A comparative study of CO₂ sorption properties for different oxides

Abstract

It is essential to capture CO₂ from flue gas stream, which is considered as one of the prime reasons of global warming. Although various CO₂ capture technologies already exist, most of these techniques are still unfit to be employed at a large scale. In the past one decade, oxides have emerged as a strong candidate to capture CO₂ for post-, pre- and oxy-combustion conditions. Oxides combine with CO₂ present in the flue gas and form carbonate, which when heated regenerates the oxides and thus liberates almost pure stream of CO₂. The selection criteria for potential oxides entail their CO₂ capture capacity, absorption rate, thermal stability, regeneration heat, cost and structural properties. Thus, here, we review the technical merit of all the suitable oxides which can be used for CO₂ capture at any fossil fuel burning plants.

Introduction

The problem of carbon dioxide (CO₂) emission from various fossil fuel burning plants (such as power, steel, cement and gasifiers) has become an important issue that may tremendously affect our future survival. Carbon dioxide, one of the greenhouse gasses, contributes to the global warming effect once released into the atmosphere [1, 2]. CO₂ is thus far regarded as the most important contributor to global warming and accounts for 64 % of the increased greenhouse effect [3]. At present, there is ample scientific evidence to support the claim that global warming is human induced [4]. Many researchers believe that if such extensive use of fossil fuels continues for another 50 years, the CO₂ concentration will rise to 580 ppm, which would trigger a severe climate change.

The atmospheric concentration of CO₂ rose from 280 ppm in 1,800 to 397.8 in early 2014. Figure 1a presents the world energy-related CO₂ emissions. According to International Energy Outlook (IEO) 2013, the emission is projected to grow from 31.2 billion metric tons in 2010 to 45.5 billion metric tons in 2040. As expected, much of the growth is attributed to the developing countries (non-OECD). Non-OECD countries heavily rely on the supply of fossil fuels to meet their energy demand. The CO₂ emissions from the use of different fuel types such as coal, liquid fuels and natural gas are shown in Fig. 1b. Effluent emissions from coal use account for 44 % in 2010 and are projected to increase to 47 % in 2020–2030, before dropping marginally to 45 % in 2040. Liquid fuels have the slowest growth, resulting in an increment of only 3.5 billion metric tons of CO₂ from 2010 to 2040. The world consumption of natural gas is growing more rapidly than that of coal or liquid fuel. As natural gas has relatively low carbon intensity, its projected contribution to world energy-related CO₂ emissions is only 22 % in 2040.

Fig. 1

World energy-related carbon dioxide emissions by a year and b fuel type

Currently, most of the world energy requirement for transportation and heating (two-thirds of the primary energy demand) is satisfied through petroleum and natural gas. These two fuels are generally chosen due to the relative ease of transportation of liquid or gaseous forms. It is noteworthy that the combustion of hydrocarbon fuels for transportation and heating contributes over half of all greenhouse gas emissions and a large fraction of air pollutant emissions. The current carbon emissions have to be reduced by at least a factor of three to maintain the CO₂ level in the atmosphere while meeting the ever-increasing global energy demand.

Therefore, it is essential to capture CO₂ from the industrial flue gas stream, one of the leading contributors of the anthropogenic gasses [5]. The development of carbon capture and sequestration (CCS) technologies can be seen as a viable solution to mitigate the vast carbon emission. The separation of a pure CO₂ stream, combined with a well-managed geological storage site, is being considered as a mitigation option for climate change. The CCS technologies could be applied using available technologies as many of the components in these systems are already in use. However, it is widely accepted that there is a large scope for cost reduction and energy efficiency improvements in CO₂ capture systems.

Currently, various CO₂ capture technologies exist including physical absorption [6, 7], chemical absorption [8, 9], adsorption [10] and membranes [11]. However, they are still far from being considered as a technologically viable solution. The obvious bottlenecks are the huge amount of flue gas which needs to be treated as well as low mass transfer rate during the processes. Among all the proposed techniques, chemical absorption using aqueous alkanolamine solutions is the most conventional way for CO₂ capture. But these amine-based solutions suffer severe drawbacks such as high energy consumption in regeneration process, high corrosive rate of equipment, essential pretreatment steps, large absorber volume and high sorbent cost. Because of these limitations of the chemical absorption techniques, an alternative way for post-, oxy- or pre-combustion CO₂ capture at fossil fuel burning power plants, mainly a solid adsorption/absorption method, can be proposed.

A schematic diagram on Fig. 2 is depicted to demonstrate the concept of cyclic CO₂ capture process by metal oxides. The metal oxides (MO), such as CaO and MgO, combine with CO₂ to form thermodynamically stable carbonates. Metal carbonates, when heated, liberate pure stream of CO₂ gas and regenerate the oxides. The combined use of exothermic carbonation reaction and endothermic regeneration reaction forms a cyclic process. Eventually, the generated pure CO₂ gas can either be sequestered underground or used for enhanced oil recovery [12]. In general, any metal oxides that can qualify for CO₂ capture application must be abundant in earth’s crust, react with CO₂ at low temperature, require low regeneration energy, should have suitable reaction kinetics and must form a carbonate that is stable in the environment at ambient conditions. Porous oxides, namely alkali and alkaline-earth metals, have been reported as promising candidates for CO₂ capture. These metal oxides possess properties such as long durability, good mechanical strength, wide availability, and low cost since they are present as natural minerals and high CO₂ absorption capacity at moderate working temperatures [13–16]. In this study, we review the CO₂ capture capacity of these different potential oxides.

Fig. 2

Cyclic CO₂ capture process for metal oxides (MO) and metal carbonates (MCO₃)

Selection criteria for metal oxides

Recently, researchers proposed a theoretical screening methodology to identify the most promising CO₂ sorbent candidates from the vast array of possible solid materials [17–19]. The methodology uses both the thermodynamic database searching with first principles density functional theory and phonon lattice dynamics calculations. The thermodynamic properties of solid materials are used for computing the thermodynamic reaction equilibrium properties of CO₂ absorption/desorption cycle based on the chemical potential and heat of reaction analysis. The selection is based on the pre- and post-combustion technologies and conditions in power plants. This includes only those solid materials that may be suitable as CO₂ sorbent candidates and further be considered for experimental validation, if they meet the following criteria: lead to lowering in energy cost for the capture and regeneration process and could operate at desired conditions of CO₂ pressure and temperature. Figure 3 illustrates the thermodynamic properties and CO₂ wt% absorbed by different metal oxides. However, the thermodynamic properties and CO₂ wt% absorbed by any sorbent could not be the only factors to select them for experimental studies. Various other important parameters such as availability or cost, regeneration temperature, kinetics, reversibility and durability must also be carefully considered before selecting any metal oxides for a large-scale application.

Fig. 3

Thermodynamic properties and CO₂ wt% for metal oxides MO(s) + CO₂ (g) = MCO₃(s) at 300 K

Based on the above figure, it can deduce that BeO has a favorable thermodynamics, but this system is not a viable candidate for CO₂ capture because of the health issues related to beryllium powder or dust. It should be noted that a high exothermic forward reaction (MO(s) + CO₂ (g) = MCO₃(s)) generally requires a large regeneration temperature for the oxide (MCO₃(s) = MO (s) + CO₂ (g)). Also, a low CO₂ wt% (~40–50) will require a large amount of the solid materials to deal with a huge volume of flue gas stream at any fossil fuel burning plants. Thus, a high regeneration temperature and low CO₂ wt% can cause a vast energy or carbon emission penalty. Moreover, such oxides can eventually increase the electricity cost by a significant factor and thus may not be suitable for CO₂ capture at any plants.

The turnover temperature represents the value above which sorbents cannot absorb CO₂ and the release of CO₂ begins. The flue gas conditions are different for pre- and post-combustion CO₂ capture at power plants. Under pre-combustion conditions, after water–gas shift reaction, the gas stream mainly contains CO₂, H₂O and H₂. The partial CO₂ pressure in this case is around 20–25 bar, and the temperature is about 300–350 °C. But for post-combustion conditions, the flue gas stream mainly contains CO₂ and N₂. In this case, the partial pressure of CO₂ is around 0.1–0.2 bar, while the temperature range can vary around 27–77 °C. The goal set up by the US Department of Energy is to capture at least 90 % CO₂ with an increase in the electricity cost of no more than 10 and 35 % for pre- and post-combustion technologies, respectively [20]. To minimize the energy consumption, the ideal sorbents should work in the above-indicated pressure and temperature ranges to separate CO₂ from H₂. A list of turnover temperatures for different oxides can be found elsewhere [17–19].

Based on the above-discussed limitations, exclusively few oxides seem to be promising and thus chosen to study for the experimental validation. The two alkaline-earth metal oxides (such as CaO and MgO) draw the tremendous attention because of accessibility and favorable thermodynamics. Recently, lithium-, sodium- and potassium-based silicates or zirconates also gained interest due to their high CO₂ absorption capacity. Moreover, FeO can also be seen as a promising material for CO₂ capture at both power and non-power sectors (iron and steel industry). In the following sections, we review the experimental studies for the most promising CO₂ sorbent candidates.

Metal oxides

Alkali metal-based oxides

Lithium zirconate (Li₂ZrO₃)

The reaction between lithium carbonate (Li₂CO₃) and zirconium dioxide (ZrO₂) is well known to synthesize lithium zirconate (Li₂ZrO₃). However, the reverse reaction was not considered for CO₂ capture until 1998 [21]. Nakagawa and Ohashi first investigated the capture of CO₂ using Li₂ZrO₃ at high temperatures (400–600 °C) [21]. Li₂ZrO₃ has great potential because it has an excellent CO₂ sorption capacity (28 wt%) as well as a small volume change during the CO₂ sorption/desorption cycles [22]. The reaction (1) occurs mainly due to the Li ion mobility in the ceramics [23, 24]. As can be seen from Fig. 4, during CO₂ sorption, the Li ions diffuse from the core of the particles to the surface and react with CO₂ to form Li₂CO₃. The diffusion of CO₂ in the solid Li₂CO₃ is recognized as the rate-limiting step.

Fig. 4

Proposed mechanisms for CO₂ sorption on Li₂ZrO₃

$$ {\text{Li}}_{2} {\text{CO}}_{3} \left( {\rm s} \right) + {\text{ZrO}}_{2} \left({\rm s} \right) \leftrightarrow {\text{Li}}_{2} {\text{ZrO}}_{3} \left({\rm s} \right) + {\text{CO}}_{2} \left({\rm g} \right)\quad \Delta H_{{298{\text{K}}}} = - 160\,{\text{kJ}}/{\text{mol}} $$

(1)

The separation of CO₂ with aid of Li₂ZrO₃ works in a temperature range of 450–700 °C. The main advantage of this new material with respect to calcium-based absorbents is its unique thermal stability that allows achieving many absorption/regeneration cycles (over 20) without loss of capacity. Previous reports suggest that only 11–13 wt% change after absorption of CO₂ using Li₂ZrO₃ at 500–700 °C in more than an hour is possible. Thus, it is evident that Li₂ZrO₃ has extremely slow absorption kinetics, and hence, recently, the addition of dopants was proposed as a solution. Hence, the addition of dopants such as K₂CO₃ and Y₂O₃ to Li₂ZrO₃ could potentially accelerate its CO₂ sorption kinetics.

In the case of the pure Li₂ZrO₃ (without K₂CO₃ additive), formation of an impervious shell of Li₂CO₃ on the surface leads to a residual unreactive Li₂ZrO₃ core (Fig. 4). This observation suggests that the diffusion resistance of CO₂ through the solid Li₂CO₃ surface layer is the rate-limiting step. At 500 °C, Li₂CO₃ and K₂CO₃ form a eutectic mixture. During the absorption process, the eutectic mixture helps to diffuse more CO₂ toward the inner unreacted particles [25]. K₂CO₃- and Li₂CO₃-containing Li₂ZrO₃ sample reacts with gaseous CO₂ at 500 °C to form an outer layer of a molten liquid and a solid interior ZrO₂ core. Consequently, CO₂ diffuses through the molten carbonate layer at a much faster rate. Such difference in CO₂ sorption rate of doped and undoped Li₂ZrO₃ can be partially attributed to the molten carbonate layer.

Another dopant, Y₂O₃, can improve the O⁻² ionic diffusion in the ZrO₂ layer that is formed during the absorption. Unfortunately, Y₂O₃ failed to increase the CO₂ absorption kinetics of Li₂ZrO₃ [26]. Other findings reveal that solid solutions of lithium and potassium meta zirconates (Li_2−xK _x ZrO₃) increase both the kinetics and CO₂ absorption capacity compared to undoped Li₂ZrO₃ [27]. In the same line, solid solutions of lithium and sodium zirconates were also explored [28]. It was concluded that the higher the Na content in the samples the faster the absorption kinetics. On the contrary, the higher the Li content in the mixture the faster the regeneration kinetics.

Lithium orthosilicate (Li₄SiO₄)

Lithium silicate-based (Li₄SiO₄) sorbent material has been considered as a promising candidate for high-temperature CO₂ removal in the recent times [29, 30]. Li₄SiO₄ could adsorb different concentrations of CO₂ in the temperature range of 450–700 °C, but it has relatively (compared to amine solutions) higher decomposition temperature (>800 °C). A high decomposition temperature implies that more heat input and costly equipment would be required to perform the process. However, compared to other solid sorbents such as CaO, the decomposition temperature is low and thus less supply of energy will be needed [31]. As can be calculated from reaction (2), the theoretical CO₂ adsorption capacity of Li₄SiO₄ is 36.7 wt% (~8.34 mmol CO₂/g Li₄SiO₄).

$$ {\text{Li}}_{4} {\text{SiO}}_{4} \left( \rm s \right) + {\text{CO}}_{2} \left( \rm g \right) \leftrightarrow {\text{Li}}_{2} {\text{SiO}}_{3} \left( \rm s \right) + {\text{Li}}_{2} {\text{CO}}_{3} \left( \rm s \right)\quad \Delta H_{298\rm K} = - 141.97\,{\text{kJ}}/{\text{mol}} $$

(2)

In a study, Wang et al. [32] reported an interesting way to synthesize Li₄SiO₄ using rice husk, which were calcined at 800 °C in the presence of Li₂CO₃. This material is ideally suited for CO₂ removal from synthesis gas (syngas) derived from gasification of carbonaceous fuels (coal, coke, natural gas, biomass, etc.) and is hence considered as a pre-combustion CO₂ sorbent.

Li₄SiO₄ has shown the ability to remove more than 90 % of the CO₂ from simulated syngas in both fixed-bed and fluidized-bed process configuration. The lithium silicate-based sorbent is highly effective at temperatures of 250–550 °C, pressures of 0–20 atm, CO₂ concentrations of 2–20 % and in the presence of contaminants such as hydrogen sulfide. Yamachui et al. [33] reported a carbonation yield of 87.2 % under 100 cm³/min flow rate of pure CO₂. The water vapor plays a major role in increasing the CO₂ absorption rates of Li₄SiO₄. About 22–26 wt% (capacities of ~5−6 mmol/g) were obtained using a gas stream containing about 15 % CO₂ and 10 % H₂O in N₂ at 550 °C [34]. Moreover, the sorbent has shown excellent regenerability and attrition resistance in thermal cycling tests. Recent analysis has shown that the lithium silicate-based sorbent has the capability not only to separate CO₂ from syngas, but also to promote the water–gas shift reaction [35]. It is reported that dopants such as Al, Fe, Na, K and Cs increase the CO₂ uptake of lithium orthosilicate [36, 37].

Researchers have attempted to improve the reaction kinetics of CO₂ adsorption by altering the synthesis routes [38]. The new methods are aimed at reducing the particle size of Li₄SiO₄. The experimental results reveal that a smaller particle size yields higher CO₂ adsorption efficiency and adsorption rate. Further studies focused on decreasing the precursor particle size [39] or choosing more sintering-resistant precursor to result in the smaller product grain size [40]. Similarly, based on the particle size effects on Li₄SiO₄, the reactivity of Li₄SiO₄ was modeled as a reaction of CO₂ at the solid surface followed by the rate-limiting diffusion of lithium [41].

A calculation based on the most optimistic assumptions illustrates that despite the fact that Li₄SiO₄-based absorbent demonstrates high CO₂ uptake capacity, it would be impractical to perform the process at the scale of a fixed-bed temperature swing process at a 500-MW coal-fired power plant [34]. However, efforts have been made to modify the Li₄SiO₄ synthesis methods such as solid-state reactions [30, 42], sol–gel method [38, 43], impregnated suspension [44], precipitation method [45], amorphous silica [40], Quartz [39], fly ashes [46], rice husk ashes [47] and diatomite [48].

Sodium zirconate (Na₂ZrO₃)

Sodium zirconate has gained attention due to its favorable thermodynamics and high theoretical CO₂ sorption capacity [23, 49, 50]. Na-based sorbents can be used as solid CO₂ acceptors at high temperatures. The use of sodium-based oxides is more economical than lithium-based oxides. Moreover, previous reports suggest that Na₂ZrO₃ can have a better reaction rate than the synthetic sorbents such as Li₂ZrO₃ and Li₄SiO₄ [51].

Na₂ZrO₃ has a lamellar structure, where sodium ions are located among the zirconate layers, which permits for sodium diffusion. The mechanism for CO₂ chemisorption on Na₂ZrO₃ has already been proposed. Initially, the surface of the alkaline ceramic particles reacts with CO₂ and forms carbonates on the periphery. Once the external layer of carbonate is completely formed, sodium atoms have to diffuse throughout this external carbonate shell to reach the surface and be able to react with CO₂.

$$ {\text{Na}}_{2} {\text{ZrO}}_{3} \left( \rm s \right) + {\text{CO}}_{2} \left( \rm g \right) \leftrightarrow {\text{Na}}_{2} {\text{CO}}_{3} \left( \rm s \right) + {\text{ZrO}}_{2} \left( \rm s \right) $$

(3)

The maximum theoretical CO₂ absorption should correspond to an increase of weight equal to 23.75 wt%. The reaction (3) is controlled by two different processes—CO₂ absorption over the surface of Na₂ZrO₃ particles and sodium diffusion. Alcerreca-Corte et al. [52] performed the kinetic analysis for chemisorption of CO₂ on the Na₂ZrO₃ particles in the temperature range of 150–700 °C. A fast kinetics was observed between 550 and 700 °C. However, at low temperatures, kinetics was relatively low and was attributed to the sintering effect as well as diffusion problems. The optimum temperature for the reaction (3) was 600 °C. All the isotherms were fitted to a double exponential model. The activation energies for the CO₂ absorption and the diffusion of sodium were calculated to be 33.87 and 48.01 kJ/mol, respectively. Therefore, it was concluded that sodium diffusion is the limiting step of the process. Similarly, Jimenez et al. reported the CO₂ sorption kinetics of Na₂ZrO₃ at similar temperatures and $P_{{{\text{CO}}_{2} }}$ = 0.4–0.8 atm. Contrary to previous study, the surface reaction is identified to be the rate-limiting step for the reaction and attributed to the fast CO₂ sorption kinetics of Na₂ZrO₃ [53].

In general, the addition of steam enhances the capture kinetics and also favors the regeneration. It is believed that the presence of steam increases the mobility of alkaline ions and therefore accelerates the reactions. Therefore, several studies have been performed to evaluate the effect of steam addition on the CO₂ capture properties of Na₂ZrO₃ [54, 55]. One such study was performed by Santillán-Reyes et al. in the presence of water vapor and low temperature range (30–70 °C). They synthesized Na₂ZrO₃ by solid-state reaction using the reagents Na₂CO₃ and ZrO₂ in the molar ratio of 1.1:1.0, respectively. Further, the reaction mechanism of Na₂ZrO₃–CO₂–H₂O was proposed and illustrated in (4). Na₂ZrO₃ reacts with water vapor to produce hydroxyl species at the particle surfaces and then absorbs 2 mol of CO₂ and 1 mol of H₂O to eventually form NaHCO₃ and ZrO₂.

$$ {\text{Na}}_{2} {\text{ZrO}}_{3} \mathop{\longrightarrow}\limits^{{{\text{H}}_{2} {\text{O}}}}{\text{Na}}_{2} {\text{ZrO}}_{{3\,{\text{superficially}}\,{\text{hydroxylated}}}} \mathop{\longrightarrow}\limits^{{2{\text{CO}}_{2} }}2{\text{NaHCO}}_{3} + {\text{ZrO}}_{2} $$

(4)

Na₂ZrO₃ was able to absorb 5.8 mmol/g of CO₂ (which has theoretical capacity is 10.8 mmol/g). Therefore, Na₂ZrO₃ can also be used as CO₂ sorbent at low temperatures and in the presence of water vapor. The reported absorption results for Na₂ZrO₃ are still higher than for other materials in the same temperature conditions (≤6 mmol/g of material) [56].

Nanosized materials are expected to increase the CO₂ capture/regeneration rates due to their improved characteristics, such as high surface area and higher surface reactivity because of unsaturated bonds on their pristine surfaces [57–59]. Therefore, Zhao et al. synthesized nanosized Na₂ZrO₃ using a soft-chemical route starting with the formation of a complex from zirconoxy nitrate and sodium citrate, followed by highly exothermic reaction between citrate and nitrate during calcination in a controlled atmosphere [49]. It was found that much faster CO₂ capture rates can be achieved on the monoclinic Na₂ZrO₃ compared to the hexagonal one, even at a very low CO₂ pressure (0.025 bar). Thus, a higher CO₂ capture rate is obtained for nanosized Na₂ZrO₃ due to the dual effect of its crystal size and structure.

Moreover, the doping effect of Li to Na₂ZrO₃ was investigated. The change in sorption/regeneration kinetics was evident from the experimental data. The greater the Li content, the lower the CO₂ sorption kinetics. This observation can be attributed to the fact that Li substitutes some of the Na in the Na₂ZrO₃ structure and results in the formation of small amount of Li₂ZrO₃ [60]. Lopez-Ortiz et al. concluded that the highest activity toward CO₂ sorption was exhibited by Na₂ZrO₃, followed by sodium antimoniate (Na₃SbO₄) and finally sodium titanate (Na₂TiO₃). Other applications for Na₂ZrO₃ may involve its use for the development of inorganic membranes for CO₂ separations [24, 61].

Sodium metasilicate (Na₂SiO₃)

Sodium metasilicate has not yet been extensively examined. One of the first studies illustrates the synthesis of Na₂SiO₃ using solid-state and precipitation methods [62]. It was noticed that Na₂SiO₃ can absorb only a small quantity (~1–2 wt%) of CO₂ in the range of room temperature and 130 °C. Further, the kinetic analysis of CO₂ absorption was also performed, and the results indicate that the sorption mechanism proceeds through a two-step process: (1) superficial chemical sorption and (2) sodium diffusion process. The calculated activation energies for surface reaction and sodium diffusion were 17.48 and 23.97 kJ/mol, respectively. The sodium diffusion is recognized as the limiting step for the carbonation process. Moreover, the CO₂ absorption increases with the decrease in the particle size of Na₂SiO₃ and can be attributed to the subsequent increase in the surface area.

Another study was performed at much lower temperatures (30–60 °C) for the Na₂SiO₃–CO₂–H₂O system [63]. Na₂SiO₃ was synthesized using two methods—(1) solid-state reaction (Na₂CO₃ + SiO₂); and (2) combustion method (NaOH, SiO₂ and urea (CO (NH₂)₂). The reactions were performed under the flow of two different carrier gasses—N₂ and CO₂. In the presence of N₂ as a carrier gas, Na₂SiO₃ traps water physically and chemically, where the water vapor adsorption and/or absorption depend on temperature and relative humidity. When CO₂ was used as the carrier gas, Na₂SiO₃ traps both water vapor and CO₂. Interestingly, the presence of water vapor promotes the higher CO₂ chemisorption (>1 wt%). It can be seen in reaction series (5) that at first, Na₂SiO₃ reacts with water superficially, producing Na–OH and Si–OH species. Further, superficially activated Na₂SiO₃ combines with CO₂ to form NaHCO₃.

$$ {\text{Na}}_{2} {\text{SiO}}_{3} \mathop{\longrightarrow}\limits^{{{\text{H}}_{2} {\text{O}}}}{\text{Na}}_{2} {\text{SiO}}_{{3\,{\text{superficially}}\,{\text{activated}}}} \mathop{\longrightarrow}\limits^{{{\text{CO}}_{2} }}{\text{Na}}_{2} {\text{CO}}_{3} + {\text{SiO}}_{2 } \mathop{\longrightarrow}\limits^{{{\text{CO}}_{2} + {\text{H}}_{2} {\text{O}}}} {\text{NaHCO}}_{3} + {\text{SiO}}_{2} $$

(5)

It has also been observed that under the condition of thermal humidity, Na₂SiO₃ absorbs up to 16.39 mmol of CO₂/gm of ceramic, twice more CO₂ than the quantity absorbed under dry conditions. The synthesis method of Na₂SiO₃ influences its CO₂ capture capacity. Higher the surface area, the better is the CO₂ absorption. Here, the surface areas were 0.5 and 1.6 m²/g for solid-state and combustion method, respectively. Hence, the latter has about 8.6 mmol/g CO₂ capture efficiency, almost twice of that of the former sample. Thus, CO₂ absorption capacity of Na₂SiO₃ is governed by the combined effect of water vapor and surface area. These recent results support the potential of Na₂SiO₃ as a CO₂ capture sorbents at moderate or environmental temperatures.

Alkali-earth metal oxides

Magnesium oxide (MgO)

MgO combines with CO₂ to form MgCO₃. MgCO₃ is thermodynamically stable at ambient conditions. A simple calculation shows that to capture 1 ton of CO₂, about 0.92 ton of MgO is required (assuming 100 % conversion). Thus, it is certain that to capture CO₂ using MgO at any power plant site, a large quantity of MgO will be needed. Moreover, the kinetics and recycling efficiency of MgO–CO₂ reaction are also the major issues to use MgO as a sorbent for practical applications. MgO is an attractive candidate for both pre- and post-combustion capture technologies due to its low regenerating temperatures: T₂ = 287 °C (post-combustion) and T₁ = 447 °C (pre-combustion) [64]. One should note that MgO when recycled between naturally occurring magnesite and dolomite can cause relatively lesser energy and carbon emission penalty, respectively [65].

Many researchers studied the heterogeneous reaction between magnesium species and CO₂ in the presence or absence of H₂O. Although a majority of research was not directly aimed at carbon capture, still the results are valuable additions to the mineral carbonation research. The CO₂ uptake capacity on different MgO sorbents at various conditions is listed in Table 1. Most of the experiments are restricted to low-temperature and ambient or low-pressure conditions. Hence, the reported sorption capacities of CO₂ on MgO are not very high. In contrast, recent work of Highfield et al. reports a conversion of about 70–80 % MgO to MgCO₃ in about 2 h with steam pressure between 5 and 10 % of the total pressure 20–40 bar and temperature around 300–350 °C.

Table 1 CO₂ uptake capacity of MgO at different conditions

One of the key factors in gas–solid reaction is the presence of water. There have been several observations when water catalyzed the reaction [80, 81]. Figure 5 depicts the absorption/adsorption model of MgO–CO₂–H₂O.

Fig. 5

Adsorption/absorption model for MgO–CO₂–H₂O reaction

MgO particles are surrounded by water vapor where CO₂ reacts to form CO₃²⁻ ions and H⁺ ions. Free Mg⁺² ions could further react with the CO₃²⁻ ions to form MgCO₃. Recently, Fagerlund et al. [78] proposed the reaction mechanism for MgO carbonation in the presence of steam:

$$ {\text{MgO}} + {\text{H}}_{2} {\text{O}} \leftrightarrow {\text{MgO}} \cdot {\text{H}}_{2} {\text{O}}^{*} $$

(6)

$$ {\text{MgO}} \cdot {\text{H}}_{2} {\text{O}}^{*} + {\text{CO}}_{2} \leftrightarrow {\text{MgCO}}_{3} + {\text{H}}_{2} {\text{O}} $$

(7)

$$ {\text{MgO}} + {\text{CO}}_{2} \leftrightarrow {\text{MgCO}}_{3} $$

(8)

However, MgCO₃ forms an impervious layer around unreacted MgO particles and hinders the further diffusion of CO₂ molecules. Thus, only increasing the amount of water vapor cannot lead to the complete carbonate conversion of MgO. Therefore, besides the amount of steam, surface properties of MgO (such as surface area, particle size and porosity) are also very crucial parameters for the carbonation process. It is observed that in the absence or low partial pressure of water vapor, MgO·2MgCO₃ is formed. It can be easily calculated that the CO₂ capture capacity of MgO·2MgCO₃ is about two-thirds that of MgCO₃. Hence, formation of MgO·2MgCO₃ can significantly reduce the further uptake of CO₂ by MgO particles. But at a high temperature (>350 °C), MgO·2MgCO₃ decomposes to MgO.

MgO–CO₂ reaction has been studied to some extent. The role of water vapor in increasing the activity of MgO is now well established. The direct relation of porosity, surface area and particle size with the carbonation yield has been examined. However, several factors such as carbonation kinetics, sorbent reversibility and durability are still not completely resolved and thus need to be investigated.

Calcium oxide (CaO)

In 1995, Silaban and Harrison proposed the method that involves the separation of CO₂ at high temperatures (>600 °C) using the reversible reaction of CaO [82].

$$ {\text{CaO}}\left( \rm s \right) + {\text{CO}}_{2} \left( \rm g \right) \leftrightarrow {\text{CaCO}}_{3} \left( \rm g \right) $$

(9)

The low price (because of naturally occurring mineral, limestone) and favorable thermodynamics of CaO have attracted a lot of interest in the past one decade, and thus, the CaO–CO₂ system has been extensively studied. A great advantage of calcium-based sorbents is that the absorption of CO₂ can occur at temperatures above 600 °C (possibility of heat recovery) [83]. The efficiency of the CO₂ capture depends on many parameters such as diffusion resistance, which depends on the size of sorbent particles, volume and the pore structure, as well as surface size and reaction kinetics [84]. These parameters influence carbonation, but many of them also affect calcination. A detailed list of recent methods used to enhance the CO₂ uptake by CaO-based sorbents can be found elsewhere [85].

Bulk CaO, as a CO₂ sorbent, has three severe limitations: (1) kinetics of the carbonation reaction, despite being highly exothermic, becomes slow after the first layer of carbonate formation because the reaction is limited by the diffusion of CO₂ through the thin surface layer of CaCO₃ formed on CaO [86, 87]. The uptake kinetics of CaO-based sorbents was improved by innovative synthesis methods or precursors [88, 89]. (2) Regeneration step is very energy intensive (>800°C for CaCO₃ decomposition), so excessive sintering and mechanical failure of the oxide occur. This leads to a drastic loss in adsorption activity after a few sorption and regeneration cycles [90]. (3) To be an effective sorbent, CaO, that is suitable for high surface area exposure, such as a powder. However, powders can be used in fluidized beds but the pressure drop associated with them is very large. Also, use of fine powders can be problematic due to entrainment in the process flow and attrition of the material [91].

The improvement in the sorbent reversibility can be seen as a major challenge for extended operation purpose. Decrease in absorption capacity is associated with sintering of the sorbent surface because of the influence of temperature and the reduction in porosity and active surface [92–95]. Other factors that reduce the activity of sorbents are the attrition of sorbent grains during the process and chemical inactivation [96]. It should be noted that the reaction with sulfur oxide (SO₂) is competitive to the carbonation reaction. Sulfurization and carbonation reactions are similar because of being heterogeneous and occurring in the porous structure of the sorbent. However, sulfurization is irreversible in the temperature range 500–900 °C, occurs in small pores and covers the surface of the sorbent [97–100]. Decrease in the absorption capacity after 45 cycles was 60 % for limestone, 40 % for dolomite and less than 20 % for huntite, and can be attributed to the differences in the sorbent structure [101]. The decrease in absorption capacity by CaO can be limited in several ways. The parameters that have impact are dopant (NaCl, Na₂CO₃, KCl, KMnO₄, MgCl₂, CaCl₂, Mg (NO₃)₂), sorbent grain size, raw material for calcium-based sorbent, thermal pretreatment of the sorbent and steam reactivation. The properties of CaO obtained from calcium acetate are better than the properties of CaO obtained from natural limestone because it has a much better conversion and greater durability [102].

Calcium silicate (CaSiO₃)

Calcium silicate received attention because it is inexpensive and has lower sorption temperature compared to CaO. Wang et al. [103] explored the sorption–desorption property of CaSiO₃. CaSiO₃ is synthesized through solid chemical reaction of CaCO₃ and SiO₂ at 800 °C. CaSiO₃ starts to absorb CO₂ at 400 °C with about 28.72 % sorption efficiency, using 15 % CO₂ and rest N₂. The observed regeneration temperature for CaSiO₃ was 800 °C. However, the CO₂ capture capacity drastically decreases from large number of cycles and could be attributed to sintering of the materials which leads to the loss of specific surface area. Similarly, Gupta and Fan reported that CaSiO₃ captured 27.85 % CO₂ at 700 °C and 1 atm [104]. Under the same conditions, Tilekar et al. reported the CO₂ capture capacity of CaSiO₃ to be 14.19 wt% [105]. The water molecules in the CaSiO₃ facilitate the capture of CO₂.

Transition metal-based oxides

Transition metal-based oxides act as an oxygen carrier in the chemical looping process. In 1983, Richter and Knoche introduced chemical looping for heat generation using metal oxides as chemical ingredients, which are now considered as a potential solution for the simultaneous power generation and CO₂ capture [106]. Metal oxide (Me _x O _y ) is reduced by the carbonaceous fuel to generate CO₂ and/or H₂O in the fuel reactor. The reduced metal (Me _x O_y−1) is then oxidized back to Me _x O _y by oxygen in the air reactor, which is an exothermic reaction. In the process, transition metal oxides (oxygen carriers) are transported within and between the fuel reactor and the air reactor.

$$ {\text{C}}_{n} {\text{H}}_{2m} + \left( {2n + m} \right){\text{Me}}_{x} {\text{O}}_{y} \to n\,{\text{CO}}_{2} + {\text{mH}}_{2} {\text{O}} + \left( {2n + m} \right){\text{Me}}_{x} {\text{O}}_{y - 1} $$

(10)

$$ 2{\text{Me}}_{x} {\text{O}}_{y - 1} + {\text{O}}_{2} \to 2{\text{Me}}_{x} {\text{O}}_{y} $$

(11)

The widely developed oxygen carriers include Ni-, Cu- and Fe-based metal oxides and their composited particles. Despite the great reactivity of Ni- or Cu-based oxygen carriers, their development is restricted because of their relatively high cost. Fe-based oxides are the strong candidates to be commercialized because of their economic feasibility. A number of reviews about the current status of the chemical looping process have been published in recent times [107–111]. Here, the focus is on the direct interaction of CO₂ with transition metal-based oxides via carbonation–calcination cycle.

Iron-based oxides

Iron oxide-based materials can be used for CO₂ capture purpose in both the power and non-power sectors. The main advantages of iron-based sorbent are as follows: (1) accessibility, (2) favorable thermodynamics and (3) slow degradation of sorption capacity.

Recently, Ohio State University developed an iron oxide-based oxygen carrier for particles suitable for operation in the coal direct chemical looping (CDCL) process. CDCL is an efficient power generation process. The process consists of a unique moving bed reactor, namely the reducer, where pulverized coal is fully converted using iron oxide.

$$ {\text{Fe}}_{2} {\text{O}}_{3} \left({ \rm s} \right) + {\text{C}}\left( {\rm s }\right) \to {\text{FeO}}\left( {\rm s }\right) + {\text{Fe}}\left({\rm s} \right) + {\text{CO}}_{2} ({\rm g}) $$

(12)

$$ 2{\text{FeO}}\left( {\rm s} \right) + 0.5{\text{O}}_{2} \left( {\rm g }\right) \to {\text{Fe}}_{2} {\text{O}}_{3} \left( {\rm s} \right) $$

(13)

Fe₂O₃, an oxygen carrier, is reduced to FeO and Fe while the exhaust gas is a stream of pure CO₂ that can be sequestered or used for enhanced oil recovery [eq. (12)]. The reduced FeO and Fe mixture can then be oxidized to Fe₂O₃ using air in the combustor reactor, liberating heat to produce steam [112]. Another recent study presents the CO₂ capture technique for iron and steel industries using iron-based raw materials easily available at the iron-making sites [113].

$$ {\text{Fe}}_{3} {\text{O}}_{4} \left( \rm s \right) + {\text{Fe}}\left( \rm s \right) + 4{\text{CO}}_{2} \left( \rm g \right) \to 4{\text{FeCO}}_{3} \left( \rm s \right) $$

(14)

$$ 3{\text{FeCO}}_{3} \left( {\rm s} \right) + 0.5{\text{O}}_{2} \left( {\rm g} \right) \to {\text{Fe}}_{3} {\text{O}}_{4} \left( {\rm s} \right) + 3{\text{CO}}_{2} \left( {\rm g} \right) $$

(15)

The global steel production is growing rapidly, from 1,248 Mt in 2006 to 1,490 Mt in 2011—an increase of about 16 % in a short span of 5 years [114]. Steel production is an energy intensive process and thus consumes huge amount of fossil fuels. On world average, each ton of steel production emits 2.2 ton of CO₂. Therefore, iron-making industry is one of the biggest targets next to power plants to curb vast emission of greenhouse gasses.

A mixture of magnetite and iron combines with CO₂ and forms siderite. The reaction generates heat and thus can be utilized to produce steam. Siderite once formed can further decompose back to magnetite in the presence of oxygen at 350 °C. Once the absorption capacity of magnetite degrades, it can be sent back to blast furnace for further processing. For example, the sorbents can be processed in a blast furnace for the production of iron or steel. Thus, the proposed method can advantageously reduce or even eliminate the loss of raw materials. This system and method can be more thermodynamically favorable and can save energy. The reversibility and suitable kinetics can make the system highly favorable for CO₂ capture for iron and steel industries.

Other transition metal oxides

It is considered that MnO can be used for both post- and pre-combustion conditions. At $P_{{{\text{CO}}_{2} }}$ = 0.1 bar, the driving force for MnO to convert in MnCO₃ was high even at a low temperature 77 °C. The CO₂ could be released at about 377°C at partial $P_{{{\text{CO}}_{2} }}$ = 10 bar. For pre-combustion, MnO will capture CO₂ at a partial pressure of about 10 bar below 327 °C and then could be regenerated by heating above 377 °C to produce a stream of pure CO₂. According to the thermodynamics, CdO can be used for both pre- and post-combustion conditions. However, the toxicity of Cd system would be a big hurdle in its practical applications.

ZnO can be used for post-combustion but not for pre-combustion. The equilibrium temperature required for carbonate formation is 77 °C at $P_{{{\text{CO}}_{2} }}$ = 0.1 bar and around 167 °C for $P_{{{\text{CO}}_{2} }}$ = 10 bars [18]. Therefore, it may be essential to cool post-combustion gasses prior to carbonation for high driving force and thus make the sorbent unfit for post-combustion capture. Few researcher groups studied the heterogeneous reaction between ZnO and CO₂ in the presence or absence of H₂O under mild conditions. Previous experiments were performed under mild conditions, mainly room temperature and $P_{{{\text{CO}}_{2} }}$ varied from 0.01 atmospheres to 1 atmosphere (atm) [115–117]. Basic zinc carbonate with chemical formula Zn₃CO₃ (OH)₄·2H₂O was reported as a predominant phase at these conditions. Chen et al. synthesized the compound, zinc carbonate hydroxide hydrate (JCPDS #11-0287), using chemical precipitation method [5.0 % Na₂CO₃ (aq.) added to 0.1 mol/L ZnSO₄(aq.)] [118]. A direct relation of heating rate of the sample to its decomposition temperature was noticed. Interestingly, only one endothermic peak for zinc carbonate hydroxide hydrate was observed and was thus attributed to the combined evolution of water vapor and CO₂. The decomposition temperature was observed in the range of 197–257 °C for various heating rate (5–20 °C/min).

$$ {\text{Zn}}_{4} \left( {{\text{CO}}_{3} } \right)\left( {\text{OH}} \right)_{6} \cdot{\text{H}}_{2} {\text{O}}\left( \rm s \right) \to 4{\text{ZnO}}\left( \rm s \right) + {\text{CO}}_{2} \left( \rm g \right) \uparrow +\, 4{\text{H}}_{2} {\text{O}}\left( \rm g \right) \uparrow $$

(16)

ZnO–CO₂–H₂O process is favored by low operating and regeneration temperature and could be ideal for post-combustion CO₂ capture. However, its application is limited due to its rare occurrence in the nature, and this could be the reason for the sparse study of ZnO–CO₂ system.

Similarly, NiO can be considered for post-combustion capture, but not for pre-combustion capture. According to thermodynamics, post-combustion absorption will occur below 77 °C and NiO could be regenerated above 187 °C. However, NiO availability is still a significant concern for its practical application. Therefore, the catalytic potential of NiO leads to its huge application in the CO₂ utilization field. Under sunlight, CO₂ can be converted into methanol in the presence of NiO/lnTaO₄ [119–121].

Perovskites

Perovskites have a stable structural framework, despite the movement of active ions into and out of the structure. Such materials have the potential to obviate the challenges associated with the other oxides. Currently, the main issues that most of the oxides face are as follows: (1) rapid degradation of their CO₂ capture capacity over repeated cycles and (2) a large volume change when oxide is converted into carbonate. Recent reported experimental results have substantially attracted the interest toward perovskites, as potential CO₂ capture materials.

One such study by Galven et al. investigates the CO₂ adsorption behavior of Li₂SrTa₂O₇, a layered perovskite family of the Ruddlesden–Popper (RP) phase (general formula, A_n−1 A′₂ B _n X_3n+1, where A, A′ and B are cations, X is an anion, and n is the number of the layers of octahedra in the perovskite-like stack) [122]. Under humid CO₂ environment, Li₂SrTa₂O₇ is transformed into LiHSrTa₂O₇ releasing LiOH, which then locks CO₂ in the form of Li₂CO₃.

$$ {\text{Li}}_{2} {\text{SrTa}}_{2} {\text{O}}_{7} + 1/2{\text{H}}_{2} {\text{O}} + 1/2{\text{CO}}_{2} \leftrightarrow {\text{LiHSrTa}}_{2} {\text{O}}_{7} + 1/2{\text{Li}}_{2} {\text{CO}}_{3} $$

(17)

It is interesting to note that the operating temperature of Li₂SrTa₂O₇ is 140 °C compared to ~400 °C of most of the other inorganic solid compounds. On heating, Li₂SrTa₂O₇ can be fully recovered at increased temperature (~700 °C) showing that the CO₂ capture is reversible and can be performed for a number of cycles. Similar CO₂ absorption capacity was observed for Li₂SrNb₂O₇. These materials, Li₂SrTa₂O₇ and Li₂SrNb₂O₇, have phases consisting of layers of corner-connected TaO₆ or NbO₆ octahedra separated by a layer of corner-connected LiO₄ tetrahedra [123]. In the presence of water at room temperature, H⁺ shifts Li⁺ in the layered structure and allows the shifted Li⁺ to react with CO₂ to form Li₂CO₃. During regeneration, Li⁺ is reinserted into the Ruddlesden–Popper structure, as in alkali garnet materials [124–127].

Another perovskite material, Ba₄Sb₂O₉, can be synthesized using a solid-state reaction in air from BaCO₃ and Sb₂O₃ [128]. 6H-Ba₄Sb₂O₉ reacts with CO₂ to form BaSb₂O₆ and BaCO₃.

$$ 6{\text{H-}}{\text{Ba}}_{4} {\text{Sb}}_{2} \rm O_{9} + 3{\text{CO}}_{2} \leftrightarrow {\text{BaSb}}_{2} \rm O_{6} + 3{\text{BaCO}}_{3} $$

(18)

6H-Ba₄Sb₂O₉, was found to be able to capture CO₂, mixed with N₂, at 600 °C while the regeneration temperature was around 950 °C under N₂ atmosphere [129]. No significant reduction in the CO₂ absorption capacity was observed for 100 cycles. After 100 cycles, the capacity [~0.1 g (CO₂)/g sorbent] is stable and translates to 73 % of the total molar capacity. The reaction between 6H-Ba₄Sb₂O₉ and CO₂ has similar equilibrium constant to that of CaO-based sorbents. However, the CO₂ absorption capacity was stable for a large number of cycles, unlike CaO that degrades rapidly over cycles under same operating conditions.

Miscellaneous

Aluminosilicate zeolite is attractive for post-combustion CO₂ capture due to its selective CO₂ adsorption and high CO₂ capacities [130]. An extensive research has been carried out on aluminosilicate zeolite, and the results exhibit high uptake capacity than pure-silica zeolites [131–133]. In general, adsorption kinetics of CO₂ on zeolites is relatively fast and can achieve equilibrium capacity within a few minutes. The temperature and pressure strongly govern the CO₂ adsorption on zeolites. As the temperature increases, the adsorption of CO₂ decreases but as the gas-phase partial pressure of CO₂ increases, CO₂ adsorption increases accordingly. However, the presence of water vapor may limit the application of zeolite sorbents by decreasing its capacity. It is clear that optimizing different factors, such as basicity, pore size of zeolites and electric field strength (due to the exchangeable cations in the cavities), can significantly influence the CO₂ adsorption capacities of zeolites.

Conclusion

It is certain that an improved and cost-efficient process for CO₂ capture obtained from flue gas streams from power plants burning fossil fuels is needed to mitigate the huge emission of greenhouse gasses. Among several different proposed processes, metal oxides are promising options for suitable sorbent and have a great potential in the future. However, these sorbents also have limitations and challenges which must be solved before these can be employed commercially for CO₂ capture purpose. Based on the current scenario, following suggestions are proposed for the future research on CO₂ capture using metal oxides:

• (a) Improvement in the potential sorbent performance, in terms of capture capacity, reversibility rate, carbonation kinetics, multi-cycle durability and sorbent reversibility;

• (b) In-depth analysis of performance check of potential sorbent under actual flue gas conditions;

• (c) A detailed techno-economic assessment of the potential sorbents;