CO 2 Capture by Accelerated Carbonation of Alkaline Wastes: A Review on Its Principles and Applications

CO 2 capture, utilization, and storage (CCUS) is a promising technology wherein CO 2 is captured and stored in solid form for further utilization instead of being released into the atmosphere in high concentrations. Under this framework, a new process called accelerated carbonation has been widely researched and developed. In this process, alkaline materials are reacted with high-purity CO 2 in the presence of moisture to accelerate the reaction to a timescale of a few minutes or hours. The feedstock for accelerated carbonation includes natural silicate-minerals (e.g., wollastonite, serpentine, and olivine) and industrial residues (e.g., steelmaking slag, municipal solid waste incinerator (MSWI) ash, and air pollution control (APC) residues). This research article focuses on carbonation technologies that use industrial alkaline wastes, such as steelmaking slags and metalworking wastewater. The carbonation of alkaline solid waste has been shown to be an effective way to capture CO 2 and to eliminate the contents of Ca(OH) 2 in solid residues, thus improving the durability of concrete blended with the carbonated residues. However, the operating conditions must be further studied for both the economic viability of the technology and the optimal conditions for CO 2 reaction.


INTRODUCTION
Human activities result in emissions of four long-lived greenhouse gases (GHGs): CO 2 , CH 4 , N 2 O and halocarbons, of which CO 2 is the most important anthropogenic GHG because it is responsible for about two-thirds of the enhanced greenhouse effect (Nyambura et al., 2011).The following are possible strategies for reducing CO 2 in the atmosphere: (i) reducing the amount of CO 2 produced (reducing energy intensity); (ii) using CO 2 (or reducing carbon intensity); (iii) capturing and storing CO 2 ; (iv) switching to less carbonintensive fuels; and (v) increasing the use of renewable energy (IPCC, 2005;Yang et al., 2008;Mikkelsen et al., 2010).The third strategy involves the development of innovative, available and cost-effective carbon capture and storage (CCS) technologies because of a 50-year estimate for the continued widespread burning of fossil fuels, the goal of reaching a 500 ppm atmospheric CO 2 concentration plateau, and the lag time needed for the development and implementation of new carbonless sources of energy (Pacala and Socolow, 2004;IPCC, 2005;Figueroa et al., 2008).
CCS technologies can be classified into three categories: pre-combustion capture, post-combustion capture, and oxycombustion (Figueroa et al., 2008).Table 1 presents the comparison of the post-combustion CO 2 capture technologies by absorption process and adsorption process using aqueous absorbents, solid adsorbent, ionic liquid (IL), and metal organic frameworks (MOFs).Other post-combustion capture technologies include membrane separation, enzymebase system, and bio-algae capture.However, in recent years, extending CCS technology to incorporate "utilization" has received dramatically global attention as a viable option for reducing CO 2 emissions.With the newly developed concept, i.e., moving CCS to Carbon Capture, Utilization and Storage (CCUS), Mainland Europe, and in particular Germany, the U.S. and Australia, are well-advanced in research and development of CCUS technologies (CSLF, 2011;Styring et al., 2011).The utilization routes of the captured CO 2 include enhanced fuel recovery (i.e., EOR and EGR), biological conversion (i.e., algae), food industry, chemicals (i.e., fertilizer and liquid fuel), refrigerant, inerting agents, fire suppression, plastics, and even mineralization as carbonates (i.e., precipitation calcium carbonates (PCC) and construction materials).CO 2 utilization by mineralization as carbonates is related to the accelerated carbonation technology (ACT).
Accelerated carbonation (chemical adsorption) is a feasible and attractive alternative method for the mitigation of CO 2 emissions in the coming decades because it involves storing CO 2 as carbonate precipitation, which is environmentally benign.The basic goal of accelerated carbonation is to mimic the natural weathering processes in which CO 2 reacts with metal oxide bearing materials to form stable and insoluble carbonates, with calcium-or magnesium-oxide being the most favourable metal oxide in reacting with CO 2 .Besides that, carbonation is an exothermal reaction, so energy consumption and costs may be reduced by its inherent properties (Lackner, 2003;Eloneva et al., 2008a;Baciocchi et al., 2009a).In all cases, carbonation must provide base ions, such as monovalent sodium and potassium, or divalent calcium and magnesium ions to neutralize the carbonic acid (Costa et al., 2007;Gerdemann et al., 2007).Other carbonate-forming elements such as iron carbonates are not practical due to their unique and precious features (Lackner, 2002).Industrial residues such as steelmaking slags, combustion residues, waste concrete, fly ashes, etc. are alkaline and also appear to be potential raw materials for CO 2 sequestration by accelerated carbonation due to the fact that these materials are generally rich in metal oxides including calcium, magnesium, aluminium, iron, and manganese oxide.The potential advantages of accelerated carbonation of applied industrial alkaline solid wastes include: (a) carbonation is an exothermal reaction and hence energy consumption and costs may be reduced by its inherent properties; (b) carbonation products such as calcium or magnesium carbonates are thermodynamically stable under ambient conditions, that is, in the absence of acidification.; (c) it offers great sequestration capacity due to the high availability of deposits; (d) it does not require transport at sites within steel-works and is therefore cost-effective; (e) products may be beneficially reused in a variety of application, such as construction materials; (f) carbonation eliminates environmental impacts due to decreased leaching of heavy metal trace elements such as Pb, Ni, and Cd from residues and stabilizing of the waste leading to an improvement of environmental quality; and (g) it could neutralise the pH of the solution as carbonate precipitations are formed if alkaline wastewater is used as the liquid agents (Lackner, 2003;Fernandez-Bertos et al., 2004a;Huijgen et al., 2005b, c;Costa et al., 2007;Eloneva et al., 2008a;Bonenfant et al., 2009;Huntzinger et al., 2009a;Gunning et al., 2010;Lim et al., 2010).
This paper focuses on accelerated carbonation of industrial alkaline wastes, e.g., steelmaking slags and metalworking wastewater.The principles of accelerated carbonation, the applications of carbonation of various wastes and the performance evaluations for CO 2 sequestration by carbonation of various alkaline wastes are reviewed to establish the state of the art for this emerging technology.

Natural Carbonation
Carbonation can occur naturally.Natural carbonation is also well known as "weathering", which eventually removes CO 2 from the atmosphere by neutralizing the acid with mineral alkalinity.Natural weathering occurs by the reaction between natural alkaline silicates and atmospheric CO 2 , as shown in Eqs. ( 1) and (2): Atmospheric CO 2 dissolves in rainwater, producing weak carbonic acid, and becomes slightly acidic by nature.Calcium and magnesium silicates are therefore leached from the mineral matrix after coming in contact with the rainwater.Rainwater thus carries the leached calcium and magnesium to rivers and subsequently to the ocean, where calcium and magnesium precipitates and forms solid carbonates (M 2+ represents alkaline earth metal element) as shown in Eq. ( 3) Chemical weathering of natural mineral ores is one of the main mechanisms contributing to the global geochemical carbon cycle.Nature's carbon cycle sequesters CO 2 as mineral carbonates by precipitation from the ocean in a geological timescale (Gerdemann et al., 2007).However, the kinetics of natural carbonation is extremely slow due to the relatively low CO 2 concentration, about 0.03-0.06%(Lackner, 2002;Costa et al., 2007).According to an investigation by Haug et al. (2010), the weathering rate of olivine is estimated to be 10 −8.5 mol/(m 2 s) using the average ground temperature in Norway of 6°C and a pH of 5.6, which corresponds to the acidity of rainwater.

Accelerated Carbonation
First proposed by Seifritz (1990), accelerated carbonation involves alkaline materials reacting with high-purity CO 2 in the presence of moisture to accelerate the reaction to a timescale of a few minutes or hours (Lim et al., 2010).Accelerated carbonation can be classified into two main types: mineral carbonation and alkaline solid waste carbonation.Mineral carbonation can be defined as the reaction between minerals and CO 2 where the product is at least one type of carbonate (Haug et al., 2011).Currently, accelerated carbonation has been primarily investigated through the direct aqueous route (Huijgen and Comans, 2006a;Costa, 2009;Uibu et al., 2009;Chang et al., 2011a).Carbonation can result in lowering of pH, affecting the solubility, and leaching of metals which are mobilized at high pH and fixed at low pH (Costa et al., 2007).
Currently, accelerated carbonation processes have been focused on assessing and maximizing the storage of CO 2 by optimizing the operating conditions including pressure, temperature, liquid-to-solid (L/S) ratio, gas humidity, gas flow rate, liquid flow rate, particle size, and solid pretreatment in numerous experimental investigations (Costa et al., 2007;Huntzinger et al., 2009b;Haug et al., 2010;Chang et al., 2011a, b).
The main process routes for accelerated carbonation of minerals are shown in Fig. 1.They are classified as: direct processes, where carbonation occurs in a single process step, and indirect processes, where alkaline earth metal is first extracted from the mineral matrix and subsequently carbonated.
The processes via which the carbonation of the minerals or alkaline solid wastes takes place in a single route step are referred to as direct carbonation.Usually, these processes can be divided into two types: gas-solid (dry) carbonation and aqueous (wet) carbonation; these processes are generally operated at an L/S ratio of less than 0.2 and more than 5, respectively (Lim et al., 2010).The direct gas-solid carbonation process, first studied by Lackner et al. (1995), consists of converting oxide or silicate minerals directly to carbonates using gaseous or supercritical CO 2 .Lackner et al. (1997) reported that the highest conversion by direct gas-solid carbonation of silicates was found to be 25% by exposing 100-μm serpentine particles to a CO 2 pressure of 340 bar and a temperature of 500°C for 2 h.Drawbacks of the gas-solid process that have prevented its widespread large-scale use to date are (i) the slow kinetics of carbonation chemistry and (ii) the significant energy input requirement (IPCC, 2005).However, the gas-solid process is much more economical than wet carbonation using aqueous solutions (Zevenhoven et al., 2008).Zevenhoven et al. (2008) investigated the gas-solid process based on MgO-based minerals, and they found that the process for MgO extraction followed by MgO hydration and Mg(OH) 2 carbonation, compared with the process for MgO extraction followed by MgO carbonation, is preferable from an exergy point of view.Also the process appeared to show more rapid carbonation kinetics.Therefore, MgO-based mineral carbonation requires a stepwise process, combining the extraction of the MgO from a mineral with carbonation of the MgO, at presumably different temperatures and pressures.
In indirect carbonation processes, e.g., pH-swing methods (Park and Fan, 2004;Eloneva et al., 2008a;Kodama et al., 2008;Doucet, 2010;Wang and Maroto-Valer, 2011), reactive alkaline earth metals are first extracted from the solid matrix and then carbonated.Fig. 2 shows the typical indirect carbonation for carbonating minerals using HCl.The drawback of this process is the high energy consumption for evaporation of the aqueous solution and the large variations in free energy resulting from the necessary formation of intermediate products (Teir, 2008).Moreover, Kakizawa et al. (2001) suggested the use of acetic acid for the extraction of calcium from a calcium-rich feedstock in two-steps.The general problem of this process is that other elements, such as heavy metals, may also leach out during the extraction step, thus leading to impure carbonate precipitation.
According to a study by Teir (2008), where the Gibbs free energy of the carbonation reactions was calculated using Outokumpu HSC 5.1, the carbonation of calcium ions proceeds at temperatures over 45°C, while the carbonation of magnesium ions should be possible only at temperatures over 144°C.The results also indicated that the dissolution reactions of calcium and magnesium are more exothermic (∆H < 0), while the carbonation reactions are less endothermic (∆H > 0), giving an exothermic net reaction.Mineral carbonates are thermodynamically stable and could theoretically store CO 2 permanently under ambient conditions, since they have a lower energy state than their reactants (CO 2 and silicates).Mineral carbonates, however, can be readily dissolved in the presence of strong acids in ambient conditions.Therefore, there is still a small risk that CO 2 gas could be released into the atmosphere if mineral carbonates make contact with acids, e.g., acid rain (pH 5-7).

Alkaline Wastes as Adsorbents
Both alkali metals (i.e., Na, K, etc.) and alkaline earth metals (i.e., Ca, Mg, etc.) can be carbonated from a chemical elements perspective and a thermodynamic view point (Huijgen and Comans, 2003;2005a).Alkali (bi)carbonates, however, are soluble in water, which would result in releasing CO 2 back into the atmosphere.This thus suggests that alkali metals are unsuitable for the long-term storage of CO 2 .In addition, a number of other metals such as Mn, Fe, Co, Ni, Cu, and Zn could potentially be carbonated, but most of these metals are impractical due to their unique and precious features.To provide significant storage of CO 2 , large amounts of raw materials are required as feedstock for carbonation, which must be abundant and cheap (Teir, 2008).
Fig. 3 shows the relationship of CaO content with the actual CO 2 capture capacity of various alkaline solid wastes in the literature.The CaO contents were significantly high in these wastes, e.g., steelmaking slag (~30-60% wt.CaO), residues from APC (Ca content up to 35%) and bauxite (4.8% Ca), cement kiln dust (~34-50% wt.CaO), oil-shale waste (CaO content up to 50%), and fly ash (53% CaO) from municipal solid waste incinerators (Huijgen et al., 2005b;Bonenfant et al., 2008;Eloneva et al., 2008a;Baciocchi et al., 2009c;Wang et al., 2010;Chang et al., CaO  2011a).Industrial alkaline solid wastes such as sources of calcium or magnesium oxide are ideal CO 2 sequestration materials due to their availability and low cost.These materials are generally rich in calcium-content and often associated with CO 2 point source emissions so no mining is needed and the consumption of raw materials is avoidable.In addition, these solid wastes tend to be chemically less stable than geologically derived minerals (Huijgen et al., 2006c).Carbonation of industrial solid waste does not generally require the extraction of reactive ions from the solid matrix due to the alkaline-containing silicates, oxides and hydroxides as the mainly reactive phase (Costa, 2009).Worldwide, the iron and steel industry accounts for 6-7% of total CO 2 emissions (Doucet, 2010).The industry is moving toward environmental sustainability through careful control of GHG emissions and appropriate management of steel manufacturing residues generated.Within this framework, a combined approach aimed at capturing CO 2 while also improving the environmental and mechanical properties of steelmaking slags, appears to be highly desirable.Therefore, CO 2 capture by carbonating the steelmaking slags could be an interesting option for reducing CO 2 emissions from the steel plant.In recent years, accelerated carbonation of steelmaking slags has been progressively tested to assess the CO 2 capture potential of this material (e.g., Huijgen et al., 2005c;Bonenfant et al., 2008;Teir, 2008;Eloneva et al., 2008a;Kodama et al., 2008;Costa, 2009;Wang and Yan, 2010;Chang et al., 2011a, b;De Windt et al., 2011).
Fig. 4 shows the main chemical composition of various alkaline wastes, e.g., steelmaking slag, cement, and fly ash in the literature.Along with GHG emissions from the steel manufacturing industry, considerable amounts of alkaline solid residues such as fly-ash (FA) slag, ultra-fine (UF) slag, basic oxygen furnace (BOF) slag, and blended hydraulic slag cement (BHC) are generated, which are either used in various applications or ultimately landfilled.BOF slag is a residue from the basic oxygen converter in the steelmaking process (about 10% of the steel production) and has the potential of capturing 6 to 10% of BOF CO 2 emissions (considering about 519 kg CO 2 production per ton steel produced) (Costa, 2009;Wang and Yan, 2010;De Windt et al., 2011).While global theoretical CO 2 emissions reduction potential of steelmaking slag carbonation is only 170 Mt/year (Eloneva et al., 2009), the reduction could be quite significant for an individual steel industry.
The chemical compositions of BOF slag, primarily in the form of Ca and Mg silicates, are highly variable due to the differing characteristics of the raw materials used and types of manufacturing and smelting processes used in steel production.On the other hand, the physical properties of the slags depend chiefly on the cooling techniques applied.The mineralogical compositions of the steelmaking slags are generally more complex than those of other types of residues such as APC ash (Costa, 2009).Hence, those mineralogical investigations are crucial in order to identify the possible CO 2 -reacting phases precisely, as well as the main reaction products contributed by the carbonation process.The physico-chemical properties of various alkaline solid wastes used in studies reported in the literature are presented in Table 3.The CaO content in the BOF slag was higher than that in the UF and BF slag.The typical chemical compositions of BOF slag consist of CaO 38-60%, SiO 2 10-15%, Al 2 O 3 1-5%, Fe 2 O 3 3-20%, MgO 3-13%, FeO 5-20%, and P 2 O 5 1-4% (Shi andQian, 2000, CSC Group, 2003).The common mineralogy of BOF slag includes olivine, merwinite, C 3 S (tricalcium silicates), larnite (C 2 S, dicalcium silicates), C 4 AF, C 2 F, RO phase (CaO -FeO -MgO solid solution), free lime (CaO), and periclase (MgO) (Shi and Qian, 2000;Costa, 2009;Birat, 2009).Free lime (as much as 6% in total) in steelmaking slag comes from two sources: residual free lime from the raw material  and precipitated lime from the molten slag (Costa, 2009;Birat, 2009).The hydrolysis of BOF slag can induce a rapid increase of alkalinity (e.g., Ca-OH), and the reactivity of steel slag increases with its alkalinity.Hence, the alkalinity A = CaO/(SiO 2 +P 2 O 5 ) can be used to evaluate the hydration activity of steel slag (Shi and Qian, 2000;Uibu et al., 2009b;De Windt et al., 2011).If alkalinity > 2.4, the steelmaking slag should be considered as highly reactive (Shi and Qian, 2000).

Process Chemistry
Mineral carbonation, a type of accelerated carbonation that utilizes natural ores such as wollastonite (CaSiO 3 ), olivine (Mg 2 SiO 4 ), and serpentine (Mg 3 Si 2 O 5 (OH) 4 ) to react with high-purity CO 2 as shown in Eqs. ( 4) to ( 9), has been extensively researched in the literature (e.g., O'Connor et al., 2002;Costa et al., 2007;Gerdemann et al., 2007).The process is based on acid-base reactions in which carbonate acid is neutralized by a base (alkaline mineral).The major challenge hindering the large-scale use of silicate minerals for carbonation is their slow conversion rate.Therefore, most research in this field has focused on identifying reaction pathways by bench-scale experiments for determining reaction rates, as well as characterization of the mineralogy of the reactants and products. (5) Fe 2 SiO 4(s) + 2CO 2(g) + 2H 2 O (l) → 2FeCO 3(s) + H 4 SiO 4(aq) ( 8) The overall carbonation reaction for magnesium and calcium silicates can be expressed as Eq. ( 10) (Teir, 2008): Fig. 5 illustrates the proposed mechanism of accelerated carbonation, which indicates that the carbonation reaction occurred in four routes.Similar results were also found in Huntzinger et al. (2009b) and Uibu and Kuusik (2009a), suggesting some mechanisms that can affect the rate and extent of carbonation: In addition, some investigations (Thiery et al., 2007;Montes-Hernandez et al., 2009) suggested a simple reaction mechanism for CO 2 sequestration by steelmaking slag in two successive steps: first, the irreversible hydration of calcium oxide or lime as shown in Eq. ( 15) and, second, the spontaneous carbonation of calcium hydroxide suspension as shown in below Eq. ( 16): Not only Ca(OH) 2 but also other hydrate compounds can react with CO 2 .For instance, the calcium-silicatehydrates (C-S-H) can react with CO 2 , and produce CaCO 3 and a silica gel (Thiery et al., 2007;Tsuyoshi et al., 2010).Uibu et al. (2011) found that Ca 2 SiO 4 and Ca 3 Mg(SiO 4 ) 2 were the main CO 2 binding components in electric arc furnace (EAF) slag according to quantitative XRD analysis, and the main product was calcite.A similar conclusion was made by Fernández-Bertos et al. (2004a), which indicated that C-S-H becomes rapidly activated by CO 2 where carbonation is accomplished by hydration at first and is followed, after some delay, by secondary carbonation as shown in Eq. ( 17).However, Tsuyoshi et al. (2010) suggested that γ-C 2 S does not react with water, and the leaching of Ca 2+ and O 2− (or OH − ) is very low at normal temperature.Hence, CaCO 3 is directly generated in the carbonation of γ-C 2 S without the formation of intermediate products such as Ca(OH) 2 , due to the catalytic action of capillary water.On the other hand, the breakdown of C 2 S and the formation of calcite can be observed by XRD (Fernández-Bertos et al., 2004b).
The actual process involving steelmaking slag should be more complicated, because most of the CaO is not present in pure form and combined with silicates or another complex oxide phase.For simplicity, we assumed that the aqueous carbonation by using steelmaking slags occurred mainly through three reaction routes: (a) the dissolution of CO 2reactive species from a solid matrix; (b) the contemporary dissolution of CO 2 into a liquid phase and conversion of carbonic acid to carbonate/bicarbonate ions; and (c) the consequent nucleation and precipitation of carbonates.The balance between dissolution and precipitation is dependent on the kinetics and solubility of the present feedstock and possible products, where carbonates are probably the most important products.
The affinity of oxides for carbonate formation principally depends on the strength of the chemisorptions of CO 2 and number of basic sites at the surface and varies as follows: basic oxides (CaO, MgO) > amphoteric oxides (Al 2 O 3 , Cr 2 O 3 , TiO 2 , MnO, iron oxides) > acidic oxides (SiO 2 ) (Bonenfant et al., 2009).Hence, the capacity of sequestered CO 2 for these alkaline residues depends directly on the proportion of binary oxide (CaO and MgO) and/or hydroxide (Ca(OH) 2 and Mg(OH) 2 ) content in the matrix, which are primarily responsible for the CO 2 sequestration potential (Montes-Hernandez et al., 2009;Bonenfant et al., 2009).Moreover, CaO offers more potential for chemisorptions of CO 2 than MgO due to its basic character (Bonenfant et al., 2009).

Ion Equilibrium in Solution
The first step of carbonation was believed to be the leaching of calcium in solid particles, and the properties of the solids may change considerably due to the leaching process.The rapid and strong increase in pH (10.88-11.88) is due to the dissolution of lime and larnite in the solid residues, as shown in Eqs. ( 18) and ( 19) (Uibu et al., 2009b;De Windt et al., 2011) However, CaO and MgO are rarely present in pure form in steelmaking slag.The alkalinemetal oxides are primary locked into the silicate, aluminate or ferrite phase (Lekakh et al., 2008).
The results reported by McCabe et al. (2005) suggest that coarse, hard, or granular feed solids should be disintegrated into pulp or mush when their soluble content is removed.Huijen and Comans (2006a) also suggested that calcium leaching and carbonation in steelmaking slag would occur simultaneously, and carbonation may reduce the leaching of alkaline earth metals (except Mg) by conversion of a Ca-phase such as portlandite, ettringite, and Ca-Fesilicates into calcite, possibly containing traces of Ba and Sr.
The second step of carbonation was assumed to be the dissolution of CO 2 into the solution.CO 2 has a critical point at 31.06°C and 73.8 bars (1,070 lb/in 2 ) and a critical density of 0.469 g/cm 3 .It could be dissolved from the atmosphere in available water, and carbonic acid can be created with a pH around 5.6 (Haug et al., 2010).In addition, diffusion of CO 2 in water is approximately 10000 times lower than in air.
CO 2 can be physically absorbed in water (or a solvent) in accordance with Henry's law.Henry's law states that the relationship between the gas solubility in pore water and the partial pressure of the gas is strictly valid only for gases that can be infinitely diluted in solution.The binding of the solvent with CO 2 occurs at high pressure, and a reduction in pressure releases the gas (Klemeš et al., 2007).The amount of CO 2 dissolution in water can be expressed by Henry's law as shown in Eq. ( 20): where C' is the concentration of CO 2 dissolved in aqueous solution (M); H ' CO2 is Henry's constant for CO 2 (= 10 −1.46 M/atm at 25°C); and P CO2 is the partial pressure of CO 2 in the gas phase (atm).Hence, the higher partial pressure of CO 2 will result in a large amount of dissolved CO 2 .In addition, Henry's constant is a function of temperature that can be modified by Eq. ( 21) (Morel and Hering, 1993): where C is the constant for all gases (2400 K for CO 2 ) and T is the temperature (K).
where K a = 10 −6.3 and K b = 10 −10.3 at 25°C.Fig. 6 shows the mole balance and equilibrium conditions for carbonation of alkaline solid wasts.The mole balances of the carbonic acid system can be expressed as : where C T is the total inorganic carbon (TIC) concentration (M).By substitution of Eq. ( 24) into Eq.( 25 [HCO 3 The fraction (α i ) of each carbon species present is dependent on the solution pH.At a low pH (~4), the production of H 2 CO 3 dominates, at a mid pH (~8) HCO 3 − dominates, and at a high pH (~12) CO 3 2− dominates (Morel and Hering, 1993).Therefore, accelerated carbonation is favoured at a basic pH due to the availability of carbonate ions.

2−
] are the activities of calcium and carbonate ions (M) in the solution, respectively.In nature, calcium carbonate occurs most commonly in hexagonal form (as calcite) but also occurs in orthorhombic form (aragonite) (McCabe et al., 2005).The super-saturation ratio of [Ca 2+ ] and [CO 3 2− ] in the interface can be expressed as shown in Eq. ( 32).A low super-saturation ratio is disadvantageous for the nucleation of CaCO 3 (Wang, 2004).
Dissolution or crystallization of precipitations does not occur instantaneously.It was suggested that some characteristic time, often longer than the time constant for the overall process, is needed to achieve a new equilibrium (Salminen and Prausnitz, 2007).The carbonate precipitation using steelmaking slag as feedstock should be mainly calcite (CaCO 3 ) (Bonenfant et al., 2008;Eloneva et al., 2008a;Costa, 2009;Wang and Yan, 2010;Chang et al., 2011a, b).The crystal volume of calcium carbonate is approximately 11.7 percent more than that of calcium hydroxide (Ishida and Maekawa, 2000).In general, the pore structures of solid wastes are thought to become finer with carbonation, thus leading to a solid of lower porosity, lower tortuosity, and lower pore area with calcite infilling the pore space after carbonation (Ishida and Maekawa, 2000;Fernández-Bertos et al., 2004a).

Reaction Kinetics Leaching of Metallic Ions from Solid Matrix
The rate and extent of calcium leaching in steelmaking slag was found to be inversely related to particle size and pH, and increased with increasing temperature, pressure, and surface area (Iizuka et al., 2004;Tai et al., 2006;Costa et al., 2007;Gerdemann et al., 2007;Lekakh et al., 2008;Teir, 2008;De Windt et al., 2011).Based on a report by Alexander et al. (2005), the effects of increased reaction temperature from 25°C to 50°C provided a 70% improvement on dissolution of the primary phase.Iizuka et al. (2004) proposed a new CO 2 sequestration process using carbonation of waste cement, which indicated that calcium extraction increased with pressure.They also found that the influence of S/L ratio was significant, with the calcium extraction rate about 50% with an S/L ratio of 0.29 wt% and 22% with the ratio of 2.9 wt%.
In addition, research by Huijegn et al. (2005c) using steelmaking slag as feedstock for carbonation suggested that the leaching of Ca 2+ -ions should result in a withdrawing Ca-silicate core surrounded by a Ca-depleted SiO 2 phase.This SiO 2 rim apparently hindered the diffusion of Ca 2+ions from the particle interior, resulting in a declining reaction rate.Hence, Ca 2+ -ions leaching rate, probably determined by diffusion through the Ca-depleted silicate rim, rather than by the boundary layer at the solid-liquid interface, seems to be the main rate-limiting step in aqueous carbonation (O'Connor et al., 2002;Huijgen et al., 2005c;Santos et al., 2009).

Dissolution of Carbon Dioxide into Solution
The kinetics of dissolution of CO 2 and dehydration of H 2 CO 3 has been studied intensively (Ishida and Maekawa, 2000).It was noted that these two reactions should occur simultaneously as shown in Eq. ( 33) (Morel and Hering, 1993).( 33) Therefore, the rate expression of the dissolutiondehydration reaction is: By substituting Eq. ( 24) into Eq.( 34), we obtain: where the overall rate constants in Eq. ( 35) were simplified to be k CO2 and k H2CO3 .The values of k CO2 and k H2CO3 at 25°C were 0.032 s −1 and 26.6 s −1 , respectively (Edsall and Wyman, 1958).However, at higher pH (e.g., pH > 9), an alternative reaction pathway would be expressed as (Morel and Hering, 1993): where k 4 (i.e., 8500 1/M-s at 25°C) and k -4 (i.e., 0.0002 s −1 at 25°C) are the rate constants (Stumm and Morgan, 1981).

Rate of Carbonate Precipitation
The carbonation reaction is regulated by solution equilibrium, and the reaction of calcium ions combining with carbonate ions is very fast.It was widely assumed that the carbonation reaction as shown in Eq. ( 29) was of the first order with respect to the concentrations of Ca 2+ and CO 3 2− (Ishida and Maekawa, 2000); therefore, the rate of carbonation could be expressed by the following differential Eq. ( 37).
where C CaCO3 is the concentration of calcium carbonate (mole/L), and k is the reaction rate coefficient (1/mol-sec).
The precipitation rate is related to the CO 3 2− concentration in the liquid phase, but not to the concentration of other species containing carbonate.In addition, the reaction rate coefficient (k) is dependent on the reaction temperature according to Arrhenius's law.On the other hand, the Boltzmann distribution law could be applied to determine the ratio of the molecular having a kinetic energy above the activation energy of the chemical reaction.The value of k reported by Ishida and Maekawa (2000) was 2.08 (L/mol/sec) at 25°C from several sensitivity analyses.
As the calcium ions are converted to metal carbonates and precipitated out, more calcium hydroxide dissolves to equalize the concentration of metal ions.Although the Ca 2+ -ion dissolution kinetics improved with increasing temperature, carbonation precipitation was retarded at higher temperatures due to reduced CO 2 solubility (Park et al., 2003;Costa et al., 2007;Gerdemann et al., 2007;Chang et al., 2011a).In addition, limited MgCO 3 formation for carbonation of steelmaking slag under ambient conditions was expected due to the relatively low magnesium oxide content in the slag, low pressure of CO 2 and short reaction times.Typical process conditions for the formation of magnesium carbonation via aqueous carbonation are p CO2 > 100 bar and a reaction time of hours (Huijgen et al., 2006c).However, MgCO 3 formation could be observed when natural ores (e.g., serpentine, olivine, etc) were selected as the feedstock for carbonation (Alexander et al., 2007;Krevor and Lackner, 2011).Furthermore, the study reported by Haug et al. (2011) indicated that the mineral carbonation of olivine (generally rich in Mg contents) at 115 bar and 185°C is limited by MgCO 3 (magnesite) precipitation and not the olivine dissolution rate.O'Connor et al. (2002) also found that conversion of the silicate to carbonate does not occur in the solid state but appears to require mineral dissolution in the aqueous phase.

Summary of Theoretical Considerations
Minerals formed in pure calcium or magnesium oxide are rarely found in nature due to their high reactivity; thus, they instead present in a variation of these oxides in a silicate matrix.In addition, vast quantities of silicate minerals are required to sequester a significant fraction of emitted CO 2 .On the other hand, it currently seems that aqueous carbonation of industrial solid wastes combined with alkaline wastewater is the most attractive route due to lower material and energy requirements.
From the view point of process chemistry, the challenge is that the dissolution of calcium species in alkaline solid waste is favoured at low pH; however, it is not favoured for the precipitation of calcium carbonate.The balance between dissolution and precipitation is dependent on the kinetics and solubility of the present feedstock and possible products, where carbonates are probably the most important species.Therefore, finding a balanced operating condition between these two mechanisms is essential for optimizing the overall carbonation process.
From the view point of reaction kinetics, many researchers have attributed the rate-determined step of aqueous carbonation to reacted (product) layer diffusion.Therefore, this suggests that a well-designed reactor to enhance the mass transfer between the gas, liquid, and solid phases is needed to facilitate the carbonation reaction and increase the carbonation conversion.

Performance Evaluation
The carbonation of alkaline solid waste has been proved to be an effective way to capture CO 2 (Uibu et al., 2009b;Wang and Yan, 2010;Chang et al., 2011a;Nyambura et al., 2011).Table 4 compares carbonation efficiency for different types of alkaline wastes operated under various conditions.In addition, summaries of factors required for effective accelerated carbonation are presented in Table 5.Multiple factors have been shown to affect the conversion of accelerated carbonation, as indicated by numerous investigations (e.g., Huijgen et al., 2006c;Costa et al., 2007;Li et al., 2007;Chang et al., 2011a).These factors must be further understood, since they determine the economic viability of the technology as well as help to identify the conditions that most favourable to reaction.
Particle size and specific surface area are the most important factors affecting the dissolution kinetics of any kind of material (Huijgen et al., 2006c;Eloneva et al., 2012).Since slag grinding is expected to be a fairly energyintensive process, it is important to find out how large a particle size can be used.Average grain particle sizes of less than 100-150 μm are already in the optimum range for efficient carbonation (Costa, 2009).Eloneva et al. (2012) also suggested that a particle size preferably of ~100 μm, or no more than 500 μm, was found to be optimal based on the kinetics studies of calcium leaching.In addition, over 90% and less than 40% of calcium species dissolved in 24 hours were observed from 45-75 and 150-250 μm sized ladle slag, respectively (Lekakh et al., 2008).The power consumption for grinding (crushing) can be calculated with Bond's equation as shown in Eq. ( 38) (Bond, 1961), which has been widely used in the literature (Huijgen et al., 2006b;Kodama et al., 2008;Eloneva et al., 2012): where W (kWh/ton) is the power consumption, W F80 (μm) and W P80 (μm) are the 80% passing size of the feed and the product, respectively, and W i (kWh/ton) refers to the work index of the slag.The work index for the slag obtained from the ball-mill test is approximately 30.4 kWh/ton (Kodama et al., 2008).
The CO 2 content of the incoming gas did not seem to have a significant effect on the degree of CaCO 3 precipitation.As expected, the reaction time was inversely proportional to the CO 2 content of the ingoing gas flow: the lower the gas CO 2 content, the longer the duration of the carbonation (Eloneva et al., 2008a).
Temperature also had a significant effect but only during the first 3 hours of the dissolution reaction according to the results by Lekakh et al. (2008).An investigation reported by Tai et al. (2006) also suggested that reaction temperature and solution composition should be the most important factors in the carbonation test.The carbonation reaction rate significantly increased with increasing reaction temperature; however, the amounts of CO 2 dissolution into the solution decreased at higher temperature.Chang et al. (2011a) suggested that the carbonation conversion of BOF slag in a slurry reactor is dependent on the reaction temperature and exhibits a maximum value at 60°C if the reaction temperature ranges between 30°C and 70°C.Kodama et al. (2008) also suggested that a reaction temperature higher than 80°C is unfavourable for the process design.
The pH value is another important factor related to the rate of calcium leaching, the rate of CO 2 dissolution, the rate of CaCO 3 nucleation, and the specific species that are precipitated (Druckenmiller and Maroto-Valer, 2005;Tai et al., 2006).According to a report by Park and Fan (2004), the optimum pH for aqueous carbonation is around 10, while the dissolution of BOF slag occurs under low pH conditions.The pH decreases continuously as carbonation proceeds due to CO 2 dissolution into solution (Santos et al., 2009).There is no change in pH after the carbonation is completed.On the other hand, it has been reported that sodium and potassium ions should cause pH fluctuations and also affect the carbonation rate (Ishida and Maekawa, 2000).Li et al. (2007) studied the influence of L/S ratio on accelerated carbonation of municipal solid waste incineration fly ashes and found that the optimum L/S ratio was 0.3 by weight of ash at ambient temperature.Fernández-Bertos et al. (2004a) made similar observations, finding that the optimum L/S ratio was 0.2-0.3 for APC residues and 0.3-0.4 for BA residues.Moreover, Chang et al. (2011a, b) suggested that the optimum L/S ratio for carbonation of steelmaking slag is 10:1-20:1 (mL/g).
Another important factor is the surface activities of the feedstock.Various studies have addressed methods to speed up the kinetics of direct aqueous carbonation including physical and chemical pretreatment (Maroto-Valer et al., 2005a, b), electrolysis and heat pretreatment (Li et al., 2009), and mechanical activation methods (McKelvy et al., 2004;Park and Fan, 2004;Li et al., 2009;Haug et al., 2010;Haug et al., 2011). O'Connor et al. (2002) investigated approaches to reduce particle size, and increase surface area, and elevate process temperature and pressure by thermal treatment or steam activation.The chemical activations were found to be more effective than physical activation at  (Maroto-Valer et al., 2005a).In addition, O'Connor et al. (2002) evaluated the influence of heat-pretreated serpentine and a supercritical CO 2 on conversion, which indicated that the maximum conversion of 78% could be achieved in 30 minutes at 155°C, P CO2 = 185 atm and 15% solids.It is possible to enhance the reactivity of the minerals or alkaline solid wastes considerably, while these methods require additional chemicals and energy.With the present technology, there is always a net demand for high-grade energy to drive the mineral carbonation process that is needed for (i) preparation of the solid minerals, including mining, transport, grinding and activation when necessary; (ii) processing, including the energy associated with recycling and possible losses of additives or catalysts; (iii) disposal of carbonates and byproducts (IPCC, 2005;Gerdemann et al., 2007).
From the view point of process modification, two-stage carbonation processes (indirect carbonation) have also been investigated in recent years to optimize the efficiency of the dissolution and carbonation processes, as well as to achieve a pure precipitate product for specific industrial applications (Eloneva et al., 2008a, b;Kodama et al., 2008;Teir et al., 2009).Kodama et al. (2008) developed a process using the pH-swing of a weak base -strong acid solution, with the finest particle size (< 63 μm) at 80°C, in which the maximum calcium extraction yield and carbonation conversion achieved were 60% and 70%, respectively.Another research group investigated the performance of enhancing dissolution using acetic acid solutions and other chemicals (Eloneva et al., 2008a, b;Teir et al., 2009).The results indicate that a maximum conversion of 86% was achieved under a CO 2 concentration of 10% vol.at 30°C, with a precipitate (CaCO 3 ) purity of up to 99.8%.
On the other hand, the developed rotating packed bed (RPB) process possessed a relatively higher conversion of BOF slag with a short reaction time compared to the slurry and autoclave reactors (Chang et al., 2011a, b;2012).According to the previous study by Chang et al. (2012), the maximum conversion of BOF slag achieved in an RPB was 93.5% at 65°C and 30 min, corresponding to a capture capacity of 0.29 kg CO 2 per kg of slag.However, the weakness of the developed RPB process was the large amount of energy requirements for the rotation of packed bed.Therefore, the operating modulus and conditions should be upgraded to minimize the energy consumption, and then maximize the overall capture capacity.

Utilization of Carbonated Residues
Table 6 summarizes the description of alkaline solid wastes for accelerated carbonation and utilization.The most common methods for disposal of MSW, for example, are: landfill (82%), recycling or composting (11%) and thermal treatment (7%) (Fernández-Bertos et al., 2004a).On the other hand, the treatment approach for BOF residue is dumping and partial reuse as an aggregate for civil engineering due to its good technical properties (Geiseler, 1996;De Windt et al., 2011).In the United States, steelmaking slag is currently sold and used for various applications, predominantly for asphalt and concrete aggregate and road base (Stolaroff et al., 2005).In Taiwan, steelmaking slag research has been done in the following fields: (a) cement industry, (b) road construction, (c) glass making, (d) concrete or asphalt aggregate, (e) ceramic tiles, (f) agriculture, soil conditioner, (g) aquaculture, and (h) land amendments (CSC Group, 2003).Some research has already been done on the use of steelmaking slags in construction material, e.g., as alkaline fillers in the papermaking industry and asphalt concrete applications (Lim et al., 2010), in bricks and tiles (El-Mahllawy, 2008), as aggregate for the production of bituminous paving mixtures (Sorlini et al., 2012), and in cement-based materials as cement/sand replacements or binder known as "alkaline cements" (Fernández-Jimenez et al., 2006;Lim et al., 2010;Zhang et al., 2011).
Nevertheless, reuse possibilities of carbonated slags are limited due to the small particle size of the feedstock needed for efficient carbonation reaction.In addition, there are both chemical and physical changes in the structure and surface characteristics of steelmaking slag properties after carbonation (Chang et al., 2011a).Fernández-Bertos et al. (2004a) found that the carbonated product is a solid of lower porosity, lower tortuosity, and lower pore area  (Rendek et al., 2006) More than 50% is used as a secondary building material, road bases, and civil engineering structures in Europe (Rendek et al., 2006) Replace the sand or gravel fraction in concrete bricks or Reuse as secondary raw construction material (Arickx et al., 2006).

Cement kiln dust
Beneficial to agriculture application; however, increased concerns over the health and environmental hazards (Huntzinger et al., 2009b) Lime (CaO); Portlandite (Ca(OH) 2 ); Calcium silicates; Gehlenite (Ca 2 Al(AlSiO 7 )) (Huntzinger et al., 2009b) Recycling and reuse as a road base material is an established process (Iizuka et al., 2004) Reuse as road base or construction material.No free-CaO was observed in the reacted samples (Huntzinger et al., 2009a) with calcite infilling the pore space.However, both of those changes can be beneficial for subsequent uses.First, most applications for steelmaking slag require that it be crushed and sieved into relatively small particles, which need extremely fine grinding for a cement additive.Therefore, this can avoid additional energy consumption and the cost of grinding specifically for CO 2 sequestration.On the other hand, the hydration of CaO and MgO in the fresh slag may be undesired for usage as concrete and asphalt aggregate, road base, and fill materials due to the tendency of fresh slag to have high water absorption and expansion properties.This property would be eliminated in the course of carbonation.Carbonation is proven to be an effective method to improve the durability of concrete blended with the carbonated slag because relatively insoluble CaCO 3 is formed from the soluble Ca(OH) 2 in the slag (Tsuyoshi et al., 2010).Furthermore, the carbonated particles became coarser due to agglomeration, which might be beneficial for use in aggregate manufacturing (Fernández-Bertos et al., 2004b).There are still some important issues for accelerated carbonation, such as the fate of the carbonated products, including calcium carbonates and magnesium carbonates.In recent years, some research has focused on the production of CaCO 3 from the carbonation of steelmaking slag because it could also be a profitable refining method for slag products if the purity requirements for commercial PCC (precipitated calcium carbonate) can be achieved (Teir, 2008;Eloneva et al., 2012).The potential for using the calcium carbonate produced from the carbonation process relies on the purity and variety of the crystal structures that can be achieved.Calcium carbonate could be used in the pulp, plastic, rubber, paint, and paper industries as a filler (instead of clay) and as coating pigments to provide opacity, high brightness, and improved printability due to its good ink receptivity (Eloneva et al., 2008b;Teir, 2008).The most important crystalline forms of calcium carbonates are the rhombohedral calcite type, the orthorhombic acicular aragonite type, and scalenohedral calcite, which are the favoured form in most applications (Imppola, 2000).

Life Cycle Assessment (LCA)
Since CCS is an energy-intensive process, it would consume additional energy and lead to further emissions of CO 2 .This is why a life cycle assessment (LCA) of the CCS process is particularly important (IPCC, 2007).The major purpose of LCA is to quantify the environmental impacts of energy consumption technologies over their lifetimes (e.g., GHG emissions, acidification, etc), using a "cradle to grave" approach (De Haes and Heijungs, 2007).LCA is also a systematic approach to estimate the performance as well as its associated environmental impacts of a product or a certain kind of service in the life cycle.The LCA process was structured by the International Standards Organization (ISO) 14040-14044, which provided fourstep guidelines for the practitioner: (1) definition of goal and scope; (2) life cycle inventory analysis (LCI); (3) life cycle impact assessment (LCIA); and (4) interpretation.Park and Fan (2004) showed that the net CO 2 emission of the aqueous carbonation of serpentine by the pH-swing method was -0.02 kg CO 2 /kg feedstock.However, prior investigations using the pH-swing method consumed a great amount of base and acid solvents, which would cause adverse effects on the environment.An investigation reported by Chang et al. (2011b) showed a capture capacity of 0.098 kg CO 2 per kg BHC in an autoclave reactor.Energy consumption including grinding, sieving, pressuring, and heating, was taken into account while carrying out the LCA.However, the process needed to be operated for at least 2 hr to achieve such a result.On the other hand, according to an investigation by Odeh and Cockerill (2008), the GHG emissions from UK fossil fuel plant stations with CCS can be reduced by 75-84% relative to a subcritical PC power plant.Therefore, CCS has the potential of decreasing GHG emissions proved by the results of LCA.On the other hand, CCS may increase other environmental impacts such as eutrophication or acidification due to the increase in the concentrations of other pollutants.Hence, the effects of CCS should be weighed and compared carefully due to changes in the environmental impacts (Odeh and Cockerill, 2008).In pulverized hard coal combustion, the energy penalty was found to range from 5-18% with application of the various CCS processes (Marx et al., 2011).

Summary of Practical Considerations
In general, the dissolution of calcium species in alkaline solid waste was favored at the low pH and high temperature, which is not favored for the precipitation of calcium carbonate.It was noted that the accelerated carbonation include a large amount of energy requirements and the high costs (IPCC, 2007).It was thus concluded that the challenges encountered are to accelerate the reaction and utilize the heat of reaction to maximize the capture capacity and minimize energy demand, as well as to determine the optimal operating conditions.Therefore, alkaline solid wastes can be used to sequester great amount of CO 2 , especially if the wastes are generated nearby the source of CO 2 , for achieving both the environmental and economic benefits.
The rate of carbonation reaction can be enhanced by adding acids or caustic alkali-metal hydroxide (Beard et al., 2004), utilizing a bicarbonate/salt (NaHCO 3 /NaCl) mixture (Maroto-Valer et al., 2005a;O'Connor et al., 2005), employing various pretreatment techniques (i.e., including physical and chemical pretreatment (Maroto-Valer et al., 2005a, b), electrolysis and heat pretreatment (Li et al., 2009), mechanical activation methods (McKelvy et al., 2004;Park and Fan, 2004;Li et al., 2009;Haug et al., 2010;Haug et al., 2011)), and using wastewater/brine as liquid agents (Chou, 2011;Liu and Maroto-Valer, 2011;Nyambura et al., 2011).However, pretreatments and addition of chemicals tend to consume additional energy and materials.As a result, the aqueous carbonation of alkaline solid wastes is suggested as a link to utilize wastewater for large-scale application of this approach.It was concluded that significant technological breakthroughs, i.e., reactor design and waste-to-resource supply chain, and system optimization, i.e., response surface methodology (RSM) and LCA will be needed before deployment can be considered.

CONCLUSIONS AND RECOMMENDATIONS
This paper reviews the state-of-the-art of the accelerated carbonation of alkaline wastes for CO 2 capture.The carbonation of alkaline solid waste has been proved to be an effective way to capture CO 2 .A combined approach was suggested for capturing CO 2 , at the same time improving the environmental and mechanical properties of industrial solid wastes and wastewater/brine, for instance, carbonation of the steelmaking slags and metalworking wastewater in steelmaking industry.The process also enabled on-site recycle and reuse of wasted materials.
Since dissolution of reactive species (e.g., calcium ions) and the water solubility of CO 2 exhibit contradictory performances in the limiting step, the optimum operating conditions of accelerated carbonation should be proposed to compromise the above conflict phenomena.In addition, CCS process would consume additional energy and generate more CO 2 emissions.Furthermore, utilization of the reacted wastes for construction materials should be critically assessed from the view-point of 3E (Engineering, Environmental, and Economic) aspects.Consequently, RSM (Response Surface Methodology) and LCA (Life Cycle Assessment) techniques should be introduced and evaluated for system optimization.

Fig. 3 .
Fig. 3. Comparison of actual CO 2 capture capacity with the CaO content in solid for different types of wastes.

Fig. 6 .
Fig. 6.Mole balance and equilibrium conditions for carbonation of alkaline solid wastes.

Table 1 .
Comparison of CO 2 capture process by absorption and adsorption.

Table 2 .
Example of alkaline solid wastes in literature.

Table 5 .
Summaries of factors required for effective carbonation.

Table 6 .
Description of alkaline solid wastes for accelerated carbonation and utilization.