A novel process of triple-passed CO₂ methanation over gradient-coated spiral-structured Ni/CeO₂ catalysts

Abstract

The objective of this study is to develop an efficient gradient spiral-structured Ni/CeO₂ catalyst for CO₂ methanation. The spiral-structured Ni/CeO₂ catalysts were prepared by the gradient coating method. The process employs a triple-passed operation for steam trapping to increase the CH₄ product at flow rates from 100 to 3000 ml/min. At a low flow rate (100 ml/min), a triple-passed reactor’s CO₂ conversion efficiency (98.5%) and CH₄ selectivity (99.9%) are slightly greater than those of a single-passed reactor (96.4% and 99.9%, respectively). When operating at a high flow rate of 3000 ml/min, a triple-passed reactor can reach up to 83.4% CO₂ conversion efficiency and 94.9% CH₄ selectivity, whilst a single-passed reactor can only achieve 64% CO₂ conversion and 87.4% CH₄ selectivity. In addition, the spiral structure can enhance heat transfer, eliminating hotspots as well as improving mass transfer by swirling flow. The development of gradient coating could be beneficial for conducting thermal energy from a thick-coated front spiral-catalyst to a thinner-coated terminal spiral-catalyst. Furthermore, the steam trapping would further increase CO₂ conversion as well as CH₄ selectivity.

Introduction

CO₂ methanation has been determined to be a successful method for the total removal of CO₂. The technique can not only reduce emissions of CO₂ and greenhouse gases, but also provide future energy with ecologically friendly effects¹. CO₂ methanation is an exothermic reaction², producing CH₄ and H₂O as presented in Eq. (1). The combination of an endothermic reversed water gas shift (RWGS) reaction and an exothermic CO methanation, as indicated by Eqs. (2) and (3), respectively, is the most accepted mechanism of the methanation reaction².

$${\text{CO}}_{{2}} + {\text{4H}}_{{2}} \to {\text{ CH}}_{{4}} + {\text{2H}}_{{2}} {\text{O}} \Delta {\text{H}}_{{{\text{298K}}}} = \, - {165} {\text{kJ}} {\text{mol}}^{{ - {1}}}$$

(1)

$${\text{H}}_{{2}} + {\text{ CO}}_{{2}} \to {\text{CO }} + {\text{ H}}_{{2}} {\text{O}} \Delta {\text{H}}_{{{\text{298K}}}} = {\text{ 41 kJ}}/{\text{mol}}$$

(2)

$${\text{3H}}_{{2}} + {\text{ CO}} \to {\text{CH}}_{{4}} + {\text{ H}}_{{2}} {\text{O}} \Delta {\text{H}}_{{{\text{298K}}}} = \, - {2}0{\text{6 kJ}}/{\text{mol}}$$

(3)

Catalysts are essential in CO₂ methanation because they accelerate reaction kinetics and improve selectivity. The catalyst should be highly active, stable, and selective for methane production while limiting unwanted by-products. A number of noble metals, including Rh³, Ru⁴ and Pd⁵, have found extensive use for CO₂ methanation because of their superior catalytic activities and low-temperature CH₄ selectivity. However, they have limited large-scale industrial applications due to their high cost. However, non-noble metals like nickel (Ni) have attracted much attention for applying in CO₂ methanation, as they exhibit high activities, and their low cost makes them scalable to industrial scale^6,7. Still, there are certain obstacles in the way of enhancing the CO₂ methanation performance of Ni-based catalysts. CO₂ methanation is an exothermic reaction that results in carbon formation and Ni particle sintering, which deactivates the catalyst during CO₂ methanation⁸. In addition, enhancing CO₂ methanation requires consideration of the catalyst supports. According to earlier research^9,10, oxygen vacancies are crucial for CO₂ methanation since CO₂ and CO dissociate on oxygen vacancies. Cerium oxide (CeO₂) has unique properties in storing and releasing oxygen with a large capacity, so it is frequently used as a catalyst support for CO₂ methanation in order to encourage both CO₂ activation and CO₂ adsorption^11,12.

As mentioned earlier, due to the highly exothermic heat of CO₂ methanation, inevitably the temperature inside the reactor rapidly rises, which causes an extremely high temperature zone, called hotspots. The spots would cause thermal sintering of catalyst particles and reduce CO₂ conversion and CH₄ yield due to the exothermic nature of the reaction. To prolong a catalyst’s lifetime and achieve high CH₄ selectivity, thermal management of hotspots is important. In general, packed-bed reactors remain difficult to control heat transfer in high-temperature zones due to the low heat conductivity of catalyst bed materials¹³. Recently, structured catalyst reactors exhibited excellent catalytic performance for CO₂ methanation, especially catalyst stability, attributed to improved heat transfer by the conduction mode of the structured substrate¹⁴. In our previous research¹⁵, honeycomb-structured Ni-based catalysts were studied for CO₂ methanation. It was found that the random-flow channels of the honeycomb structure can increase the catalytic performance, resulting in a high CO₂ conversion. We also reported that spiral-structured catalysts with Ni/CeO₂ loaded on spiral aluminum substrates presented a high methanation activity because the swirling flow of spiral structures improved heat and mass transfer abilities¹⁶.

Herein, we develop spiral-structured Ni/CeO₂ catalysts by applying the gradient coating, which consists of a thick-coated front spiral catalyst and a thinner-coated terminal spiral catalyst, to enhance heat and mass transfer and to prevent catalyst deactivation from the hotspots in methanation. Furthermore, a triple-passed operation of steam trapping techniques is employed to improve the CH₄ production by removing the by-product of water.

Experimental

Materials and chemicals

Cerium oxide (CeO₂, JRC-CEO-2) was purchased and used as received for catalyst support. Nickel nitrate hexahydrate (Ni(NO₃)₂.6H₂O, ≥ 98%, Sigma-Aldrich) was used as the nickel precursor for the catalyst preparation. The aluminum plate (JIS A1100P84 H18) was used as a spiral plate (50 mm × 7 mm) for catalyst coating. Isopropanol (IPA, AR grade, QReC) was used as a dispersal solution during the wash-coat method. Sodium hydroxide (NaOH, GR, Merck) and hydrochloric acid (HCl, 37%, AR grade, QReC) were used for cleaning and activating, respectively, the surface of spiral Al plates.

Preparation of gradient-coated spiral-structured Ni/CeO₂ catalysts

CeO₂ was used as the catalyst support to prepare the Ni catalysts by wetness to dryness impregnation, which were denoted as Ni/CeO₂. In detail, 9.0 g of CeO₂ was dispersed into the Ni precursor solution, which contains 4.96 g of Ni(NO₃)₂.6H₂O and 50 ml of DI water. The mixture was then stirred at 80°C to evaporate the DI water. When the solution turned into a slurry mixture, it was physically mixed with a Teflon stick to make the mixture homogenous and dried. After the mixture dried, it was then calcined at 500°C for 3 h (heating rate of 10°C/min). Photographs of each step of Ni/CeO₂ catalyst preparation are shown in Fig. 1a.

Fig. 1

The illustration of (a) Ni/CeO₂ catalyst preparation by wetness to dryness impregnation method, (b) Preparation of activated spiral Al plate, and (c) Preparation of spiral-structured Ni/CeO₂ catalyst by wash-coat method.

The preparation of spiral Al plates is illustrated in Fig. 1b. The aluminum plate was cut to the size of 50 mm × 7 mm and twisted at 360° by using the bench vise to obtain the spiral Al substrate. The spiral aluminum substrates were then washed and activated with 5-min of 0.8 N NaOH and 30-min of 2.8 N HCl solutions, respectively.

To coat Ni/CeO₂ on the activated surface of the spiral substrate, the wash-coat method was applied¹⁶ as shown in Fig. 1c. The activated spiral substrate was immersed in a slurry mixture of prepared Ni/CeO₂ and IPA. Then it was blown and dried by a dryer. The cycle of dipping and drying was further continued until the spiral substrate was coated with the desired weight of Ni/CeO₂. In this work, gradient-coated spiral catalysts were used in CO₂ methanation; hence, the Ni/CeO₂ was coated on the spiral Al plates at different weights, that is, 120 mg and 180 mg.

Characterization of Ni/CeO₂ catalyst

The as-prepared Ni/CeO₂ catalysts were characterized comprehensively using several techniques, including X-ray diffraction (XRD, BENCHTOP, D2 PHASER, BRUKER), H₂ temperature-programmed reduction (H₂-TPR, BEL-CAT-B), Raman spectroscopy (PerkinElmer® Spectrum™GX), field emission scanning electron microscope and energy dispersive x-ray spectrometer (FESEM-EDS, JEOL JSM-7610F, Oxford X-Max 20 model), transmission electron microscopy (TEM, JEOL, JEM-2100 Plus), N₂ adsorption-desorption (TriStar II 3020 Version 3.02), inductively coupled plasma-optical emission spectrometry (ICP-OES, Avio 550 Max) and atomic force microscopes (AFM, Park system Fx40).

Catalytic test of CO₂ methanation

A single-passed reactor

There were two types of experimental setups for CO₂ methanation in this study: a single-passed reactor and a triple-passed reactor, as presented in Fig. 2. Regarding the single-passed reactor (Fig. 2a) and the triple-passed reactor (Fig. 2b), the CO₂ methanation was carried out in one reactor constructed from a quartz tube (i.d. = 8 mm, length = 900 mm). The as-prepared spiral-structured Ni/CeO₂ catalysts were placed inside the reactor in the order of 180 mg, followed by 120 mg (the total catalyst weight was 300 mg). It should be noted that using 180 mg and 120 mg catalysts for gradient coating in this research is intended to provide a suitable balance of catalytic activity while preventing the creation of hotspots as we proposed. The gradient coating promotes heat convection, lowering the risk of overheating and ensuring catalytic performance and longevity. In brief, the 180 mg catalyst produces higher catalytic activity and heat, whereas the 120 mg catalyst serves to reduce the intensity of the reaction in areas that exhibit excessive heat.

Fig. 2

The experimental setup of CO₂ methanation for (a) A single-passed and (b) A triple-passed reactor over gradient-coated spiral-structured Ni/CeO₂ catalysts.

The quartz tube reactor, which carried pieces of spiral catalyst, was positioned horizontally within a tube furnace and fitted with a digital temperature controller (KP1000 series, Chino). A type K thermocouple was used to track the temperatures of the spiral catalyst bed. Prior to the CO₂ methanation process, the as-prepared spiral catalysts were reduced in situ with a 100 vol% H₂ flow at 100 ml/min at 500°C (10°C/min) for an hour. The feed gas (CO₂/H₂ = 1/4 molar ratio) was then injected into the reactor at a total flow rate of 100 ml/min, corresponding to a weight hourly space velocity (WHSV) of 20,000 ml/g·h. CO₂ methanation was performed at 200–500°C under atmospheric pressure. The outlet gas had steam as a byproduct, so it was cooled and trapped by a condenser before being manually injected into a Gas Chromatography (GC-TCD 8A, Shimadzu) by a micro syringe (MS-GAN025, ITO).

A triple-passed reactor

Three quartz tube reactors were set up in a tube furnace. Each reactor contained pieces of gradient-coated spiral-structured Ni/CeO₂ catalysts. The conditions of H₂ reduction and CO₂ methanation were similar to those of a single-passed reactor. However, the outlet gas of the R1 was used as the inlet gas for the R2 reactor, and the outlet gas of the R2 reactor was used as the inlet gas for the R3 reactor. Furthermore, as illustrated in Fig. 2b, the steam in each reactor’s outlet gas was removed by a cooling trap before passing through the sampling port and the following reactor.

The CO₂ methanation performances of these two configurations of single-passed and triple-passed reactors were investigated at high reactant feed (CO₂/H₂) flow rates of 500, 1000, 1500, 2000, 2500, and 3000 ml/min at 350°C. In addition, the pressure drop between the gas inlet and outflow locations was measured using a manometer (DT-8820) in order to investigate the flow behavior during methane production at different flow rates.

The catalytic performance in terms of CO₂ conversion ($\text X_{{\text {CO}_{2}}}$), CH₄ selectivity ($\text S_{{\text {CH}_{4}}}$), CO selectivity (S_CO), and carbon balance was calculated following Eqs. (4–7)¹⁷, respectively.

$$X_{{CO_{2} }} (\% ) = \frac{{[CO_{2} ]_{in} \times F_{in} - [CO_{2} ]_{out} \times F_{out} }}{{[CO_{2} ]_{in} \times F_{in} }} \times 100$$

(4)

$$S_{{CH_{4} }} (\% ) = \frac{{[CO]_{out} }}{{[CO]_{out} + [CH_{4} ]_{out} }} \times 100$$

(5)

$$S_{CO} (\% ) = \frac{{[CH_{4} ]_{out} }}{{[CO]_{out} + [CH_{4} ]_{out} }} \times 100$$

(6)

$$C(\% ) = \frac{{[CH_{4} ]_{out} + [CO]_{out} }}{{[CO_{2} ]_{in} }} \times 100$$

(7)

where [CO₂]_in and [CO₂]_out are the CO₂ concentrations (mol/min) at the input and output gases. The concentrations (mol/min) of CH₄ and CO outlets are denoted by [CH₄]_out and [CO]_out, respectively. F_in and F_out are the total molar flow rates of the input and output gases, respectively.

The mass-normalized reaction rate (r_m) was determined following Eq. (8).

$$r_{m} (molCO_{2} /g.s) = \frac{{[CO_{2} ]_{in} \times X_{{CO_{2} }} }}{{60 \times 100 \times m_{cat} }}$$

(8)

where [CO₂]_in is the CO₂ inlet concentration (mol/min). $\text X_{{\text {CO}_{2}}}$ represents CO₂ conversion. m_cat represents the amount of catalyst in grams.

Results and discussion

Investigation of pristine CeO₂ support and Ni/CeO₂ catalyst

The crystal structure investigation

Figure 3 shows the employing of X-ray diffraction (XRD) to investigate the crystal phase information of pristine CeO₂ supports along with Ni/CeO₂ catalysts in both calcined and reduced forms. The diffraction characteristic peaks of 2θ found at 28.6°, 33.1°, 47.5°, 56.4°, 59.2°, 69.5°, 76.8°, and 79.2° corresponded to the CeO₂¹⁸. These peaks can be indexed to the lattice planes of a typical cubic fluorite structure as follows: (111), (200), (220), (311), (222), (400), (331), and (420), respectively, indicating that eight oxygen atoms surround each metal cation¹⁹. After loading Ni on the CeO₂ support (calcined Ni/CeO₂), the specific peaks of 2θ were presented at 37.2°, 43.3°, and 62.9°. These peaks corresponded to the NiO form with the weak diffraction peaks at lattice planes of (100), (200), and (220), respectively²⁰. Before methanation, the calcined Ni/CeO₂ catalyst must be reduced by H₂ gas to transform NiO to metallic Ni, and the catalyst after H₂ reduction will be called reduced Ni/CeO₂. The distinctive metallic Ni diffraction peaks in the reduced Ni/CeO₂ catalyst were located at 44.7° and 52.1° (JCPDS no. 01-071-4655). There are no NiO peaks left after H₂ reduction, which means that the catalyst is completely reduced.

Fig. 3

XRD patterns of (a) Pristine CeO₂ supports, (b) Reduced Ni/CeO₂ catalyst, (c) Calcined Ni/CeO₂ catalyst.

The crystallite sizes of CeO₂, NiO, and Ni before and after H₂-reduction are shown in Table 1. The Scherrer equation was used to compute the crystallite sizes of CeO₂, NiO, and Ni, which were found at the 2θ peaks of 28.6°, 43.3°, and 44.7°, respectively. The pristine CeO₂ shows a small crystallite size at 7.4 nm. After calcination to form calcined Ni/CeO₂, the CeO₂ crystallite size increased from 7.4 to 8.1 nm. This is because calcination is a thermal treatment process, as it was proceeded at 500°C for 3 h at atmosphere, causing CeO₂ particles to migrate and coalesce, resulting in a larger crystallite size²¹. Similarly, when the calcined Ni/CeO₂ was then reduced by H₂ at 500°C for another 1 h to form reduced Ni/CeO₂, the crystallite size of CeO₂ kept rising to 9.9 nm due to this high temperature can also induce sintering effects, leading to the coalescence of CeO₂ crystallites and an increase in their size. In addition, the H₂ reduction process generates oxygen vacancies in the CeO₂ lattice, which can enhance atomic mobility and facilitate crystallite growth. This can be confirmed by Raman spectra (see Fig. 6b and Table 3), showing that oxygen vacancies (the ratio between peak intensities of the D and F_2g mode, ID/IF_2g) of reduced Ni/CeO₂ are a bit higher than that of the calcined one. The prior research²² also affirmed that Ni²⁺ species can enter the CeO₂ lattice to replace Ce⁴⁺ and create oxygen vacancies to balance charges.

Table 1 Crystallite sizes of NiO, Ni, and CeO₂ using XRD analysis. The samples are pristine CeO₂, calcined Ni/CeO₂, and reduced Ni/CeO₂.

The crystallite size of NiO was 21.6 nm before H₂ reduction. After H₂ reduction, the NiO transformed to metallic Ni. The metallic Ni from reduced Ni/CeO₂ presents a crystallite size of 28.3 nm, which is a bit larger than that of NiO. This is because of Ni atom diffusion from NiO to Ni at high temperatures during the H₂ reduction^23,24. After the H₂ reduction process, although the Ni and CeO₂ crystallite sizes of the reduced catalyst were slightly larger, causing a loss of some surface area, they still perform well in CO₂ methanation²⁵.

The morphology of pristine CeO₂ support and Ni/CeO₂ catalyst

As shown in Fig. 4, the morphologies of the CeO₂ support and Ni/CeO₂ catalyst were examined using SEM and TEM techniques. The Ni particle size distribution (nm) is measured from ImageJ software with a sample number of 50 (N = 50). Since the pristine CeO₂ support was produced commercially, it was used exactly as given. The SEM image in Fig. 4a showed the agglomeration of CeO₂, which consisted of aggregating CeO₂ nanoparticles. Following Ni impregnation, known as calcined Ni/CeO₂, the SEM image in Fig. 4b shows that the Ni particles were distributed over the surface of CeO₂ supports. High-magnification SEM image that was placed into Fig. 4b further demonstrated the presence of spherical Ni particles. As we may know, the SEM technique is used for surface morphological characterizations; hence, to study the internal structure of the reduced form of Ni/CeO₂, the TEM technique was employed. TEM micrographs in Fig. 4c show the dark spots of spherical Ni particles distributed over CeO₂ supports. However, some locations exhibited agglomeration of black dots due to the development of crystals from NiO to Ni during the H₂ reduction process^23,24. The Ni particle size distribution, measured from reduced Ni/CeO₂, is presented in Fig. 4d. This finding illustrates that Ni particle sizes are in between ~20 and ~60 nm. The average size of Ni particles is 32.1 ± 10.0 nm, and most Ni particle sizes are typically greater than that of the Ni crystallite size of reduced Ni/CeO₂ (28.3 nm) that is measured from the XRD technique (see Table 1). As we know, Ni particles are the entirely physical entity or aggregate of Ni that can be made from one or more crystallites. In addition to the SEM and TEM results, the ICP-OES analysis demonstrated that the actual Ni loading amount was 10.3% (mass), as indicated in Table 2, confirming that Ni species were successfully distributed on the CeO₂ surface.

Fig. 4

Micrographs of (a) SEM image of pristine CeO₂ support, (b) SEM images of calcined Ni/CeO₂, (c) TEM image of reduced Ni/CeO₂, and (d) Ni particle size distribution (nm).

Table 2 Physicochemical properties of pristine CeO₂ and Ni/CeO₂ catalysts in different forms, including Ni loading amount.

Porosity of pristine CeO₂ and its Ni/CeO₂ catalysts

To study the porosity of the pristine CeO₂ and Ni/CeO₂ catalysts (calcined form and reduced form), N₂ adsorption–desorption isotherms, BET surface areas (S_BET), total pore volumes, and pore size distribution (PSD) are shown in Fig. 5 and Table 2.

Fig. 5

(a) N₂ adsorption–desorption isotherms, and (b) Pore size distributions of pristine CeO₂ and Ni/CeO₂ catalysts in different forms (calcined and reduced forms).

Figure 5a illustrates that pristine CeO₂ exhibits type IV isotherms with a hysteresis loop, which may indicate that it contains mesoporous structures²⁶. Incorporating Ni on pristine CeO₂ to form Ni/CeO₂ catalyst influences the textural properties, which can be observed through changes in N₂ adsorption-desorption isotherms, total pore volumes, and PSDs. Calcined Ni/CeO₂ (prepared from wetness to dryness impregnation) and reduced Ni/CeO₂ (obtained from H₂ reduction) show similar N₂ adsorption–desorption isotherms; that is, type IV isotherms with a hysteresis loop. However, N₂ adsorption–desorption isotherms of pristine CeO₂ display higher adsorption capacity (cm³/g STP) than those of calcined Ni/CeO₂ and reduced Ni/CeO₂. This is due to loading Ni onto pristine CeO₂, which could affect the pore structure and S_BET, leading to decreased N₂ adsorption capacity. Table 2 shows that after Ni impregnation, S_BET of pristine CeO₂ significantly decreases, from 155.7 to 89.1 m²/g (for calcined form) and 87.5 m²/g (for reduced form). Total pore volumes also exhibit the same decreasing trend. The decrease in S_BET and total pore volumes indicates that the addition of Ni to the CeO₂ surface resulted in mesopore blockage, suggesting that Ni particles are distributed over the CeO₂ surface and partially block the porosity of the CeO₂²⁷. On the other hand, PSDs of calcined Ni/CeO₂ (6.9 nm) and reduced Ni/CeO₂ (7.0 nm) are larger than that of pristine CeO₂ (5.5 nm) as presented in Fig. 5b together with Table 2. To compare PSDs between pristine CeO₂ and calcined Ni/CeO₂, it was found that during the calcination of the as-prepared Ni/CeO₂, the Ni precursor decomposes to form nickel oxide (NiO) particles, which are dispersed on the surface and within the pores of CeO₂, leading to a larger pore size. After H₂ reduction of the calcined Ni/CeO₂ catalyst, the PSD of reduced Ni/CeO₂ is a bit larger than that of calcined Ni/CeO₂. This is due to NiO particles being transformed into metallic nickel (Ni), which can influence the pore structure. The H₂ reduction process could lead to the creation of additional pore volume or the enlargement of existing pores due to the migration and sintering of Ni particles²⁸. It can be concluded that the Ni/CeO₂ catalysts still perform an excellent CO₂ methanation even if loading Ni to the pristine CeO₂ affects the catalyst’s porosity.

Metal-support interaction and oxygen vacancies

The interactions between Ni and CeO₂ in NiO–CeO₂ and its reduction characteristics are examined by H₂-TPR characterization at 50–900°C, as shown in Fig. 6 and Table 3. The temperature range of 200–600°C is usually where the reduction peaks of NiO catalysts supported by CeO₂ are observed²⁰. Regarding the H₂-TPR result in Fig. 6a, three reduction peaks were denoted as α, β, and γ. The α peak, which is detected below 250°C, is attributed to the reduction of surface oxygen species. These species may be linked to the oxygen vacancies created by the partial substitution of Ni²⁺ ions for Ce⁴⁺ ions²⁹. The Ni/CeO₂ catalyst displayed two distinct α peaks at 243.7°C and 259.1°C. These two peaks shifted to a high reduction temperature in the α zone, demonstrating a strong relationship between metal and support on the surface of the catalyst¹⁷. The main reduction peak of Ni/CeO₂ catalysts was seen at a temperature between 250°C and 400°C, which corresponds to β-NiO. The β-NiO peak is ascribed to the interaction between NiO and CeO₂, also known as the Ni–Ce–O³⁰. The temperature of the β peak over Ni/CeO₂ was 312.1°C, indicating a strong interaction between NiO and CeO₂³¹, reported in Table 1. The temperature of the γ peak of Ni/CeO₂ was at 762.6°C, which is responsible for the bulk CeO₂ support at the high reduction temperature (650–900°C)³². In this case, Ni/CeO₂ was reduced by H₂ at 500°C for an hour to convert NiO peaks in α and β forms to metallic Ni since the metallic Ni is more active for CO₂ methanation.

Fig. 6

(a) H₂-TPR of Ni/CeO₂ catalyst and (b) Raman spectra of calcined and reduced forms of Ni/CeO₂.

Table 3 Reduction temperatures of Ni/CeO₂ catalyst and the amount of H₂ consumption.

Table 3 shows the H₂ consumption over the Ni/CeO₂ catalyst as determined by the H₂-TPR pattern. The catalyst displayed a total H₂ consumption of 2.836 mmol/g, with the greatest amount of H₂ consumption (2.090 mmol/g) occurring in the major β peak. To compare the actual amount of H₂ consumption of the Ni/CeO₂ catalyst with the theory, the theoretical H₂ consumption can be computed from NiO at 1.755 mmol/g of 10% (mass) of Ni. It was found that the actual amount of H₂ in the Ni/CeO₂ catalyst was higher than the theoretical amount. This was because of the H₂ spillover effect, which occurs when CeO₂ support in the γ region is also reduced during the H₂-reduction step³³.

The oxygen vacancy of the Ni/CeO₂ catalyst was examined using Raman characterization. Since oxygen vacancies provide active sites for CO₂ activation, they have the potential to improve the performance of CO₂ methanation³⁴. Oxygen vacancy content was determined by the ratio between the peak intensities of the D and F_2g modes, as seen in Fig. 6b and Table 4. The F_2g Raman activation mode of CeO₂ was detected by the peaks at 454 and 456 cm^-1, while the D mode of stretching vibration from the defect site was observed at 554 cm^-135. It can be seen that after the H₂-reduction process, the ID/IF_2g ratio of Ni/CeO₂ slightly increased, from 0.47 to 0.48. The reason for this could be that oxygen vacancies were abundant in the structures formed by NiO and CeO₂ nanocrystals even prior to reduction. The finding was consistent with the results of the previous study³⁶. Furthermore, the H₂-TPR result in Fig. 6a demonstrated a strong interaction between the Ni and the CeO₂ support, resulting in a high oxygen vacancy concentration of the catalyst³⁷.

Table 4 The oxygen vacancy content of the calcined and reduced forms of the Ni/CeO₂ catalyst measured from ID/IF_2g using Raman spectroscopy.

The surface roughness of the samples (Untreated Al spiral, Activated Al spiral, and Spiral-structured Ni/CeO₂ catalysts)

The AFM technique was used to identify the topographical change of the Al spiral after the surface treatments and catalyst coating, as shown in Fig. 7. Each sample was randomly selected and measured to compare the surface roughness information. The surface of untreated Al spiral without chemical treatment showed slit patterns, with a depth ranging from 250-320 nm. After chemical etching, the surface of the Al spiral became corroded, causing it to lose its pattern and develop an irregular roughness instead. Chemical corrosion also removed the shine from the Al spiral, confirmed by the digital photos in Fig. 7.

Fig. 7

AFM analysis using non-contact mode over (a) Untreated Al spiral, (b) Activated Al spiral, and (c) Spiral-structured Ni/CeO₂ catalyst (sample of 120 mg), including the corresponding digital photos.

In comparison to the untreated Al spiral substrate, the depth of the etched surface increased by 3–4 times (from 700 to 1,250 nm), improving the surface area accessible for the catalyst coating. Following Ni catalyst coating, the surface of the coated Al spiral remained inconsistent in height, ranging from 220  to 560 nm. However, it should be noted that the catalyst was strongly coated and adhered to the Al spiral surface due to its surface roughness. This promotes the strength of the spiral-structured Ni/CeO₂ catalyst as well as increases the surface area, enhancing the catalytic activities of CO₂ methanation.

CO₂ methanation

The influence of temperatures over gradient-coated spiral-structured Ni/CeO₂ catalysts in a single-passed reactor

Figure 8 shows the CO₂ methanation performance, CO₂ conversion, product selectivity, carbon balance, and temperature profile of gradient-coated spiral-structured Ni/CeO₂ catalysts in a single-passed reactor. The reaction was performed at 200-500°C with a reactant feed flow rate of 100 ml/min. During CO₂ methanation, the spiral catalysts recorded temperatures from front side to back side of the catalyst position, including 3 cm before the front side catalyst.

Fig. 8

Influence of temperature on CO₂ methanation over gradient-coated spiral-structured Ni/CeO₂ catalysts in a single-passed reactor at 200-500°C, CO₂/H₂ = ¼ (100 ml/min), WHSV = 20,000 ml/g·h, as following (a) CO₂ conversion, (b) Product selectivity, (c) Carbon balance, and (d) Temperature profile.

Figure 8a and 8b show that the CO₂ conversion and CH₄ selectivity, respectively, decreased in the 400–500°C range because of thermodynamic limitations of the methanation reaction, which can occur at high temperatures and create unwanted products like CO³⁸. In order to determine the catalytic activity of CO₂ methanation, reaction temperatures between 200 and 350°C were considered. The gradient-coated spiral-structured Ni/CeO₂ catalysts in a single-passed reactor achieved the best catalytic performance for CO₂ methanation at a temperature of 350°C, 96.4% CO₂ conversion, and 99.9% CH₄ selectivity.

The temperature of the inlet gas before reaching the spiral catalyst is close to the set value; however, when the reactant feed (CO₂/H₂) was introduced to the thick-coated front spiral catalyst (F), the temperature slightly rises because CO₂ methanation is an exothermic reaction. The gas then passes to the thinner-coated spiral catalyst behind (B), where the temperature gradually decreases, as presented in Fig. 8d. This is due to the fact that the swirling flow enhanced convective heat transfer, which increases the efficiency of CO₂ methanation by potentially reducing hotspots that can cause catalyst deactivation³⁹. Spiral-shaped catalysts also demonstrated low pressure drop values between 0.02 and 0.07 mbar, which indicates that the spiral form encourages efficient gas flow. According to the temperature effect study, the carbon balance in Fig. 8c was within a reasonable range of 94–97%.

Aside from the spiral structure and gradient coating of the catalyst, there are other reasons why the gradient-coated spiral-structured Ni/CeO₂ catalyst exhibits high performance of CO₂ methanation. One of the reasons is that even after H₂ reduction, the catalyst (reduced form) retains small Ni crystallite sizes of 28.3 nm, as shown in Table 1, thus improving CO₂ methanation performance²⁵. A further reason is the presence of oxygen vacancies (0.48 of ID/IF_2g), as indicated in Raman spectra in Fig. 6b and Table 4, which enhances CH₄ production by acting as active sites for CO₂ adsorption and activation³⁴, thereby encouraging CO₂ methanation. Another reason is that the catalyst has a rough surface, as illustrated in Fig. 7c, which increases surface area of active sites, promoting CO₂ methanation.

Comparison of catalytic performance between a single-passed and a triple-passed reactor at different flow rates

Figure 9 and Table 5 provide a comparison of the catalytic performance between a single-passed and triple-passed reactor over a spiral shape with a gradient-coated Ni/CeO₂ catalyst at flow rates ranging from 100 ml/min to 3000 ml/min at a set temperature of 350°C. Relevant insights, such as pressure drop (PD), reaction rate (r_m), and temperature profile, were also observed. As mentioned in Fig. 8, a gradient-coated spiral catalyst in a single-passed reactor presented excellent catalytic activity (96.4% CO₂ conversion, and 99.9% CH₄ selectivity) under a feed flow rate of 100 ml/min at 350°C. However, when operating a single reactor at high flow rates, Fig. 9a showed that the CO₂ conversion of a single passing reactor tended to decline as the flow rates increased. This was due to high flow rates during CO₂ methanation over Ni/CeO₂ catalysts with a spiral shape, which can shorten the duration that CO₂ species spend on the catalyst surface³⁹. As a result, high flow rates limit the interaction between CO₂ molecules and active sites, which lowers conversion rates and causes CO intermediates. For example, when increasing the reactant flow rate by 10 times, from 100 to 1000 ml/min, CO₂ conversion dropped from 96.4% to 78.3%, as did CH₄ selectivity, from 99.9% to 98.5% (see Fig. 9b). Catalytic activity in a single-passed reactor showed the same tendency, further declining at higher flow rates (see Table 5). To lower the production of CO intermediate when operating at high flow rates, the steam produced by the CO₂ methanation was eliminated in order to further shift the equilibrium towards complete conversion to CH₄⁴⁰. To improve the efficiency of producing CH₄ in this study, a method of steam trapping from the reaction between reactor stages was employed, with the reactant gas passing through three reactors in a series. In order to compare changes in catalytic performance when removing water, data on CO₂ conversion and CH₄ selectivity were recorded individually for each reactor in the triple-passed reactor, together with the overall pressure drop (see Fig. 9c) and temperature profiles. As seen in Eq. (1), when conducting CO₂ methanation, the water is also produced together with CH₄, and the presence of water vapor can hinder the catalytic process by blocking active sites on the catalyst surface⁴¹. By effectively trapping steam, the catalyst surface remains free from water vapor, allowing for better interaction between the reactants and the catalyst, thus improving the r_m⁴²_. This is consistent with the findings shown in Fig. 9d and Table 5, when eliminating steam from gas products, the r_m in a triple-passed reactor always increased following R3 > R2 > R1 at all flow rates (R1, R2, and R3 represent gradient-coated spiral Ni/CeO₂ catalysts operated in the first, second, and third reactor tubes, respectively, for a triple-passed reactor). The catalytic efficiency of CO₂ methanation and CH₄ selectivity also showed the same trend, R3 > R2 > R1, as shown in Fig. 9a and Table 5, since the effectiveness of steam removal. Sayed et al.⁴² confirmed that eliminating steam can cause the Sabatier reaction’s thermodynamic equilibrium to move to the product side, increasing the conversion of CO₂.

Fig. 9

The CO₂ methanation performance of gradient-coated spiral-structured Ni/CeO₂ catalysts were investigated in single-passed and triple-passed reactors at a set temperature of 350°C, with flow rates varying from 100 to 3000 ml/min as following: (a) CO₂ conversion, (b) Product selectivity, (c) Pressure drop, (d) r_m, and temperature profiles of (e) Single-passed and (f) Triple-passed reactors.

Table 5 List of CO₂ methanation performances of gradient-coated spiral Ni/CeO₂ catalysts in single-passed and triple-passed reactors.

When comparing the efficiency of catalytic performances between single-passed and triple-passing, it was found that at a low feed flow rate (100 ml/min), both types of these two configurations have insignificant differences in catalytic efficiency for CO₂ conversion (96.4% for S, 98.5% for R3) and CH₄ selectivity (99.9% for both S and R3), as shown in Table 5. Furthermore, the temperature profiles of a single-passed and triple-passed reactors are not significantly different at a flow rate of 100 ml/min, as the maximum temperature of spiral catalysts in a single-passed reactor was 364°C, while that of a triple-passed reactor was 372°C.

When CO₂ methanation catalytic activity is operated at higher flow rates (500-3000 ml/min), the temperature profiles of single-passed and triple-passed reactors showed that the thick-coated spiral catalyst at the front side has higher temperatures due to more intense exothermic reactions, compared to the spiral catalyst at the back side (see Table 5). The results of the temperature profile are consistent with previous research, as Najari et al.⁴³ used kinetic models to study the impact of in situ water removal from a methanation reactor. It was demonstrated that eliminating water on-site enhanced CO₂ conversion efficiency; however, this also raises the possibility of hotspots forming within the reactor, which might potentially affect the performance of the catalyst. The heat generated from the front catalyst was carried to the back spiral catalyst by swirling flow, leading to heat convection (see Fig. 9e and 9f). This helped reduce hotspots on the catalysts. Apparently, the temperature profile of a triple-passed reactor was much higher than that of a single-passed reactor. This is due to the exothermic reaction caused by CO₂ methanation in a triple-passed reactor, which is placed side by side in the furnace and promotes a significantly greater temperature than a single reactor at all flow rates. For example, at 3000 ml/min, the maximum temperature of spiral catalysts in a single reactor was 400°C, while that of a triple-passed reactor was up to 499°C. Even operating CO₂ methanation at high flow rates can cause high temperatures. A triple-passed reactor can achieve a CO₂ conversion efficiency of up to 83.4% with CH₄ selectivity at 94.9%, whereas a single-passed reactor achieved 64% with CH₄ selectivity of 87.4%. This is because steam trapping via triple-passing promotes a further shift in equilibrium towards increased CH₄ production and eliminates the issue of insufficient residence time.

It is worth noting that the CO₂ conversion and CH₄ selectivity in R1 of a triple-passed reactor are lower than those in a single-passed reactor, while they should be similar. This is due to the fact that R1 was surrounded by higher temperatures. As mentioned in Fig. 8, high temperatures can cause RWGS (Eq. (2)), reducing CO₂ conversion but promoting CO selectivity. In addition to the high temperature, there is also a pressure drop occurring in the triple-passed reactor that causes R1 to have less catalytic performance than a single reactor. The triple-passed reactor has a larger pressure drop since the feed gas passes through three reactors in series. Meanwhile, a single reactor’s pressure drop progressively rises as the feed flow rate increases as it operates with an open-ended reactor. Table 5 shows that at high feed flow rates between 500 and 3000 ml/min, the r_m of R1 in a triple-passed reactor is always slower than the r_m of a single reactor because of the pressure drop effect in the experimental system. Despite having a higher total pressure drop, the triple-passed reactor demonstrated higher catalytic activity than a single reactor due to improved water removal effectiveness and the elimination of insufficient residence time.

Investigation of spent gradient-coated spiral-structured Ni/CeO₂ catalysts after CO₂ methanation

Figure 10 presents XRD pattern of spent gradient-coated spiral-structured Ni/CeO₂ catalysts obtained from a single-passed reactor and a triple-passed reactor after conducting CO₂ methanation. These spent catalysts were obtained from operating with CO₂/H₂ (1/4) feed flow rates from 100 to 3000 ml/min at 350°C. It should be mentioned that the thick-coated front spiral catalyst (F) from the single-passed and triple-passed reactors was chosen to represent the spent catalyst for each reactor. Regarding the triple-passed reactor, R1, R2, and R3 represent gradient-coated spiral Ni/CeO₂ catalysts operated in the first, second, and third reactor tubes, respectively. The XRD pattern findings indicates no significant phase change in the CeO₂ support structure under reaction conditions, as the CeO₂ peaks remain prominent and stable across all conditions. The CeO₂ crystallite sizes for all spent catalysts are within the range of 9.1–9.4 nm, but they are a bit smaller than that of the reduced catalyst (9.9 nm), as seen in Table 6. This is due to the thermal stability of CeO₂⁴⁴, maintaining its crystallite sizes and preventing collapse of the lattice during CO₂ methanation.

Fig. 10

XRD patterns of spent gradient-coated spiral-structured Ni/CeO₂ catalysts from single-passed reactor and triple-passed reactor after CO₂ methanation (CO₂/H₂ (1/4) feed flow rates ranging from 100 ml/min to 3,000 ml/min at 350°C).

Table 6 Comparison of NiO, Ni, and CeO₂ crystallite sizes between calcined, reduced, and spent gradient-coated spiral-structured Ni/CeO₂ catalysts after CO₂ methanation (CO₂/H₂ (1/4) feed flow rates ranging from 100 to 3000 ml/min at 350°C).

Prior to H₂ reduction, the calcined Ni/CeO₂ catalyst exhibits diffraction characteristic NiO peaks. However, NiO peak is reduced by H₂ to form metallic Ni after H₂ reduction, as no diffraction characteristic peaks are observed in the reduced Ni/CeO₂ catalyst, seen in XRD results in Fig. 3. Apparently, the NiO peak at 28.5° of 2θ appeared again in all spent Ni/CeO₂ catalysts after conducting CO₂ methanation (see Fig. 10). This is because CO₂ methanation produces water as a byproduct at high temperatures, water can oxidize metallic Ni to NiO, leading to the appearance of NiO peak in the post-reaction. This oxidation process can affect the catalyst’s performance, as NiO is less active in CO₂ methanation compared to metallic Ni⁴⁵. However, Table 6 shows that all spent catalysts of both single-passed and triple-passed reactors displayed NiO crystallite sizes at the range between 12.0 and 13.9 nm, which are not that large if compared to that of the calcined Ni/CeO₂ (21.6 nm). To compare metallic Ni crystallite sizes of spent catalysts after CO₂ methanation, it was found that Ni crystallite sizes from a triple-passed reactor (32.0 nm for spent-R1 and 33.4 nm for spent-R3) are greater than those from a single-passed reactor (29.0 nm), as seen in Table 6. This is due to Ni sintering at high temperatures, especially in the triple-passed reactor (see temperature profiles in Fig. 9e, f). When conducting CO₂ methanation at the set temperature of 350°C, the actual temperatures are much higher than that set temperature as CO₂ methanation is an exothermic reaction.

Figure 11a, b presents N₂ adsorption–desorption isotherms, and pore size distributions of pristine CeO₂ and Ni/CeO₂ catalysts (calcined and reduced forms), together with spent-R1 Ni/CeO₂. To compare the porosity of spent Ni/CeO₂ with that of calcined and reduced forms, it should be noted that after CO₂ methanation (at feed flow rates of CO₂/H₂ (1/4) from 100 to 3000 ml/min at 350°C), the thick-coated front spiral catalyst (F) from the first reactor tube in a triple-passed reactor was selected to represent the spent Ni/CeO₂ catalyst in this study for Fig. 11.

Fig. 11

(a) N₂ adsorption–desorption isotherms, and (b) pore size distributions of pristine CeO₂ and Ni/CeO₂ catalysts in different forms (calcined and reduced forms), compared with spent-R1 Ni/CeO₂ catalyst after performing CO₂ methanation in triple-passed reactor (CO₂/H₂ (1/4) feed flow rates ranging from 100 to 3000 ml/min at 350°C).

As mentioned earlier in Fig. 5 and Table 2, The N₂ adsorption–desorption isotherms of pristine CeO₂ have a higher adsorption capacity than those of calcined and reduced Ni/CeO₂ because adding Ni to pristine CeO₂ may change the pore structure and S_BET, leading to lower N₂ adsorption capacity. Similarly, Fig. 11a shows that the isotherm of the spent Ni/CeO₂ catalyst shows the lowest adsorption capacity, with the lowest S_BET at 69.24 m²/g, (see Table 7). The significant loss of adsorption capacity and S_BET can be attributed to sintering of Ni particles and carbon deposition during the CO₂ methanation when operating at high flow rates. Figure 11b shows that the spent Ni/CeO₂ catalyst has a significantly larger average pore size of 8.5 nm when compared to other Ni/CeO₂ forms. This reveals structural damage, which could be caused by Ni particle sintering, pore collapse, or carbon formation during CO₂ methanation⁸.

Table 7 Physicochemical properties of pristine CeO₂ and Ni/CeO₂ catalysts in different forms (reduced, calcined, and spent forms).

Conclusion

The gradient coating approach was effectively used for fabricating spiral-structured Ni/CeO₂ catalysts. The gradient coating transferred heat energy from a thick-coated front catalyst to a thinner-coated back catalyst, reducing hotspots as well as improving mass transfer through swirling flow. Single-passed and triple-passed operations were used to compare the catalytic performance of gradient-coated spiral-structured Ni/CeO₂ catalysts with and without steam trapping. When operating at a low input flow rate (100 ml/min), a triple-passed reactor achieved slightly better CO₂ conversion efficiency of 98.5% and CH₄ selectivity of 99.9% than a single-passed reactor, which showed 96.4% CO₂ conversion and 99.9% CH₄ selectivity, respectively. It should be noted that both types of these two configurations exhibited insignificantly different catalytic performances. This was due to the fact that a low flow rate produced similar temperature profiles, a small amount of pressure drop, and no residence time limits. On the other hand, CO₂ methanation is carried out at higher flow rates (500-3000 ml/min), and a triple-passed reactor performed better in terms of CO₂ conversion and CH₄ selectivity compared to a single-passed reactor. For instance, at 3000 ml/min, a triple-passed reactor can reach up to 83.4% CO₂ conversion efficiency along with 94.9% CH₄ selectivity, while a single-passed reactor only achieved 64% CO₂ conversion with 87.4% CH₄ selectivity. Increased CO₂ conversion and CH₄ selectivity are results of effectively trapping steam in a triple-passed reactor, which keeps the catalyst surface free from water vapor and allows for more contact between the reactants and the catalyst. A triple-passed reactor with steam trapping at high flow rates not only eliminates the insufficient residence time but also promotes a further shift in equilibrium toward increased CH₄ production.