Ab initio thermodynamic approach to identify mixed solid sorbents for CO2 capture technology

Because the current technologies for capturing CO₂ are still too energy intensive, new materials must be developed that can capture CO₂ reversibly with acceptable energy costs. At a given CO₂ pressure, the turnover temperature (T_t) of the reaction of an individual solid that can capture CO₂ is fixed. Such T_t may be outside the operating temperature range (ΔT_o) for a practical capture technology. To adjust T_t to fit the practical ΔT_o, in this study, three scenarios of mixing schemes are explored by combining thermodynamic database mining with first principles density functional theory (DFT) and phonon lattice dynamics calculations. Our calculated results demonstrate that by mixing different types of solids, it's possible to shift T_t to the range of practical operating temperature conditions. According to the requirements imposed by the pre- and post- combustion technologies and based on our calculated thermodynamic properties for the CO₂ capture reactions by the mixed solids of interest, we were able to identify the mixing ratios of two or more solids to form new sorbent materials for which lower capture energy costs are expected at the desired pressure and temperature conditions.

Introduction

Today, fossil fuels are still the main energy sources for the world's economy. One consequence of the use of these fuels is the emission of huge quantities of CO₂ into the atmosphere, creating environmental problems such as climate change (White et al., 2003; Aaron and Tsouris, 2005; Allen et al., 2009; Haszeldine, 2009; Li et al., 2013). To mitigate such problems, CO₂ emissions into the atmosphere must be reduced by being captured and stored (Ochoa-Fernandez et al., 2005; Pfeiffer and Bosch, 2005; Li et al., 2013). Current technologies for capturing CO₂, including solvent-based (amines) and CaO-based materials, are still too energy intensive. Hence, development of new materials that can capture and release CO₂ reversibly with acceptable energy costs are critical. In particular, solid oxide sorbent materials have been proposed for capturing CO₂ through a reversible chemical transformation leading primarily to formation of carbonate products. Solid sorbents containing alkali and alkaline earth metals have been reported in previous studies to be promising candidates for CO₂ sorbent applications due to their high CO₂ absorption capacity at moderate working temperatures (Duan and Sorescu, 2009, 2010; Duan et al., 2012a).

During the past few years, NETL developed a theoretical methodology to identify promising solid sorbent candidates for CO₂ capture by combining thermodynamic database searching with ab initio thermodynamics obtained based on the first-principles density functional theory (DFT) and lattice phonon dynamics (Duan and Sorescu, 2009, 2010; Duan and Parlinski, 2011; Duan et al., 2011, 2012a,b,c; Zhang et al., 2012). As shown in Figure 1, the primary outcome of our screening scheme is a list of promising CO₂ sorbents with optimal energy usage. After screening a given material databank, we selected only a short list of candidates for further experimental validations.

FIGURE 1

Figure 1. Schematic of our screening methodology.

A practical CO₂ capture technology has optimal operating conditions, such as the absorption/desorption of CO₂ at the necessary pressure and operating temperature range (ΔT_o). As a good sorbent, its CO₂ capture/release temperature should fit into such a range. However, at a given CO₂ pressure, the turnover temperature (T_t) of an individual solid capture CO₂ reaction is fixed. Such T_t may be outside the operating temperature range (ΔT_o) for a particular capture technology. In order to adjust T_t to fit the practical ΔT_o, its corresponding thermodynamic property must be changed by changing its structure by reacting (mixing) it with other materials or doping it with other elements. In this study, we demonstrate that by mixing different types of solids in three scenarios, it's possible to shift T_t to the range of practical operating conditions.

Calculation Methods for Mixed Solid Sorbents

A complete description of the computational methodology together with relevant applications can be found in our previous publications (Duan and Sorescu, 2009, 2010; Duan, 2011; Duan and Parlinski, 2011; Duan et al., 2011, 2012a,b,c; Zhang et al., 2012). The CO₂ capture reactions of solids can be expressed generically in the form (for convenient description, we normalized the reaction to 1 mole of CO₂)

$\begin{array}{ll}{\displaystyle \sum _{Ri}}{n}_{Ri}Solid\text{\_}{R}_i+C{O}_2\leftrightarrow {\displaystyle \sum _{Pj}}{n}_{Pj}Solid\text{\_}{P}_j& (a)\end{array}$

where n_Ri, n_Pj are the numbers of moles of reactants (R_i) and products (P_j) involved in the capture reactions. As discussed in the following section, the reactants R_i can simply be mixed solids or a newly formed solid by mixing different kinds of solids with certain mixing ratios. We treat the gas phase CO₂ as an ideal gas. By assuming that the difference between the chemical potentials (Δμ⁰) of the solid phases of reactants (R_i) and products (P_j) can be approximated by the difference in their total energies (ΔE^DFT), obtained directly from DFT calculations, and their vibrational free energies of phonons dynamics and by ignoring the PV contribution terms for solids, the variation of the Gibbs free energy (ΔG) for reaction (a) with temperature and CO₂ pressure can be written as (Duan and Sorescu, 2009, 2010; Duan, 2011; Duan and Parlinski, 2011; Duan et al., 2011, 2012a,b,c; Zhang et al., 2012)

$\begin{array}{rc}\Delta G(T,P)=\Delta {\mu }^0(T)-RTln\frac{P_{C{O}_2}}{P_0}& (1)\end{array}$

where

$\begin{array}{rc}\Delta {\mu }^0(\text{T})\approx \Delta {\text{E}}^{\text{DFT}}+\Delta {\text{E}}_{\text{ZP}}+\Delta {\text{F}}^{\text{PH}}(\text{T})-{\text{G}}_{\text{C}{\text{O}}_2}^0(\text{T})& (2)\end{array}$

Here, ΔE^DFT is the DFT energy difference between the reactants and products of reaction (a), ΔE_ZP is the zero point energy difference between the reactants and products and can be obtained directly from phonon calculations. ΔF^PH is the phonon free energy change excluding zero-point energy (which is already counted into the ΔE_ZP term) between the solids of products and reactants. P_CO2 is the partial pressure of CO₂ in the gas phase and P₀ is the standard state reference pressure taken to be 1 bar. The heat of reaction [ΔH^cal(T)] can be evaluated through the following equation

$\begin{array}{rc}\Delta {\text{H}}^{\text{cal}}(\text{T})=\Delta {\mu }^0(\text{T})+\text{T}\left[\Delta {\text{S}}_{\text{PH}}(\text{T})-{\text{S}}_{\text{C}{\text{O}}_2}(\text{T})\right]& (3)\end{array}$

where ΔS_PH(T) is the difference of entropies between product solids and reactant solids. The free energy of CO₂ (${\text{G}}_{\text{C}\text{O}2}^0$) can be obtained from standard statistical mechanics (Duan and Sorescu, 2009, 2010; Zhang et al., 2012), and its entropy (S_CO2) can be found in the empirical thermodynamic databases (Chase, 1998). Equation (1) provides the relationships of Gibbs free energy change of reaction (a) vs. temperature and CO₂ pressure. Obviously, when set ΔG = 0, the P-T relationship (van't Hoff plot) is obtained to determine the T_t:

$\begin{array}{rc}{T}_t=\frac{\Delta {\mu }^0(T)}{Rln\frac{P_{C{O}_2}}{P_0}}& (4)\end{array}$

Based on this equation, at giving CO₂ pressure P_CO2, the T_t can be determined for each CO₂ capture reaction.

Results and Discussion

For a given CO₂ capture process, its optimal working conditions [CO₂ pressures of pre- and after-capture, absorption/desorption temperature range (ΔT_o), etc.] were fixed. However, at a given CO₂ pressure, the T_t of an individual solid capture CO₂ reaction is also fixed. Such T_t may be outside the operating temperature range ΔT_o for a particular capture technology. To adjust T_t to fit the practical working range through reversible chemical transformations, the chemical properties (such as structure, phase, etc.) of solids must be modified to change the ΔG(T, P) in Equation (1). If we want to increase the T_t to a higher temperature range, the ΔG(T, P) should be more negative. To achieve it, we can either destabilize the reactants (sorbents), stabilize the products, or do both. Conversely, if we want to decrease T_t to a lower temperature range, the ΔG(T, P) should be less negative, which can either stabilize the reactant (sorbents), destabilize the products, or do both. In other words, mixing stabilizer/destabilizer with a solid could change the thermodynamic properties of their CO₂ capture reactions to shift T_t. Some mixing examples are given in Table 1. As one can see that the mixed sorbent could be a new formed solid (e.g., lithium silicates) or just a simple mixture (e.g., MgO + Na₂CO₃) to change the chemical properties of reactants and products. Depending on the main captor A and the direction of T_t shifts, a different mixed solid B and mixing ratio could be determined. Although one can mix any number of solids to form a new CO₂ sorbent, to focus on exploring the nature of mixtures, here we restrict ourselves with cases of two and three mixed solids.

TABLE 1

Table 1. Mixing schemes and examples of mixed solids and their effects on CO₂ capture.

As shown in Table 1, as effective main CO₂ captors through chemical reactions to form carbonates, alkali and alkaline earth metal oxides are of interest due to their ease in reacting with CO₂ and low costs. The problem is that they can strongly react with CO₂ to form carbonates, but their T_t are very high, and they can only be regenerated at very high temperatures, which are unsuitable for many CO₂ capture technologies. Hence, mixing with other solids to shift their T_t becomes important for their suitability as CO₂ sorbents. Generally, when we mix two solids A and B to form a new sorbent C, the T_t of the new system (T_C) is located between the A and B (T_A, T_B). Here, it was assumed that A is a strong CO₂ sorbent, while B is a weak CO₂ sorbent and T_A > T_B. Also, we assumed that the desired operating temperature T_O is between T_A and T_B (T_A > T_O > T_B). Depending on the properties of A and B, as shown in Table 1, we typically have three scenarios to synthesize the mixing sorbent C.

T_A >> T_B and the A Component is Key to Capturing CO₂

In this case, by mixing B into A, a new solid C is formed. Because B is not involved in the CO₂ capture, it serves as a stabilizer to stabilize A to form solid C, and the reaction possesses lower energy than A and B individually. Therefore, the energy state of reactant is lower than pure A. The Δμ⁰ in Equation (2) will become higher (less negative). According to Equation (4), the corresponding T_C should be lower than T_A. In other words, the mixed sorbent B shifts T_A to a lower temperature range. The amount of shifted temperature depends on the mixing ratio.

An example of this case is represented by Li₂O. As we know, Li₂O is a very strong CO₂ sorbent that forms Li₂CO₃. However, its regeneration from Li₂CO₃ can only occur at very high temperatures (T_A > 1000 K). In order to move its T_A to lower temperatures, one can mix in some weak CO₂ sorbents (such as SiO₂, ZrO₂). With different mixing ratios of Li₂O/SiO₂ (or ZrO₂), different stable lithium silicates (or zirconates) can be formed, such as Li₈SiO₆, Li₄SiO₄, Li₆Si₂O₇, Li₂SiO₃, Li₂Si₂O₅, Li₂Si₃O₇; Li₈ZrO₆, Li₆Zr₂O₇, Li₂ZrO₃, etc. The crystal structures of these lithium silicates and zirconates can be found in the literature. By performing ab initio thermodynamic property calculations on these solids capturing CO₂ reactions (Duan and Sorescu, 2009; Duan, 2011, 2012, 2013; Duan and Parlinski, 2011; Duan et al., 2013), Figure 2 shows the calculated relationships of Gibbs free energy (ΔG), P_CO2, and T of the CO₂ capture reactions by the mixed Li₂O/SiO₂ and Li₂O/ZrO₂ solids with different mixing ratios. Figure 3 shows the turnover temperatures and the CO₂ capture capacities of Li₂O/SiO₂ and Li₂O/ZrO₂ mixtures vs. the ratio of Li₂O/SiO₂ or Li₂O/ZrO₂ (Duan and Sorescu, 2009, 2010; Duan, 2011, 2012, 2013; Duan and Parlinski, 2011; Duan et al., 2012a; Duan and Lekse, 2015).

FIGURE 2

Figure 2. Plots of the calculated free energy vs. CO₂ pressures and temperatures for the CO₂ capture reaction by Li₂O+SiO₂ or Li₂O+ZrO₂ mixtures. Y-axis plotted in logarithm scale. The line shows ΔG(T, P) = 0. For each reaction, above its ΔG(T, P) = 0 curve, its ΔG < 0, which means the solids absorb CO₂ and the reaction goes forward, whereas below the ΔG(T, P) = 0 curve, its ΔG > 0, which means the CO₂ starts to release and the reaction goes backward to regenerate the sorbents.

FIGURE 3

Figure 3. The calculated turnover temperatures and CO₂ capture capacity vs. molar percentage of SiO₂ or ZrO₂ in the lithium silicates or zirconates. T₁ are the turnover temperatures under pre-combustion conditions with CO₂ partial pressure at 20 bars, while T₂ are the turnover temperatures under post-combustion conditions with CO₂ partial pressure at 0.1 bar.

From Figures 2, 3, one can see that after mixing SiO₂ (or ZrO₂) in Li₂O with different Li₂O/SiO₂ (or Li₂O/ZrO₂) ratios, the T_C of the newly formed C compound (silicate or zirconate) is lower than T_A of pure Li₂O and could be close to the ΔT_o range to fit the practical needs. Although SiO₂ and ZrO₂ do not capture CO₂, they can exothermically react with Li₂O to form silicates and zirconates (as shown in Figure 2). Such reactions bring the energy levels of sorbents (reactants) down to more negative values, but do not affect the products (carbonates). Therefore, the heat of reaction and free energy change of the CO₂ capture reactions by these lithium silicates and zirconates will be increased to less negative values compared to pure Li₂O reacting with CO₂, and in turn, the turnover temperatures are shifted to a lower temperature range. When the SiO₂/Li₂O or ZrO₂/Li₂O ratios are increased by adding more SiO₂ or ZrO₂ into Li₂O, as shown in Figures 2, 3, the turnover temperatures of mixed sorbents (T_t) are shifted to a lower temperature range. Therefore, by controlling the mixing ratio, it is possible to move the CO₂ capture temperature down to the range required by certain capture technology.

T_A >> T_B and B Component is Key to Capturing CO₂

Opposite to the previous case, in this case, we want to increase the T_t. Because T_B is lower than T_O, mixing A into B will increase the turnover temperature T_C of the C solid to a value closer to T_o. In this way, A will either destabilize solid B or stabilize the captured products. For example, pure MgO has a very high theoretical CO₂ capture capacity. However, its T_t (250°C) is lower than the required temperature range of 300–470°C used in warm gas clean up technology and its practical CO₂ capacity is very low. Therefore, pure MgO cannot be used directly as a CO₂ sorbent in such capture technology (Zhang et al., 2013, 2014; Chi et al., 2014). Figure 4 shows an example of mixing Na₂O (or Na₂CO₃) with MgO to move the ΔG(T, P) = 0 curve of MgO (red line in Figure 4) to a higher temperature range (green line in Figure 4).

FIGURE 4

Figure 4. Plots of the calculated free energy vs. CO₂ pressures and temperatures for the CO₂ capture reaction by MgO + Na₂CO₃ (or Na₂O) to form the double salt Na₂Mg(CO₃)₂. Y-axis plotted in logarithm scale. The line shows ΔG(T, P) = 0. For each reaction, above its ΔG(T, P) = 0 curve, its ΔG < 0, which means the solids absorb CO₂ and the reaction goes forward, whereas below the ΔG(T, P) = 0 curve, its ΔG > 0, which means the CO₂ starts to release and the reaction goes backward to regenerate the sorbents.

One can see that in Figure 4, Na₂O (or Na₂CO₃) does not react with the reactant MgO directly, but it reacts exothermically with the carbonate MgCO₃ to form the double salt Na₂Mg(CO₃)₂ and, hence, stabilizes the product. Therefore, the heat of reaction and free energy change of CO₂ capture reaction by MgO + Na₂CO₃ are lower (more negative) compared with pure MgO reacting with CO₂. According to Equations (1) and (4), the T_t of the Na₂CO₃-promoted MgO sorbent capturing CO₂ is higher than that of pure MgO. Figure 5 shows the changes of MgO turnover temperatures by adding different oxides or carbonates. Generally, by mixing alkali metal oxides, M₂O (M = Na, K, Cs, Ca) or their carbonates (M₂CO₃) into MgO, the corresponding newly formed mixing systems have higher turnover temperatures, making them useful as CO₂ sorbents through the reaction MgO + CO₂ + M₂CO₃ = M₂Mg(CO₃)₂ (Zhang et al., 2013; Duan et al., 2014).

FIGURE 5

Figure 5. The shifts of MgO turnover temperatures by adding several oxides (carbonates) with a molar ratio of 1:1. T₁ are the turnover temperatures under pre-combustion conditions with CO₂ partial pressure of 20 bars, while T₂ are the turnover temperatures under post-combustion conditions with CO₂ partial pressure of 0.1 bar.

From Figure 5, one can see that, in these CO₂ capture reactions, the Na₂CO₃-promoted MgO possesses a larger T_t shift to a higher temperature range than the K₂CO₃ + MgO and CaCO₃ + MgO do. Because the desired T_t of the mixture is lower than the dissociation temperature of promotors (Na₂CO₃, K₂CO₃, CaCO₃), during the CO₂ capture sorption process, these promoters act as stabilizers to react with MgCO₃ to form a double salt, while during the CO₂ desorption, the double salt dissociates to MgO and the promoter.

Both A and B are Active for CO₂ Capture

In this case, we want both A and B components active to capture CO₂, and the CO₂ capture capacity of the mixture is the summation of those of A and B. As we know, another potential advantage of mixing solids is to increase the surface area of the solids to have a faster reaction rate. Such a mixing scenario does not show too much advantage in shifting the capture temperature, but it may enhance the kinetics of the capture process and eventually make the mixtures more efficient for capturing CO₂. Although, up to now, no such report has appeared in literature, we think such an attempt is worthwhile, and we are working on several doped systems. Here we show a case of Li-/K- doped Na₂ZrO₃ capture for CO₂ through the following reaction (b): (Duan, 2014; Duan and Lekse, 2015; Duan et al., 2015)

$\begin{array}{l}{\text{M}}_{\text{x}}{\text{N}}_{2-\text{x}}{\text{ZrO}}_3+{\text{CO}}_2\leftrightarrow \frac{\text{x}}{2}{\text{M}}_2{\text{CO}}_3+\frac{2-\text{x}}{2}{\text{N}}_2{\text{CO}}_3+{\text{ZrO}}_2\\ (\text{M},\text{N}=\text{Li},\text{Na},\text{K};\text{x}=0,0.5,1,1.5,2)\hfill & (\text{b})\end{array}$

Figure 6 demonstrated the calculated relationships among the Gibbs free energy change, CO₂ pressure and temperature for CO₂ capture reactions by Na_2−xM_xZrO₃ (M = Li, K, x = 0.0, 0.5, 1.0, 1.5, 2.0). Figure 7 showed the dependence of T_t and CO₂ capture capacity on Li/K doping molar percentage in Na_xM_2−xZrO₃ (M = Li, K). One can see from Figure 6 that when Na₂ZrO₃ is doped with a different molar ratio of Li or K, the thermodynamic properties of the doped systems are quite different from pure Na₂ZrO₃.

FIGURE 6

Figure 6. The contour plotting of calculated Gibbs free energy vs. CO₂ pressures and temperatures of the CO₂ capture reactions by Na_xM_2-xZrO₃ and M₂ZrO₃(M = Li, Na, K, x = 0, 0.5, 1.0, 1.5, 2.0) through reaction Na_xM_2−xZrO₃ + CO₂ = ^x/₂Na₂CO₃ + ^2−x/₂M₂CO₃ + ZrO₂. Y-axis plotted in logarithm scale. The line shows ΔG(T, P) = 0. For each reaction, above its ΔG(T, P) = 0 curve, its ΔG < 0, which means the solids absorb CO₂ and the reaction goes forward, whereas below the ΔG(T, P) = 0 curve, its ΔG > 0, which means the CO₂ starts to release and the reaction goes backward to regenerate the sorbents.

FIGURE 7

Figure 7. The dependence of the turnover temperatures and CO₂ capture capacity on Li/K doping molar percentage in Na_xM_2−xZrO₃ (M = Li, K). T₁ is the turnover temperatures under pre-combustion conditions with CO₂ partial pressure at 20 bars, while T₂ is the turnover temperatures under post-combustion conditions with CO₂ partial pressure at 0.1 bar.

Based on the calculated relationships among the Gibbs free energy change, CO₂ pressure and temperature for CO₂ capture reactions by Na_2−xM_xZrO₃ are shown in Figure 6. Compared to pure Na₂ZrO₃, overall, the Li- and K-doped mixtures Na_2−xM_xZrO₃ have lower T_t, shown in Figure 7. The calculated results show that the shift in T_t depends not only on the doping element, but also on the doping level. As one can see from Figure 7, the Li-doped systems have larger T_t decreases than the K-doped systems. When increasing the Li-doping level x, the T_t of the corresponding mixture Na_2−xLi_xZrO₃ decreases further to a low temperature range. However, in the case of K-doped systems Na_2−xK_xZrO₃—although initial doping of K into Na₂ZrO₃ can shift its T_t to a lower temperature range—further increasing the K-doping level x results in an increase in T_t. Therefore, compared to K-doping, lithium inclusion into Na₂ZrO₃ structure has a larger influence on its CO₂ capture performance.

All of these obtained results may be of great interest in the development of specific CO₂ capture applications. As shown in Figures 6, 7, the Na_2−xLi_xZrO₃ and Na_2−xK_xZrO₃ compositions can produce modifications in the CO₂ capture temperatures, which may be used in the design of a specific composition, depending on the temperature range that industry requires. These results have identified that the capture of CO₂ in zirconate materials is not simply a matter of a substitutional element but also the doping level. This insight will need to be considered during future sorbent development. The obtained results have also demonstrated that computational methods can be used to accurately predict aspects of CO₂ capture and have the potential to drive future work by leading to researchers identifying and designing the most promising candidate materials.

In addition, the Na_2−xM_xZrO₃ materials can be stoichiometrically regarded as a mixture of three oxides: Na₂O, M₂O, and ZrO₂ in the ratio (2-x):x:1. These results provide ways that mixing/doping more than two oxides to form new sorbents that can fit the industrial needs to capture CO₂ with better performances and proper working conditions. More complicated doping systems, such as N_2−xM_xMeSiO₄ (N, M = Li, Na, K; Me = Fe, Co, Ni, Mn, Zn, Mg, etc.), can be regarded as a mixture of four oxides: N₂O, M₂O, MeO, SiO₂ in the ratio (2-x):x:1:1. Their CO₂ capture reactions could be written as:

$\begin{array}{l}{\text{M}}_{\text{x}}{\text{N}}_{2-\text{x}}{\text{MeSiO}}_4+{\text{CO}}_2\leftrightarrow \frac{\text{x}}{2}{\text{M}}_2{\text{CO}}_3+\frac{2-\text{x}}{2}{\text{N}}_2{\text{CO}}_3+{\text{MeSiO}}_3\\ (\text{M},\text{N}=\text{Li},\text{Na},\text{K};\text{Me}:\text{transition}\text{metal})\hfill & (\text{c})\end{array}$

By analyzing their CO₂ capture behaviors with different doping/mixing ratios, researchers will draw more general conclusions. Analysis of the results is underway and will appear in future reports.

From the mixing systems mentioned, one can see that after mixing/doping solid B into A, the theoretical maximum of CO₂ capture capacity of the mixture is decreased compared with pure A or B. However, it does not mean the practical CO₂ capacity will be decreased. The CO₂ capacity of solid sorbents should be above 3 mole CO₂ per kilogram solid (~15 wt. %) to meet the industrial requirements and have a chance of providing energy reductions of 30–50% or more compared to the optimum aqueous-monoethanolamine(MEA)-based process (Gray et al., 2008). As shown in Figures 3, 5, 7, the theoretical CO₂ weight percentage maxima of all these mixtures are greater than this minimum requirement (>15 wt. %). Therefore, from the CO₂ capture capacity point of view, all of these systems could meet this criterion to be used as CO₂ sorbents. In addition, after mixing another solid, the structure of the sorbent is changed, and more active sites could be contacted by CO₂ to enhance the capture kinetics, and in turn, to increase its practical capture capacity.

Conclusions

At a given CO₂ pressure, the T_t of an individual solid capture CO₂ reaction is fixed. Such T_t may be outside the operating temperature range (ΔT_o) for a practical capture technology. To adjust T_t to fit the practical ΔT_o, in this study, by combining thermodynamic database mining with first principles DFT and phonon lattice dynamics calculations, our calculated results demonstrate that by mixing different types of solids, it is possible to shift T_t to the range of practical operating temperature conditions.

The obtained results showed that by changing the mixing ratio of solid A and solid B to form mixed solid C, it is possible to shift T_t of the newly formed solid C to fit the practical CO₂ capture technologies. In this study, we investigated three scenarios of mixing schemes: (i) T_A > > T_B, and the A component is the key component for capturing CO₂. In this case, because T_A is higher than T_O, mixing B into A will decrease the turnover T_A of the A solid to values closer to T_o. For example, Li₂O is a very strong CO₂ sorbent that forms Li₂CO₃. However, its regeneration from Li₂CO₃ only can occur at very high temperatures (T_A). To move its T_A to lower temperatures, Li₂O can be mixed with some weak CO₂ sorbents (such as SiO₂, ZrO₂). Our results showed that, in this way, the turnover T_t and the theoretical CO₂ capture capacity of mixtures decrease as the ratios of Li₂O/SiO₂ or Li₂O/ZrO₂ decrease. (ii) T_A > > T_B and B component is the key part to capturing CO₂. In this case, because T_B is lower than T_O, mixing A into B will increase the T_B of the B solid to values closer to T_o. For example, pure MgO (as B component) has a very high theoretical CO₂ capture capacity. However, its T_B (250°C) is lower than the required temperature range of 300–470°C used in warm gas clean up technology. The obtained results showed that by mixing alkali metal oxides M₂O (M = Na, K, Cs, Ca) or their carbonates (M₂CO₃) into MgO, the corresponding newly formed mixed systems have higher turnover temperatures by forming double salts through the reactions MgO + CO₂ + M₂CO₃ = M₂Mg(CO₃)₂. (iii) Both A and B components are active to capture CO₂. In this case, the CO₂ capacity of the mixture is the summation of those of A and B. Li₂MSiO₄ (M = Mg, Ca, etc.) and M_2−xN_xZrO₃ (M, N = Li, Na, K) belong to this category. Those doped systems can be treated as the mixing of three solids (Li₂O:MO:SiO₂, M₂O:N₂O:ZrO₂). This study summarized the results of Na_2−xM_xZrO₃ (M = Li, Na, K, x = 0, 0.5, 1, 1.5, 2) doped sorbents. The results showed that when capturing CO₂, the K-/Li- doped Na₂ZrO₃ have lower T_t compared to pure Na₂ZrO₃.

The obtained results can be used to provide insights for designing new CO₂ sorbents. Therefore, although one single material taken in isolation might not be an optimal CO₂ sorbent to fit the particular needs to operate at specific temperature and pressure conditions, by mixing or doping two or more materials to form a new solid, the calculated results showed that it is possible to synthesize new CO₂ sorbent formulations that can fit the industrial needs. Our results also show that computational modeling can play a decisive role for identifying materials with optimal performance.

It should be pointed out that in this study we only focused on the thermodynamic properties of the CO₂ capture reactions, which are essential and critical to determine whether the sorbents can capture CO₂. Once the capture reaction is thermodynamically favorable, other properties (such as kinetics, mechanical resistance, toxicity, sulfur poisoning resistance, cost, etc.) also play important roles to select proper sorbent candidates for experimental validations. Generally speaking, further simulations can be performed on sorbents to optimize their performance for capturing CO₂. For example, by calculating the transition states with DFT, the kinetic properties of CO₂ capture reactions can be evaluated; by conducting mechanical and chemical engineering modeling with finite element method, the mechanical and sintering behaviors of sorbents can be obtained; by performing process modeling, the overall energetic and material costs can be estimated, and then some comparisons with amine-based solution capture technology can be drown. Poisoning gases (such as SO_x, NO_x, H₂S, etc.) have big effects on the sorbent performance for CO₂ capture. Further exploring the mechanisms of poisoning gases interacting with sorbents will provide useful information for designing new CO₂ capture technologies.

Disclaimer

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference therein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed therein do not necessarily state or reflect those of the United States Government or any agency thereof.

Conflict of Interest Statement

The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

The results in this report were presented at AIChE Spring and ACEME-15 conferences. The author wishes to acknowledge Drs. D. C. Sorescu, H. Pennline, K. Johnson, H. Pfeiffer, B. Y. Li, K. L. Zhang, D. L. King, X. S. Li, X. F. Wang, D. Luebke, J. Lekse, J. W. Halley, and B. Alcantar-Vazquez for their collaborations and fruitful discussions, and C. Wamsley (internal technical writer) for proofreading the manuscript.

References

Aaron, D., and Tsouris, C. (2005). Separation of CO₂ from flue gas: a review. Sep. Sci. Technol. 40, 321–348. doi: 10.1081/SS-200042244

CrossRef Full Text | Google Scholar

Allen, M. R., Frame, D. J., Huntingford, C., Jones, C. D., Lowe, J. A., Meinshausen, M., et al. (2009). Warming caused by cumulative carbon emissions towards the trillionth tonne. Nature 458, 1163–1166. doi: 10.1038/nature08019

PubMed Abstract | CrossRef Full Text | Google Scholar

Chase, M. W. J. (1998). NIST-JANAF themochemical tables, 4th edition. J. Phys. Chem. Ref. Data 9(Monogr.), 1–1951.

Chi, J. L., Zhao, L. F., Wang, B., Li, Z., Xiao, Y. H., and Duan, Y. (2014). Thermodynamic performance assessment and comparison of IGCC with solid cycling process for CO₂ capture at high and medium temperatures. Int. J. Hydrogen Energy 39, 6479–6491. doi: 10.1016/j.ijhydene.2014.02.005

CrossRef Full Text | Google Scholar

Duan, Y. (2011). Electronic structural and phonon properties of lithium zirconates and their capabilities of CO₂ capture: a first-principle density functional approach. J. Renew. Sust. Energy 3, 013102. doi: 10.1063/1.3529427

CrossRef Full Text | Google Scholar

Duan, Y. (2012). A first-principles density functional theory study of the electronic structural and thermodynamic properties of M₂ZrO₃ and M₂CO₃ (M = Na, K) and their capabilities for CO₂ capture. J. Renew. Sust. Energy 4, 013109. doi: 10.1063/1.3683519

CrossRef Full Text | Google Scholar

Duan, Y. (2013). Structural and electronic properties of Li₈ZrO₆ and its CO₂ capture capabilities: an ab initio thermodynamic approach. Phys. Chem. Chem. Phys. 15, 9752–9760. doi: 10.1039/c3cp51101d

PubMed Abstract | CrossRef Full Text | Google Scholar

Duan, Y. (2014). Electronic, structural, phonon dynamical, and CO₂ capture properties of LiMZrO₃ (M = Na, K) by ab initio thermodynamic investigation. Sci. Lett. J. 3:56.

Duan, Y., and Lekse, J. (2015). Regeneration mechanisms of high-lithium content zirconates as CO₂ capture sorbents: experimental measurements and theoretical investigations. Phys. Chem. Chem. Phys. 17, 22543–22547. doi: 10.1039/C5CP03968A

PubMed Abstract | CrossRef Full Text | Google Scholar

Duan, Y., Lekse, J., Wang, X. F., Li, B. Y., Alcantar-Vazquez, B., Pfeiffer, H., et al. (2015). Electronic structural, phonon dynamic properties, and CO₂ capture capabilities of Na_2−xM_xZrO₃ (M = Li, K): density functional theory calculations and experimental validations. Phys. Rev. Appl. 3:044013. doi: 10.1103/PhysRevApplied.3.044013

CrossRef Full Text | Google Scholar

Duan, Y., Luebke, D., and Pennline, H. W. (2012a). Efficient theoretical screening of solid sorbents for CO₂ capture applications. Int. J. Clean Coal Energy 1, 1–11. doi: 10.4236/ijcce.2012.11001

CrossRef Full Text | Google Scholar

Duan, Y., Luebke, D. R., Pennline, H. W., Li, B. Y., Janik, M. J., and Halley, J. W. (2012b). Ab initio thermodynamic study of the CO₂ capture properties of potassium carbonate sesquihydrate, K₂CO₃ · 1.5H₂O. J. Phys. Chem. C 116, 14461–14470. doi: 10.1021/jp303844t

CrossRef Full Text | Google Scholar

Duan, Y., and Parlinski, K. (2011). Density functional theory study of the structural, electronic, lattice dynamical, and thermodynamic properties of Li₄SiO₄ and its capability for CO₂ capture. Phys. Rev. B 84:104113. doi: 10.1103/PhysRevB.84.104113

CrossRef Full Text | Google Scholar

Duan, Y., Pfeiffer, H., Li, B., Romero-Ibarra, I. C., Sorescu, D. C., Luebke, D. R., et al. (2013). CO₂ capture properties of lithium silicates with different ratios of Li₂O/SiO₂: an ab initio thermodynamic and experimental approach. Phys. Chem. Chem. Phys. 15, 13538–13558. doi: 10.1039/c3cp51659h

PubMed Abstract | CrossRef Full Text | Google Scholar

Duan, Y., and Sorescu, D. C. (2009). Density functional theory studies of the structural, electronic, and phonon properties of Li₂O and Li₂CO₃: Application to CO₂ capture reaction. Phys. Rev. B 79:014301. doi: 10.1103/PhysRevB.79.014301

CrossRef Full Text

Duan, Y., and Sorescu, D. C. (2010). CO₂ capture properties of alkaline earth metal oxides and hydroxides: a combined density functional theory and lattice phonon dynamics study. J. Chem. Phys. 133, 074508. doi: 10.1063/1.3473043

CrossRef Full Text | Google Scholar

Duan, Y., Zhang, B., Sorescu, D. C., and Johnson, J. K. (2011). CO₂ Capture properties of M-C-O-H (M = Li, Na, K) systems: a combined density functional theory and lattice phonon dynamics study. J. Solid State Chem. 184, 304–311. doi: 10.1016/j.jssc.2010.12.005

CrossRef Full Text | Google Scholar

Duan, Y., Zhang, B., Sorescu, D. C., Karl Johnson, J., Majzoub, E. H., and Luebke, D. R. (2012c). Density functional theory studies on the electronic, structural, phonon dynamical and thermo-stability properties of bicarbonates MHCO₃, M = Li, Na, K. J. Phys. 24:325501. doi: 10.1088/0953-8984/24/32/325501

PubMed Abstract | CrossRef Full Text | Google Scholar

Duan, Y., Zhang, K., Li, X. S., King, D. L., Li, B., Zhao, L., et al. (2014). Ab initio thermodynamic study of the CO₂ capture properties of M₂CO₃ (M = Na, K)- and CaCO₃-promoted MgO sorbents towards forming double salts. Aerosol Air Qual. Res. 14, 470–479. doi: 10.4209/aaqr.2013.05.0178

CrossRef Full Text | Google Scholar

Gray, M. L., Champagne, K. J., Fauth, D., Baltrus, J. P., and Pennline, H. (2008). Performance of immobilized tertiary amine solid sorbents for the capture of carbon dioxide. Int. J. Greenhouse Gas Control 2, 3–8. doi: 10.1016/S1750-5836(07)00088-6

CrossRef Full Text | Google Scholar

Haszeldine, R. S. (2009). Carbon capture and storage: how green can black be? Science 325, 1647–1652. doi: 10.1126/science.1172246

PubMed Abstract | CrossRef Full Text | Google Scholar

Li, B. Y., Duan, Y., Luebke, D., and Morreale, B. (2013). Advances in CO₂ capture technology: a patent review. Appl. Energy 102, 1439–1447. doi: 10.1016/j.apenergy.2012.09.009

CrossRef Full Text | Google Scholar

Ochoa-Fernandez, E., Rusten, H. K., Jakobsen, H. A., Ronning, M., Holmen, A., and Chen, D. (2005). Sorption enhanced hydrogen production by steam methane reforming using Li₂ZrO₃ as sorbent: sorption kinetics and reactor simulation. Catal. Today 106, 41–46. doi: 10.1016/j.cattod.2005.07.146

CrossRef Full Text | Google Scholar

Pfeiffer, H., and Bosch, P. (2005). Thermal stability and high-temperature carbon dioxide sorption on hexa-lithium zirconate (Li₆Zr₂O₇). Chem. Mater. 17, 1704–1710. doi: 10.1021/cm047897+

CrossRef Full Text | Google Scholar

White, C. M., Strazisar, B. R., Granite, E. J., Hoffman, J. S., and Pennline, H. W. (2003). Separation and capture of CO₂ from large stationary sources and sequestration in geological formations-coalbeds and deep saline aquifers. J. Air Waste Manag. Assoc. 53, 645–715. doi: 10.1080/10473289.2003.10466206

PubMed Abstract | CrossRef Full Text | Google Scholar

Zhang, B., Duan, Y., and Johnson, J. K. (2012). First-principles density functional theory study of CO₂ capture with transition metal oxides and hydroxides. J. Chem. Phys. 136, 064516. doi: 10.1063/1.3684901

CrossRef Full Text | Google Scholar

Zhang, K. L., Li, X. H. S., Duan, Y., King, D. L., Singh, P., and Li, L. Y. (2013). Roles of double salt formation and NaNO₃ in Na₂CO₃-promoted MgO absorbent for intermediate temperature CO₂ removal. Int. J. Greenhouse Gas Control 12, 351–358. doi: 10.1016/j.ijggc.2012.11.013

CrossRef Full Text | Google Scholar

Zhang, K. L., Li, X. H. S., Li, W. Z., Rohatgi, A., Duan, Y., Singh, P., et al. (2014). Phase transfer-catalyzed fast CO₂ absorption by MgO-based absorbents with high cycling capacity. Adv. Mater. Interfaces 1:1400030. doi: 10.1002/admi.201400030

CrossRef Full Text | Google Scholar