Bifunctional application of sodium cobaltate as a catalyst and captor through CO oxidation and subsequent CO₂ chemisorption processes

Abstract

The potential bifunctional mechanism of sodium cobaltate (NaCoO₂) in the catalysis of CO oxidation and subsequent CO₂ chemisorption was systematically analysed. Different catalytic and gravimetric experiments were performed dynamically and isothermally at multiple temperatures. Initially, the CO oxidation process was evaluated using a catalytic reactor connected to a gas chromatograph. Once the production of CO₂ was confirmed, its chemisorption capacity with NaCoO₂ was studied gravimetrically. Catalytic and gravimetric analysis products were studied by XRD, FTIR and SEM to elucidate the double reaction mechanism. Sodium cobaltate exhibited interesting catalytic properties over a wide temperature range, although the NaCoO₂ crystalline structure and chemical composition changed during the CO₂ capture process. Furthermore, all the experiments were theoretically supported by first-principles density functional theory thermodynamic calculations. The calculated thermodynamic properties of the CO oxidation and CO₂ capture reactions with NaCoO₂ under different oxidation conditions were in good agreement with the experimental measurements.

---

Introduction

Humanity would not be able to survive in the modern era without utilizing large amounts of energy, and fossil fuels have become essential for providing this energy. However, the use of carbonaceous fuels has resulted in huge amounts of CO₂ emissions, which brings with it a slew of known environmental implications. Thus, one of the most important scientific challenges today concerns the development of alternatives for satisfying energy demands while protecting the environment.

Within this context, over the last several years, important research progress has been made in the development of solid CO₂ capture materials. Based on the CO₂ capture temperature, these materials can be divided in three main groups: (i) low temperature captors (T < 200 °C), (ii) moderate temperature sorbents (200–400 °C) and (iii) high temperature captors (T > 400 °C).^1–3 Among the high temperature captors studied are various alkaline ceramic materials.^4,5 Some of these binary-metal oxides, containing at least one alkaline element, have been considered potential CO₂ captors, as their chemical basicity renders them able to chemisorb CO₂. Since 1998, the most studied ceramics have included lithium silicates (Li₄SiO₄ and Li₈SiO₆), pentalithium aluminate (Li₅AlO₄), lithium zirconates (Li₂ZrO₃, Li₆Zr₂O₇, and Li₈ZrO₆), lithium cuprate (Li₂CuO₂) and sodium zirconate (Na₂ZrO₃), as well as others.^4–19

However, as previously mentioned, equally important to CO₂ capture are improvements in energy efficiency, conversion processes and alternative energy sources. In this sense, there are several studies where hydrogen is produced as an alternative energy vector using materials that are similar to those previously mentioned for the CO₂ capture process.^20–27 Some of these mechanisms correspond to different reforming reactions^{21–23,25,26} and biomass reactions.²⁴ Of course, most of these reactions produce synthetic gas, also known as syngas, which ideally is composed of H₂ and CO or CO₂. Therefore, hydrogen enrichment depends on the successful separation of H₂ from the carbon oxides present. Here, alkaline thermal treatment has been proposed as a novel process for H₂ enrichment using biomass under moderate thermal conditions.^28,29 Additionally, in steam reforming (SR), the H₂/CO ratio can be varied by manipulating the relative concentrations of CO₂ and O₂ in the feed, which in turn can be achieved by selectively adsorbing the co-generated CO₂ on an appropriate sorbent and/or by carbon monoxide oxidation.³⁰ Noble metal catalysts, especially gold-based catalysts, have shown high activity for CO oxidation.^31,32 However, due to the high prices and scarcity of noble metals, the less costly transition metal oxides have been considered promising catalysts for eliminating the CO from reformed fuels.^33–37 Among them, cobalt oxide catalysts have been extensively studied. Depending on the synthesis method and the reaction conditions employed, Co₃O₄ shows different activity in CO oxidation and oxygen evolution.^33,36,37

Additionally, it has been reported recently that Na₂ZrO₃ and NaCoO₂ seem to have the capacity to act as bifunctional materials, acting as a catalyst during CO oxidation and subsequently as a CO₂ chemisorbent.^38,39 As such, all these materials and processes may be of great interest for use in next-generation H₂ purification technologies. Unfortunately, due to the complexity of the reactions and systems involved, they are poorly understood. The aim of this work was to analyse the bifunctional ability of sodium cobaltate for use as a CO oxidation catalyst and as a CO₂ chemisorbent. Both processes were evaluated dynamically and isothermally at different temperatures, varying the gas mixtures, to determine the catalytic and chemisorption reaction processes, as well as to determine the rate limiting step of the whole process. Additionally, the experiments were theoretically supported by first-principles density functional theory (DFT) thermodynamic calculations.

Experimental and computational methods

Sodium cobaltate (NaCoO₂) was synthesized by a solid–state reaction, and its composition and microstructure have been previously reported.³⁸ Elemental analysis showed that the real chemical composition of the sample was equal to Na_0.89CoO_1.95, although the sample is labelled as NaCoO₂ here.

The CO oxidation reaction was carried out and analysed in a Bel-Rea catalytic reactor (Bel Japan), using NaCoO₂ (200 mg) and a gas mixture of 5 vol% O₂ (Praxair, grade 2.6), 5 vol% CO and N₂ (Praxair, certificate mixture) as balance with a total flow rate of 100 mL min⁻¹. For the dynamic analysis, the sample was heated from 30 to 900 °C at a heath rate of 5 °C min⁻¹. Then, the NaCoO₂ sample was isothermally analysed in the same flux at different temperatures (from 300 to 700 °C). Prior to all the analyses, the sample was cleaned under a 50 mL min⁻¹ N₂ (Praxair, grade 4.8) flow for 15 minutes. The samples were heated to the desired temperature under a N₂ flow. Once the corresponding temperature was reached, the flow gas was switched from N₂ to the mixture mentioned above. The gas products were analysed in a GC 2014 gas chromatograph (Shimadzu) with a Carboxen-1000 column and with an Alpha Platinum FTIR (Bruker) connected to a ZnS gas flow cell.

To analyse the in situ chemisorption of the CO₂ generated by the CO oxidation reaction, different dynamic and isothermal experiments were performed on a Q500HR thermobalance (TA Instruments). Initially, the sample was dynamically heated from 30 to 850 °C at a heath rate of 5 °C min⁻¹ using 60 mL min⁻¹ of CO₂, CO or CO–O₂ (all diluted in N₂). Then, the NaCoO₂ sample was isothermally analysed in the same gas flow at different temperatures (from 300 to 700 °C). For the isothermal analysis, the samples were heated to the desired temperature under a N₂ flow (Praxair, grade 4.8). Once the corresponding temperature was reached, the flow gas was switched from N₂ to the gas mixture.

The structural and microstructural evolution of the products after CO oxidation and subsequent CO₂ chemisorption analysis were studied by XRD using a diffractometer (Siemens, D5000) with a Co-Kα radiation at 34 kV and 30 mA. The phases were identified using the Joint Committee Powder Diffraction Standards (JCPDS). Additionally, the products were analysed by FTIR using Alpha Platinum equipment from Bruker connected to a diamond attenuated total reflectance (ATR) cell. The microstructural evolution was analysed in a JEOL JMS-7600F scanning electron microscope.

To better understand the experimental results, different theoretical calculations on the thermodynamic properties of the relevant reactions were performed. Because the thermodynamic properties of the capture reactions between NaCoO₂ and CO₂ or CO and O₂ are not available in the literature, the ab initio thermodynamic calculations were performed on the following reactions by combining density functional theory (DFT) with lattice phonon dynamics. More detailed descriptions of the calculation method can be found in previous studies.^40,41 The mechanism by which NaCoO₂ captures CO₂ or CO can be expressed with the following three reactions (for convenience, the reaction was normalized to 1 mole of CO₂ or CO):

2NaCoO₂ + CO₂ → Na₂CO₃ + Co₂O₃ (1)

2NaCoO₂ + CO → Na₂CO₃ + 2CoO (2)

2NaCoO₂ + CO + ½O₂ → Na₂CO₃ + Co₂O₃ (3)

In the above reactions, CO₂, CO and O₂ were treated as ideal gases. Therefore, the difference between the Gibbs free energy (ΔG°) of the solid phases of reactants and products may be approximated by the difference in their total energies (ΔE_DFT), obtained directly from DFT calculations and the vibrational free energy of the phonons. Ignoring the PV contribution terms for solids, the variation of the Gibbs free energy (ΔG) for these reactions as a function of temperature and pressure can be written as:

ΔG(T, P) = ΔG⁰(T) − RT ln(P_gas/P₀) (4)

where

ΔG⁰(T) ≈ ΔE_DFT + ΔE_ZP + ΔF_PH(T) − G⁰_gas(T) (5)

ΔE_DFT is the DFT energy difference between the reactants and products of the capture reactions ((1) to (3)); ΔE_ZP is the zero point energy difference between the reactants and products and can be obtained directly from phonon calculations; and ΔF_PH is the phonon free energy change excluding zero-point energy (which is already counted into the ΔE_ZP term) between the reactants and products solids phases. P_gas is the partial pressure of CO₂ or CO in the gas phase (in reaction (3), CO and O₂ are involved, so P_gas = P_CO × P_O2^0.5) 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:

ΔH^cal(T) = ΔG⁰(T) + T(ΔS_PH(T) − S_gas(T)) (6)

where ΔS_PH(T) is the difference in entropies between the solid products and reactants. The CO₂, CO and O₂ free energies (G⁰_gas) can be obtained by way of standard statistical mechanics,^41,42 while their entropies (S_gas) can be found in empirical thermodynamic databases.⁴³ The DFT calculations with plane-wave basis sets and pseudopotential approximations were performed to describe the structural, energetic and electronic properties of the solids considered in this study. All calculations were performed using the Vienna ab initio simulation package (VASP).^44,45 In this study, the PAW pseudo-potentials and PW91 exchange–correlation functional were used in all of the calculations.⁴⁶ Plane wave basis sets were used with a kinetic energy cutoff of 520 eV and an augmentation charge cutoff of 605.4 eV. The k-point sampling grids of m × n × l, obtained using the Monkhorst–Pack method,⁴⁷ are used for these bulk calculations, where m, n, and l are determined with a spacing of approximately 0.028 Å⁻¹ along the reciprocal axes of their unit cells. In the phonon calculations, for each generated supercell, displacements of 0.03 Å of non-equivalent atoms were generated. The solids with hexagonal structures were built in orthorhombic supercells with the option of restoring the symmetry of the crystal space group. Then, for each supercell, the DFT calculations were performed again to obtain the force on each atom due to the displacements. These forces were carried back to PHONON package⁴⁸ to calculate the phonon dispersions and densities from which the partition function could be applied and used to obtain free energies and entropies as shown in eqn (4)–(6).

Results and discussion

The most stable crystal structures of NaCoO₂, Co₂O₃ and CoO are shown in Fig. 1, and their corresponding space groups are R3̄m (no. 166), R3̄cH (no. 167), and Fm3̄m (no. 225), respectively.^49–51 Table 1 summarizes the optimized structure parameters and calculated energetic properties of each crystal structure, which are further used for evaluating the capture reactions as described in the following sections.

Fig. 1 NaCoO₂ (A), Co₂O₃ (B) and CoO (C) crystal structures of solids involved in this study. Red balls stands for O, purple for Na, blue for Co, and c axis is vertical.

Table 1 Comparison of the experimental and the DFT structural parameters and energies for the compounds in the reactions studied. The zero-point energy and entropy calculated from phonon density of states, as well as the available experimental data are also listed

Compound	Space group	Structural parameters		Calculated energy (eV per f.u.)		Entropy (J mol^−1 K)
		Experimental	Calculated	E^DFT	EZP	Phonon (T = 300 K)	Exp.^a (T = 298.15 K)
a Obtained from HSC chemistry package.^52
NaCoO2	R3̄mH (no. 166), Z = 3	a = 2.8897 Å, c = 15.609 Å	a = 2.930265, b = 15.50044	−23.62017	0.27544	63.53
CoO	Fm3̄m (no. 225), Z = 4	a = 4.273 Å	a = 4.232817	−12.97780	0.06990	60.25	52.85
Co2O3	R3̄cH (no. 167), Z = 6	a = 4.782Å, c = 12.96 Å	a = 4.806947, c = 13.03805	−34.19846	0.40625	64.72
Na2CO3	C12/m1 (no. 12)	a = 9.01029, b = 5.23116, c = 6.34548, β = 96.062°	a = 8.95180, b = 5.33507, c = 6.13861, β = 102.21°	−37.29272	0.49152	122.534	138.783
CO2 molecule	P1		a = 20 (fixed)	−22.99409	0.31598		213.388
CO molecule	P1		a = 20 (fixed)	−14.81380	0.13412		197.66
O2 molecule	P1		a = 20 (fixed)	−8.73325	0.09763		205.15

---

Initially, the CO conversion to CO₂ was evaluated dynamically and isothermally in a catalytic reactor; as expected, the degree of CO conversion increases as a function of temperature and mass/flow ratio. In fact, the point of 100% efficiency of CO conversion was shifted from 570 to 490 and 450 °C, using 50, 200 and 250 mg, respectively (data not shown). When using 200 mg of NaCoO₂, the CO conversion to CO₂ was evidenced beginning at 115 °C, and total conversion was obtained between 450 and 900 °C. However, the CO₂ and O₂ concentrations varied between 740 – 900 °C and 820–900 °C, respectively. Both concentrations tended to increase, although CO conversion was complete. At the highest temperatures, the increase in CO₂ concentration may be associated with NaCoO₂ decarbonation, which occurs in this temperature range. Therefore, the CO₂ quantified corresponds to the amount of CO catalytically converted at that specific temperature as well as to the fraction of the produced CO₂ that was previously chemisorbed on the NaCoO₂. On the other hand, the increase in oxygen concentration must be associated with the cobalt oxide reduction process. According to previous investigations, the Co₂O₃ produced during carbonation can be reduced to Co₃O₄ and consecutively to CoO, releasing oxygen.³⁸

The results show that using higher amounts of sodium cobaltate did not reduce the CO conversion temperature significantly. Therefore, all subsequent experiments were performed using 200 mg of NaCoO₂ and the same CO–O₂ mixture and flow rate. Different isothermal experiments were performed, and the products obtained were re-characterized by XRD and ATR-FTIR to confirm the CO oxidation and the consecutive CO₂ chemisorption. Fig. 2 shows the CO conversion isotherms obtained at different temperatures. The initial isothermal experiment was performed at 300 °C. This isotherm exhibited decreasing exponential behaviour, in which equilibrium was not reached after 5 hours, and the conversion efficiency decreased from 46 to 25%. As previously described, NaCoO₂ is able to convert CO to CO₂, but it is also able to trap CO₂ chemically. Thus, NaCoO₂ carbonation may partially inhibit CO conversion due to the formation of a Na₂CO₃–Co₃O₄ external shell (as discussed below). When the isotherms were performed between 350 and 450 °C, the same tendency to decrease was observed, although the CO conversion increased. In fact, at 450 °C, the CO conversion efficiency was approximately 75%. Finally, at higher temperatures (500–700 °C), CO was completely converted throughout the whole isothermal experiment.

Fig. 2 CO conversion isothermal analyses using 200 mg of catalysts at different temperatures.

All the isothermal products were analysed by XRD and ATR-FTIR, as is shown in Fig. 3, to probe the NaCoO₂ carbonation process. The XRD patterns of the products mainly showed the formation of Co₃O₄, the rate of which increased with increasing temperature, while the amount of NaCoO₂ decreased. The formation of this cobalt oxide may be associated with the CO₂ carbonation process. Nevertheless, Na₂CO₃ formation was not totally proven by this technique, it may be due to the difference in cobalt and sodium diffraction scatter factors. Thus, ATR-FTIR spectra of these samples were obtained, and the results of this spectroscopic analysis confirmed the formation of carbonate species (CO₃²⁻), which must correspond to the presence of Na₂CO₃. Therefore, although CO is catalytically converted to CO₂, part of it must be chemically trapped to produce Na₂CO₃ and Co₃O₄ until the sodium cobaltate is saturated.³⁸

Fig. 3 XRD patterns (A) and ATR-FTIR spectra (B) of different CO oxidation isothermal products.

The catalytic activity was evaluated over 24 hours (data not shown) at 650 °C. The isotherm curve obtained showed a small degree of deactivation of 3.5 ± 1% during the first three hours, but, after five hours, the reaction tended to stabilize at ∼99% CO conversion, which suggests a good thermal stability of the NaCoO₂ as a catalyst for the CO oxidation process. This result strongly suggests that initially CO is oxidized by NaCoO₂ and latter it is continued by the different cobalt oxides produced after the carbonation process.

The calculated thermodynamic properties of NaCoO₂ capturing CO₂ or CO are shown in Fig. 4. As can be observed, the mechanisms of CO₂ or CO capture by NaCoO₂ possess Gibbs free energy differences that are not excessively large. This is an interesting result considering that, in the CO reaction, more oxygen atoms have to be taken from the NaCoO₂ structure. However, if O₂ is introduced into the CO capture reaction, it causes a huge variation in the thermodynamic properties of the reaction. This can be explained by the fact that the CO–O₂ oxidation reaction produces large amounts of heat. Fig. 5 shows the calculated vant Hoff plots of the relationships among the free energy (ΔG), temperature (T) and gas pressure (P in logarithmic scale). In this figure, only ΔG = 0 curves are shown. Table 2 summarizes the calculated thermodynamic properties of these reactions and the corresponding turnover T. From Fig. 5 and Table 2, it can be observed that NaCoO₂ would capture CO₂ or CO (especially if O₂ is introduced) up to a very high temperature, but it has to be taken into account that Na₂CO₃ decomposes at 852 °C. For example, when the P_CO2 = 0.1 bar, corresponding to the post-combustion condition, the turnover T is approximately 450 °C. Above this temperature, NaCoO₂ does not capture any additional CO₂, and it starts to release CO₂.

Fig. 4 The calculated thermodynamic properties of reactions of NaCoO₂ capturing CO₂ or CO versus temperature: Heat of reaction (A) and free energy change (B).

Fig. 5 Calculated vant Hoff plots of the relationships among the free energy (ΔG), temperature (T) and gas pressure (P in logarithmic scale). Only ΔG = 0 curves are shown explicitly.

Table 2 Calculated thermodynamic properties of CO₂ and CO captured by NaCoO₂. T₁ and T₂ are the turnover temperatures of the CO₂ capture reactions at P_CO2 = 0.1 bar for post-combustion, P_CO2 = 20 bar for pre-combustion condition. For NaOH, assuming P_H2O = 1 bar^a

Reactions	CO2 or CO (wt%)	ΔE^DFT (kJ mol^−1)	ΔH (kJ mol^−1)	ΔG (kJ mol^−1)	T1 (K)	T2 (K)
a Taken from ref. 43.b hT stands for the temperature out of the range of 1500 K.
2NaCoO2 + CO2 = Na2CO3 + Co2O3	19.31	−121.260	−117.934	−75.432	725	985
2NaCoO2 + CO = Na2CO3 + 2CoO	12.29	−115.222	−112.839	−91.954	hT^b	hT
2NaCoO2 + CO + ½O2 = Na2CO3 + Co2O3	12.29	−489.227	−476.121	−407.672	hT	hT

---

To further analyse and evidence the in situ production and chemisorption of CO₂, different thermogravimetric analyses were performed. Fig. 6A shows the dynamic thermograms of NaCoO₂ using three different gases: CO₂, CO and CO–O₂, all of which were diluted in N₂, with a P_CO or P_CO2 of 0.05. When the NaCoO₂ sample was thermally heated in flowing CO₂, it exhibited a small weight decrease (∼2 wt%) between 30 and 85 °C, which was associated with a superficial dehydration process. Then, between 120 and 420 °C, the sample presented practically no superficial CO₂ chemisorption, a behaviour typically observed in alkaline ceramics. This observation must also be related to the P_CO2 value of 0.05 used, which modifies the sorption–desorption dynamic equilibrium. On the other hand, the bulk CO₂ chemisorption was observed between 450 and 745 °C. After the weight increase, a weight loss due to the sample decarbonation and decomposition processes was observed. Likewise, it had been already shown that a NaCoO₂–CO–O₂ system³⁸ presented a similar trend to that observed with CO₂, confirming the occurrence of CO oxidation and subsequent CO₂ chemisorption. In this case, the superficial and bulk chemisorption processes were observed between 280 and 440 °C and between 440 and 805 °C, respectively. It must be noted that both processes are shifted to higher temperatures in comparison with the CO₂ case, which may be related to a different reaction mechanism because, in this case, a double process is taking place (that is, both CO oxidation and the subsequent CO₂ chemisorption are taking place).

Fig. 6 Dynamic thermograms of NaCoO₂ in CO₂ (5 vol% diluted in N₂), CO (5 vol% diluted in N₂) and CO–O₂ (5 and 5 vol% diluted in N₂) flows (left) and NaCoO₂–CO gas evolution into the catalytic reactor (right).

Conversely, the NaCoO₂ dynamic thermogram obtained merely with CO and in the absence of O₂ showed a different behaviour. The NaCoO₂–CO thermogram did not show any significant increase in weight, as in the previous cases. This thermogram initially showed a weight decrease related to dehydration (≤100 °C) and dehydroxylation (350–400 °C) processes. Then, between 500 and 630 °C, the thermogram showed a slight weight increase of approximately 0.5 wt%. In the same temperature range, the NaCoO₂–CO₂ and NaCoO₂–CO–O₂ systems chemisorbed CO₂, which suggests that NaCoO₂ is partially trapping CO₂. Fig. 6B supports these results, showing the catalytic dynamic evolutions of CO consumption and CO₂ formation in comparison to the thermogravimetric analysis and using NaCoO₂ in the absence of oxygen. Between room temperature and 430 °C, the CO concentration did not vary, and CO₂ was not detected, indicating negligible CO conversion. From this temperature to 630 °C, the presence of CO₂ was observed, which is in agreement with the decrease in CO present. Thus, CO is being partially converted to CO₂ even in the absence of oxygen, which must be released from the NaCoO₂ structure. This behaviour has been reported for other transition metal oxides where the CO adsorbed on the metal sites is oxidised by the lattice oxygen atoms.^33,53,54 This result is also in accord with the thermogravimetric curve, which shows a partial increase in weight. Therefore, the degree of catalytic oxidation of CO and the subsequent chemisorption of CO₂ by NaCoO₂ is significantly reduced in the absence of oxygen. It can be assumed from this that the CO oxidation process is a conditioning step for the CO oxidation and CO₂ chemisorption bifunctionality.

Different isothermal experiments were performed to further analyse the chemisorption functionality of NaCoO₂. Fig. 7 shows the NaCoO₂ isotherms obtained at different temperatures in a CO–O₂ flux. The initial isothermal experiment was performed at 300 °C and exhibited an exponential behaviour, failing to reach equilibrium after 3 hours and exhibiting a weight increase of only 1.9 wt%. Subsequent isotherms were performed between 400 and 700 °C. All of these isotherms exhibited a similar exponential behaviour where the weight increased as a function of temperature from 4.8 wt% at 400 °C to a maximum of 9.8 wt% at 700 °C. Expressed another way, the CO₂ capture efficiency increased from 10.7 to 55.4% by increasing the temperature from 300 to 700 °C. It is important to note here that the catalytic results indicate CO is totally oxidized at T ≥ 500 °C.

Fig. 7 CO–O₂ thermogravimetric isothermal analyses of NaCoO₂ at different temperatures.

When the NaCoO₂–CO–O₂ isotherms are compared with the NaCoO₂–CO₂ isotherms,³⁸ some interesting features are noted (Fig. 8). At short times, CO₂ chemisorption is faster in the NaCoO₂–CO₂ system than in the NaCoO₂–CO–O₂ system. However, at T ≤ 600 °C, the final weight increase of the NaCoO₂–CO–O₂ system is higher than that of the NaCoO₂–CO₂ system. The behaviour presented at short times may be related to a gas concentration effect because the rate of CO₂ production by CO oxidation is as high as 3 mL min⁻¹ and solid–gas interactions are favoured in the NaCoO₂–CO₂ system, resulting in a higher weight increase. On the other hand, the final weight gains may be explained as follows: for T ≤ 400 °C, the carbonate shell is formed faster in the NaCoO₂–CO₂ system, partially inhibiting the chemisorption process on NaCoO₂. At T ≥ 600 °C, however, the chemisorption process in both systems is mainly dependent on intercrystalline diffusion processes. Hence, the final weight gain is similar for the isotherms of both systems.

Fig. 8 Isotherms comparison of the NaCoO₂–CO–O₂ (blue) and NaCoO₂–CO₂ (black) systems at different temperatures (A), and inset of the same isothermal curves at the first times (B). The NaCoO₂–CO₂ isotherms were taken from ref. 38.

To analyse the microstructural evolution of NaCoO₂, SEM images of the material were obtained (Fig. 9). The pristine sample analysed was comprised of polyhedral flake-like particles, as described previously.³⁸ The CO oxidation products, obtained at 400 and 700 °C, possessed the same flake-like microstructure, with particle sizes ranging between 1 and 6 μm. For the 400 °C products, the particles formed non-dense agglomerates, and, in the case of 700 °C, the agglomerates were slightly denser than in the pristine sample and were approximately 20–60 μm in size. Additionally, the backscattered electron images of both products showed the presence of two different phases, which correspond to Na₂CO₃ (dark phase) and Co₃O₄ (light phase), as previously identified by XRD. At 400 °C, the Na₂CO₃ and Co₃O₄ phases seem to be homogeneously mixed, whereas, in the sample treated at 700 °C, these phases tended to segregate. This segregation may enhance the ability of the CO oxidation process to continue in the cobalt oxide phase even when the carbonation process is completed.

Fig. 9 Scanning electron images of the NaCoO₂ pristine sample (A) and the isothermal CO oxidation products at 400 (B) and 700 °C (C).

Summarizing, NaCoO₂ is able to be used into a bifunctional application; where NaCoO₂ acts as catalyst and captor through CO oxidation and subsequent CO₂ chemisorption processes. This bifunctional mechanism is represented in the following reaction system:

where the cobalt oxide composition changes as a function of temperature, varying the cobalt oxidation state and consequently producing oxygen.

Conclusions

Sodium cobaltate was analysed as a possible bifunctional material for CO oxidation and subsequent chemisorption of the CO₂ produced. Initially, catalytic CO oxidation to CO₂ was analysed using gas chromatography in-line with the reactor. The results obtained confirmed that CO can be oxidized by NaCoO₂ at different temperatures depending on the mass/flow ratio. On the other hand, the in situ production and chemisorption of CO₂ was analysed using thermogravimetric analysis and was found to follow a similar reaction mechanism to those already reported for alkaline ceramics.

The catalytic and gravimetric results, as well as analysis of the NaCoO₂ products, indicated that NaCoO₂ is able to catalyse CO to CO₂ and subsequently chemisorb the latter. Of course, the chemisorption of CO₂ changed the NaCoO₂ crystalline structure and its composition, transforming it to Na₂CO₃ and different cobalt oxides. Nevertheless, the catalytic effect was not reduced within a certain temperature range, even when an experimental catalytic evaluation was performed during 24 hours. Therefore, CO oxidation occurs on NaCoO₂ initially, but, once Na₂CO₃ is produced, the catalytic reaction must be performed on the cobalt oxides, and it is not limited by the Na₂CO₃ formation.

Finally, all of these results were corroborated with theoretical thermodynamic data calculated for the NaCoO₂–CO, NaCoO₂–CO–O₂ and NaCoO₂–CO₂ reaction systems, for which different oxidation conditions were modelled. The ΔH and ΔG values clearly showed the different thermal stabilities for each reaction process depending on temperature. Although CO₂ and CO capture by NaCoO₂ have similar ΔH and ΔG values, in the CO reaction, interestingly, more oxygen atoms have to be taken from the NaCoO₂ structure. However, when O₂ was added into the CO capture reaction, the ΔH and ΔG thermodynamic values changed significantly. In fact, the CO–O₂ oxidation reaction produces large amounts of heat, which may explain the ΔH and ΔG variations.

Acknowledgements

This work was financially supported by the projects SENER-CONACYT and PAPIIT-UNAM. E. Vera thanks to CONACYT for financial support. The authors thank to O. Novelo, J. Romero and A. Tejeda for technical help.

References

1. S. D. Kenarsari, D. Yang, G. Jiang, S. Zhang, J. Wang, A. G. Russell, Q. Wei and M. Fan, RSC Adv., 2013, 3, 22739–22773.
2. X. Lu, D. Jin, S. Wei, Z. Wang, C. An and W. Guo, J. Mater. Chem. A, 2015, 3, 12118–12132.
3. J. Wang, L. Huang, R. Yang, Z. Zhang, J. Wu, Y. Gao, Q. Wang, D. O'Hare and Z. Zhong, Energy Environ. Sci., 2014, 7, 3478–3518.
4. L. K. G. Bhatta, S. Subramanyam, M. D. Chengala, S. Olivera and K. Venkatesh, J. Cleaner Prod., 2015, 103, 171–196.
5. S. Kumar and S. K. Saxena, Materials for Renewable and Sustainable Energy, 2014, 3, 1–15.
6. M. B. I. Chowdhury, M. R. Quddus and H. I. deLasa, Chem. Eng. J., 2013, 232, 139–148.
7. J. Ortiz-Landeros, T. L. Ávalos-Rendón, C. Gómez-Yáñez and H. Pfeiffer, J. Therm. Anal. Calorim., 2012, 108, 647–655.
8. M. R. Quddus, M. B. I. Chowdhury and H. I. deLasa, Chem. Eng. J., 2015, 15, 347–356.
9. S. Wang, C. An and Q. H. Zhang, J. Mater. Chem. A, 2013, 1, 3540–3550.
10. S. Y. Shan, Q. M. Jia, L. H. Jiang, Q. C. Li, Y. M. Wang and J. H. Peng, Ceram. Int., 2013, 39, 5437–5441.
11. Z. Qi, H. Daying, L. Yang, Y. Qian and Z. Zibin, AIChE J., 2013, 59, 901–911.
12. K. Wang, P. Zhao, X. Guo, Y. Li, D. Han and Y. Chao, Energy Convers. Manage., 2014, 81, 447–454.
13. K. Oh-ishi, Y. Matsukura, T. Okumura, Y. Matsunaga and R. Kobayashi, J. Solid State Chem., 2014, 211, 162–169.
14. S. M. Amorim, M. D. Domenico, T. L. P. Dantas, H. J. José and R. F. P. M. Moreira, Chem. Eng. J., 2016, 283, 388–396.
15. L. Martinez-dlCruz and H. Pfeiffer, J. Phys. Chem. C, 2012, 116, 9675–9680.
16. H. A. Lara-García, B. C. Alcántar-Vázquez, Y. Duan and H. Pfeiffer, RSC Adv., 2015, 5, 34157–34165.
17. C. Wang, B. Dou, Y. Song, H. Chen, Y. Xu and B. Xie, Ind. Eng. Chem. Res., 2014, 53, 12744–12752.
18. A. Sanna, I. Ramli and M. M. Maroto-Valer, Energy Procedia, 2014, 63, 739–744.
19. F. Fujishiro, Y. Kojima and T. Hashimoto, J. Am. Ceram. Soc., 2012, 95, 3634–3637.
20. R. Raskar, V. Rane and A. Gaikwad, Water, Air, Soil Pollut., 2013, 224, 1569–1584.
21. D. Y. Aceves-Olivas, M. R. Baray-Guerrero, M. A. Escobedo-Bretado, M. M. Silva-Paula, J. Salinas-Gutiérrez, V. Guzmán-Velderrain, A. López Ortiz and V. Collins-Martínez, Int. J. Hydrogen Energy, 2014, 39, 16595–16607.
22. M. H. Halabi, M. H. J. M. de Croon, J. van der Schaaf, P. D. Cobden and J. C. Schouten, Chem. Eng. J., 2011, 168, 872–882.
23. M. H. Halabi, M. H. J. M. de Croon, J. van der Schaaf, P. D. Cobden and J. C. Schouten, Chem. Eng. J., 2011, 168, 883–888.
24. K. Wang, P. Zhao, X. Guo, D. Han and Y. Chao, Environ. Prog. Sustainable Energy, 2015, 34, 526–532.
25. H. K. Rusten, E. Ochoa-Fernández, H. Lindborg, D. Chen and H. A. Jakobsen, Ind. Eng. Chem. Res., 2007, 46, 8729–8737.
26. E. Ochoa-Fernández, H. K. Rusten, H. A. Jakobsen, M. Rønning, A. Holmen and D. Chen, Catal. Today, 2005, 106, 41–46.
27. B. Tsuchiya, K. Morita and S. Nagata, Surf. Interface Anal., 2012, 44, 717–720.
28. M. R. Stonor, T. E. Ferguson, J. G. Chen and A.-H. A. Park, Energy Environ. Sci., 2015, 8, 1702–1706.
29. S. N. Reddy, S. Nanda, A. K. Dalai and J. A. Kozinski, Int. J. Hydrogen Energy, 2014, 39, 6912–6926.
30. Z. Guo, B. Liu, Q. Zhang, W. Deng, Y. Wang and Y. Yang, Chem. Soc. Rev., 2014, 43, 3480–3524.
31. A. Sandoval, C. Louis and R. Zanella, Appl. Catal., B, 2013, 140–141, 363–377.
32. P. Rodriguez, D. Plana, D. J. Fermin and M. T. M. Koper, J. Catal., 2014, 311, 182–189.
33. X. Xie, Y. Li, Z. Q. Liu, M. Haruta and W. Shen, Nature, 2009, 458, 746–749.
34. M. Y. Guo, F. Liu, J. Tsui, A. A. Voskanyan, A. M. C. Ng, W. K. Chan, K. Y. Chan, C. Liao, K. Shih and C. Surya, J. Mater. Chem. A, 2015, 3, 3627–3632.
35. J. Q. Lu, C. X. Sun, N. Li, A. P. Jia and M. F. Luo, Appl. Surf. Sci., 2013, 287, 124–134.
36. B. Han, D. Qian, M. Risch, H. Chen, M. Chi, Y. S. Meng and Y. Shao-Horn, J. Phys. Chem. Lett., 2015, 6, 1357–1362.
37. M. Zhou, L. Cai, M. Bajdich, M. García-Melchor, H. Li, J. He, J. Wilcox, W. Wu, A. Vojvodic and X. Zheng, ACS Catal., 2015, 5, 4485–4491.
38. E. Vera, B. Alcántar-Vazquez and H. Pfeiffer, Chem. Eng. J., 2015, 271, 106–113.
39. B. Alcántar-Vazquez, E. Vera, F. Buitrón-Cabrera, H. A. Lara-García and H. Pfeiffer, Chem. Lett., 2015, 44, 480–482.
40. Y. Duan, H. Pfeiffer, B. Y. Li, I. C. Romero-Ibarra, D. C. Sorescu, D. R. Luebke and J. W. Halley, Phys. Chem. Chem. Phys., 2013, 15, 13538–13558.
41. Y. Duan and D. C. Sorescu, J. Chem. Phys., 2010, 133, 074508.
42. B. Zhang, Y. Duan and J. K. Johnson, J. Chem. Phys., 2012, 136, 064516.
43. M. W. J. Chase, NIST-JANAF Themochemical Tables, J. Phys. Chem. Ref. Data, Monograph, 4th edn, 1998, vol. 9, p. 1.
44. G. Kresse and J. Hafner, Phys. Rev. B: Condens. Matter Mater. Phys., 1993, 47, 558–561.
45. G. Kresse and J. Furthmuller, Comput. Mater. Sci., 1996, 6, 15–50.
46. J. P. Perdew and Y. Wang, Phys. Rev. B: Condens. Matter Mater. Phys., 1992, 45, 13244–13249.
47. H. J. Monkhorst and J. D. Pack, Phys. Rev. B: Solid State, 1976, 13, 5188–5192.
48. K. Parlinski, Software PHONON, Computing for Materials, Krakow, Poland, 2006.
49. J. Chenavas, J. C. Joubert and M. Marezio, Solid State Commun., 1971, 9, 1057–1060.
50. J. F. Liu, S. Yin, H. P. Wu, Y. W. Zeng, X. R. Hu, Y. W. Wang, G. L. Lv and J. Z. Jiang, J. Phys. Chem. B, 2006, 110, 21588–21592.
51. Y. Takahashi, Y. Gotoh and J. Akimoto, J. Solid State Chem., 2003, 172, 22–26.
52. HSC Chemistry, version 6.1, Outotec Research Oy, Pori, Finland, 2006, http://www.outotec.com/hsc.
53. F. Mariño, C. Descorme and D. Duprez, Appl. Catal., B, 2005, 58, 175–183.
54. P. Mars and D. van Krevelen, Chem. Eng. Sci., 1954, 3, 41–59.

---