Hybrid Plasma-Catalytic CO₂ Dissociation over Basic Metal Oxides Combined with CeO₂

Abstract

The problem of CO₂ waste in the atmosphere is a major concern, and methods of CO₂ utilization are being currently developed. In the present work, a plasma-catalytic process is applied for CO₂ dissociation. A series of MgO and CeO₂-containing catalysts were synthesized, and the samples were characterized by: a low-temperature N₂ adsorption-desorption analysis, X-ray diffraction, X-ray photoelectron spectroscopy, temperature-programmed desorption of CO₂, and X-ray fluorescence spectroscopy. It was stated that under dielectric barrier discharge conditions, the catalyst surface, composition, and phase content remain unchanged. The superior catalytic activity of the MgCe-Al sample is attributed to the combination of weak basic sites and oxygen vacancies on the catalyst surface.

1. Introduction

The increasing concentration of carbon dioxide (CO₂) in the atmosphere is recognized as one of the major contributors to global warming and climate change. The current levels of CO₂ are unprecedented in human history, and they have been increasing at an alarming rate over the past few decades [1]. The consequences of this trend are already visible in the form of rising sea levels, more frequent extreme weather events, and biodiversity loss, among others [2]. Therefore, finding efficient ways to reduce CO₂ emissions and mitigate its harmful effects has become a major challenge for scientists globally.

One promising approach is to use CO₂ as a feedstock to produce valuable chemicals, fuels, and materials through various chemical transformations. This concept, known as carbon capture and utilization (CCU), has gained significant attention in recent years due to its potential to address both the environmental and economic challenges associated with CO₂ emissions.

Traditional methods for CO₂ utilization include chemical and biological processes, such as chemical absorption [3], catalytic hydrogenation [4], and microbial conversion [5]. However, these methods suffer from low conversion efficiency, high consumption of energy, and limited scalability. Therefore, there is a need to explore new and innovative approaches that can overcome these limitations and enable the sustainable utilization of CO₂.

Plasma-catalytic decomposition of CO₂ is a promising alternative, which combines the advantages of both plasma and catalysis to achieve high conversion rates and selectivity. Plasma can provide the high-energy electrons, ions, and radicals needed to activate CO₂ molecules [6,7,8], while catalysts can enhance the adsorption, activation, and reaction of CO₂ on their surfaces [9,10,11]. Moreover, plasma-catalytic systems can be operated at ambient conditions, making them more energy-efficient and environmentally friendly than traditional methods.

Barrier discharge or dielectric barrier discharge (DBD) is commonly used for the plasma-assisted catalytic decomposition of CO₂ due to the simplicity of the reactor and low operating temperature (from 20 to 150 °C) [12,13,14]. Such a plasma-catalytic setup consists of two coaxial electrodes, between which the electric discharge is generated [15,16]. The catalytic material is packed between the electrodes (in the discharge zone), so the plasma and catalyst may interact with each other. The micro-discharges, which are generated between the catalyst particles, may modify the catalyst properties [17]. To achieve efficient plasma-catalytic decomposition of CO₂, various catalysts and dielectric materials have been applied to barrier discharge. For instance, CO₂ conversion was enhanced in the presence of BaTiO₃ or CaTiO₃ [18,19,20,21], which is explained by the higher plasma electron density inside the packed bed. In fact, such materials play a more “discharge modifier” role rather than “catalytic”. In a traditional meaning, two main groups of substances are attractive as potential catalysts. One of them is a group of metal oxides possessing oxygen vacations in the structure, e.g., CeO₂, which has gained much attention as a catalyst due to its crystal defects and oxygen storage capacity [22]. Such properties make it possible to conduct various catalytic reactions involving CeO₂, such as methanol oxidation [23], CO₂ reduction [24], water splitting [25], various photocatalytic reactions [26], diesel soot abatement [27], and many more. When introduced inside the CeO₂-packed discharge zone during plasma-catalytic CO₂ dissociation, the carbon dioxide molecule is activated by plasma electrons. Then the molecule is adsorbed by the oxygen vacancy V_o of the CeO₂ catalyst with consequent C=O bond cleavage producing CO. The oxygen atom adsorbed on the vacancy is then desorbed from the catalyst surface; the catalytic cycle is repeated [28].

Another type of catalytic substance for plasma-catalytic CO₂ dissociation is metal oxides with basic properties (e.g., MgO, CaO). Since CO₂ is an acidic oxide, it reacts readily with molecules possessing basic properties. Basic sites of MgO and CaO may enhance the CO₂ chemisorption on the catalyst surface and therefore raise the plasma-catalytic CO₂ conversion [29,30,31].

A combination of two types of oxides (CeO₂ with oxygen vacancies and basic metal oxides) in a single catalyst may be useful, as it would enhance the CO₂ adsorption on the catalyst surface with subsequent decomposition. However, no study on the synthesis and plasma-catalytic properties has been reported in the literature currently. To reveal the possible synergetic effect for plasma-catalytic CO₂ decomposition, we studied combined MgO–CeO₂ and CaO–CeO₂ catalysts. Our group has conducted a preliminary study, which showed the advantage of such a catalyst compared with a mono-oxide catalyst [32]. In the present work, we aim to investigate the plasma-catalytic CO₂ decomposition process in the presence of MgO and CaO-promoted CeO₂ catalysts. The main goal of this study is to establish the key characteristics, which determine the activity and stability of such catalysts.

2. Materials and Methods

2.1. Materials

All reagents were purchased from commercial suppliers and used without further purification. Cerium nitrate (LLC “Tsentr Tekhnologyi Lantan”, 99%), magnesium acetate (JSC “Vekton”, 99%), and calcium nitrate (JSC “Lenreaktiv”, 98.5%) were used as metal oxides precursors. Metal salt solutions were impregnated on a γ-Al₂O₃ support (0.63–1.00 mm fraction), which was obtained by the extrusion of Pural SB boehmite (Sasol, 99%). CO₂ (JSC “Moscow Gas Refinery Plant”, 99.5%) was used as a feed, and Ar (JSC “Moscow Gas Refinery Plant”, 99.993%) was used as a carrier gas in the gas chromatograph.

2.2. Catalysts Preparation

Catalysts were prepared using the wetness impregnation technique. In a typical synthesis, 2.29 g of Ce nitrate was dissolved in 7.3 cm³ of distilled water and added to the 8.1 g of γ-Al₂O₃ support. The impregnated support was dried at 60 °C for 2 h, 80 °C for 2 h, 90 °C for 2 h, 100 °C for 2 h, and then calcined in a muffle furnace at 400 °C for 3 h. Catalysts with two metal oxides (MgO–CeO₂ and CaO–CeO₂) were prepared using a two-step impregnation procedure. In the first step, CeO₂/Al₂O₃ was prepared similarly to the technique described above. Then, CeO₂/Al₂O₃ was impregnated with a solution of a second metal salt with further drying and calcination. The catalysts prepared were denoted as follows (

Table 1. The denotation of prepared catalysts.

Catalyst Composition	Denotation
Al2O3	Al
CeO2 (10 wt.%)/Al2O3	Ce–Al
CeO2 (20 wt.%)/Al2O3	Ce20–Al
MgO (10 wt.%)/Al2O3	Mg–Al
MgO (10 wt.%)–CeO2 (10 wt.%)/Al2O3	MgCe–Al
CaO (10 wt.%)–CeO2 (10 wt.%)/Al2O3	CaCe–Al

).

2.3. Catalysts Characterization

The surface characteristics (S_BET, V_pores, d_pores) were determined using a Belsorp miniX (Microtrac MRB) instrument. Before analysis, the samples were degassed at 300 °C and 10 Pa for 8 h. To calculate the surface area, the BET method was applied with adsorption data in the range of relative pressures (P/P₀) of 0.05−0.20. The total pore volume was calculated based on the amount of adsorbed nitrogen at a relative pressure P/P₀ = 0.95.

Powder X-ray diffraction (XRD) was used to determine the phase composition of the catalysts. The X-ray diffractograms were obtained for a range of 10–100° 2θ by using a Rigaku Rotaflex RU-200 diffractometer (CuK_α radiation) equipped with a Rigaku D/Max-RC goniometer (a rotation speed of 1°/min; a step 0.04°). The identification of diffractograms was carried out using the PDF-2 ICDD database of powder diffraction patterns.

The average size (D) of crystallites was calculated by the Scherrer equation:

$$D=\frac{K\lambda }{\beta \cos \mathsf{\theta }}$$

(1)

where D is the crystallite size, λ is the wavelength of the Cu–Kα radiation, K is a constant and its value is taken as 0.9, θ is the diffraction angle [rad], and β is the full-width at half maximum (FWHM) [rad].

The X-ray photoelectron spectroscopy (XPS) measurements were performed using a «PREVAC EA15» electron spectrometer. In the current work, AlKα (hν = 1486.74 eV, 150 W) was used as a primary radiation source. The pressure in the analytical chamber did not exceed 5 × 10⁻⁹ mbar during the spectra acquisition. The binding energy scale was pre-calibrated using the positions of Ag3d5/2 (368.3 eV) and Au4f7/2 (84.0 eV) from silver and gold foils, respectively. The peaks were deconvoluted using PeakFit software set to the Shirley background subtraction, followed by peak fitting to Voigt functions with an 80% Gaussian and 20% Lorentzian character.

The content of elements (Al, Si, Ce, Mg, Ca) was measured with an X-ray fluorescence spectrometer: ARL Perform’x Sequential XFR (Thermo Fisher Scientific, Waltham, MA, USA) equipped with 2500 W X-ray tube. Before analysis, the samples were ground and pressed into a tablet with H₃BO₃.

The basicity of the catalysts was determined by the CO₂-temperature programmed desorption (TPD-CO₂) method using USGA-101 (LLC “Unisit”) equipment. The catalyst sample was placed into a quartz reactor and treated in a flow of He (JSC “Moscow Gas Refinery Plant”, 99.995%) at 512 °C for 40 min to remove water and oxygen from the catalyst surface. A saturation was performed with CO₂ (5% CO₂–95% He, LLC “NII KM”) at 60 °C for 24 min. The physically adsorbed CO₂ was removed at 102 °C in the flow of He (30 mL/min) for 60 min. An analysis was performed in the flow of He in the temperature range of 100–800 °C (heating rate 7 °C/min). The registration of desorbed CO₂ was carried out with a thermal conductivity detector.

2.4. Plasma-Catalytic Experiments

The investigation of catalytic activity was carried out using an experimental plasma-catalytic unit (Figure 1). A quartz tube (16 mm diameter, 2 mm wall thickness) was used as a reactor. A steel rod with a screw thread (8 mm diameter) was applied as an inner (high voltage) electrode. A stainless steel mesh (0.5 mm mesh size, 80 mm length) was placed on the outer wall of the quartz tube and served as a ground electrode. The discharge gap was 4 mm. The catalyst was placed into the reactor and fixed with mineral wool on both ends of the catalyst bed. The ceramic beads were used as an inert packing material for comparison with the catalysts prepared. Carbon dioxide was injected into the reactor using a mass flow gas controller RRG-20 (LLC “Eltochpribor”, Russia). The CO₂ flow rate was kept at 17 mL/min for all experiments. A high voltage generator with 23 kHz frequency was used as a power source. The discharge voltage and current were registered with a TDS 2012B oscilloscope (Tektronix, Beaverton, OR, USA). Plasma-absorbed power was calculated from the area of the Lissajous figure and was 5–9 W depending on the catalyst type being put inside the reactor. Gaseous products were analyzed using a gas chromatograph PIA (LLC “NPF MEMS”, Samara, Russia), with a thermal conductivity detector, and equipped with Hayesep N adsorbent column (l = 2 m) and molecular sieves 13Å column (l = 2 m).

The conversion of CO₂ (X) was calculated as

$$\mathrm{X}\left({\mathrm{CO}}_2\right)\left(\%\right)=\frac{\mathsf{\nu }{\left({\mathrm{CO}}_2\right)}_{\mathrm{inlet}}-\mathsf{\nu }{\left({\mathrm{CO}}_2\right)}_{\mathrm{outlet}}}{\mathsf{\nu }{\left({\mathrm{CO}}_2\right)}_{\mathrm{inlet}}}\times 100\%$$

(2)

where ν(CO₂)_inlet is the quantity of CO₂ which is put into the reactor, ν(CO₂)_outlet is the quantity of CO₂ in the gas sample.

Plasma-absorbed power was calculated from the Volt-Coulomb characteristic (Lissajous figure) of the discharge as [33,34]:

$$P=fW=f{{\displaystyle \int }}_{t_0-\frac{T}{2}}^{t_0+\frac{T}{2}}u\left(t\right)i\left(t\right)dt=f{C}_n{{\displaystyle \int }}_{t_0-\frac{T}{2}}^{t_0+\frac{T}{2}}u\left(t\right)d{u}_c\left(t\right)=f{C}_{n}A$$

(3)

where u(t) is the discharge voltage, i(t) is the discharge current, C_n is the value of the capacitor included in series with the discharge tube, u_c(t) is a voltage on the C_n, T is the period of the applied voltage, f is a frequency of the applied voltage, A is the area of the Lissajous figure.

The energy efficiency of the process was calculated as

$$\eta \left(\%\right)=\frac{\mathrm{X}\left(\%\right)G_{in}^{{\mathrm{CO}}_2}\Delta H_{298,K}^0}{P}\times 100\%,$$

(4)

where $G_{in}^{{\mathrm{CO}}_2}$ is inlet CO₂ flow rate [mol/s], $\Delta H_{298,K}^0$ is the enthalpy of C=O bond cleavage, which is 283 kJ/mol.

3. Results and Discussion

3.1. Catalyst Characterization

The catalysts obtained were characterized by physicochemical methods. The composition of the catalysts was analyzed using XRF spectroscopy (

Table 2. The content of oxides in the samples and the textural characteristics of the prepared catalysts.

Sample	Oxide Content, wt.%					Textural Characteristics
	Al2O3	MgO	CeO2	SiO2	CaO	SBET, m2/g	Vpores, cm3/g	dpores, nm
Al	99	-	-	1	-	202	0.50	8.6
Ce–Al	88.7	-	10.5	0.8	-	190	0.43	8.5
Mg–Al	87.2	12	-	0.8	-	181	0.43	8.1
MgCe–Al	75.7	11.8	11.7	0.8	-	161	0.33	7.6
Ce20–Al	79.3	-	19.7	1	-	169	0.37	8
CaCe–Al	78.5	-	11.8	0.5	9.2	140	0.30	7.5

). It is stated from the analysis results that the content of the oxide in the samples is close to the calculated data.

The textural characteristics were determined by a low-temperature N₂ adsorption-desorption analysis. The adsorption isotherms (Figure 2a) correspond to type IV and possess a hysteresis loop (in the range of p/p₀ = 0.6–0.9), which means mesopores are present in the sample. The surface area of the catalysts decreases when the metal oxides are introduced, and so does the pore volume (Table 2). This can be attributed to partial pore blockage with metal oxide particles. It should be noted that the catalysts possess the unimodal narrow pore size distribution (Figure 2b), which is preserved with different oxide loading.

The phase composition of the catalysts was determined using an X-ray diffraction analysis. It is seen from Figure 3 that the CeO₂ phase in the samples is of a high crystallinity degree (narrow diffraction peak at 2θ = 28.6°). There are no mixed or double oxides present in the samples.

The surface of the Ce–Al and MgCe–Al samples was also analyzed by X-ray photoelectron spectroscopy (Figure 4). Raw spectra were deconvoluted to obtain information about the Ce oxide composition. The peaks assigned u‴ (917 eV), u″ (907 eV), u (900.5 eV), v‴ (898 eV), v″ (888 eV), and v (882.1 eV) correspond to the CeO₂, while u′ (903 eV) and v′ (885 eV) correspond to Ce₂O₃ [35]. It is stated from the XPS analysis that the catalysts prepared contained mostly CeO₂ oxide (83% for Ce–Al and 84% for MgCe-Al). Deconvoluted peak characteristics (peak position, full width at half maximum, and peak area) are demonstrated in Table S1.

3.2. The Plasma-Catalytic Performance

The plasma-catalytic CO₂ decomposition tests were carried out in a continuous gas flow mode. The input power was calculated based on the Lissajous figures obtained (Figure 5). It can be seen from Figure 5 that in the absence of any packing material inside the discharge zone, the Lissajous figure is broad, and it shrinks when the inert beads or catalyst are packed inside. This can be related to the difference in dielectric properties of the materials.

At the beginning of each plasma-catalytic process, some amount of condensed liquid was observed at the outlet end of the quartz tube. This liquid is attributed to the water which is absorbed by hygroscopic MgO and Al₂O₃ substances and then evaporated during the first 60 min of run (Figure 6). When all of the adsorbed water is evaporated, the steady-state CO₂ decomposition process is observed.

The dissociation of the CO₂ molecule can be accomplished either with CO formation or C formation as the main carbon product. However, under dielectric barrier discharge plasma conditions, solid carbon formation is quite unlikely, as this reaction requires extremely high temperatures (6000–7000 °C) [36]. To prove that CO was the main product of CO₂ decomposition, the [CO]/[O₂] ratio was calculated from the gas chromatography results. The calculated ratio was close to 2 (1.9–2.1) which corresponds to only CO formation during CO₂ dissociation.

During the plasma-catalytic experiments, the catalysts were compared by the characteristics of the CO₂ conversion degree (X) and energy efficiency (η) (Figure 7a,c). It could be seen from the results that the lowest CO₂ conversion and energy efficiency values are achieved in the absence of any packing material in the reactor. The introduction of inert ceramic beads changes the electric properties of the barrier discharge (Figure 5) but does not lead to a significant increase in CO₂ conversion (Figure 7a). The highest CO₂ conversion values are reached in the presence of Ce–Al, Ce20–Al and MgCe–Al samples. The catalyst stability test is performed using the MgCe–Al sample (Figure 7d). It is seen that the conversion of CO₂ reaches a maximum at 75 min of run and does not decrease significantly up to 120 min. The energy efficiency was calculated based on the CO₂ conversion/input power ratio. In the presence of MgCe–Al, the highest energy efficiency is achieved which means less energy input is needed for a higher conversion value. The highest performance of CO₂ decomposition in the presence of the MgCe–Al sample could be proof of the initial hypothesis of MgO and CeO₂ properties combined in a single catalyst. However, in the presence of CaCe–Al (which also contains basic oxide CaO), the conversion of CO₂ is even lower than that in the presence of Al₂O₃, without any additional metal oxides. To explain this, additional results were obtained that will be discussed in the next section.

To investigate the reusability of the catalysts, the MgCe-Al sample was chosen as the catalyst, which showed the highest CO₂ conversion activity. The reusability tests were carried out in the same reactor for five runs without any pretreatment or regeneration of the catalyst before each run. The results (

Table 3. The reusability tests of the MgCe-Al sample during five cycles.

No. of Run	X (CO2), %	Energy Efficiency, %
1	18.3	14.4
2	17.8	11.8
3	18.6	13.0
4	18.0	13.3
5	18.3	13.5

) show that the activity of the catalyst does not change significantly during five tests, so the catalyst can be used several times without any regeneration.

The plasma-catalytic experiment results obtained were compared with previously published data on CO₂ decomposition (

Table 4. A comparison of the DBD plasma-catalytic CO₂ decomposition using Ce and Mg catalysts.

Catalyst	Input Power, W	Energy Efficiency, %	X (CO2), %	Source
Mo/CeO2	13.5	14.3	23.2	[28]
MgO	60	6.1	24	[31]
MgAl–LDH	60	5.8	20
CaO	60	4.8	19
ZrO2–CeO2	50	2.4	64	[37]
5 wt.% Fe2O3–5 wt.% CeO2/Al2O3	15	10.2	24.5	[38]
10 wt.% CeO2/Al2O3	15	15.7	28.5
CeO2 coated on frosted quartz	26.5	2.1	26.3	[39]
CeO2-cubes	15–20	3.8	16	[40]
CeO2-rodes	15–20	3.7	16
CeO2-hexagons	15–20	3.1	20
Ce–Al	6.6	8.5	15.8	This work
Ce20–Al	5.6	10.6	16.6	This work
MgCe–Al	5.1	13	18.6	This work
CaCe–Al	8.9	5.2	13.2	This work

). It should be noted that a direct comparison of the experiment results is quite difficult because the operating parameters (power, CO₂ flow rate) and reactor design (discharge gap, electrode materials, etc.) are dissimilar in each study. However, it can be seen that under significantly lower input power (5.1–8.9 W), comparable conversions are achieved in the presence of our catalytic systems. It should be noted that in the work [31] the CaO sample showed an inferior plasma-catalytic CO₂ performance, which is consistent with our study. However, no explanation of this phenomenon was provided in that study.

3.3. Catalyst Characterization after the Plasma-Catalytic Reaction

The catalysts after the reaction (spent catalysts) were characterized by physicochemical methods to estimate how the plasma-catalytic CO₂ decomposition process affects the properties of the catalysts. When comparing the adsorption isotherms (Figure 8), it is clear that the textural characteristics do not change after the reaction. The values of S_BET, V_pores, and d_pores of the spent catalysts (

Table 5. The oxide content and textural characteristics in the spent catalysts.

Sample	Oxide Content, wt.%					Textural Characteristics
	Al2O3	MgO	CeO2	SiO2	CaO	SBET, m2/g	Vpores, cm3/g	dpores, nm
Ce–Al	88.5	-	11	0.5	-	184	0.43	8.6
Mg–Al	87	12.5	-	0.5	-	178	0.40	8.1
MgCe–Al	77	12.8	9.4	0.8	-	171	0.35	7.7
Ce20–Al	79	-	20.5	0.5	-	165	0.36	7.1
CaCe–Al	78	-	12.1	0.5	9.4	127	0.24	6.7

) are close to their initial ones (Table 2) and are within the error of the method. The oxide content (Table 5) corresponds to those of the catalysts prepared, which means no carbon deposition occurred during the CO₂ decomposition.

The phase content also remains unchanged after the reaction (Figure 9). As can be seen from the diffractograms, spent catalysts did not contain any new phases; the intensity of the diffraction patterns is comparable with that of the catalysts before the reaction. The average size of the crystallites in the catalysts before and after the reaction was calculated using the Scherrer equation [41]. It is seen from

Table 6. The average crystallite size of the fresh and spent catalysts was determined from the XRD analysis.

	Sample
	Mg–Al		Ce–Al		MgCe–Al		Ce20–Al
	Fresh	Spent	Fresh	Spent	Fresh	Spent	Fresh	Spent
MgO crystallite size, nm	1.5 ± 0.2	1.4 ± 0.1	-	-	2.1	2.2	-	-
CeO2 crystallite size, nm	-	-	3.3 ± 0.3	3.6 ± 0.5	3.1 ± 0.4	3.1 ± 0.5	4.5 ± 0.8	4.6 ± 0.7

that the average size of the crystallite remained unchanged for all of the phases. It indicates that the crystallites did not aggregate under mild DBD conditions.

The oxidation state of the Ce in the Ce–Al and MgCe–Al spent samples was analyzed by Ce 3d XPS (Figure 10). It was observed that the CeO₂ state is still predominant, but the Ce₂O₃ state quantity is enhanced compared to that found in the catalysts before the reaction (peak position, full width at half maximum, and peak area of the deconvoluted spectra are provided in Table S2). Based on the Ce³⁺ peak area proportion in the spectrum, the oxygen vacancies content was calculated (Ce³⁺/(Ce³⁺ + Ce⁴⁺) [42]. It was revealed that the oxygen vacancies content in the fresh Ce-Al and MgCe-Al catalysts were 0.17 and 0.16, correspondingly, and were enhanced after the reaction (0.27 for spent Ce–Al and 0.29 for spent MgCe–Al). These results indicate the oxygen state modification during the plasma-catalytic process, which helps efficient CO₂ adsorption with subsequent decomposition.

To investigate the influence of the MgO or CaO addition to the CeO₂/Al₂O₃ samples, a basicity analysis was carried out using the TPD-CO₂ technique. It is seen in Figure 11 that TPD-CO₂ patterns include two desorption peaks. The low-temperature peak corresponds to the weak basic sites, while the high-temperature peak corresponds to the strong basic sites. From

Table 7. The quantity of the basic sites, as determined by TPD-CO₂.

Sample	Basic Sites Concentration, mkmol/g
	Weak (50–250 °C)	Strong (500–700 °C)	Total
Al	35	-	35
Ce–Al	32	53	85
Mg–Al	69	96	165
MgCe–Al	66	52	118
Ce20–Al	51	-	51
CaCe–Al	83	190	273

, it is clear that Mg-Al and CaCe–Al are considered as the samples possessing the highest basicity. However, their plasma-catalytic activity is the lowest, which is inconsistent with their basicity. The probable explanation can be made based on the strong basic site concentration. The desorption of CO₂ from the strong basic sites occurs at a high temperature, which cannot be reached during the dielectric barrier discharge conditions. It means that under plasma-catalytic conditions better CO₂ conversion is achieved using catalysts with moderate basicity. This observation is per the work [43] where MgO-based sorbents for CO₂ capture are demonstrated as better candidates compared to CaO, due to the lower capture temperature.

Thus, from a practical point of view, combined MgO–CeO₂ catalysts may be applied for CO₂ decomposition using plasma catalysis. However, this process certainly needs further research and development.

4. Conclusions

The present work shows that the combination of MgO and CeO₂ in a single catalyst is beneficial for plasma-catalytic CO₂ decomposition under mild conditions. The dominant role in the CO₂ dissociation process is assigned to the CeO₂ rather than the MgO. When the single oxide catalysts are compared (Mg–Al vs. Ce–Al), the catalyst with CeO₂ shows better plasma-catalytic performance (CO₂ conversion is 11.1% and 15.8% in the presence of Mg–Al and Ce–Al, correspondingly). Despite this, the MgO and CeO₂ combination shows superior CO₂ conversion due to the increase in weak basic sites, which facilitate CO₂ adsorption on the catalyst surface. The CaO-CeO₂ combination in a catalyst leads to a decrease in CO₂ conversion and energy efficiency, which can be related to the presence of strong basic sites from which CO₂ cannot be desorbed under reaction conditions.

The results obtained in this work may help in further developing efficient plasma-catalytic CO₂ decomposition methods. The practical impact of this work consists of a search for appropriate catalyst design. The MgO–CeO₂ combination in the catalyst for such a process seems promising and needs to be more deeply investigated. For example, different MgO–CeO₂ molar ratios will be tested in CO₂ decomposition, as well as bulk metal oxides (without Al₂O₃ support) in future work.