite is known to form in nature from dolomite/silica assemblages in the presence of water, but its formation kinetics are slow and we do not expect its formation to occur spontaneously in an industrial carbonation plant. Thus, we propose that the carbonation of readily available forsterite is a viable route for carbon sequestration.
From the computational perspective, a CO 2 sequestration system with mineral carbonation can be treated as a fully-coupled problem between rock deformation, pore-fluid flow, heat transfer, mass transport, and chemical reaction processes. The results obtained from this study establish the viability of a geochemical model that can be used in to simulate the dynamic processes involved in CO 2 sequestration in the Mount Tawai peridotite, Malaysia.