Promotion effect of CO on methanation of CO₂-rich gas­­ over nanostructured modified NiO/γ-Al₂O₃ catalysts

XRD spectra of the catalysts (figure 3) showed the characteristic peaks of NiO at 2θ = 37.04°, 43.03°, 62.78°, 75.4°, and 79.4° (JCPDS cards No.71-1179), CeO₂ at 2θ = 28.5° (JCPDS cards No.34-394), CaCO₃ at 2θ = 29.4° (JCPDS cards No.47-1743), the weak peak of γ-Al₂O₃ at 2θ = 67.37° (JCPDS cards No.82-1399), and very weak diffraction peaks for MgO [28]. In addition, diffraction peaks for NiO in XRD pattern of 6CeNiAl and 0.1Pt11CaNiAl were considerably less intense than that of NiAl and 3MgNiAl. This result indicated that CeO₂ and Pt increased the dispersion of NiO particles on γ-Al₂O₃. The NiO particle size values of the catalysts, calculated from the Scherrer equation [30], varied from 16 to 28 nm (shown in table 1).

Zoom In Zoom Out Reset image size

According to data in table 1, loading NiO and promoters accounted for falling of specific surface area from 225 m².g⁻¹ of γ-Al₂O₃ to 80–95 m².g⁻¹ due to a low specific surface area of them and covering the carrier pore by NiO and promoter particles.

SEM images of NiAl, 3MgNiAl, and 6CeNiAl catalysts (figures 4(a)–(c)) showed uniform particles and almost sphere shape (tens nm in size), ascribed to NiO and big blocks (hundreds of nm in size), which was attributed to carrier. The big blocks on non-promoted sample had larger size than that on promoted one, indicating that the additives improved the thermal stability of the catalyst.

Zoom In Zoom Out Reset image size

There were significant differences in the shape of NiO particles on NiAl, 3MgNiAl, and 6CeNiAl samples compared to 0.1Pt11CaNiAl one (figure 4(d)). The NiO particles on the first three samples had a smooth, rounded shape, while on the fourth one the fine, spongy and sharp edges particles were presented. This showed that the first three catalysts were sintered to a certain degree when were heated at 600 °C while the last one was not. From this observation it can be concluded that CaO exhibited higher ability to increase the thermal stability of the NiAl catalyst than other additives.

It can be derived from the TEM analysis (figure 5) that nanoscale NiO particles have been successfully prepared although the NiO content in catalysts is as high as 37.7 wt% thanks to the high specific surface area (225 m² g⁻¹) and large pore diameter (>4 nm) of γ-Al₂O₃. The NiO particles of several nm inside the pore (bright colored spots) and particles of dimensions from several nm to several tens' nm on the outer surface of γ-Al₂O₃ (dark spots) can be observed. Two samples modified with MgO and CeO₂ showed smaller and discrete particles than others. On the 3MgNiAl sample besides very small dark-colored particles, 50–80 nm blocks covering the surface of the carrier were also observed. Therefore this sample showed the smallest specific surface area (table 1). Compared to MgO promoted sample, 6CeNiAl and especially 0.1Pt11CaNiAl catalysts showed higher densities of dark spots on the surface. This is explained by the fact that in these two samples CeO₂ and CaCO₃ crystals coexist with NiO crystalline particles while MgO exists in the amorphous state (from the XRD analysis). The additional surface covering of CaCO₃, as evidenced from the XRD spectra and CO₂-TPD profiles (figure 7), also contributed to a slight reduction in the surface area of the PtCaNiAl catalyst compared to non-promoted one (table 1). The CeO₂-modified catalyst is the most evenly distributed and uniform NiO particles compared to the others, resulting in the highest surface area of this sample. In particular, specific surface area of 6CeNiAl catalyst (95.0 m² g⁻¹) is greater than that of the non-promoted catalyst NiAl, which may be explained that CeO₂ enhanced the dispersion of NiO on surface and into pore of the support, it also prevents NiO particles from sintering.

Zoom In Zoom Out Reset image size

H₂-TPR diagram of the NiAl catalyst (figure 6) indicated a single reduction peak (α₂) at 380 °C, attributed to the reduction of NiO species weakly interacted with support. The strong interaction of NiO with the MgO additive was demonstrated by shifting the reduction peak α₂ to α₃ with the higher temperature (415 °C) and the lower reduction intensity on H₂-TPR diagram of 3MgNiAl catalyst. In contrast, 6CeNiAl and 0.1Pt11CaNiAl profiles displayed similar outcomes, there were four reduction peaks between α₁ and α₃ (327 °C–415 °C). According to Liu et al [31] the peak appeared in the range of 360 °C–640 °C on H₂-TPR diagrams of the CeO₂-promoted NiO/γ-Al₂O₃ catalysts is corresponding to the reduction of NiO species. It was reported by Zhou et al [32] that for NiO/CeO₂ sample a peak at temperature range of 280 °C–330 °C characterized the reduction for NiO species on the subsurface of the NiO/CeO₂, while a peak at 380 °C can be attributed to the reduction of bulk NiO species. Thus, it can be assumed that the reduction peaks at 373, 390 and 396 °C on the 6CeNiAl sample were attributed to the reduction of the NiO small particles interacting with CeO₂, NiO with γ-Al₂O₃ and NiO blocks interacting with CeO₂, respectively.

Zoom In Zoom Out Reset image size

Compared to NiAl, the maximum reduction temperature of buck NiO (α₄) in the 11CaNiAl sample was higher (around 550 °C versus 380 °C) [16], meaning that CaO made interaction NiO-Al₂O₃ becomes stronger and NiO harder to be reduced. On the TPR diagram of 0.1Pt11CaNiAl sample the reduction peak α₄ appeared very weak instead of it there were three reduction peaks appearing in the lower temperature region. According to [33], for high dispersed platinum state, the hydrogen reduction took place in the temperature range of 250 °C–400 °C. In our previous work [34], the reduction peaks in the TPR diagram of Pt/γ-Al₂O₃ catalysts at 230 °C–250 °C and 420 °C–430 °C, characteristic for the reduction of Pt⁴⁺ → Pt²⁺ and Pt²⁺ → Pt° were proposed. Subramanian [35] demonstrated that, in the system Pt/γ-Al₂O₃ the reduction process took place in the temperature range of 300 °C–500 °C. Therefore, it can be suggested that two first peaks on TPR diagram of PtCaNiAl sample at 310 °C and 330 °C between α₁ and α₂ characterized the reduction of the highly dispersed Pt⁴⁺ in Pt²⁺ and Pt²⁺ in Pt° overlapped with the reductions of dispersed NiO particles (α₁) and NiO species weakly interacted with support (α₂), while the third peak (351 °C), with the most intense, ascribed to the reduction of NiO species. The reduction peaks of 0.1Pt11CaNiAl sample appeared at lower temperatures with stronger intensity than others one, indicating that this catalyst had the best reduction properties, promising high activity. Sample 6CeNiAl took second place after 0.1Pt11CaNiAl in terms of the amount of reduced NiO.

From the TPR results, it can be seen that the CeO₂ and Pt additives increased the reducibility of NiO, while alkali metal oxides CaO and MgO reduced this quantity. The results revealed that NiO in 3MgNiAl catalyst can be effectively reduced to Ni° in a hydrogen atmosphere at temperatures of 450 °C and the rest catalysts did it at 400 °C. So, the reduction conditions of catalysts before testing their activity displayed in the experimental were consisted.

It can be seen in figure 7, NiAl, 3MgNiAl and 6CeNiAl catalysts showed similar CO₂-TPD diagrams. In detail, there were several CO₂ desorption peaks appearing in a temperature range of 70 °C–250 °C, in which the highest intense peaks are recorded from 110 to 200 °C. The desorption peak was attributed to the weak basic sites [36]. According to BET specific surface area, 3MgNiAl sample had the lowest one; however, the amount of CO₂ desorption of this sample was slightly higher than that of two others. This is because the presence of MgO increased Lewis basicity leading to improve CO₂ chemisorption. As for 0.1Pt11CaNiAl, there was a complex CO₂-TPD diagram including four considerable CO₂ desorption peaks in a wide range of temperature. In detail, two CO₂ desorption peaks centered at 118 °C and 143 °C attributed to weak Lewis-basic sites and a large peak documented at 345 °C can be assigned to the medium strength basic sites [36]. It is noticed that there is a giant CO₂ desorption peak at 550 °C on CO₂-TPD diagram of the 0.1Pt11CaNiAl catalyst, which can be attributed to the formation of ${{\rm{HCO}}}_{3}^{-}$ or ${{\rm{CO}}}_{3}^{2-}$ leading to a strong interaction between CO₂ and the support as mentioned in [37]. So, promoting by the high CaO loading content (11 wt%) caused enhancement of basicity and adsorption of CO₂ of NiAl catalyst.

Zoom In Zoom Out Reset image size

It was clear from figure 8 that all promoters greatly enhanced the activity of catalyst in hydrogenation of CO₂. On the modified catalysts, the reactions began to take place at temperatures of 225 °C–250 °C, in contrast, at temperatures under 300 °C, both reactions on non-promoted NiAl did not happen.

Zoom In Zoom Out Reset image size

It has been shown in figure 8, CO₂ conversion in the co-hydrogenation of carbon oxides mixture (case I) was much higher than those in CO₂ solo-hydrogenation (case II). In term of the NiAl catalyst, the gap between CO₂ conversions of two cases of hydrogenation increased markedly corresponding to the rise of temperature in a range of 300 °C–400 °C. The 3MgNiAl catalyst produced a substantial CO₂ conversion. As for CO₂ methanation in co-hydrogenation, the started reaction temperature is 250 °C which is lower than 300 °C of the NiAl catalyst. This is because MgO prefers a high CO₂ adsorption capacity. Consequently, CO₂ conversion at 400 °C for both cases of CO₂ methanation on the 3MgNiAl catalyst was somewhat satisfied, the former finished at 88.1% of CO₂ conversion, the latter ended at nearly 4% lower. Observably, CO₂ conversion graph of the co-hydrogenation reaction was always above that of the CO₂ solo-hydrogenation.

Paradoxically, at 250 °C, while the hydrogenation reaction on the 6CeNiAl catalyst in the second case occurred, the first case did not happen. This phenomenon may be due to the adsorption competition between CO and CO₂ on the catalyst. In a higher temperature range of 275 °C–400 °C, CO₂ conversion in the first case was always greater than that in the second case because at that temperature region CO has completely converted, as seen in figure 9. In general, comparing the conversion of CO₂ (figure 8) and CO (figure 9) in co-hydrogenation showed that CO₂ started to be converted at the temperature, where CO was almost completely consumed, it is 250 °C for MgNiAl, 275 °C for CeNiAl and PtCaNiAl and finally 325 °C for NiAl catalyst. According to Inui et al [21] this phenomenon indicated that the adsorption of CO on catalysts in the reaction condition is much stronger than that of CO₂ that inhibit the CO₂ methanation. However, it should be noted that, even in the absence of CO of CO₂ solo-hydrogenation case, the conversion of CO₂ is lower than that in the co-hydrogenation reaction. Thus, CO adsorption may inhibit the CO₂ hydrogenation in the low temperature region, while in the high temperature zone the difference in the conversion of CO and CO₂ is mainly related to the nature of the reactants.

Zoom In Zoom Out Reset image size

In the reaction product mixture, besides the main product CH₄, the by-product is only CO, other hydrocarbons are not detected. Figures 8 and 10 show that, in the CO₂ solo-hydrogenation reaction, the CH₄ selectivity is low at low CO₂ conversion, increases with increasing CO₂ conversion except catalyst modified by MgO. The existence of CO as a by-product of the CO₂ solo-hydrogenation reaction is an evidence of occurrence of water-gas shift reaction in condition of CO₂ hydrogenation reaction [22]. CO, which is more easily hydrogenated as showed in figure 9, in a mixture with CO₂, hydrogenated at a lower temperature and the generated heat in this reaction providing to the endothermic stage converting CO₂ into CO that promoted the main reaction of CO₂ hydrogenation.

Zoom In Zoom Out Reset image size

On MgNiAl catalyst in CO₂ solo-hydrogenation CH₄ selectivity reached 100% even when CO₂ conversion was very low (2.5% at 250 °C), means no CO was found in the reaction product mixture. This may be related to the fact that this catalyst possesses the highest activity in CO hydrogenation, which completely converts CO to methane at temperatures as low as 250 °C, as shown in figure 9. It was reported [22] that the occurrence of WGS reaction in carbon oxides hydrogenation depends on the nature of the active metal and carrier.

The promotion effect of CO in CO₂ hydrogenation is expressed most impressively on 0.1Pt11CaNiAl catalyst thanks to it being modified simultaneously by both Pt and Ca additives. The former has a strong affinity to H₂, and the latter raises the support's basicity causing the high CO₂ adsorption. At 275 °C, CO₂ conversion of both cases of hydrogenation was analogous, at around 6.3%. At the next temperature level (300 °C), although CO₂ conversion in the CO₂ solo-hydrogenation was three-fold, that in the co-hydrogenation was approximately 10 times as large as its CO₂ conversion at 275 °C. Finally, at 400 °C, the first finished at 72.3% of CO₂ conversion and the second ended at 90.2%. Apparently, CO₂ conversion rapidly increases under the addition of CO into reactants.

Comparing the results in figures 8 and 9 indicated that the activity of catalysts in hydrogenation of CO was much higher than that of CO₂. Even at 225 °C the CO conversion on catalysts has reached a value of several tens of percent while CO₂ hydrogenation has begun at 275 °C. Similar to hydrogenation of CO₂, the improvement in activity of NiAl catalysts by adding of promoters was also impressively expressed in hydrogenation of CO. In detail, at 250 °C CO conversions in hydrogenation of carbon oxides mixture over 3MgNiAl, 6CeNiAl and 0.1Pt11CaNiAl catalysts reached 100%, 94.0% and 57.1% respectively compared to 25.6% on non-promoted catalyst NiAl. The temperature of 100% CO conversion were 250 °C and 275 °C respectively for 3MgNiAl and 6CeNiAl, 0.1Pt11CaNiAl catalysts, compared to 325 °C for the non-modified catalyst.

In general, the order in the activities of catalysts, evaluated in CO₂ conversion in CO₂ solo-hydrogenation (case II) and CO conversion in co-hydrogenation of mixture of CO and CO₂ (case I), was observed as follows: 3MgNiAl ≈ 6CeNiAl > 0.1Pt11CaNiAl ≫ NiAl. It indicated that hydrogenation of carbon oxides takes place on the one kind of active center. The results obtained in the study confirm that the presence of CO in feedstock activated the CO₂ hydrogenation and in this case the order in the activities of the promoted catalysts was 0.1Pt11CaNiAl ≈ 3MgNiAl > 6CeNiAl ≫ NiAl. However, the mechanism of CO promotion for CO₂ methanation is still not clear because the mechanism and intermediate compounds of CO₂ hydrogenation although have been discussed extensively in the references but there is still no agreement. There are two perspectives on the mechanism of hydrogenation of CO₂ into methane, namely the direct conversion of CO₂ to CH₄ without the formation of CO and the second mechanism is indirect methanation, through CO formation [38].

In terms of surface compounds, some argue that CO is intermediary of CO₂ methanation and others reported that CO was not a key intermediary. Some studies confirm that CO₂ directly dissociated into CO, which is the main intermediate of the reaction. For example, adsorbed CO₂ dissociated to adsorbed CO and O on the surface of the catalyst Ni/γ-Al₂O₃, which continues to hydrogenate into CH₄ [39, 40]. On Ru/TiO₂ it was proposed that, through the reverse water gas shift reaction H₂ facilitates the dissociation of CO₂ to adsorbed CO, and formation adsorbed CO is the rate determining step of methanation [41]. Other studies suggest that CO is formed from already formed species, resulting from the reaction of adsorbed CO₂ with H₂ dissociated on metal sites [42]. For example, this was proposed for Ru catalysts [43, 44], CeO₂-contained [45, 46] and Ni/MgAl-MMO [47] catalysts.

It seems that the intermediate compound depends on the activity phase and catalyst carrier. However, it was shown that the energy barrier for CO₂ hydrogenation to formate followed by decomposition to CO is similar to that of direct dissociation of CO₂ to CO on Ru nanoclusters supported on TiO₂ [48]. It is likely that the reaction mechanism is highly dependent on the nature of the active metal and support of the catalyst. From the results presented above, on studied catalyst the reaction takes place under the second mechanism, through the creation of intermediate compounds CO.

In methanation three main reactions take place in the process: mildly endothermic reverse water gas shift, highly exothermic methanation of CO and CO₂. It can be argued that the promotion of CO addition in CO₂ methanation thanks to its prior hydrogenation provides heat for endothermic reverse water gas shift, facilitating CO formation and making the reaction to achieve the same CO₂ conversion extent at a lower reaction temperature than the solo CO₂-hydrogenation reaction and the consumption of generated heat of hydrogenation reaction keep the methanation of CO₂ within an ideal temperature range at the optimum value, balancing between the thermodynamic and kinetic properties of the reaction.

Generally, CH₄ selectivity in the co-methanation was higher and more stable than that in the CO₂ solo-methanation (figure 10). Indeed, CH₄ selectivity of the CO₂ solo-methanation on NiAl catalyst at 325 °C was approximately 31% and gradually rose to 73 °C–77% when the reaction temperature was increased to 400 °C. In co-hydrogenation, on this catalyst there was a dramatic improvement in CH₄ selectivity, it was ranged from 93% to 95% in the reaction temperature range of 350 °C–400 °C. There were incredible tendencies for CH₄ selectivity of CO₂ methanation over 3MgNiAl, it almost remained stable at 100% in both cases as depicted in figure 10 thanks to its very high activity in CO hydrogenation as seen in figure 9. As seen, over the 6CeNiAl and 0.1Pt11CaNiAl catalysts CH₄ selectivity in CO₂ solo-hydrogenation gradually increased respectively from 70% and 60% at 250 °C to approximately 97% at 350 °C or higher, while in co-hydrogenation this quantity was always higher than 99% at all temperatures.