CO₂ Absorption of Powdered Ba₂Fe₂O₅ with Different Particle Size

Sample preparation

The XRD result of Ba₂Fe₂O₅ is presented in Figure 1, from which it can be seen that the pure Ba₂Fe₂O₅ can be prepared by BaCO₃ and Fe₂O₃ at the temperature of 1200 ℃ using solid-state method. And the reaction kinetics of Fe₂O₃ and BaCO₃ to prepare Ba₂Fe₂O₅ was studied in our previous paper [14]. SEM photograph of the Ba₂Fe₂O₅ sample is shown in Figure 2(a), from which it can be seen that the product is built by unregularly polyhedron-shaped and dense particles with sizes ranging within 1.0–3.0 μm. For the CO₂ absorption measurement, three kinds of powdered Ba₂Fe₂O₅ samples were prepared from the Ba₂Fe₂O₅ samples by the procedure mentioned under the experimental part. In order to measure average particle size of the three kinds of powered samples, particle size distribution measurement was performed on the powdered samples. Accumulative distribution and measured frequency distribution of the powdered samples are shown in Figure 3. According to the figure, it could be recognized that the frequency distributions of the samples look almost symmetry, and median particle sizes of the samples were measured that 52.5 μm, 77.5 μm and 100.0 μm, respectively. Because of symmetry of the frequency distribution, the median particle size would be equal to average particle size.

Figure 1:

X-ray diffraction patterns of Ba₂Fe₂O₅ and reaction products of Ba₂Fe₂O_5.

Figure 2:

SEM photographs of the Ba₂Fe₂O₅ sample and reaction products of Ba₂Fe₂O_5.

Figure 3:

Accumulative and frequency distributions of the powdered samples, (a) 52.5 μm, (b) 77.5 μm (c) 100.0 μm.

Figure 1 also showed the XRD of reaction products of CO₂ and Ba₂Fe₂O₅ that median particle size was 100.0 μm. From the figure, it can be seen that only BaCO₃ and BaFe₂O₄ appeared at the end of reaction. And the SEM of reaction products is shown in Figure 2(b). It was found that secondary particles with diameters of 3–4 μm, which consisted of BaCO₃ and BaFe₂O₄, were observed on large raw particles, and they were dense and homogeneous.

Rate analysis for the CO₂ absorption

Weight change of the powdered Ba₂Fe₂O₅ samples was measured with reaction time t at a target reaction temperature T by the thermogravimetric instrument. As a typical example of the estimation, reaction extent (Δmt/Δmo, Δmt is the quality increase at time t, Δm0 is the theory maximum quality increase) was plotted to the reaction time t at 1000 ℃ for the samples with 52.5 μm, 77.5 μm and 100.0 μm of the particle sizes in Figure 4. From the figure, it is found that, first, reaction extent linearly increases with a constant slope in the first 180 s and the slopes of the three curves are almost equal, then from 180 s, the slopes of the three curves began to appear different. So it can be concluded that the reaction rates begin to differ after 180 s, and the reaction extent of the powdered samples with smaller particle size rises in shorter reaction time, that is, the CO₂ absorption reaction seems to progress in shorter reaction time. But the slopes decrease with the increase of reaction time. The difference of slopes means that progression of the CO₂ absorption reaction seems to be disturbed by some kind of factor. The longer the reaction time of Ba₂Fe₂O₅ and CO₂ gas is, the thicker the products (BaCO₃ and BaFe₂O₄) phases formed on Ba₂Fe₂O₅ particle are, which will disturb the progression of the CO₂ absorption reaction. Therefore, the reaction is more and more difficult to proceed with the CO₂ absorption reaction proceed gradually. And a tendency is found that the slops decrease with the increasing particle size.

Figure 4:

CO₂ adsorption curves under various particle sizes at 1000 ℃.

Chou model

The reaction between CO₂ and Ba₂Fe₂O₅ firstly begins from the surface of the Ba₂Fe₂O₅ particle, and the layer of BaCO₃ and BaFe₂O₄ becomes thicker and thicker. Therefore, the rate of the whole reaction will be controlled by the diffusion in the product layer.

Chou et al. [15, 16, 17] have succeeded in establishing a RPP (Real Physical Picture) model to describe the kinetics of reaction controlled by the diffusion process in the solid-product layer, as shown in eq. (2), which is a form of explicit function, so the RPP model is easy to use and enable the easy performance of theoretical analysis.

and

where Ro is the radius of the original whole particle, BRois a function of T, PCO2and PCO2eq, in which PCO2eqis the carbon dioxide partial pressure in equilibrium with oxide and should be related to the temperature; KCO2oβand DCO2o are constant independent of temperature; BRo will be constant as the temperature and the carbon dioxide partial pressure are fixed. Equation (2) is the formula representing the effect of particle size on the reacted fraction from which it may be seen that the smaller the particle size, the larger the reacted fraction. It is because the particle with a small radius has a larger surface and thus a faster reaction rate. Note that particle size has the most effect on the reacted fraction in comparison with the other two factors, temperature and oxygen partial pressure.

The reaction kinetics of CO₂ and Ba₂Fe₂O₅ in isothermal can be described by eq. (2). Figure 5 shows the fitting results, and it can be seen that the experimental data and theoretical predicted values are in good agreement which validates the present theoretical approach. The model describing the effects of particle size and time on the reacted fraction is given as follows:

Figure 5:

A comparison between experimental data and theoretical predictions from the RPP model.

Multicycle property of Ba₂Fe₂O₅

The multicycle performance of Ba₂Fe₂O₅ in the conditions of adsorption in CO₂ gas (1000 ℃) and desorption (1200 ℃) in N₂ gas is shown in Figure 6. It can be seen from Figure 6, Ba₂Fe₂O₅ exhibits a good recycling performance though the adsorption capacity is reduced slightly during the cycle processes.

Figure 6:

The multicycle performance of Ba₂Fe₂O₅ in the conditions of adsorption in CO₂ gas and desorption in N₂ gas.

The produced powders of different stages of reaction were characterized SEM analyses by Fujishiro [18]. The morphologies of the specimens at various reaction stages were observed to evaluate the CO₂ absorption mechanism. After heating for different temperatures, three types of specimens with reaction ratio of 0.3, 0.5, and 0.9 were obtained by Fujishiro. It can be seen that the as-prepared Ba₂Fe₂O₅ was composed of particles with diameters of 50–70 μm. As the reaction proceeded, secondary particles with diameters of 5–7 μm, which consisted of BaCO₃ and BaFe₂O₄, were observed on large raw particles. Figure 7(a) and 7(b) show the sample SEM images after 1 and 5 times adsorption–desorption cycles, respectively. Compared with the SEM images of adsorbed samples, it can be noticed that the morphologies of the samples after adsorption-desorption test are quite different, and it can be observed that the particles are flawed and loosed, which due to the release of CO₂ in desorption stage. Based on the above analysis, the pathway of the CO₂ adsorption–desorption on Ba₂Fe₂O₅ can be obtained and shown in Figure 8. At the initial stage of adsorption, the BaCO₃ and BaFe₂O₄ external shell are formed on the surface of Ba₂Fe₂O₅, and as the reaction proceeded, the unreacted center Ba₂Fe₂O₅ is smaller and smaller until completely react. At the desorption stage, BaCO₃ reacts with BaFe₂O₄ to generate Ba₂Fe₂O₅ and CO₂, which makes Ba₂Fe₂O₅ flawed and loosed. The multicycle processes of adsorption–desorption take place between the flawed and loosed Ba₂Fe₂O₅ and BaCO₃, BaFe₂O₄.

Figure 7:

The sample SEM images of 1 (a) and 4 (b) times adsorption-desorption cycle.

Figure 8:

The CO₂ adsorption-desorption pathway of Ba₂Fe₂O_5.