https://doi.org/10.1016/j.ccst.2022.100096
”
A single batch of Li-Ru/A DFM was used to perform a parametric study of the integrated CO2 capture and methanation process aiming to elucidate the individual effects of O2 and H2O when added to a feed gas containing 5% CO2 in N2. Fig. 3 presents the typical transient CO2, CH4, and CO concentration traces as well as the temperature profiles recorded at the exit of the DFM bed during standard cycles run at a fixed preheating (280 °C). As soon as admitted to the reactor, CO2 was quickly captured by the DFM (Fig. 3a) so that its concentration dropped to zero after ca 25 s independently from the presence of O2 or H2O in the feed stream; thereafter, it started to raise progressively until the overall capacity was mostly saturated (within 4 min). The contribution from the reactor hold-up can be visualized by the dashed line in Fig. 3a. Some weakly bonded CO2 was spontaneously desorbed from the DFM during the intermediate purge phase under N2 flow (required to avoid gas mixing) before H2 was admitted to the reactor to start the hydrogenation phase. At that point, CH4 formation occurred with an apparent initial rate that was not affected by the eventual presence of O2 or H2O during the previous stage. The peak production of CH4 (up to 3.5% by volume) was achieved within 25s, being slightly lower when the simulated feed gas contained some H2O, due to the lower amount of CO2 stored on the DFM. Simultaneously, a very low amount of CO was formed (≤40ppm, Fig. 3c), and it became almost undetectable for the humid flue gas case. A limited thermal desorption of CO2 was observed at the beginning of the hydrogenation phase (Fig. 3a), driven by the heat released by the exotherm of the catalytic reaction, which indeed caused a temperature increase recorded at the exit of the DFM bed (Fig. 3d). The maximum and minimum values of ΔTexit were measured when the flue gas contained O2 or H2O, respectively.
Fig. 3. Integrated CO2 capture and methanation cycles on Li-Ru/A DFM operating at 280 °C with 3 different feed gas compositions during adsorption: 5% CO2 in N2 (black lines), with the addition of +0.25% O2 (red lines) or +1.5% H2O (green lines). Temporal outlet concentrations of CO2 (a) CH4 (b) and CO (c), and temperature profiles as measured at the exit of the catalytic bed (d). Step duration: 18 min CO2 capture, 2 min N2 purge, 14 min hydrogenation (15% H2 in N2). Dashed line in (a) represents the reactor hold-up.
Interestingly, also the CO2 capture process caused similar temperature increases due to the exothermic nature of the surface reactions involved, which increased in the presence of O2 in the flue gas (Fig. 3d). In particular, at temperatures exceeding 200 °C Ru can be easily oxidized during adsorption and then reduced back to its metal form during the methanation step: both the oxidation of Ru by O2 and the reduction of RuOx by H2 are exothermic reactions (Porta et al., 2021) which therefore contribute to enhancing the heat release during each half cycle (Fig. 3d). It can be argued that the Ru content in the DFM (as well as any other reducible metal oxide such as NiOx or CeOx) should be kept as low as possible to limit the extent of the parasitic H2 consumption to form water (Abdallah and Farrauto, 2022).
The parametric study was further extended to investigate the effect of the average reaction temperature in the 260–320 °C range. A minimum of three adsorption-reaction cycles were run for each specific test condition giving highly repeatable results: the average values are summarized in Fig. 4 reporting the CH4, CO and CO2 released during the methanation, the resulting CO2 conversion, and the CO2 lost during the intermediate purge. All relevant data, including the C-balance, selectivity to CH4 and maximum temperature gradients recorded during adsorption and reaction half-cycles are also listed in Table S1.
Fig. 4. Effect of the feed gas composition and the reaction temperature on the average values of CH4 and CO produced, CO2 desorbed and purged (bars), CO2 conversion (dots) during 3 consecutive cycles of integrated isothermal CO2 capture and methanation with Li-RuA DFM. Adsorption conditions: 5% CO2 in N2 (reference case) with the eventual addition of 1.5% H2O, 0.25% O2, or both. Experimental details can be found in Section 2.6.
The Li-RuA DFM displayed a cyclic CO2 capture capacity of ca 350 μmol/g under reference conditions (CO2 in dry N2) that was poorly affected by the operating temperature in the range explored herein. The presence of water vapour in the feed gas reduced it by ca 16–19% with respect to dry conditions, whereas the addition of O2 showed a marginally positive effect (ca 2–4%). The amount of CO2 released during the purge phase was almost constant, but the CO2 desorbed during the methanation increased at higher temperatures as well as in the presence of O2, and decreased under humid conditions. Accordingly, the CO2 conversion during methanation showed a decreasing trend with temperature passing from 90.6% at 260 °C to 85.3% at 300 °C under reference conditions (Fig. 4). Moreover, it was higher with humid rather than with O2-containing feed (Table S1). CO production was always very small but it became negligible under humid conditions and for temperatures up to 300 °C (Table S1). Water retained on the surface of the DFM during adsorption can inhibit the formation of CO upon the injection of H2 by shifting the water gas shift equilibrium reaction (H2 +CO2 ↔ CO +H2O).
Eventually, when O2 and H2O were simultaneously present in the simulated flue gas, their individual thermal and chemical effects partially compensated, so that, for example, the associated penalty in the CO2 capture capacity was favourably limited to only ca 5% with respect to the reference case. This resulted in a maximum methane production of 247±3 μmol/gDFM achieved at as low as 280 °C, with an outstanding 99.98±0.01% selectivity and a CO2 conversion exceeding 90%; similar performances were also achieved at 260 °C (Table S1). While further improvements of the specific methane production are possible by increasing the Li loading in the DFM (at least up to 5% wt, (Cimino et al., 2022)), this was deliberately kept low to better highlight durability and poisoning issues during reaction ageing with SO2-loaded flue gas (see Section 3.2.4). For comparison purposes, a similar Na-RuA DFM (same metal and alkali loadings) tested by us under identical rector configuration and experimental conditions returned a maximum CH4 production equal to 183±2 μmol/gDFM (@300 °C) with 99.73±0.03% selectivity and 89% CO2 conversion (Cimino et al., 2022). Notably, the favourable and unique synergy existing between Li and Ru phases in the DFM guarantees a 20–60 °C reduction for the optimal operating temperature of the integrated process with respect to state-of-the-art Na-Ru DFMs (Merkouri et al., 2021): this is particularly important in view of the long term durability of the DFM (Jeong-Potter et al., 2022; Bermejo-López et al., 2022).
“