Oxygen Vacancy-Mediated Selective H₂S Oxidation over Co-Doped LaFe_xCo_1−xO₃ Perovskite

Abstract

Compared to the Claus process, selective H₂S catalytic oxidation to sulfur is a promising reaction, as it is not subject to thermodynamic limitations and could theoretically achieve ~100% H₂S conversion to sulfur. In this study, we investigated the effects of Co and Fe co-doping in ABO₃ perovskite on H₂S selective catalytic oxidation. A series of LaFe_xCo_1−xO₃ (x = 0, 0.2, 0.4, 0.6, 0.8, 1.0) perovskites were synthesized by the sol-gel method. Compared to LaFeO₃ and LaCoO₃, co-doped LaFe_xCo_1−xO₃ significantly improved the H₂S conversion and sulfur selectivity at a lower reaction temperature. Nearly 100% sulfur yield was achieved on LaFe_0.4Co_0.6O₃ under 220 °C with exceptional catalyst stability (above 95% sulfur yield after 77 h). The catalysts were characterized by XRD, BET, FTIR, XPS, and H₂-TPR. The characterization results showed that the structure of LaFe_xCo_1−xO₃ changed from the rhombic phase of LaCoO₃ to the cubic phase of LaFeO₃ with Fe substitution. Doping with appropriate iron (x = 0.4) facilitates the reduction of Co ions in the catalyst, thereby promoting the H₂S selective oxidation. This study demonstrates a promising approach for low-temperature H₂S combustion with ~100% sulfur yield.

1. Introduction

Natural gas processing/utilizing processes, petroleum refining processes, and coal chemistry produce H₂S-containing waste gas [1,2,3].The removal of H₂S from industry waste gas is crucial as H₂S is a toxic and corrosive gas, leading to severe pollution and equipment/pipeline corrosion [4]. Numerous efforts have been conducted to remove H₂S and the most widely used technology is called the Claus process, which recovers elemental sulfur from H₂S-containing gas [5,6]. However, due to thermodynamic limitations, the Claus process cannot achieve 100% H₂S conversion, leaving 2–5% unconverted H₂S gas remaining after the treatment [7]. Selective H₂S catalytic oxidation could be added as a secondary process after the Claus process apparatus to oxidize the residual, low-concentration H₂S-containing gas to elemental sulfur with increased overall sulfur yield. The design of a catalyst with high H₂S conversion and high sulfur selectivity is the key to the selective catalytic oxidation. So far, H₂S selective oxidation catalysts mainly include active carbon (AC)-based catalysts [8], carbon nanotube-based catalysts [9], clay-supported catalysts [10], and metal oxide-based catalysts [11]. Nevertheless, most of these H₂S selective oxidation catalysts face challenge on the needs of excessive oxygen feed, which is produced from economically expensive air separation units. For instance, Fang et al. [8] found that the Mn/AC catalyst can achieve optimal catalytic activity at 180 °C, but it needs excess amounts of oxygen to participate in the H₂S oxidation process. Ba et al. [9] found that the H₂S conversion over nitrogen-doped carbon nanotubes dropped from 100% to 92% as the O₂/H₂S ratio decreased from 2.5 to 0.6. Soriano et al. [11] observed that VNa-0.1 had high sulfur selectivity at 200 °C under the H₂S/air/He molar ratio of 1.2/5.0/93.8, with an excess of O₂. Moreover, the durations of the catalyst stability in the literature were usually not long, less than 20 h [12,13,14]. Thus, a catalyst with remarkable catalytic activity, excellent selectivity, outstanding stability, and desired stoichiometric H₂S/O₂ operating conditions is urgently needed.

Perovskite oxides such as ABO₃ have been extensively investigated. The typical structure of perovskite oxides is cubic, with one A-site cation coordinated by 12 anions and eight B-site cations (belonging to 1/8 to one specific unit cell) coordinated by six anions [15,16]. For (distorted) perovskites with lower symmetry, the first shell coordination numbers are smaller. It is well-recognized that the tuning of A and B site cations can affect the catalytic activity and stability of perovskite oxides. Moreover, structural modifications related to the generation of the oxygen vacancies and/or changes in the valence states of the original cations can be achieved by the partial substitution of the ions at A site or B site [17,18,19]. Therefore, the flexibility and chemical versatility of ABO₃ perovskite oxides can be used to design highly active, selective, and stable catalysts [20]. Among these, LaCoO₃ and LaFeO₃ stand out as redox catalysts with satisfactory catalytic performance reported for CO oxidation [21] and NO oxidation [22]. Moreover, Yang et al. [23] found that LaCoO₃ showed high H₂S conversion and high selectivity to sulfur (98.2%) at 260 °C. Zhang et al. [24] found that LaFeO₃ exhibited outstanding selectivity (100%) under relatively low reaction temperature and various H₂S/O₂ ratios. It is desired to obtain catalysts with ~100% H₂S conversion and ~100% sulfur selectivity, and the influence of co-doping Fe and Co into the B-site of perovskite could be further explored.

This work investigates the influence of Fe substitution on LaCoO₃ in H₂S selective oxidation. LaFe_xCo_1−xO₃ (x = 0, 0.2, 0.4, 0.6, 0.8, 1) perovskite oxides were synthesized by the sol-gel method and their catalytic behaviors in the H₂S selective oxidation were investigated. It was observed that the incorporation of iron enhanced the catalytic activity of the LaFe_xCo_1−xO₃ catalysts. In addition, the crystal structure and redox ability were comprehensively investigated by various characterization techniques such as XRD, BET, FTIR, XPS, and H₂-TPR. On the basis of these results, the catalytic and deactivation mechanisms of H₂S selective oxidation over LaFe_xCo_1−xO₃ are discussed.

2. Experiment

2.1. Catalyst Preparation

Powder samples of LaFe_xCo_1−xO₃ (x = 0, 0.2, 0.4, 0.6, 0.8, and 1.0) perovskite oxides were synthesized by the sol-gel method using citric acid as a complexing agent. The starting materials were La(NO₃)₃∙6H₂O (AR, Macklin, Shanghai, China), Fe(NO₃)₃∙9H₂O (AR, Macklin, Shanghai, China), and Co(NO₃)₂∙6H₂O (AR, Macklin, Shanghai, China). The required amounts of nitrate samples were first dissolved in distilled water. Citric acid (AR, Aladdin, Shanghai, China) was added so that the number of moles of citric acid was equal to the number of moles of total metal cations. The obtained solution was evaporated at 80 °C and then PEG-20000 (AR, Macklin, Shanghai, China) was added into the solution. Stirring was continued until a viscous gel was formed. Finally, the gel was dried in an oven at 105 °C overnight and calcined at 650 °C for 5 h in a furnace to obtain the catalyst samples.

2.2. Catalyst Characterization

XRD (X-ray diffraction) patterns were determined using the X’Pert PRO powder diffraction system (Empyrean, San Jose, California, USA) with Cu Kα radiation (λ = 0.15418 nm, 40 kV/40 mA) in the 2θ range of 10–80°.

BET (Brunauer–Emmett–Teller) surface areas and textural properties of the catalysts were determined by nitrogen adsorption–desorption isotherms using an ASAP-2020 apparatus (Micromeritics, Norcross, GA, USA). The adsorption process was carried out at liquid nitrogen temperature. The BET method was adopted to calculate the specific surface area.

FTIR (Fourier transform infrared) spectra were measured using a Nicolet Model iS-50 instrument (Thermo Scientific, Waltham, MA, USA) in the region from 400 to 4000 cm⁻¹. The sample was taken in pellet form in the KBr matrix for measurement.

X-ray photoelectron spectroscopy (XPS) was performed on an ESCALAB 250Xi (Thermo Scientific, USA) with Al Kα radiation as the radiation source at 300 W. The spectra of C 1s, O 1s, La 3d, Co 2p, and Fe 2p were recorded. The binding energies were calibrated using the C 1s peak of contaminant carbon (B.E. = 284.8 eV) as the standard.

H₂-TPR (H₂ temperature-programmed reduction) experiment was measured on an Auto Chem 2920 chemical adsorption instrument (Micromeritics, Norcross, GA, USA). The TPR profiles were obtained by passing a 10% H₂/Ar flow (50 mL/min) through the pretreated catalyst (about 60 mg). The temperature was increased from room temperature to 700 °C with a rate of 10 °C/min, and the H₂ concentration was measured by a thermal conductivity detector. Before the determination, the catalyst was heated to 300 °C under He atmosphere for 60 min.

2.3. Catalytic Performance Tests

All catalytic tests were performed in a continuous flow fixed-bed quartz reactor with an internal diameter of 10 mm at atmospheric pressure, as shown in Figure 1. A certain amount of catalyst was placed in the reactor. A simulant gas was flown into the reactor, with 0.5% H₂S, 0.25% O₂, and balance gas of N₂. The total gas flow rate was fixed at 150 mL/min and the reaction temperature ranged from 180 °C to 260 °C. A sulfur condenser was attached to the bottom of the fixed-bed quartz reactor. The tail gas was analyzed by gas chromatography (GC2060, Ruimin, China) equipped with an FPD (flame photometric detector) and a TCD (thermal conductivity detector). We note that in H₂S oxidation, the formation of S is considered a selective reaction and the formation of SO₂ is considered an unselective reaction. The conversion of H₂S, sulfur selectivity, and sulfur yield were determined by the following equations:

$${\mathrm{H}}_2\mathrm{S} \mathrm{conversion}(\%)=\frac{{{[\mathrm{H}}_2\mathrm{S}]}_{\mathrm{in}}{-[\mathrm{H}}_2{\mathrm{S}]}_{\mathrm{out}}}{{{[\mathrm{H}}_2\mathrm{S}]}_{\mathrm{in}}}\times 100\%$$

(1)

$$\mathrm{Sulfur} \mathrm{selectivity}(\%)=\frac{{{[\mathrm{H}}_2\mathrm{S}]}_{\mathrm{in}}{-[\mathrm{H}}_2{\mathrm{S}]}_{\mathrm{out}}{-[\mathrm{SO}}_2{]}_{\mathrm{out}}}{{{[\mathrm{H}}_2\mathrm{S}]}_{\mathrm{in}}{-[\mathrm{H}}_2{\mathrm{S}]}_{\mathrm{out}}}\times 100\%$$

(2)

$${\mathrm{Sulfur} \mathrm{yield}(\%)=[\mathrm{H}}_2\mathrm{S} \mathrm{conversion}]\times [\mathrm{sulfur} \mathrm{selectivity}]\times 100\%$$

(3)

3. Results and Discussion

3.1. Catalytic Performances of LaFe_xCo_1−xO₃ Catalysts

3.1.1. Effect of Reaction Temperature

The effect of reaction temperature on the catalytic performance of the LaFe_xCo_1−xO₃ catalysts for H₂S selective catalytic oxidation is investigated in Figure 2. Figure 2A shows the H₂S conversion for LaCoO₃, LaFeO₃, and Fe-doped LaCoO₃. At 180 °C, LaCoO₃ can only achieve H₂S conversion of 75%, while LaFeO₃ achieved slightly higher H₂S conversion of 83%. With Fe co-doping in the B-site, H₂S conversions were significantly higher, and all samples achieved conversions higher than 88%. This indicates that B-site Fe doping significantly increases the activity of the pristine perovskite catalysts. Temperature effects were also investigated. For LaCoO₃, it was shown that H₂S conversion increased with temperature and eventually reached 93% at 260 °C. For co-doped LaFe_0.4Co_0.6O₃, LaFe_0.6Co_0.4O₃, and LaFe_0.8Co_0.2O₃, H₂S conversion increased to 99% at only 200 °C. This again shows that LaFe_xCo_1−xO₃ with Fe doping has notably higher activities. For LaFeO₃, the H₂S conversion increased and then decreased with the further increasing temperature. This could be ascribed to lattice oxygen uncoupling from the perovskite structure at higher temperatures, which caused a loss of activity of perovskite [25].

Figure 2B compares the selectivities towards sulfur. It was observed that LaFeO₃ exhibited almost perfect selectivities (~100%) at all temperatures investigated, while LaCoO₃ exhibited lower selectivity, with ~95% at 180 °C. A small amount of Fe doping did not increase the selectivity, and 94% sulfur selectivity was achieved for LaFe_0.2Co_0.8O₃ at 180 °C. However, a further increase in Fe contents significantly increased the selectivities, and >98% selectivities were achieved for LaFe_0.4Co_0.6O₃, LaFe_0.6Co_0.4O₃, and LaFe_0.2Co_0.8O₃. This again shows that larger Fe co-doping is beneficial. Temperature effects were also investigated. For LaCoO₃ and LaFe_0.2Co_0.8O₃, the selectivities increased with respect to temperature. For LaFe_0.4Co_0.6O₃, LaFe_0.6Co_0.4O₃, and LaFe_0.2Co_0.8O₃, a volcanic shape was present and the highest selectivities (~100%) were obtained at 220 °C. Sulfur yield was further compared in Figure 2C. As expected, LaFe_0.4Co_0.6O₃, LaFe_0.6Co_0.4O₃, and LaFe_0.2Co_0.8O₃ achieved ~100% sulfur yield at 220 °C, higher than LaFe_0.2Co_0.8O₃, LaFeO₃, and LaCoO₃. This shows that perovskites with Co and Fe co-doping are highly effective in H₂S selective oxidation to sulfur.

3.1.2. Effect of GHSV (Gas Hourly Space Velocity)

LaFe_0.4Co_0.6O₃ was selected as one of the best catalysts for selective H₂S oxidation. The influence of GHSV on the performance of the LaFe_0.4Co_0.6O₃ catalyst was investigated at 220 °C. As shown in Figure 3, the H₂S conversion, sulfur selectivity, and sulfur yield were all above 99% when the GHSV was in the range of 3000 h⁻¹–6000 h⁻¹. However, it was found that as the GHSV further increased, the catalytic activity of LaFe_0.4Co_0.6O₃ catalyst was greatly reduced. H₂S conversion decreased from 99% to 82% with the increase in GHSV from 6000 h⁻¹ to 15,000 h⁻¹. Therefore, LaFe_0.4Co_0.6O₃ presented excellent catalytic performance in the GHSV range of 3000 h⁻¹–6000 h⁻¹.

3.1.3. Effect of the H₂S/O₂ Molar Ratio

Figure 4 depicts the effect of the H₂S/O₂ molar ratio on H₂S selective catalytic oxidation over the LaFe_0.4Co_0.6O₃ catalyst at 220 °C. As shown in Figure 4, as the H₂S/O₂ molar ratio decreased from 3:1 to 2:1 (i.e., from oxygen-deficient to stoichiometric), H₂S conversion increased from 75% to nearly 100% with sulfur selectivity close to 100%. As H₂S/O₂ further reduced to 1 (i.e., oxygen-excess), H₂S conversion decreased slightly, but sulfur selectivity decreased drastically to ~70%. This is attributed to the further oxidation of the generated sulfur to sulfur dioxide under excess oxygen. Therefore, as opposed to previous studies that required an oxygen-excess environment [9,11], co-doped LaFe_0.4Co_0.6O₃ catalyst only requires stoichiometric H₂S/O₂ condition, showing process advantages.

3.1.4. Durability of LaFe_xCo_1−xO₃ Catalysts

Catalyst stability is an important criterion for industrial applications. Figure 5 illustrates the durability behavior of the LaFe_0.4Co_0.6O₃ catalyst at 220 °C. The selectivity of sulfur can be maintained at ~100% with the reaction proceeding within 77 h. For catalyst activity, during the first 64 h, the H₂S conversion of the LaFe_0.4Co_0.6O₃ catalyst decreased slightly from 99.9% to 98.1%. The deactivation accelerated after 64 h; however, H₂S conversion was still above 95% at 77 h. Longer reaction time was not attempted given the limitation of the lab equipment. Although deactivation was still observed, the co-doped LaFe_0.4Co_0.6O₃ catalyst exhibited excellent stability when compared with several other reported catalysts using a 5% decrease in sulfur yield as a criterion. The comparisons with previously reported carbon materials, pillared clay materials, and metal oxides are shown in Figure 6 [2,7,13,26,27].

3.2. Characterization of LaFe_xCo_1−xO₃ Catalysts

3.2.1. Phase Identifications

The phase structures of all catalysts were examined by XRD. Figure 7 shows the XRD patterns of LaFeO₃, LaCoO₃, and LaFe_xCo_1−xO₃ catalysts. It was observed that co-doped LaFe_xCo_1−xO₃ generally maintained a perovskite-phase structure. With increasing amounts of Fe (III) dopant, a small shift to lower 2-theta angles was observed, indicating that Fe³⁺ is substituted into the perovskite structure. For x = 0, 0.2, and 0.4, there is a clear reflection peak splitting at 59.4°, consistent with the spitting of the main peak at 33°. For x = 0.6 and x = 0.8, the peak splitting effect has become less apparent, but the peak shape is still asymmetric. It is noted that the peak splitting effect is minimal on LaFeO₃, with no Co-doping. This behavior is mainly due to the transformation of the phase structure from a rhombohedral structure of LaCoO₃ (PDF 84-0848) to a cubic structure of LaFeO₃ (PDF 75-0541) with Fe³⁺ dopant.

3.2.2. FTIR Spectrum

FTIR has the advantage of being very sensitive to structural distortions [28]. Figure 8 shows the FTIR spectra of LaFeO₃, LaCoO₃, and LaFe_xCo_1−xO₃ catalysts. The ~3400 cm⁻¹ band may be the water vapor background error caused by the instrument state. Moreover, the catalysts were calcined at 650 °C and any remaining CH_X shall not appear. LaCoO₃ has the characteristic bands at 418 cm⁻¹ (stretching La-O vibrations) [29], 553 cm⁻¹ (stretching Co-O vibrations) [30], and 595 cm⁻¹ (bending O-Co-O vibrations) [31] that ascribed to the vibration of the metal–oxygen band. As a comparison, the FTIR spectrum of LaFeO₃ shows stretching Fe–O vibration at 555 cm⁻¹, which corresponds to the stretching Fe–O vibration in FeO₆ octahedra [32,33]. For LaFe_xCo_1−xO₃, with the increase in Fe doping, two bands at 553 cm⁻¹ and 595 cm⁻¹ gradually changed to a single band at 555 cm⁻¹, and no new band was found. This indicates that the structure distortion of LaFe_xCo_1−xO₃ changed from LaCoO₃-like to LaFeO₃-like with an increasing amount of Fe doping. This result is consistent with XRD.

3.2.3. Textural Properties

The specific surface area and pore structure of the catalyst were studied by N₂-physisorption analysis.

Table 1. Textural properties of the LaFe_xCo_1−xO₃ catalysts.

Catalysts	BET Surface Area (SBET) (m2/g)	Average Pore Diameter (DP) (nm)	Pore Volume (VP) (cm3/g)
LaCoO3	9.26	20.85	0.048
LaFe0.2Co0.8O3	10.43	18.30	0.048
LaFe0.4Co0.6O3	11.39	19.92	0.057
LaFe0.6Co0.4O3	10.70	20.67	0.055
LaFe0.8Co0.2O3	9.62	22.06	0.053
LaFeO3	8.02	24.22	0.049

summarizes the BET surface area, pore volume, and average pore diameter of all investigated catalysts. The specific surface area increased from 9.26 m²/g of LaCoO₃ to 11.39 m²/g of LaFe_0.4Co_0.6O₃ with the increase in Fe doping amount, and further decreased to 8.02 m²/g of LaFeO₃ with excess Fe in the B-site. This trend is also consistent with the increase in H₂S conversion, indicating that the increasing specific surface area of the catalyst may be favorable for the exposure of active sites, leading to higher catalytic activity [24].

The low temperature data at 180 °C are used to illustrate this effect, as all high temperature points have close to 100% H₂S conversion, which cannot reflect the intrinsic activity of the oxygen vacancies. We set the lowest surface area as 1 and scale other samples based on that. It is shown in Figure 9 that after normalization, except for LaFeO₃ and LaFe_0.8Co_0.2O₃, all other samples have similar conversions between 64% and 68%. This reflects that the intrinsic nature of oxygen vacancies is similar and a larger surface area with more oxygen vacancies exposed can lead to higher catalyst activity. This is consistent with the conclusion of this manuscript, where oxygen vacancy is an important factor to determine catalyst activity. As for the increased intrinsic activity of oxygen vacancies for LaFeO₃ and LaFe_0.8Co_0.2O₃, this could be due to Fe-enriched perovskite having oxygen vacancies with higher activity. It would be an interesting topic in the future to synthesize Fe-enriched perovskite with higher surface areas to further boost catalyst activity.

3.2.4. H₂-TPR Spectrum

H₂-TPR was conducted to determine the reducibility of the catalysts, as shown in Figure 10. It can be seen that the LaCoO₃ sample has reduction peaks at 380 °C and 405 °C, which can be attributed to the reduction of Co³⁺ to Co²⁺ [34], and a higher-temperature reduction peak at 546 °C belonging to the reduction of Co²⁺ to Co⁰ [35,36]. The addition of Fe in the LaCoO₃ leads to reduction peaks with lower temperatures. For example, LaFe_0.2Co_0.8O₃ catalyst has reduction peaks at 367 °C and 390 °C. With the further increase in the Fe doping amount, LaFe_0.4Co_0.6O₃, LaFe_0.6Co_0.4O₃, LaFe_0.8Co_0.2O₃, and LaFeO₃ show reduction peaks in the range of 300 °C–400 °C. These indicate that the addition of Fe promotes the reduction of Co³⁺. As the reduction of Co³⁺ to Co²⁺ is also associated with the creation of lattice oxygen vacancies, it can be concluded that Co and Fe co-doping facilitates oxygen vacancy formation in perovskite. This is consistent with the trend of H₂S conversion, which indicates that the ease of oxygen vacancy formation leads to increased catalyst activities for H₂S oxidation. This result is also consistent with other literature reports [37,38]. The high-temperature reduction peak (546 °C in LaCoO₃) was pushed to a much higher temperature (660 °C in LaFe_0.2Co_0.8O₃), and completely eliminated in LaFe_0.4Co_0.6O₃, LaFe_0.6Co_0.4O₃, and LaFe_0.8Co_0.2O₃ [39,40]. The higher-temperature reduction peak is associated with further Co²⁺ reduction to Co⁰, which indicates that the Co and Fe co-doped samples are highly stable even under a high-temperature reducing environment, explaining the exceptional catalyst stability observed in experiments. The reduction peak area also increased with the increase in Fe doping amount, which also indicates that Fe-doped LaCoO₃ has stronger oxidation ability and a higher amount of active lattice oxygen [41].

3.2.5. The Surface Chemical States Revealed by XPS

XPS analysis was conducted to elucidate the chemical states of surface Co, Fe, O, and S species of the fresh and used LaFe_0.4Co_0.6O₃ catalysts, shown in Figure 11. From the high-resolution spectrum of Co 2p (Figure 11a), it can be seen that there are peaks at bonding energies of 795.1 eV and 780.0 eV, corresponding to 2p_1/2 and 2p_3/2 of Co³⁺, respectively [42,43,44]. Two peaks appeared at 710.4 eV and 723.8 eV in Figure 11b and can be ascribed to Fe 2p_3/2 and Fe 2p_1/2 levels, respectively. These binding energies agreed well with an Fe³⁺ oxidation state assignment [45,46]. The O 1s spectrum was shown in Figure 11c, which showed two peaks at 531.2 eV and 529.0 eV. The peak at 529.0 eV is attributed to lattice oxygen (O_lat), while the peak at 531.2 eV corresponds to adsorbed oxygen (O_ads) on oxygen vacancies [42,47,48]. It was reported that oxygen vacancy plays an important role in the reaction, and can not only accelerate the decomposition of oxygen molecules, but also increase the mobility of lattice oxygen [49]. It can be seen that the content of lattice oxygen decreased sharply, which confirms that the lattice oxygen of the catalyst may participate in the H₂S oxidation reaction to form sulfate species. Simultaneously, the oxygen peaks were shifted towards higher B.E., with a shift value of 0.3 eV and 0.9 eV for lattice oxygen species and oxygen vacancies. As a shift to higher B.E. indicates lower electron density, this shows that the oxygen vacancies in used samples have much lower electron density. It was reported that oxygen vacancies with lower electron density lead to significantly lower ethylene yield in oxidative dehydrogenation of ethane [50]. The shift could also be caused by the formation of OH⁻ and SO₄²⁻ groups during the catalytic reaction. In this study, it is hypothesized that the deactivation of the H₂S selective oxidation reaction is also mainly related to the lower electron density of oxygen vacancies. Figure 11d shows the S 2p XPS spectrum of the used LaFe_0.4Co_0.6O₃ catalyst to investigate the deactivation process of the catalyst. Tiny peaks around 163.3 eV were detected, which can be ascribed to the presence of the conjugated sulfur (S_n) [51], indicating a sulfur deposit on the sample. The main peaks of S 2p_1/2 and S 2p_3/2 at binding energies of 170.3 eV and 169.0 eV indicate that S exists as a sulfate species [52]. It should be noted that deposited sulfur could also lead to catalyst deactivations.

3.3. Catalytic and Deactivation Mechanisms

Based on the results of experiments and characterizations, the possible reaction mechanism of H₂S selective oxidation over LaFe_0.4Co_0.6O₃ can be proposed as follows: H₂S was first adsorbed on the oxygen vacancies on the LaFe_0.4Co_0.6O₃ catalyst surface, then the adsorbed H₂S dissociated into HS^- or S²⁻ ions. Subsequently, the HS⁻ or S²⁻ ions were oxidized by the surface-active oxygen and/or lattice oxygen of the LaFe_0.4Co_0.6O₃ catalyst, and Fe³⁺ was reduced to Fe²⁺ or Co³⁺ to Co²⁺ and oxygen vacancies were formed. Then, O₂ diffused and adsorbed on the catalyst surfaces, and oxidized Fe²⁺ to Fe³⁺ or Co²⁺ to Co³⁺, thereby replenishing the lattice oxygen. The possible reaction mechanism of H₂S selective oxidation over the LaFe_0.4Co_0.6O₃ is proposed as illustrated in Figure 12. Compared to LaFeO₃ and LaCoO₃, co-doped LaFe_0.4Co_0.6O₃ has high oxygen vacancy concentration, leading to increased catalyst activities.

The S 2p XPS spectrum of used LaFe_0.4Co_0.6O₃ revealed the deactivation process of the catalyst. The sulfate accounted for 92% of the total molar amount of sulfur and the elemental sulfur accounted for 8%. The Fe 2p energy value of the used catalyst shifted slightly to a higher value, an index of an increase in positive polarity, indicating that the interaction between sulfate anion and iron cation generates Fe₂(SO₄)₃ [24]. It can be concluded that the catalyst deactivation is mainly due to the formation of iron sulfate, which will destroy the redox process of Fe³⁺/Fe²⁺, resulting in a decrease in catalytic activity. The deposition of elemental sulfur also leads to a reduction in catalytic activity.

4. Conclusions

In summary, a series of LaFe_xCo_1−xO₃ catalysts were synthesized by the sol-gel method and investigated for H₂S selective oxidation. XRD, N₂-adsorption, H₂-TPR, and XPS analyses were conducted. The conclusions of this study can be summarized as follows:

(1)

The incorporation of Fe into LaCoO₃ had a great influence on the conversion and selectivity of H₂S selective oxidation reaction. It was shown that Fe doping not only enhanced catalytic activity for the H₂S selective oxidation reaction, but also reduced the optimal reaction temperature.

(2)

LaFe_0.4Co_0.6O₃ had the highest catalytic activity and the catalyst deactivation was not obvious before 77 h, and its stability is better than some types of catalysts. Analysis of XRD and FTIR showed that the LaFe_xCo_1−xO₃ catalyst existed in a single perovskite structure. H₂-TPR indicates that oxygen vacancy creation is more feasible on a co-doped LaFe_xCo_1−xO₃ catalyst.

(3)

The reaction mechanism and inactivation mechanism were predicted. The presence of lattice oxygen in the LaFe_0.4Co_0.6O₃ catalyst and the oxygen vacancies generated in the reaction are beneficial to the H₂S selective oxidation reaction. In addition, the deposition of elemental sulfur and the formation of oxygen vacancies with lower electron density could lead to catalyst deactivation.