https://doi.org/10.1016/j.fuel.2022.123842
The following content was copied from the above open access paper. It is noted that using this reported LaNiO3-derived DFMs, the capacity of CO2 capture is quite low (<1 mmol/g). Thus the yield of CH4 is also very low. In addition, CO formation was observed during the CO2 capture stage. This could be an issue if flue gas is the CO2 source, producing new pollutant into flue gas.
“Aiming to introduce the basic principles of the operation, Fig. 6 displays the evolution with time of the outlet concentration of CO2, CH4, CO and H2O during an entire CO2 adsorption and hydrogenation cycle at 400 °C. Although these results correspond to the DFM derived from 30% LaNiO3/CeO2 formulation, the global reaction evolution is similar for alumina-supported samples and bulk perovskite.”
“Fig. 6. CO2, CH4, H2O and CO concentration profiles during a complete CO2 adsorption and hydrogenation to CH4 cycle at 400 °C for the DFM obtained after the controlled reduction of the 30% LaNiO3/CeO2 precursor.”
“During the adsorption cycle (1 min) a gas stream composed of 1.4% CO2/Ar is fed. To estimate the amount of CO2 adsorbed on the catalyst, the CO2 concentration profile when the reactor is bypassed is also included. As can be observed, CO2 concentration is almost negligible at the beginning of the adsorption period; in fact, no CO2 signal at the reactor outlet is detected during the first 35 s. Following, it increases rapidly achieving almost the inlet concentration at the end of the storage period. This trend reveals the progressive CO2 adsorption on storage sites, mainly La2O3 phase [47] and, in minor extent, on NiO–CeO2 interface, up to their total saturation through the following reaction:
(12)La2O3+ CO2⇄La2O2CO3
Few seconds delayed, an increasing H2O signal is detected at the reactor outlet. The identification of this compound during the adsorption period reveals that CO2 is progressively displacing pre-adsorbed H2O due to its competitive adsorption on La2O3 storage sites through Eq. (13). However, it can be concluded that the CO2 adsorption preferentially occurs onto free La2O3 sites, since H2O is detected quite delayed with respect to CO2 identification. Once La2O3 adsorption sites are completely carbonated (Eq. (12)), the storage of CO2 is transferred to La(OH)3 sites (Eq. (13)).
(13)2LaOH3+ CO2⇄La2O2CO3+ 3H2O
Note that the desorption of a small fraction of H2O stored on ceria or alumina supports in form of hydroxyls cannot be ruled out, which can conform bicarbonates during CO2 adsorption period. However, it is well-known that their stability is limited at working temperatures, which makes this adsorption route minority with respect to that expressed by Eqs. (12–13), especially for the ceria-supported sample [38].
From these data, the amount of CO2 adsorbed onto the catalyst is calculated by Eq. (2) (Table 4). In order to assess the stable behavior of the DFM, the corresponding values to 3 consecutive cycles is included, which results in values between 86.4 and 90.7 µmol CO2 g−1. Furthermore, an almost negligible CO peak is observed during the adsorption period, which value is determined by Eq. (4) and summarized in Table 4 (around 9 µmol g−1), which is related to the incomplete hydrogenation of adsorbed CO2 with H2 chemisorbed on the Ni0 sites during the previous hydrogenation period following the reverse water gas shift reaction (RWGS, Eq. (14)). Alternatively, other authors related CO formation to the progressive decomposition of adsorbed formate species [48].
(14) H2+ CO2⇄CO + H2O “
“Once the adsorption period is completed, the CO2 is removed from the feed stream and a constant Ar flow rate is fed during 2 min, in order to purge the catalyst as well as the reaction system. As a result, CO2 and H2O signals progressively decrease practically to zero during this period.
Then, the hydrogenation period (2 min) begins with the admission of a gas stream composed of 10% H2/Ar. Immediately after the injection of 10% H2/Ar mixture, a sudden CH4 production is observed with a long tail extended during the rest of the period. Besides, H2O formation is detected around 10 s delayed from CH4 detection. This process can be described by the following reaction scheme:(15)Step1:La2O2CO3⇄La2O3+ CO2(1)Step2:4H2+ CO2⇄CH4+ 2H2O(16)Step3a:La2O3+ 3H2O⇄2LaOH3(17)Step3b:Ce2O3+ 3H2O⇄2CeOH3
Firstly, lanthanum oxide carbonate is decomposed to form gaseous CO2 (Eq. (15)). Then, the CO2 released reacts with hydrogen to form methane and water following Sabatier reaction (Eq. (1)). Taking into account the stoichiometry of Eq. (1), 2 mol of H2O should be detected per mol of CH4; nevertheless, the experimental ratio during hydrogenation period ranges between 1.38 and 1.40, which reveals that part of H2O is stored on the surface La2O3 sites (Eq. (16)) or ceria support (Eq. (17)). As we already reported in previous work for conventional Ni/CeO2 catalysts [49], the water adsorption on ceria sites is limited due to its high oxygen mobility, which favours water desorption during the hydrogenation period. Indeed, the H2O/CH4 ratio is significantly higher than that observed for conventional Ru-CaO/Al2O3 and Ru-Na2CO3/Al2O3 DFMs (H2O/CH4 < 1.14) [13]. Finally, a small fraction of CO (around 1 µmol g−1) is also detected during the hydrogenation period due to RWGS reaction (Eq. (14)).
If the entire CO2 adsorption and hydrogenation cycle is considered, a H2O/CH4 ratio ranging between 2.00 and 2.03 is obtained, that is close to the stoichiometry value (H2O/CH4 = 2) defined by Sabatier reaction (Eq. (1)). With the aim of giving more reliability to the results obtained, the carbon balance was also determined (Eq. (7)). As can be observed in Table 1, the amount of CO2 stored during the adsorption period is around 88 µmol g−1, whereas around 80 µmol g−1 of CH4 and 8 µmol g−1 of CO are produced during CO2 hydrogenation period. Thus, carbon balance closed within ± 5% since CH4 and CO are the only products detected by FTIR during the reaction.”