Catalytic Reduction of SO₂ with CO over LaCoO₃ Perovskites Catalysts: Effect of Fe Doping and Pre-Sulfurization

Abstract

SO₂ emissions are a major source of air pollution, and the catalytic reduction of SO₂ to elemental sulfur by CO represents a promising solution. This study investigates the effects of Fe doping and pre-sulfurization on the catalytic performance of LaCoO₃ perovskite catalysts. A series of Fe-doped LaCoO₃ perovskites were synthesized via the sol–gel method and evaluated for the catalytic reduction of SO₂ by CO. The results showed that LaCo_0.8Fe_0.2O₃ exhibited the highest catalytic performance, achieving 84.0% SO₂ conversion at 500 °C. The oxygen-free sulfurization (OFS) treatment compared with oxygen-assisted sulfurization (OAS) treatment significantly enhanced the activity, reaching a SO₂ conversion of 95.9% from 80.0% at 450 °C with the lower byproduct generation. Characterization analyses demonstrated that the OFS treatment facilitated the formation of active sulfur species and oxygen vacancies on the catalyst surface while also enhancing the adsorption capacity of the catalyst for the reactant gases. These factors were identified as key contributors to the improved catalytic performance, driven by the combination of redox and carbonyl sulfide (COS) intermediate mechanism. The findings suggest that the OFS treatment is an effective strategy to improve the catalytic reduction of SO₂ by CO, offering a more environmentally friendly solution for SO₂ emission control through resource utilization.

1. Introduction

Due to the characteristics of China’s energy structure, sulfur dioxide (SO₂) emissions from coal-fired flue gas represent the primary source of SO₂ pollution [1,2]. Despite the widespread use of conventional desulfurization technologies, like absorption, adsorption, and catalysis, the reduction of SO₂ emissions continues to face significant challenges, including secondary pollution and operational complexity [3,4]. Catalytic reduction of SO₂ to sulfur in the reducing atmosphere (e.g., CO, CH₄) of flue gas not only facilitates effective SO₂ removal but also converts it into sulfur, a valuable chemical raw material for various industrial productions, like fertilizers, chemicals, and pharmaceuticals [5,6]. The technology, which promotes the resource utilization of SO₂, supports both flue gas desulfurization and the recycling of valuable resources, making it of critical importance for sustainable development [7,8]. Furthermore, the investigation of the single-stage catalytic converter is a key step in advancing the technology. Catalytic materials derived from metal and rare-earth are extensively investigated in SO₂ reduction. Ng reviewed dry-based and wet-based catalytic SO₂ reduction technologies [8]. Lum summarized SO₂ reduction for environmental sustainability and circular economy [9]. Sepehrian recently reviewed SO₂ dry-based catalytic removal from flue gas leading to elemental sulfur production, encompassing various reductants employed, reduction mechanisms, catalyst types, methods for catalyst improvement, and diverse operational conditions [10]. Up to date, various reducing agents, such as carbon monoxide, hydrogen, methane, and carbon materials, have been investigated for SO₂ reduction reactions [10,11,12,13]. CO, due to its low cost and widespread application in various industries, is a common gas used as a reducing agent in the catalytic conversion of SO₂. Converting SO₂ to sulfur with high efficiency, achieving a pure product, producing fewer byproducts, and requiring lower temperatures compared to other reductants are some of the main features of using CO reduction [10]. Catalytic SO₂ reduction to sulfur via transition metal-based catalysts (like Cu, Ni, Ce loaded SiO₂, γ-Al₂O₃, TiO₂, and ZrO₂) [6,14,15] and sulfided metal-based catalysts (like FeS, ZnS, Co₉S₈^-N@NC) [7,16,17] appeared to be highly promising for high SO₂ conversion while maintaining high sulfur selectivity. Ren reported a Fe-loaded catalyst derived from blast furnace slag (BFS) for the CO-SO₂ reaction to produce sulfur due to the strong adsorption of BFS and the production pathway of H₂S [18]. Cao investigated that the highest Ce³⁺ content, oxygen vacancies content, and Ir⁰ content over several Ir/CeO₂ catalysts exhibited the strongest Ir-O-Ce interaction and caused better catalytic activity on the reduction of SO₂ by CO [11]. Tian found the effect of temperature on the reaction path of pyrite (FeS₂)-based catalyst catalyzed CO reduction of SO₂ to sulfur due to the S-vacancy formation [7]. The activity selectivity of sulfur over the above-reported catalysts was mainly tested with the absence of O₂, which may have shown better catalytic performance. However, conventional catalytic desulfurization for flue gas normally would turn to inactivation with oxidation by the existence of O₂.

Owing to the diverse chemical properties and high electron mobility, metal oxide perovskite materials with an ABO₃ structure have been widely studied in various catalytic reactions, like environmental pollution remediation [19,20,21,22,23,24], including the catalytic reduction of SO₂ by CO [25,26]. Partial substitution of A- or B-site cations can effectively promote lattice distortion and enhance the mobility of oxygen species, thereby improving the catalytic activity toward reaction gas, like H₂S oxidation and SO₂ reduction [26,27]. Moreover, the unique structure of perovskites facilitates the formation of sulfide species during the SO₂ reduction process, which generates a synergistic effect [28]. For instance, Hibbert observed that during the reduction of SO₂ with La_1−xSr_xCoO₃, the catalyst was fully converted into various metal sulfides when high catalytic activity was stabilized, indicating that sulfides serve as the active phase [28]. Additionally, catalyst sulfurization has also been shown to improve catalyst performance. For example, Ma found that the sulfides formed during the sulfurization process of LaCoO₃ play a crucial role in promoting CO reduction of SO₂ [29]. Wang investigated Cu-doping LaCoO₃ in a pre-sulfurized process, given better SO₂ reduction performance in the presence of oxygen due to the decreasing formation of carbonyl sulfide (COS) intermediate product [26]. A metal sulfide phase (MeS) is formed on the metal oxide catalysts, which were subjected to the sulfur-containing gases flow under certain temperature conditions. The performance of various pre-sulfided metal oxide catalysts such as Fe, Mo, Ni, and Co demonstrated that the formation of FeS₂, MoS₂, NiS₂, and CoS₂ metal sulfide active phases in the catalyst matrix would accelerate the catalytic process [10]. Furthermore, the sulfidation of the Fe catalyst can form Fe₇S₈, FeS_2, and FeS mixed phases, which have higher access to the edge and corner unsaturated active sites on the surface with SO₂ and reduction gas, together with the formation of the most important actives sites of S-vacancy [7,10].

Based on the extensive reported work, it is evident that perovskite-type catalysts show promising potential for application in SO₂ catalytic reduction technology. However, the function of the doped LaCoO₃ catalysts for catalytic activity and the pre-sulfurized process, together with the adjustment for catalytic performance and the mechanism in the presence of oxygen, are still undetermined. In the present work, a series of Fe-doped LaCoO₃ perovskites were synthesized by the sol–gel method for the catalytic reduction of SO₂ by CO under O₂. Various parameters of the catalysts were then investigated, including the physical properties, surface chemical state, morphology, catalytic activity, and gas adsorption capacity to elucidate the effects of the doping and sulfurization process, together with the structure–activity relationship of the catalysts.

2. Results and Discussion

2.1. Physical Properties of XRD for Catalysts

Figure 1A presents the XRD patterns of the synthesized Fe-doped LaCoO₃ and oxygen-assisted and oxygen-free sulfurization of LaCo_0.8Fe_0.2O₃. The diffraction peaks corresponded to the orthorhombic structure of LaCoO₃ (PDF#48-0123) and the cubic structure of LaFeO₃ (PDF#75-0541) [27]. As the Fe content in LaCoO₃ increased, the XRD peaks shifted slightly to lower angles, which could be attributed to the substitution of the larger Co³⁺ ions by the smaller Fe³⁺ ions, leading to an increase in the lattice constant [30]. Ko also reported that La_0.7Ce_0.1Co_xNi_yTi_0.6O₃ perovskites with tuned Co and Ni ratios reflect a structural relaxation of the lattice, which facilitates increased spacing between lattice planes as decreasing the larger ionic radii of Co content [31]. Notably, all diffraction peaks of LaCo_1−xFe_xO₃ were consistent with the perovskite phase, confirming the successful synthesis of stoichiometric LaCo_1−xFe_xO₃ [27]. Oxygen-assisted and oxygen-free sulfurization (OAS and OFS) experiments were conducted on LaCo_0.8Fe_0.2O₃ to evaluate the impact of the sulfurization on the phase structure of catalysts. The results show a significant reduction in the intensity of the XRD diffraction peaks in Figure 1A. Although OAS-LaCo_0.8Fe_0.2O₃ maintained relatively better phase stability compared to OFS-LaCo_0.8Fe_0.2O₃, both processes indicate a decrease in the crystallinity of the catalysts. Furthermore, no new peaks corresponding to additional phases were detected in the XRD patterns of both sulfurized samples, suggesting that the sulfurization weakened the phase integrity without forming new phases. The observed loss of crystallinity in the XRD results for OAS and OFS-LaCo_0.8Fe_0.2O₃ samples may be due to the effects of SO₂ poisoning on the structure of the LaCo_0.8Fe_0.2O₃ catalyst. The holding diffraction peaks of the perovskite structure were assumed to be due to the lesser SO₂ sulfurization time. Zhu also reported that the perovskite structure of LaCoO₃ was destroyed by SO₂ and became amorphous due to a high temperature of 500 °C and long SO₂ exposure [32].

The XRD patterns of LaCo_1−xFe_xO₃, OAS-LaCo_0.8Fe_0.2O₃, and OFS-LaCo_0.8Fe_0.2O₃ catalysts after catalytic testing under oxygen-assisted reaction (OAR) conditions are shown in Figure 1B. The decrease in intensity indicates a significant reduction in the crystallinity of the catalysts, while the main perovskite phase remained intact, accompanied by the emergence of weak peaks corresponding to new phases. Specifically, the peaks at 2θ = 25.6° and 28.5° were attributed to La₂O₂S (PDF 026-0825), which has proved the active sulfides formed in the reduction of SO₂ [33]. The peaks at 2θ = 36.6° and 42.8° corresponded to CoO (PDF 01-1227), and the peaks at 2θ = 28.5° and 30° were assigned to La₂O₃ (PDF 74-2430). In addition, OFS-LaCo_0.8Fe_0.2O₃ exhibited distinct characteristic peaks around 36.4° and 38.0°, which were not observed in other samples. These peaks can be attributed to FeS₂ (PDF 03-0795), a high-activity species that has also been proven to promote SO₂ treatment [7], indicating that OFS treatment effectively achieves structural change of the catalyst. Hibbert summarized the treatment of SO₂ on perovskite oxides for the structural changes [34]. For example, the La_0.5Sr_0.5Co₃ catalyst showed a mixture of perovskite, CoS₂, SrS, and several unknown phases after SO₂ adsorption flow reactor experiments [34]. Another example of higher concentrations of SO₂ (10%) and CO (20%) reaction, the perovskite LaCoO₃ decomposed, and the active catalyst was a mixture of La₂O₂S and CoS₂ [34]. So, the presence of metal sulfide phases in the spent catalysts for OAR-OFS-LaCo_0.8Fe_0.2O₃ may be due to the reaction of SO₂ and the perovskite phase.

2.2. Physical Properties of BET for Catalysts

The Brunauer–Emmett–Teller (BET) surface area Barrett–Joyner–Halenda (BJH) pore size distribution of LaCo_1−xFe_xO₃ samples were investigated through N₂ adsorption–desorption isotherms, as the results are shown in Figure 2. All samples exhibited a typical type IV adsorption isotherm with an H1 hysteresis loop in the relative pressure range of P/P₀ = 0.6–1.0, indicating the presence of irregular mesopores resulting from the aggregation of nanocrystals. The BET surface area and pore structure are similar to the reported perovskite-based systems [35].

Table 1. Textural data of the samples.

Samples	SBET (m2/g)	Pore Volume (cm3/g)	Pore Size (nm)
LaCoO3	12.86	0.10	23.23
LaCo0.8Fe0.2O3	14.67	0.10	21.21
LaCo0.5Fe0.5O3	16.04	0.10	19.05
LaCo0.2Fe0.8O3	14.82	0.10	20.44
LaFeO3	14.48	0.10	21.88

summarizes the specific surface area, pore volume, and pore size of the samples. The increased Fe content may lead to a change in the crystal structure and particle growth during the synthesis process, resulting in a different particle size distribution varied from 23.23 to 19.05 nm. The specific surface area of all samples ranged from 12 m²/g to 16 m²/g, with pore sizes around 20 nm, which indicates that the textural properties of the catalysts have a negligible impact on the differences in their catalytic performance.

2.3. Morphological Characteristics

SEM images of the LaCo_0.8Fe_0.2O₃ catalysts, the pre-sulfurization samples, and their catalytic reaction samples are presented in Figure 3. As shown in Figure 3A, the morphology of pristine LaCo_0.8Fe_0.2O₃ consisted of large particles composed of tightly bound small particles. Similarly, the samples exhibit well-defined, small particles with an approximate diameter of 100 nm after sulfurization with OFS-LaCo_0.8Fe_0.2O₃ and OAS-LaCo_0.8Fe_0.2O₃, as depicted in Figure 3B,C, which could be expected to enhance the contact between the catalyst and the reactants. Heidinger reported the morphology evolution of Fe-LaCoO₃ perovskite synthesized by reactive grinding, which also exhibited agglomerating particles with low-energy ball milling [23]. After undergoing SO₂ catalytic testing under oxidative conditions, the sulfurized samples of OFS-LaCo_0.8Fe_0.2O₃ and OAS-LaCo_0.8Fe_0.2O₃ after SO₂ treatment exhibited varying degrees of sintering, with the originally small particles turning to greater agglomeration (Figure 3D,E)_. EDS mapping was further employed to investigate the elemental distribution on the catalyst surface, with the results for OFS-LaCo_0.8Fe_0.2O₃ shown in Figure 3F. The presence of Fe on the surface confirms the successful doping of Fe into the catalyst, while the S distribution indicates that sulfurization led to a significant accumulation of sulfur species on the surface. The atomic percentage data obtained from the EDS mapping for all samples are summarized in

Table 2. The atomic percentage data on the surface of the catalysts detected by SEM-EDX.

Sample	O (%)	S (%)	Fe (%)	Co (%)	La (%)
LaCo0.8Fe0.2O3	70.31	0	4.18	13.50	12.01
OFS-LaCo0.8Fe0.2O3	75.45	3.85	3.48	7.91	9.31
OAS-LaCo0.8Fe0.2O3	71.41	1.12	3.96	11.22	12.29
OAR-OFS-LaCo0.8Fe0.2O3	73.72	12.45	1.36	5.43	7.06
OAR-OAS-LaCo0.8Fe0.2O3	39.41	31.73	2.68	10.59	15.59

. The S atomic percentage on the surface of OFS-LaCo_0.8Fe_0.2O₃ was higher than that of OAS-LaCo_0.8Fe_0.2O₃. However, the OFS-LaCo_0.8Fe_0.2O₃ catalyst exhibited a lower sulfur enrichment than OAS-LaCo_0.8Fe_0.2O₃ after the oxygen-assisted reaction test, which suggests that the OFS treatment contributed to the formation of active sulfur species on the surface of LaCo_0.8Fe_0.2O₃, thereby maybe enhancing its catalytic activity for SO₂ conversion. The atomic percentage data on the surface of the catalysts were also detected by XPS and shown in

Table 3. The atomic percentage data on the surface of the catalysts detected by XPS.

Sample	O (%)	S (%)	Fe (%)	Co (%)	La (%)
LaCo0.8Fe0.2O3	53.70	0	3.74	6.14	12.29
OFS-LaCo0.8Fe0.2O3	44.81	13.20	2.88	5.21	6.76
OAS-LaCo0.8Fe0.2O3	51.32	5.77	3.13	4.23	9.22
OAR-OFS-LaCo0.8Fe0.2O3	50.03	15.90	1.93	2.78	6.31
OAR-OAS-LaCo0.8Fe0.2O3	50.57	13.14	2.14	4.47	6.39

. The phenomena above the general trend are similar to EDS but have slight differences. OFS-LaCo_0.8Fe_0.2O₃ still owned a high S atomic percentage on the surface in comparison with OAS-LaCo_0.8Fe_0.2O₃. OAR-OFS-LaCo_0.8Fe_0.2O₃ catalyst showed a slightly higher sulfur enrichment than OAS-LaCo_0.8Fe_0.2O₃, which was opposite to EDS. This may be due to the difference in depth of the sample being analyzed.

2.4. Catalytic Performance

The catalytic activities in the SO₂ reduction over LaCo_1−xFe_xO₃ perovskites were first evaluated as a function of temperature under the oxygen-assisted reaction (OAR), and the results are presented in Figure 4A. As the reaction temperature increased, the SO₂ conversion rate generally showed an upward trend for all catalysts except for LaFeO₃, which exhibited deactivation, suggesting SO₂ poisoning during the reaction. Among all the catalysts, LaCo_0.8Fe_0.2O₃ demonstrated the highest activity, achieving an SO₂ conversion rate of 84.0% at 500 °C, and showed lower COS production than that of Fe-doped catalysts, which is presented in Figure 5A. The temperatures up to 550 °C used in our experiments were indeed above the boiling point of sulfur (444.6 °C), which is essential to ensure the effective removal of sulfur from the catalyst surface.

Consequently, the catalyst was subjected to both OFS and OAS treatments, followed by tests under oxygen-free reaction (OFR) and oxygen-assisted reaction (OAR), as shown in Figure 4B. The results indicate that the OFS catalyst exhibited higher activity and lower COS production than those of OAS catalyst (Figure 5B), and the presence of oxygen during the reaction further enhanced the conversion rate, which reached 95.9% at 450 °C, indicating that the OFS treatment enhanced the performance of LaCo_0.8Fe_0.2O₃ in SO₂ catalytic reduction. The SO₂ conversion for OFS-LaCo_0.8Fe_0.2O₃ was even close to many reported catalysts mainly based on oxygen-free reactions. Xu reported that 15% Gd-Ce-O_x exhibited superior catalytic activity, achieving an impressive 97% conversion of SO₂ toward sulfur by CO at 400 °C, but with the absence of O₂ [1]. Li investigated that iron oxide supported on activated coke for SO₂ to S by CO reaches 90.9% SO₂ conversion at 400 °C [15]. Jin observed SO₂ reduction for S production by CO over Ce-Al-O_x composite oxide catalyst reaching 98% beyond 425 °C [5]. At an inlet SO₂ concentration of 5%, the byproduct COS concentration was below 700 ppm, indicating a very low level of byproducts. For the OAS-LaCo_0.8Fe_0.2O₃ catalyst at 500 °C, a relatively higher COS concentration was observed compared to other catalysts. This difference might be perceived due to the overall low concentration of COS. Subsequently, OFS-LaCo_0.8Fe_0.2O₃ underwent a thermal stability test at 550 °C for 28 h, as well as a water resistance test with the introduction of theoretical 20% and 40% water during the reaction, both of which demonstrate the excellent stability of the catalyst, as shown in Figure 4C,D. The weight of the collected production solid sulfur was further measured. The results showed that the amount of sulfur collected corresponded well with the SO₂ conversion rates, indicating a high degree of material balance. To gain deeper insights into the activation effect of the OFR process on LaCo_0.8Fe_0.2O₃, further characterization and analyses were conducted.

2.5. TPD Analysis

TPD measurements were conducted on LaCo_0.8Fe_0.2O₃, sulfurized LaCo_0.8Fe_0.2O₃, LaCoO₃, and LaFeO₃ catalysts, with the results shown in Figure 6. It is generally accepted that the basic sites on the catalyst surface are crucial for SO₂ adsorption [14]. The CO₂-TPD method was used to investigate the change in surface basicity of the varied prepared perovskite catalysts. The CO₂-TPD profiles (Figure 6A) revealed a prominent peak around 550 °C for OFS-LaCo_0.8Fe_0.2O₃, suggesting the presence of strong basic sites that were absent in the other catalysts, thereby providing substantial support for SO₂ adsorption. The CO₂-TPD profiles were similar to the reported LaFe-based and LaCo_1−xGa_xO₃ perovskites [36,37].

The finding was further corroborated by the SO₂-TPD results, as shown in Figure 6C, where a clear SO₂ desorption peak was observed around 550 °C on OFS-LaCo_0.8Fe_0.2O₃, highlighting its excellent SO₂ adsorption capability. In comparison to the LaCoO₃ catalyst, Wu reported that the desorption peaks in the high-temperature region of the Ce-LaCoO₃ catalyst were mainly attributed to the preferential adsorption of SO₂ by Ce species as sacrificial sites [38]. The desorption peaks in the high-temperature section were mainly attributed to the metal sulfates on the surface of the catalysts, which were related to their strong ability to adsorb SO₂ [38].

The CO-TPD result (Figure 6B) indicates that the sulfurization process significantly improved the CO adsorption capacity of the catalyst, which is essential for the reduction of SO₂. Moreover, the higher desorption peak observed for OFS-LaCo_0.8Fe_0.2O₃ suggests a stronger CO adsorption capacity, which contributes to its superior catalytic activity. Teng investigated that the total adsorption amount of CO on LaCoO₃ nanowires is larger than that on the LaCoO₃ particles, indicating that there may be more active centers on LaCoO₃ nanowires than those on the latter, which would exhibit higher catalytic activity for CO oxidation, compared with the nanoparticles [39].

2.6. XPS Analysis

In order to investigate the surface states of the catalysts, XPS was employed to analyze Co 2p, O 1s, Fe 2p, and S 2p, with the corresponding spectra presented in Figure 7. In the Co 2p spectra (Figure 7A), peak deconvolution reveals two distinct peaks corresponding to Co³⁺ and Co²⁺. The peaks at 779.0–780.0 eV are attributed to octahedrally coordinated Co³⁺, while the peaks at 780.6–781.6 eV correspond to tetrahedrally coordinated Co²⁺ [40]. Additionally, the S₁ and S₂ peaks are satellite features of Co²⁺. Notably, the Co 2p peak of the OFS-LaCo_0.8Fe_0.2O₃ shifts 0.9 eV toward lower binding energy compared to other catalysts, suggesting a strong interaction between Fe and Co in the catalyst. Electrons escaping from Fe ions are captured by Co³⁺, which is subsequently reduced to Co²⁺ [27]. The data obtained from peak deconvolution are summarized in

Table 4. XPS results of LaCoO₃, LaCo_0.8Fe_0.2O₃, OAS-LaCo_0.8Fe_0.2O₃, and OFS-LaCo_0.8Fe_0.2O₃.

Sample	Co2+/Co3+	Fe2+/Fe3+	Oads/Ototal	Binding Energy (eV)
				Co2+(2p3/2)	Co3+(2p3/2)	Fe2+(2p3/2)	Fe3+(2p3/2)
LaCoO3	0.80	-	0.59	781.6	780.0	-	-
LaCo0.8Fe0.2O3	0.61	0.74	0.47	781.6	780.0	710.5	713.7
OAS-LaCo0.8Fe0.2O3	0.82	0.84	0.59	781.4	779.8	710.6	713.8
OFS-LaCo0.8Fe0.2O3	1.29	1.25	0.74	780.6	779.0	710.8	714.0

. After sulfurization, the more pronounced Co²⁺/Co³⁺ observed in the OFS-LaCo_0.8Fe_0.2O₃ suggests that the OFS treatment facilitates electron transfer between Co ions and Fe ions.

Figure 7B presents the Fe 2p spectra, where two broad peaks are observed in the binding energy ranges of 720–730 eV and 706–718 eV, corresponding to Fe 2p_1/2 and Fe 2p_3/2, respectively. After peak deconvolution, the peaks at 710.5–710.8 eV are assigned to Fe²⁺, while the peaks at 713.7–714.0 eV correspond to Fe³⁺, and the peaks at 718.0–718.8 eV are satellite peaks [41,42,43]. The shift of Fe species to higher binding energy in the OFS-LaCo_0.8Fe_0.2O₃ further confirms the interaction between Co and Fe in the catalyst. The data in the table show that the Fe²⁺/Fe³⁺ ratio for the different samples follows the following trend: OFS-LaCo_0.8Fe_0.2O₃ > OAS-LaCo_0.8Fe_0.2O₃ > LaCo_0.8Fe_0.2O₃, indicating that the OFS treatment enhanced the reduction of Fe³⁺. The resulting charge imbalance is expected to promote the formation of oxygen vacancies, as evidenced by the O 1s spectra.

O 1s spectra are shown in Figure 7C. The peaks at 528.9–529 eV are assigned to lattice oxygen (O_lat) [44], the peaks at 531.3–531.7 eV to surface adsorbed oxygen (O_ads) [45], and the peaks at 533.1–533.3 eV to hydroxyl groups (OOH) on the catalysts surface [46]. Kustov reported similar O 1s spectra over LaMO₃ perovskite-like oxide materials (M: Fe, Co, Ni) [47]. Analysis of the peak areas after deconvolution reveals that the O_ads/O_total ratio is highest for the OFS-LaCo_0.8Fe_0.2O₃, which reflects the result of charge compensation. This suggests that the OFS treatment facilitates the formation of oxygen vacancies on the catalyst surface, thereby enhancing the activation of both reactants and gaseous oxygen, ultimately improving catalytic activity.

Figure 7D presents the S 2p spectra for OFS-LaCo_0.8Fe_0.2O₃, OAS-LaCo_0.8Fe_0.2O₃, OAR-LaCo_0.8Fe_0.2O₃, and the oxygen-assisted reaction (OAR) for the spent catalysts of the sulfurization of LaCo_0.8Fe_0.2O₃. All the samples showed the binding energy of S 2p was about 168–170 eV, which can be mainly attributed to sulfate and sulfite, implying that sulfate and sulfite were formed on the surface due to both the pre-sulfurization and oxygen-assisted reaction [32]. Furthermore, according to the binding energy of the peak, it can be judged that the strong S 2p peaks of 162.4 eV existed only for OFS-LaCo_0.8Fe_0.2O₃ and OAR-OFS-LaCo_0.8Fe_0.2O₃ correspond to S²⁻, which may belong to FeS₂ on the catalyst surface [7]. Tian investigated the effect of temperature on the reaction path of pyrite (FeS₂)-based catalyst catalyzed CO reduction of SO₂ to sulfur, in which they found that sulfate, sulfite, and S²⁻ together with FeS₂ by applying S 2p and Fe 2p to characterize the catalyst and reaction process [7]. The FeS₂ detected by the XPS result perhaps was consistent with the XRD results for OAR-OFS-LaCo_0.8Fe_0.2O₃ with the existence of the FeS₂ phase.

2.7. Mechanism Explanation

XRD demonstrated that the crystallinity of OFS-LaCo_0.8Fe_0.2O₃ was significantly reduced, as evidenced by the decreased intensity of the diffraction peaks. Moreover, after the oxygen-assisted reaction (OAR), the formation of the La₂O₂S phase was observed. The atomic percentage data obtained from SEM-EDX and XPS analyses indicated sulfur enrichment on the surface of OFS-LaCo_0.8Fe_0.2O₃. The CO/CO₂/SO₂-TPD results further revealed that OFS-LaCo_0.8Fe_0.2O₃ exhibited the strongest surface basicity, CO, and SO₂ adsorption capacity among the catalysts. XPS also showed that OFS-LaCo_0.8Fe_0.2O₃ had the highest O_ads/O_total ratio and surface active S²⁻ species. These structural, physical, and chemical properties collectively suggested that OFS-LaCo_0.8Fe_0.2O₃ possesses the optimal conditions for high SO₂ conversion. Based on these characterization results, we further speculated the reaction mechanism responsible for the enhanced catalytic performance.

The SO₂ catalytic reduction mechanism over transitional metal oxides by reduction gas, like CO, is normally related to the COS and redox intermediate mechanisms, which are ascribed to the active sites and oxygen vacancies. There is a need for a clear discussion of the reaction mechanism, particularly the identification of active sites and oxygen vacancies (O_v) during the catalytic process for the LaCo_xFe_1−xO₃ perovskite system under Fe doping and sulfurization conditions. In the redox mechanism, the reaction pathway can be expressed as Equations (1)–(4), SO₂ adsorption on the perovskite surface, which leads to the S formation through the SO₂ direct reduction by CO [10]. So, the high O_v content may induce high SO₂ conversion. The catalytic activities in the SO₂ reduction show that LaCo_0.8Fe_0.2O₃ owns higher SO₂ conversion than those of LaCo_xFe_1−xO₃ (Figure 4A). However, LaCo_0.8Fe_0.2O₃ had less O_ads/O_total than that of LaCoO₃ (0.47 vs. 0.59), so, besides the redox mechanism, there may be a related synergistic effect of Fe and LaCoO₃. Fe pyrite (FeS₂)-based catalysts were normally reported to promote the CO reduction of SO₂ to S through the COS mechanism [7]. The phenomenon was identical to the observation from Figure 5A, which can find there is COS formation for LaCo_0.8Fe_0.2O₃, but the absence of COS formation for LaCoO₃ within 450–550 °C. Over LaCoO₃, there may only be related to the redox mechanism, which almost zero COS production as some of the most important features at the relatively high reaction temperature with high capacity and mobility of O_v [10].

Oxygen-free sulfurization of LaCo_0.8Fe_0.2O₃ perovskite owns the highest SO₂ conversion than that of oxygen-assisted sulfurization and LaCo_xFe_1−xO₃ (Figure 4A,B). One of the reasons for the phenomenon may be due to the OFS-LaCo_0.8Fe_0.2O₃ owning the highest O_ads/O_total with 0.74 (Table 4). The other reason may be due to the promotion of the COS mechanism. In the COS mechanism, the reaction pathway can be expressed as Equations (5)–(7). COS can form through the reaction between CO and metal sulfide, then turn to S with SO₂ [10]. OFS-LaCo_0.8Fe_0.2O₃ exhibited the FeS₂ phase detected from XRD (Figure 1B) and S 2p XPS (Figure 7D), which were unanimous with the optimal SO₂ reduction performance. Moreover, the detection of La₂O₂S for OFS-LaCo_0.8Fe_0.2O₃ and OAS-LaCo_0.8Fe_0.2O₃ under oxygen-assisted reaction (OAR) catalytic testing indicated that the formation of La₂O₂S during the SO₂ reduction, which was also reported to be the reactive sites based on the COS mechanism [29,33]. Based on the above analysis, LaCo_0.8Fe_0.2O₃ by Fe doping exhibited increased SO₂ conversion, maybe due to the main redox intermediate mechanism, together with the COS mechanism by the synergistic effect of Fe and LaCoO₃. OFS-LaCo_0.8Fe_0.2O₃ exhibited much higher SO₂ conversion compared with OAS-LaCo_0.8Fe_0.2O_3, perhaps due to both redox and COS intermediate mechanisms.

Cat-O + CO → Cat-□ + CO₂

(1)

Cat-□ + SO₂ → Cat-O + SO_cat

(2)

Cat-□ + SO_cat → Cat-O + S_cat

(3)

S_cat → Cat + 1/x S_x

(4)

M-S_cat + CO → M + COS

(5)

SO₂ + 2COS → 3S + 2CO₂

(6)

S + M → M − S_cat

(7)

3. Materials and Methods

3.1. Materials

La(NO₃)₃·6H₂O (CAS: 100587-94-8), Co(NO₃)₂·6H₂O (CAS: 10026-22-9), Fe(NO₃)₃·9H₂O (CAS: 10421-48-4), citric acid, ammonium hydroxide, polyethyleneglycol, and quartz sands were purchased from Shanghai Macklin Biochemical Technology Co., Ltd. (Shanghai, China). All these chemicals with analytical purity were used as obtained without further purification.

3.2. Synthesis of LaCo_xFe_1−xO₃ Catalysts

LaCoO₃ was synthesized as follows: La(NO₃)₃·6H₂O and Co(NO₃)₂·6H₂O were dissolved in deionized water under magnetic stirring at a stoichiometric molar ratio of 1:1, resulting in a homogeneous solution. Subsequently, citric acid was added to the above solution in an amount equimolar to the total metal nitrates. Ammonia solution was then gradually added to adjust the pH of the mixture to a range of 6–7. After adding a little polyethyleneglycol, the resulting solution was stirred continuously at 75 °C until a stable gel was formed. The porous dry gel was obtained after the solution was dried overnight at 130 °C. The dry gel was subjected to self-combustion ignition and further calcined in a tubular furnace. Calcination was carried out at a heating rate of 2 °C·min⁻¹, with the temperature maintained at 600 °C for 2 h.

LaCo_xFe_1−xO₃ perovskites were synthesized as follows: Fe(NO₃)₃·9H₂O was introduced in varying concentrations based on the molar ratio, while all other experimental conditions were kept identical to those used for preparing LaCoO₃. The final samples were named as LaCo_xFe_1−xO₃ (x = 0, 0.2, 0.5, 0.8).

3.3. Sulfurization of LaCo_0.8Fe_0.2O₃ Catalyst

Catalysts were sulfurized as follows: 2 g of the catalysts (30–40 mesh) were placed in a tubular furnace and heated to 350 °C under a nitrogen flow of 100 mL/min. Subsequently, the nitrogen was replaced with a mixture of either 5% SO₂/16% CO/3% O₂/N₂ or 5% SO₂/10% CO/N₂, and the temperature was maintained for 30 min. The catalysts were named OAS catalyst (oxygen-assisted sulfurization) and OFS catalyst (oxygen-free sulfurization) based on the presence or absence of oxygen in the sulfurization process.

3.4. Catalysts Characterization

The crystal structure of the catalysts was analyzed using a Rigaku SmartLab SE X-ray diffractometer (XRD, Tokyo, Japan), employing Cu-Kα radiation as the diffraction source (incident wavelength λ = 1.54 Å). The scanning range was typically set to 5–90°, with a scanning speed of 10°·min⁻¹. The N₂ adsorption–desorption isotherms were determined by a Micromeritics ASAP 2460 analyzer (Norcross, GA, USA), with BET surface area, BJH desorption cumulative volume of pores, and BJH desorption average pore diameter for analysis. The JSM-7800 (Prime, JEOL, Tokyo, Japan) scanning electron microscope (SEM) equipped with energy-dispersive X-ray spectroscopy (EDS) was used to investigate the morphology of samples. The temperature-programmed desorption (TPD) of CO₂, CO, and SO₂ (CO₂-TPD, CO-TPD, and SO₂-TPD) experiments were performed using a Micromeritics Autochem 2920II chemisorption analyzer. All the catalysts were pretreated in a He stream at 300 °C for 1 h before the measurement; subsequently, the adsorption of target gas was conducted on the catalysts. A thermal conductivity detector (TCD) detector was used to record the signal of gas desorbed through heating from 50 to 800 °C. The chemical states of the surface elements were characterized using X-ray photoelectron spectroscopy (XPS) with a Thermo Scientific K-Alpha instrument (Waltham, MA, USA). Al K-Alpha radiation (1486.6 eV) served as the excitation source, and the binding energy was calibrated to C 1s at 284.8 eV.

3.5. Catalytic Performance Tests

The catalyst evaluation experiment for CO reduction of SO₂ was studied with 2 g (30–40 mesh) catalysts packed in the middle part of the quartz tube in a fixed-bed reactor system (Figure 8). The total gas flow rate was confirmed to be 100 mL/min adjusted with a mass flow controller (MFC), with a gas composition of 5% SO₂, 16% CO/3% O₂/N₂ or 5% SO₂ and 10% CO/N₂ represented oxygen-assisted reaction (OAR) and oxygen-free reaction (OFR). The electric furnace was employed to regulate the temperature of the reactor. The reaction was run from 250 °C to 550 °C at an interval of 50 °C with a 15 min continuous test at each temperature. In addition, a cooling unit was incorporated into the sulfur collection process, and a filter was installed at the outlet of the condensation unit to prevent sulfur powder from being carried out with the reaction gas. The concentrations of SO₂, COS, H₂S, and CS₂ in the outlet gas at each temperature interval were determined online by gas chromatography (Agilent 8860, Wilmington, DE, USA, columnHayeSep N) with TCD. The standard curve for SO₂ was obtained using the peak area from a two-point method with zero point and 5vol% SO₂ (N₂ balance gas) detected by TCD. The conversion rate of SO₂ (X_SO2) are calculated as follows:

$$X_{{SO}_2}={\displaystyle \frac{\left[{SO}_2\right]in-\left[{SO}_2\right]out}{\left[{SO}_2\right]in}}\times 100\%$$

(8)

where [SO₂]in is the concentration of SO₂ in the inlet gas, while [SO₂]out is the concentration of SO₂ in the exhaust gas after the reaction.

4. Conclusions

This study provides significant insights into the catalytic reduction of SO₂ to elemental sulfur by CO over Fe-doped LaCoO₃ perovskite catalysts, together with pre-sulfurization treatment. The incorporation of Fe into the LaCoO₃ structure enhanced the catalytic activity, with LaCo_0.8Fe_0.2O₃ achieving the highest performance through oxygen-free sulfurization (OFS) under oxygen-assisted reaction. OFS-LaCo_0.8Fe_0.2O₃ perovskite presented the highest SO₂ conversion with 95.9% at 450 °C than that of LaCo_0.8Fe_0.2O_3, reaching 84.0% SO₂ conversion at 500 °C. XRD structure, CO/CO₂/SO₂-TPD, and XPS surface analysis revealed that OFS treatment promoted the formation of active sulfur species and oxygen vacancies, as well as the adsorption capacity of reactants, which were crucial for improving catalytic activity. Additionally, the catalysts exhibited excellent thermal stability and water resistance, making them suitable for practical applications in industrial SO₂ reduction. These findings highlight the critical role of sulfurization treatments and surface properties in enhancing the catalytic performance of perovskite-based materials, offering insights for the development of more efficient and sustainable flue gas desulfurization technologies.