The effect of rare earth modification on the denitration performance of Fe₂O₃–MnO₂/TiO₂ catalysts

Abstract

High denitration efficiency and strong adaptability to flue gas temperature fluctuations are essential properties for NH₃ selective catalytic reduction (SCR) catalysts. In this study, rare earth oxide modification of the catalyst 8Fe₂O₃–40MnO₂/TiO₂ via the addition of CeO₂ during the preparation process is explored. The results show that the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst has excellent denitration properties, with a denitration efficiency of greater than 90%. The operational temperature range reaches 129 to 390 °C, and its N₂ selectivity, anti-SO₂ performance and anti-H₂O performance are good, with the denitration performance being significantly improved. The addition of 6% CeO₂ promotes the lattice shrinkage of TiO₂, improves its dispersion, refines the grain size, and increases the specific surface area of the catalyst. At the same time, CeO₂ increases the chemical adsorption of oxygen on the catalyst surface and the proportion of low-valence metal ions, promotes electron transfer between active elements, generates more surface reactive oxygen species, and increases the content of oxygen vacancies and adsorption sites for NO_x and NH₃. It also significantly improves the redox performance of the catalyst, which is especially conducive to the generation of weak and medium-strong acidic sites on the catalyst surface. The NH₃-SCR reaction on the surface of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst follows both L–H and E–R mechanisms, with the L–H mechanism being dominant.

---

1 Introduction

Nitrogen oxide (NO_x) pollutants have become one of the main sources of air pollution. Among industrial NO_x emission sources, the power industry accounts for the largest proportion.^1,2 Excessive NO_x in the atmosphere can lead to disasters such as acid rain, photochemical smog, eutrophication of water bodies, and the greenhouse effect.³ In recent years, China's regulations on industrial NO_x emissions have become increasingly stringent, placing significant environmental pressure on metallurgical, building materials, glass, and other industrial enterprises. The selective catalytic reduction (SCR) method is currently the most widely used flue gas denitration technology for controlling NO_x emissions from industrial sources.⁴ The most mature catalyst for this method is V₂O₅-WO₃(MoO₃)/TiO₂, which belongs to the category of medium-temperature catalysts.⁵ To avoid SO₂ poisoning, the V₂O₅-WO₃(MoO₃)/TiO₂ catalyst is placed after the desulfurization and dust removal units in the flue gas treatment process. As a result, the flue gas must be reheated before denitration, which imposes greater economic pressure on enterprises. The vanadium component in the catalyst is toxic, and the disposal and recycling of waste catalyst can easily cause environmental pollution. Therefore, non-toxic SCR catalysts capable of denitration at low temperatures have broad application prospects.

Among the many transition metal oxides, MnO_x has the advantages of multivalent states, strong acidity, and low cost. Fe₂O₃ can further expand the active temperature range of the SCR catalyst. Therefore, low-temperature SCR denitration catalysts with MnO_x and Fe₂O₃ as the main active components or additives have become research hotspots in the field of denitration catalysts.⁶ Liu et al.⁷ used the sol–gel method to prepare a series of Mn–Ti oxide catalysts. Through experiments, it was found that such catalysts exhibit high low-temperature SCR activity. This excellent low-temperature NH₃-SCR activity can be attributed to their appropriate structural properties, amorphous manganese oxides with equal proportions of Mn³⁺/Mn⁴⁺, good low-temperature reducibility, and abundant surface acidic sites. Mu et al.⁸ prepared a Fe₂O₃–MnO₂/TiO₂ catalyst using an ethylene-glycol-assisted impregnation method; the catalyst showed excellent low-temperature activity, a low apparent activation energy, and strong resistance to sulfur poisoning. Characterization results indicate that the catalyst has high dispersion, small particle size, and partial doping of Fe into the TiO₂ lattice to form a Fe–O–Ti structure, which enhances the electronic inductive effect and increases the proportion of surface-chemisorbed oxygen. The NO oxidation is thus enhanced through the “fast SCR” process, which is conducive to low-temperature SCR activity.

China is the country with the richest rare earth ore resources in the world.⁹ The combination of rare earth oxides with other metal oxides for the research and development of denitration catalysts can achieve efficient denitration of industrial flue gas at a low cost; this catalyst preparation technology route is in alignment with China's national conditions. Many studies have shown that rare earth modification can improve the denitration performance and sulfur resistance of SCR catalysts. Zhai et al.¹⁰ used the coprecipitation method to modify an MnO_x catalyst with Pr and found that the Pr-modified MnO_x catalyst had significantly improved tolerance to SO₂. The characterization results showed that the Pr-modification of the MnO_x catalyst improves the specific surface area, dispersion, reducibility, surface Mn⁴⁺ content, and relative percentage of chemically adsorbed oxygen (O_α) of the catalyst. Additionally, Pr modification increases the adsorption of NH₃ on the surface of the catalyst, resulting in more Lewis acidic sites on the surface of the MnPrO_x-0.1 catalyst. These factors contributed to the good catalytic properties of the Pr-modified MnO_x catalyst. Research by Liu et al.¹¹ found that the addition of Ce can promote the dispersion of MnO_x on the TiO₂ carrier. Doping Ce into MnO_x can also increase the content of Mn⁴⁺ species. These species play a vital role in the NO oxidation reaction, which can trigger the “fast SCR” reaction and promote the conversion of NO_x. Gao et al.¹² found that the introduction of Sm can effectively inhibit the crystallization process of MnO_x and TiO₂, increase the specific surface area, enhance the surface concentration of chemically adsorbed oxygen, and increase the number of strong acid sites on the surface, thereby optimizing the catalytic performance of the catalyst. Huang et al.¹³ added 3% of a rare earth element (Nd) to a 30% Mn/TiO₂ catalyst to improve its low-temperature catalytic performance. The addition of Nd increases the specific surface area of the catalyst, decreases the average pore size, and enhances the dispersion of MnO_x, which are the main reasons for the enhanced catalytic activity.

Through previous research and early experimental demonstrations by this group, it has been concluded that the denitration performance of the Fe₂O₃–MnO₂/TiO₂ catalyst is superior when MnO₂ and Fe₂O₃ account for 0.4 and 0.08 of the overall mass of the catalyst, respectively.⁸ CeO₂ exhibits acidity, which can enhance the surface acidity and redox properties of the catalyst.^14,15 At the same time, the interactions among Ce, Mn, and Fe are helpful in increasing the number of active sites of the catalyst and improving the dispersion state of the active components on the carrier.^16,17 In order to clarify the mechanism by which CeO₂ modification influences the denitration performance of the catalyst 8Fe₂O₃–40MnO₂/TiO₂, this study employs XRD, BET, XPS, SEM, TEM, H₂-TPR, NH₃-TPD, and in situ DRIFTS to investigate the effects of modification with CeO₂ on the denitration performance, catalytic material morphology, surface acidity, redox capacity, and catalytic mechanisms of the 8Fe₂O₃–40MnO₂/TiO₂ catalyst. This study aims to provide theoretical and technical support for the research, development, and application of rare-earth-modified denitration catalysts.

2 Materials and methods

2.1 Catalyst preparation

The Fe₂O₃–CeO₂–MnO₂/TiO₂ catalyst was prepared by the co-precipitation method. The preparation steps were as follows: manganese acetate tetrahydrate ((CH₃COO)₂Mn·4H₂O), cerium chloride heptahydrate (CeCl₃·7H₂O), iron chloride hexahydrate (FeCl₃·6H₂O), and anatase (TiO₂) were added to a beaker containing deionized water in a specific weight ratio. The mixture was continuously stirred for 0.5 hours, and ammonia was then added dropwise to the catalyst solution until the pH reached 9–10. After that, the mixture was stirred for an additional 3 hours and then filtered, and deionized water was added to the filter cake. The mixture was stirred and washed for 0.5 hours and then filtered again. These operations were repeated three times. The resulting filter cake was dried at 75 °C for 24 hours. Finally, the dried filter cake was placed in a muffle furnace and calcined for a certain period of time to obtain the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalyst. The value y represents the mass percentage of CeO₂ in the entire catalyst (×100), and the values used were 0, 3, 6, and 9. The actual y values were determined by ICP and rounded to the nearest integer. The preparation process is shown in Fig. 1.

Fig. 1 The 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalyst preparation process.

2.2 Catalyst characterization

X-ray diffraction (XRD) analysis was conducted using an X-ray diffractometer, model JSM-7800F, manufactured by the Japanese Electronics Company. The test conditions included a Cu target, Kα radiation, an analysis voltage of 40 kV, an analysis current of 40 mA, a scanning range of 5° to 90°, and a scanning rate of 2° per minute.

The N₂ adsorption–desorption curve was measured using an ASAP-2020 physical adsorption instrument from Micromeritics Company in the United States. Prior to the test, the sample was vacuum treated at 400 °C for 4 hours. The Brunauer–Emmett–Teller (BET) method was used to calculate the specific surface area of the sample, and the Barrett–Joyner–Halenda (BJH) method was employed to determine the pore size distribution.

For scanning electron microscope (SEM) analysis, a Hitachi SU8020 ultra-high-resolution field emission scanning electron microscope was used, with a magnification range from 300 000× to 800 000×. The X-ray spectrometer attached to the SEM has an elemental analysis range from B5 to U92 and an energy resolution of 130 eV.

A JEM-2100F transmission electron microscope (TEM, Japan) was used, which has a point resolution of 0.23 nm, a magnification range of 50 to 1 000 000, and an acceleration voltage of 200 kV. Using this instrument, we observed the particle size and microstructure of the catalyst and conducted nanoscale analysis of the catalyst through high-resolution lattice fringe images.

X-ray photoelectron spectroscopy (XPS) was carried out using the Thermo Fisher Scientific Escalab 250Xi X-ray Photoelectron Spectrometer. The excitation source was Al Kα (1486.6 eV), and the acceleration voltage was 150 W. The binding energy of C 1s (284.8 eV) was used as the internal standard. Peak fitting was performed using XPS Peak software.

Temperature-programmed reduction (TPR) analysis was conducted using a Micromeritics Autochem II 2920 chemical adsorption instrument (USA). The test conditions were as follows: approximately 70 mg of the sample was placed in a quartz tube, heated to 300 °C for drying and pretreatment, purified with He gas for 1 hour, and then cooled to room temperature. A 10% H₂/Ar mixture was passed through it for 30 minutes until the baseline was stable. Finally, it was heated to 800 °C in a 10% H₂/Ar atmosphere for TPR. The gas flow rate was 50 mL min⁻¹, and the heating rate was 10°C min⁻¹. The detector used was a thermal conductivity detector (TCD), which detects the consumption of the reducing gas.

Temperature-programmed desorption (TPD) tests were carried out on an AutoChem II 2920 Chemisorption Analyzer. The specific conditions were as follows: approximately 70 mg of the sample was weighed out and placed into a quartz tube; the sample was then heated to 300 °C for drying and pretreatment under a flowing He gas atmosphere. After 1 hour, it was cooled to room temperature. Subsequently, a 10% NH₃/He gas mixture was introduced for 1 hour to achieve saturation; the gas was then switched to He, and the sample was purged for 1 hour to remove physically adsorbed NH₃. Finally, the sample was heated to 800 °C at a rate of 10 °C min⁻¹ under a He gas flow (50 mL min⁻¹), and the desorbed gas was detected using a TCD.

In situ diffuse reflectance infrared Fourier transform spectroscopy (in situ DRIFTS) was performed using a Thermo Scientific Nicolet IS50 Fourier Transform Infrared (FTIR) spectrometer. The measurement procedure was as follows: approximately 15 mg of sample was placed flat in a small crucible and fixed inside a high-temperature in situ cell. The sample was pretreated with N₂ at 400 °C for 1 hour, then cooled to room temperature. Background spectra (32 scans, 4 cm⁻¹ resolution) were collected periodically as references. Subsequently, the corresponding reaction gas (1000 ppm NH₃, 1000 ppm NO, and 5% O₂ by volume) was introduced for in situ infrared adsorption studies. For NH₃ or NO + O₂ adsorption, the gas was introduced at 200 °C, and data were collected over time. For the experiment involving initial NH₃ pre-adsorption followed by NO + O₂ exposure, saturated NH₃ adsorption was first conducted at 200 °C, after which the NO + O₂ mixture was introduced, and data were collected over time. Similarly, for the experiment involving initial NO + O₂ pre-adsorption followed by NH₃ exposure, saturated NO + O₂ adsorption was performed at 200 °C, followed by NH₃ introduction, and data were collected over time. At the end, the background spectrum was subtracted to obtain the infrared spectra at each time point.

2.3 Measurement of catalyst activity

The laboratory micro-reaction device shown in Fig. 2 was used to simulate the components of flue gas. The test conditions were as follows: a 5 mL catalyst sample, a gas hourly space velocity (GHSV) of 36 000 h⁻¹, and a simulated flue gas composition consisting of 600 ppm NO, 300 ppm O₂, 600 ppm NH₃, and 60 ppm SO₂ (when required), with N₂ as the balance gas. Five milliliters of the catalyst to be tested were placed in a fixed-bed reactor, the flow control system was turned on, and the gas parameters were set. The pipeline was purged with N₂, and the air-tightness of the system was checked after 5 minutes. Then, the final temperature was set, the temperature control switch was turned on, other gases were introduced, and the NO concentration detected by the analyzer was recorded as the temperature changed. After the test was completed, the other gas switches were turned off, N₂ continued to flow, and the pipeline was purged with N₂ until the NO detection indicator dropped to zero.

Fig. 2 A schematic diagram of the fixed-bed catalytic reactor.

The equations for calculating the denitration efficiency and N₂ selectivity of the catalyst are as follows:

where [NO]_in represents the input concentration of NO, [NO]_out denotes the output concentration of NO, [NH₃]_in signifies the input concentration of NH₃, [NH₃]_out refers to the output concentration of NH₃, and [N₂O] is the output concentration of N₂O.

3 Results and discussion

3.1 Morphology and structure of the catalysts

3.1.1 XRD, BET surface area and pore morphology. The phase composition analysis results for the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalyst are shown in Fig. 3.

Fig. 3 XRD diffraction patterns of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts.

The characteristic peaks of the catalyst in the figure are mainly the diffraction peaks of anatase TiO₂ and MnO₂. There are no obvious Fe₂O₃ or CeO₂ diffraction peaks, indicating that these form an amorphous structure or are evenly distributed on the surface of the support. According to the Bragg equation, 2d sin θ = nλ (wheredis the lattice spacing, θrepresents the diffraction angle, n denotes the reflection series, and λ is the wavelength), the diffraction angle is inversely proportional to the lattice spacing of the crystal. As the CeO₂ content increases, the TiO₂ diffraction peak at 25.3° shifts toward higher angle. The 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst exhibits the largest shift and the lowest intensity at 25.3°. At this point, the unit cell parameters of TiO₂ in the catalyst are the smallest. The lattice spacing of TiO₂ in each catalyst is listed in Table 1. As the CeO₂ content increases from 0% to 6%, the lattice spacing of TiO₂ decreases from 0.351 nm to 0.345 nm. With further increase of the CeO₂ content, the lattice spacing of TiO₂ no longer decreases, indicating that an appropriate amount of amorphous CeO₂ inhibits the lattice expansion of the TiO₂ support and enhances the dispersion of TiO₂. At the same time, the low intensity of the MnO₂ diffraction peak may be due to interaction between CeO₂ and MnO₂ that reduces the crystallinity of MnO₂ and increases its dispersion on the surface of the catalyst. The improved dispersion of the active components can provide better mass transfer and diffusion properties, which helps to accelerate the acid cycle and the redox cycle.¹⁸

Table 1 Specific surface area and pore size parameters of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts

Catalyst	Specific surface area/m^2 g^−1	Pore capacity/cm^3 g^−1	Pore size/nm	TiO2 (101) crystal plane spacing/nm
8Fe2O3–40MnO2/TiO2	84.13	0.222	15.36	0.351
8Fe2O3–3CeO2–40MnO2/TiO2	85.26	0.243	13.59	0.349
8Fe2O3–6CeO2–40MnO2/TiO2	87.73	0.268	12.22	0.345
8Fe2O3–9CeO2–40MnO2/TiO2	87.48	0.259	12.47	0.346

---

The NH₃-SCR reaction mainly occurs on the surface of the catalyst. Micropores and mesopores can provide a larger internal surface area and pore volume during the SCR reaction, resulting in a higher BET surface area, which is conducive to improving the denitration performance of the catalyst.¹⁹ Therefore, in this study, the pore structure characteristics of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalyst were tested using the N₂ adsorption–desorption measurement method, and the results are shown in Table 1. The specific surface area and pore capacity of the 8Fe₂O₃–40MnO₂/TiO₂ catalyst are the smallest, at 84.13 m² g⁻¹ and 0.222 cm³ g⁻¹, respectively. As the CeO₂ content increases, the specific surface area and pore capacity of the catalyst gradually increase, while the average pore size decreases. This phenomenon indicates that more microporous structures are formed on the surface of the catalyst, which is conducive to providing more active sites and adsorbing more NH₃. The specific surface area (87.73 m² g⁻¹) and pore capacity (0.268 cm³ g⁻¹) of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst are the largest, and its average pore size (12.22 nm) is the smallest. The main reason is that an appropriate amount of CeO₂ inhibits the growth of the TiO₂ support crystal particles, thereby increasing the specific surface area and pore capacity of the catalyst, which is consistent with the XRD results.

3.1.2 Morphological analysis. The SEM image of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalyst is presented in Fig. 4.

Fig. 4 SEM images of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts: (a) 8Fe₂O₃–40MnO₂/TiO₂; (b) 8Fe₂O₃–3CeO₂–40MnO₂/TiO₂; (c) 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂; and (d) 8Fe₂O₃–9CeO₂–40MnO₂/TiO₂.

It can be seen from Fig. 4 that the catalyst mainly exists in the form of particles with uneven sizes, and these particles exhibit varying degrees of agglomeration. In Fig. 4(a), it can be clearly observed that the 8Fe₂O₃–40MnO₂/TiO₂ catalyst has a higher content of coarse particles, a more uneven particle size distribution, and a more severe agglomeration phenomenon. As the CeO₂ content increases, the agglomeration of the catalyst particles gradually decreases. Fig. 4(c) shows that the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst has a more uniform particle size and less agglomeration compared to the other catalysts, indicating that it can provide more active and reactive sites for the denitration reaction. This is consistent with the denitration performance test results presented in Section 3.6. Fig. 4(d) shows that further increasing the CeO₂ content inhibits the dispersion of catalyst particles, resulting in an increase in surface aggregation.

The TEM images of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalyst are shown in Fig. 5. Most of the catalysts exist as particles of different sizes, and the dark fine particles are generally adsorbed on the surface of the coarse particles (i.e., the TiO₂ carrier). Most of the TiO₂ grain sizes in Fig. 5 are between 10 and 50 nm. As the CeO₂ content increases, the TiO₂ grain size tends to decrease, indicating that the addition of CeO₂ can inhibit the growth of TiO₂ grains, which is consistent with the XRD analysis results. At the same time, with increasing CeO₂ content, the number of fine particles the of active substances loaded on the surface of the carrier increases, and their distribution gradually becomes denser and more uniform. The number of active substance particles exposed on the surface of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst was significantly increased, which provides more active reaction sites for the catalytic reaction and is conducive to promoting the NH₃-SCR reaction.

Fig. 5 TEM images of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts: (a) 8Fe₂O₃–40MnO₂/TiO₂; (b) 8Fe₂O₃–3CeO₂–40MnO₂/TiO₂; (c) 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂; and (d) 8Fe₂O₃–9CeO₂–40MnO₂/TiO₂.

Fig. 6 presents the high-resolution lattice fringe images of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts. In the figure, the lattice fringe spacings of approximately d = 0.351 nm correspond to the TiO₂ (101) crystal plane, while lattice fringe spacings of d = 0.236 nm and d = 0.287 nm correspond to the MnO₂ (004) and MnO₂ (200) crystal planes, respectively. As the CeO₂ content increases, the lattice fringe spacing corresponding to the TiO₂ (101) crystal plane gradually decreases, indicating that an increase in CeO₂ content can inhibit the crystal growth of TiO₂, which is consistent with the XRD analysis results mentioned above. Additionally, no lattice fringes associated with Fe₂O₃ or CeO₂ are observed in the figure, suggesting that the active components Fe₂O₃ and CeO₂ may exist in the catalyst in the form of a solid solution, while MnO₂ is evenly dispersed on the surface of the TiO₂ support in the form of fine particles. This is consistent with the XRD results, in which the diffraction peaks of Fe₂O₃ and CeO₂ disappeared, and only the characteristic diffraction peaks of TiO₂ and MnO₂ were present.

Fig. 6 Lattice fringe patterns of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts: (a) 8Fe₂O₃–40MnO₂/TiO₂; (b) 8Fe₂O₃–3CeO₂–40MnO₂/TiO₂; (c) 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂; and (d) 8Fe₂O₃–9CeO₂–40MnO₂/TiO₂.

3.2 Surface properties of the catalysts

In order to explore the influence of the addition of CeO₂ on the surface elemental distribution and redox performance of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalyst, XPS tests were performed on catalysts prepared using different amounts of CeO₂, and the results are shown in Fig. 7 and Table 2.

Fig. 7 XPS spectra of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts: (a) O 1s; (b) Mn 2p; (c) Ce 3d; and (d) Fe 2p.

Table 2 Surface element content of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts (mass%)

Catalyst	Fe	Ce	Mn	Ti	O
8Fe2O3–40MnO2/TiO2	6.37	1.53	11.72	12.34	68.04
8Fe2O3–3CeO2–40MnO2/TiO2	5.46	3.24	11.07	11.36	68.87
8Fe2O3–6CeO2–40MnO2/TiO2	4.96	4.51	9.98	10.30	70.25
8Fe2O3–9CeO2–40MnO2/TiO2	5.03	5.65	9.35	10.84	69.13

---

Fig. 7(a) shows the O 1s spectrum of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts. Two characteristic peaks located around 532.2 eV and 530.1 eV appear in the figure, which correspond to the chemisorbed oxygen O_α and lattice oxygen O_β, respectively. As the amount of CeO₂ added was increased, the ratio of O_α to the total oxygen (O_α/(O_α + O_β)) on the catalyst surface gradually increased. As presented in Table 2, the total amount of surface oxygen changed relatively little, indicating that part of the O_β was converted to O_α as the amount of CeO₂ added was increased. Since O_α has a higher mobility than O_β, it is more active, and the increase in the proportion of O_α is conducive to NO oxidation and the subsequent “fast SCR” reaction.²⁰ The maximum O_α/(O_α + O_β) value of 45.7% was achieved by the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst surface. The main reason is that as amount of CeO₂ increases, the Ce³⁺ content on the catalyst surface increases, causing the catalyst surface to produce more oxygen vacancies and O_α.

Fig. 7(b) and (d) show the Mn 2p and Fe 2p profiles of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts, respectively. As the amount of CeO₂ was changed, the position of each of the characteristic peaks and the peak areas remained relatively constant. That is, the ratios of Fe²⁺/Fe³⁺ and Mn³⁺/Mn⁴⁺ on the catalyst surface did not change significantly, indicating that the variation in the amount of CeO₂ added did not cause a major change in the valence states of the elements Mn and Fe on the surface of the catalyst.

The Ce 3d map of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalyst is shown in Fig. 7(c). The Ce 3d orbital includes the Ce 3d_3/2 spin–orbit component (u) and the Ce 3d_5/2 spin–orbit component (v). The Ce species on the surface of the catalyst mainly exist in the form of Ce₂O₃ and CeO₂. The peaks u′ and v′ in the figure correspond to Ce³⁺, and the peaks u, u″, u‴, and v, v″, and v‴ belong to Ce⁴⁺. It can be found that as the amount of CeO₂ increases, the ratio of Ce³⁺ to total Ce (Ce³⁺/(Ce³⁺ + Ce⁴⁺)) on the catalyst surface also increases, and the proportion of Ce³⁺ on the catalyst surface of 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ reaches 20.58%, indicating that increasing addition of CeO₂ promotes the forward reaction: Ce⁴⁺ ↔ Ce³⁺ + Cat–[O]. Studies have shown that the presence of Ce³⁺ on the surface of the catalyst is conducive to improving its catalytic performance, because the Ce³⁺/Ce⁴⁺ redox cycling can generate charge imbalance, oxygen vacancies, and unsaturated chemical bonds, among others.²¹ Therefore, a higher proportion of Ce³⁺ can increase the number of active sites on the surface of the catalyst and enhance its redox performance.

3.3 Raman spectroscopic analysis of the catalysts

In metal oxides, the formation of oxygen vacancies reduces the symmetry of the crystal structure, thereby affecting the Raman spectrum. Raman spectral characterization was performed on the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts with different amounts of added CeO₂, and the results are shown in Fig. 8. The characteristic peak at 148 cm⁻¹ corresponds to the vibration of the Ti–O bond in anatase TiO₂, while the characteristic peaks at 393 cm⁻¹, 509 cm⁻¹, and 642 cm⁻¹ correspond to CeO₂, MnO₂, and Fe₂O₃, respectively. As the amount of CeO₂ added increases, the intensity of the characteristic peak at 393 cm⁻¹ gradually increases, and the intensity of the characteristic peak at 148 cm⁻¹ also changes significantly. The intensity of the characteristic Ti–O bond peak of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst increases significantly, indicating that with the addition of 6% CeO₂, the number of oxygen vacancies on the catalyst surface increases. Combined with the previous XPS detection results, this indicates that an increase in the ratio of Ce³⁺ can promote an increase in oxygen vacancies on the catalyst surface.

Fig. 8 Raman spectra of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts at 532 nm.

3.4 Particle size analysis of the catalysts

The particle size of SCR denitrification catalysts is an important factor affecting their catalytic activity. Generally, the smaller the particle size of a catalyst, the larger its specific surface area and the greater its contact area with NO_x, NH₃, and other components in flue gas, resulting in more reaction sites, which can significantly enhance the rate and efficiency of the denitrification reaction. Particle size analysis was conducted on the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts with different amounts of added CeO₂, and the results are shown in Fig. 9. From Fig. 9, it can be seen that the particle size distribution of the catalysts is not uniform. As the amount of CeO₂ added increases, the number of particles with a size less than 10 µm significantly increases, while there is no significant change in the D90 particle size. The 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst has the finest particles with a size of less than 45 µm, indicating that the addition of 6% CeO₂ leads to refinement of the particles of the catalyst, which is consistent with the BET results and is beneficial for the NH₃-SCR reaction.

Fig. 9 Particle size analysis of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts: (a) 8Fe₂O₃–40MnO₂/TiO₂; (b) 8Fe₂O₃–3CeO₂–40MnO₂/TiO₂; (c) 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂; and (d) 8Fe₂O₃–9CeO₂–40MnO₂/TiO₂.

3.5 Acidic site distribution of the catalysts

In the NH₃-SCR reaction, the adsorption of NH₃ at the active sites on the surface of the catalyst is the primary reaction. This can take two forms: one in which NH₃ is adsorbed at Brønsted acid sites to form NH₄⁺, and another in which NH₃ is adsorbed at Lewis acid sites to form coordinatively bound NH₃ species. The key to Lewis acid sites is the “electron-deficient center”, and there are unfilled electronic orbitals (such as d-orbitals of metal ions and p-orbitals of non-metal atoms) in the support that can accept lone pair electrons from base molecules to form coordination bonds. The core of this structural model is “the composition of electron-deficient center atoms and their electronic states”. The key to Brønsted acid sites is the “weak interaction between protons and the support”, whereby protons (H⁺) can bind to negatively charged groups on the support through hydrogen bonds or coordination bonds, and can dissociate and transfer to base molecules (such as NH₃ and pyridine) under the reaction conditions. This structural model must meet two conditions: (1) the support must have groups that can stably hold negative charges; (2) the H⁺ must be able to dissociate after binding. Therefore, it is necessary to study the acid density, acid strength, and acid strength distribution on the surface of catalysts.²² The 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalyst was tested using NH₃-TPD, and the results are depicted in Fig. 10 and Table 3.

Fig. 10 NH₃-TPD patterns of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts.

Table 3 NH₃ adsorption capacity of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts

Catalyst	NH3 adsorption at the weak acid sites/µmol g^−1	NH3 adsorption at the medium-strong acid sites/µmol g^−1	NH3 adsorption at the strong acid sites/µmol g^−1
8Fe2O3–40MnO2/TiO2	175.46	295.37	983.54
8Fe2O3–3CeO2–40MnO2/TiO2	168.72	247.18	1022.35
8Fe2O3–6CeO2–40MnO2/TiO2	187.15	318.66	1108.59
8Fe2O3–9CeO2–40MnO2/TiO2	182.52	256.35	1107.28

---

As can be seen from Fig. 10, each group of catalysts exhibits four obvious desorption peaks in the range from 50 to 800 °C. Among them, the desorption peak with the peak center between 25 and 200 °C corresponds to weak acid sites and is attributed to NH₃ adsorbed at Lewis acid sites or physically adsorbed NH₃; the desorption peak with the peak center between 200 and 400 °C corresponds to medium-strong acid sites and is attributed to NH₃ adsorbed at Lewis acid sites. The desorption peak with the peak center above 400 °C corresponds to strong acid sites and is attributed to NH₃ adsorbed at Brønsted acid sites.²³ As the content of CeO₂ increases, all of the desorption peaks of the catalyst shift toward lower temperatures, which indicates that the NH₃ adsorption capacity of the acidic sites on the surface in the low-temperature zone is enhanced, thereby improving the low-temperature denitration capacity of the catalyst. As shown in Table 3, when the content of CeO₂ is increased to 5%, the adsorption of NH₃ at weak acid sites, medium-strong acid sites, and strong acid sites on the surface of the catalyst increases to a certain extent, indicating that the Ce–O–Ce and Ce=O species in CeO₂ provide part of the Lewis and Brønsted acid sites for the catalyst.

3.6 Redox performance of the catalysts

In the NH₃-SCR reaction, the redox performance of the catalyst is essential to the denitration reaction cycle. H₂-TPR is an effective method for evaluating the reduction performance of NH₃-SCR catalysts. The 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts were characterized by H₂-TPR, and the results are presented in Fig. 11 and Table 4.

Fig. 11 H₂-TPR patterns of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts.

Table 4 H₂ consumption of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts

Catalyst	Fe^3+ → Fe^2+ H2 consumption/mmol g^−1	Mn^4+ → Mn^3+ H2 consumption/mmol g^−1	Ce^4+ → Ce^3+ H2 consumption/mmol g^−1	Total consumption of H2/mmol g^−1
8Fe2O3–40MnO2/TiO2	2.65	4.65	0.83	8.13
8Fe2O3–3CeO2–40MnO2/TiO2	2.83	5.94	1.01	9.78
8Fe2O3–6CeO2–40MnO2/TiO2	4.18	5.73	1.34	11.25
8Fe2O3–9CeO2–40MnO2/TiO2	3.23	4.71	2.46	10.40

---

As shown in Fig. 11, three reduction peaks appeared in each catalyst sample. Among them, the reduction peak at 250–350 °C corresponds to the reduction of Fe³⁺ → Fe²⁺; the reduction peak at 350–520 °C corresponds to the reduction of Mn⁴⁺ → Mn³⁺; and the reduction peak at 520–800 °C corresponds to the reduction of Ce⁴⁺ → Ce³⁺.^24–26 With the increase in CeO₂ content, most of the reduction peaks shift toward lower temperatures, indicating a decrease in the reduction temperature of the catalyst. This suggests that the surface chemical adsorption of oxygen is enhanced, which is conducive to the reduction of surface species. When the CeO₂ content is increased from 3% to 6%, the Mn⁴⁺ → Mn³⁺ reduction peak area decreases. The main reason for this is that there is an interaction between CeO₂ and MnO₂, Ce³⁺ + Mn⁴⁺ ↔ Ce⁴⁺ + Mn³⁺, which makes it more difficult for some Mn⁴⁺ to be directly reduced. From the perspective of H₂ consumption (Table 4), the H₂ consumption of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst during the entire reduction process is increased significantly compared to that of the other catalysts, indicating that an appropriate CeO₂ content can enhance the overall reducibility of the catalyst, which is beneficial for promoting the NH₃-SCR reaction.

3.7 Catalytic mechanism analysis of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalysts

The above analysis indicates that the catalytic activity of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts is influenced by a variety of factors, and that one of the main reasons is the variation in acidic sites. The SCR reaction is a heterogeneous gas–solid reaction catalyzed by a solid catalyst, which requires the catalyst to have sufficient active sites to adsorb the reaction gases. In order to study the changes in the adsorption characteristics of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst during the low-temperature NH₃-SCR reaction and to explore the catalytic reaction mechanism on the catalyst surface, in situ DRIFTS was used to investigate the interaction between the reaction gases (NH₃ and NO) and the catalyst surface.

3.7.1 Adsorption of NH₃. In order to further clarify the properties of the NH₃(ad) species on the surface of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst, in situ DRIFTS of NH₃ adsorption at different temperatures was conducted, and the results are shown in Fig. 12. Based on the spectrum of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst, the bands that appear at 1768 cm⁻¹, 1603 cm⁻¹, 1342 cm⁻¹, and 1141 cm⁻¹ correspond to NH₃ molecules coordinated to Lewis acid sites, while the band at 1420 cm⁻¹ corresponds to NH₄⁺ ions adsorbed at Brønsted acid sites.^27,28 As the temperature increases, the intensity of the absorption peaks associated with each acid site decreases due to the desorption and decomposition of adsorbed NH₃. Comparatively, the absorption peaks corresponding to the Brønsted acid sites weaken more rapidly, indicating that the Lewis acid sites have higher thermal stability than the Brønsted acid sites. Therefore, at low temperatures, NH₃ adsorption in the catalytic reaction involves both Lewis and Brønsted acid sites, whereas at high temperatures, NH₃ adsorption mainly occurs at Lewis acid sites.

Fig. 12 In situ DRIFTS spectra of NH₃ adsorption on the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst.

3.7.2 Co-adsorption of NO + O₂. To understand the microscopic process of the NH₃-SCR reaction on the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst, it is necessary to distinguish the adsorbent species on the surface of the catalyst. Therefore, an in situ NO + O₂ co-adsorption experiment was carried out on the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst, and the results are shown in Fig. 13. The 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst was exposed to a NO + O₂ gas mixture (500 ppm NO + 5% O₂, N₂ balance) at room temperature, and multiple infrared vibration signal peaks appeared between 1000 and 2000 cm⁻¹. Among them, the bands located at 1408 cm⁻¹ and 1593 cm⁻¹ can be attributed to bidentate nitrates; the band located at 1153 cm⁻¹ is related to monodentate nitrates; the band located at 1342 cm⁻¹ corresponds to M–NO₂ nitro compounds; the band located at 1615 cm⁻¹ corresponds to adsorbed NO₂; and the band located at 1749 cm⁻¹ corresponds to the physical adsorption of NO₂.^29,30 As can be seen from Fig. 11, as the temperature rises, the intensity of the absorption peaks corresponding to monodentate nitrate (1153 cm⁻¹) gradually decreases with increasing temperature, while the bidentate nitrate absorption peaks (1408 cm⁻¹ and 1593 cm⁻¹) appear at lower temperatures and are not easily decomposed. This is due to the low thermal stability of the monodentate nitrate and a certain degree of desorption on the surface of the catalyst at high temperatures, and is consistent with the fact that bidentate nitrates are more thermally stable than monodentate nitrates. The peak intensity corresponding to the adsorbed NO₂ (1615 cm⁻¹) gradually decreases with increasing temperature, which is related to the effect of NO oxidation and the accelerated desorption of NO₂ after the temperature increase.

Fig. 13 In situ DRIFTS spectra of NO + O₂ adsorption on the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst.

3.7.3 The microscopic reaction process of the catalyst. The 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst was pre-adsorbed with NH₃ and subjected to in situ NO + O₂ co-adsorption tests in order to explore the microscopic process of the NH₃-SCR reaction on the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst at 250 °C. The results are shown in Fig. 14. At 250 °C, when NH₃ adsorption was saturated, signals corresponding to Lewis acid sites (1165 cm⁻¹, 1352 cm⁻¹, and 1603 cm⁻¹) and Brønsted acid sites (1420 cm⁻¹) were observed. Then, the NH₃ gas supply was turned off, and a NO + O₂ gas mixture (500 ppm NO + 5% O₂, balanced with N₂) was introduced. The following phenomena were observed over time: after 20 minutes of NO + O₂ exposure, the signals related to NH₃ adsorption completely disappeared, indicating that almost all the NH₃ species adsorbed on the catalyst surface had participated in the reaction. Subsequently, absorption peaks corresponding to bidentate nitrates (1408 cm⁻¹), monodentate nitrates (1153 cm⁻¹), M–NO₂ nitro compounds (1342 cm⁻¹), and adsorbed NO₂ (1602 cm⁻¹) appeared. The intensity of the bidentate nitrate peak at 1408 cm⁻¹ further increased with prolonged reaction time, likely due to the accumulation of bidentate nitrate species after the surface-adsorbed NH₃ species were fully consumed. These results indicate that the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst can adsorb NO + O₂ and convert it into various nitrate species. These nitrate species then react with the NH₃ species previously adsorbed on the Lewis and Brønsted acid sites to complete the denitration reaction cycle. The NH₃-SCR reaction on the catalyst surface follows the L–H mechanism, as shown in eqn (3.1)–(3.9).³¹

NO(g) → NO(ad) (3.2)

≡Mⁿ⁺=O + NO(ad) → ≡M⁽ⁿ⁻¹⁾⁺–O–NO (3.3)

M⁽ⁿ⁻¹⁾⁺–O–NO + NH₃(ad) → ≡M⁽ⁿ⁻¹⁾⁺–O–NO–NH₃→ ≡M⁽ⁿ⁻¹⁾⁺–OH + N₂(g) + H₂O (3.4)

O₂(g) → 2O(ad) (3.7)

NO(g) + O(ad) → NO₂(ad) (3.8)

NH₄⁺(ad) + e⁻ + NO₂(ad) → NH₄NO₂(ad) → N₂ + 2H₂O (3.9)

Fig. 14 In situ DRIFTS spectra of NO + O₂ adsorption on the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst with pre-adsorption of NH₃.

Additionally, the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst was pre-adsorbed with NO + O₂ and subjected to in situ NH₃ adsorption at 250 °C, as shown in Fig. 15. When the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst adsorbed NO + O₂ at 250 °C and reached saturation, the absorption peaks of bidentate nitrate (1408 cm⁻¹), monodentate nitrate (1153 cm⁻¹), bridged nitrate (1269 cm⁻¹), M–NO₂ nitro compounds (1342 cm⁻¹), and adsorbed NO₂ (1602 cm⁻¹) were detected. Then, the NO + O₂ gas mixture was turned off and NH₃ gas was introduced. Over time, the following phenomena occurred: after the NH₃ gas had been introduced for 10 minutes, the signals related to the adsorption of NO + O₂ disappeared, indicating that the nitrate species adsorbed on the surface of the catalyst had fully participated in the reaction. After 10 minutes, the Lewis acid sites (1165 cm⁻¹, 1352 cm⁻¹, and 1603 cm⁻¹) and Brønsted acid sites (1269 cm⁻¹, 1420 cm⁻¹, and 1530 cm⁻¹) had adsorbed NH₃ species, giving rise to the corresponding absorption peaks. The intensity of the absorption peaks did not change significantly with extension of the reaction time, indicating that the reaction is complete at this point. The above results show that the NO_x species adsorbed on the surface of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst can react with gaseous NH₃ molecules, indicating that the NH₃-SCR reaction on the catalyst surface also follows the E–R mechanism as shown in reaction eqn (3.10)–(3.14).³¹ Combining the information in Fig. 12 and 13, it can be concluded that the NH₃-SCR reaction on the surface of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst follows both the L–H and E–R mechanisms, with the L–H mechanism being dominant.

NH₃(g) → NH₃(ad) (3.10)

NH₃(ad) + Mⁿ⁺=O → NH₂(ad) + M⁽ⁿ⁻¹⁾⁺–O–H (3.11)

NH₂(ad) + Mⁿ⁺=O → NH(ad) + M⁽ⁿ⁻¹⁾⁺–O–H (3.12)

NH₂(ad) + NO(g) → N₂ + H₂O (3.13)

NH(ad) + NO(g) → N₂O + H⁺ (3.14)

Fig. 15 In situ DRIFTS spectra of NH₃ adsorption on the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst with pre-adsorption of NO + O₂.

3.8 Catalytic activity of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts

3.8.1 Denitration performance of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts. The results of the denitration performance tests of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts with different CeO₂ contents are shown in Fig. 16(a). The denitration performance of the catalyst with no added CeO₂ was the worst. The temperature at which its denitration efficiency reached 80% was 121 °C, and the temperature ranges for 80% and 90% denitration efficiencies were 121–385 °C and 133–355 °C, respectively. As the CeO₂ content was increased, the denitration performance of the catalyst also improved. When the content was increased to 6%, the catalyst showed the best denitration performance. The initial temperature at which it achieved 80% denitration efficiency was reduced to 115 °C, the temperature ranges for denitration efficiencies greater than 80% and 90% were widened to 115–425 °C and 129–390 °C, respectively, with the highest denitration efficiency reaching 99.17%. When the CeO₂ content was further increased to 9%, the denitration performance of the catalyst began to decline.

Fig. 16 Denitration performance of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts: (a) NO conversion; and (b) N₂ selectivity.

The N₂ selectivity of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts is shown in Fig. 16(b). At temperatures below 200 °C, all the catalyst groups maintain 100% N₂ selectivity. As the temperature rises, the N₂ selectivity shows a downward trend for all the catalysts, indicating that some NH₃ is over-oxidized at higher temperatures and the by-product N₂O begins to form. Among them, the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst exhibits the best N₂ selectivity, with the lowest rate of decline, maintaining a selectivity of 80.3% at 450 °C. At this temperature, the N₂ selectivity of the other catalysts drops below 78%.

3.8.2 Anti-SO₂ and anti-H₂O properties of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts. The results of the SO₂ and H₂O resistance tests of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts are shown in Fig. 17. From the above denitration performance test results, it can be seen that the denitration efficiency of each group of catalysts reaches its maximum value near 200 °C and tends to stabilize, indicating that the denitration performance of the catalysts is best at this temperature. As shown in Fig. 17(a), after each group of catalysts was operated stably at 200 °C for 6 hours, 60 ppm SO₂ was introduced into the simulated flue gas. At this time, the denitration efficiency of each group of catalysts decreased significantly to a certain value and then stabilized. After 18 hours, the introduction of SO₂ was stopped. Subsequently, the denitration efficiency gradually recovered to a certain value and remained stable, but remained lower than the initial value, indicating that the effect of SO₂ on the denitration performance of the catalyst is irreversible. Among the catalysts, the 8Fe₂O₃–40MnO₂/TiO₂ catalyst is most affected by SO₂, followed by the 8Fe₂O₃–3CeO₂–40MnO₂/TiO₂ catalyst and the 8Fe₂O₃–9CeO₂–40MnO₂/TiO₂ catalyst. The 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst was least affected and almost recovered to the level before the introduction of SO₂ after the SO₂ was removed. The improved denitration efficiency is mainly due to the fact that an appropriate amount of CeO₂ can inhibit the generation of (NH₄)₂SO₄ and reduce the stability of (NH₄)₂SO₄ and NH₄HSO₄, thereby improving the resistance of the catalyst to SO₂.

Fig. 17 SO₂ and H₂O resistance of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts: (a) SO₂; and (b) SO₂ and H₂O.

As shown in Fig. 17(b), after constant temperature denitration at 200 °C for 6 hours, 60 ppm SO₂ and water vapor with a volume fraction of 5% were introduced into the mixed flue gas simultaneously. The denitration efficiencies of the catalysts with CeO₂ contents of 1%, 3%, 5%, and 7% decreased to 74.5%, 79.5%, 81.6%, and 80.1%, respectively, and then gradually stabilized. After the reaction had been carried out for 18 hours, the introduction of SO₂ and H₂O was stopped. The denitration efficiency improved to some extent as the reaction continued, although it failed to recover to the value prior to the introduction of SO₂ and water vapor. Among the catalysts, the 8Fe₂O₃–40MnO₂/TiO₂ catalyst was most affected by SO₂ and H₂O. The main reason is that when the CeO₂ content is low, it is difficult for the catalyst to inhibit its own sulfation, and H₂O competes with NH₃ for adsorption on the surface of the catalyst, thereby affecting the surface adsorption performance of the catalyst. The 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst maintained a denitration efficiency of more than 80% after the introduction of SO₂ and H₂O. After the introduction of SO₂ and H₂O was stopped, its denitration efficiency gradually increased to greater than 90%, showing excellent anti-SO₂ and anti-H₂O properties.³²

3.8.3 Kinetic analysis of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts. To further investigate the influence of rare earth modification on the catalytic reaction rate of Fe₂O₃–MnO₂/TiO₂, the dynamics of the denitration reaction were studied. Differences in the apparent activation energy of the catalytic reaction can help to determine the different catalytic paths and the adsorption and desorption behaviors of species on the catalyst surface.³³ The calculation method for the reaction rate constant is shown in eqn (3.15):³⁴where k represents the reaction rate constant (cm³ g⁻¹ s⁻¹), V is the gas flow rate (cm³ s⁻¹), W is the catalyst mass (g), and X is the NO conversion rate. The value of k is calculated using eqn (3.15). Subsequently, based on the R and T values, the Arrhenius linear diagram is constructed through three-point linear fitting.

ln k = ln A − E_a/(RT) (3.16)

where A refers to the pre-exponential factor (cm³ g⁻¹ s⁻¹), E_a represents the apparent activation energy (J mol⁻¹), R is the universal gas constant, which has a value of 8.314 (J K⁻¹ mol⁻¹), and T is the reaction temperature in kelvin (K). In addition, 1/T is used as the abscissa, and the natural logarithm of the reaction rate, ln k, is plotted as the ordinate. There is a linear relationship between 1/T and ln k, and the reaction activation energy E_a can be obtained through linear regression.

Combined with the denitration performance test results of the catalysts, the denitration reaction kinetics of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalyst were studied using eqn (3.15) and (3.16). The results are shown in Fig. 18. From eqn (3.16), it can be seen that ln k has a linear relationship with T⁻¹. The smaller the absolute value of the slope, the smaller the reaction activation energy E_a, and the more easily the denitration reaction proceeds. The absolute value of the slope of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst is the smallest; thus, the catalyst with 5% added CeO₂ has the smallest activation energy, E_a = 55.78 kJ mol⁻¹. In addition, as the reaction temperature increases, the reaction rate of the catalyst also increases. At a given temperature, the reaction rates follow the order 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ > 8Fe₂O₃–9CeO₂–40MnO₂/TiO₂ > 8Fe₂O₃–3CeO₂–40MnO₂/TiO₂ > 8Fe₂O₃–40MnO₂/TiO₂, which is consistent with the denitration performance test results shown in Fig. 16.

Fig. 18 Kinetic diagrams of the 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalysts.

4 Conclusions

In this study, a new type of catalyst, 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂, which was modified using the rare earth oxide (CeO₂) with Fe–Mn/TiO₂ catalyst as the main body, was investigated. The denitration properties and reaction dynamics of NH₃ reducing NO were also analyzed. The results show that the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst modified with 6% CeO₂ exhibits the best denitration performance. The catalyst has the smallest reaction activation energy (E_a = 55.78 kJ mol⁻¹). Its denitration efficiency exceeds 80% in the temperature range from 115 to 425 °C and 90% in the range from 129 to 390 °C. The highest denitration efficiency (99.17%) is achieved within the temperature range of 220–305 °C, and the catalyst demonstrates good N₂ selectivity as well as strong anti-SO₂ and anti-H₂O performance. The 8Fe₂O₃–yCeO₂–40MnO₂/TiO₂ catalyst was characterized using XRD, BET, SEM, TEM, XPS, TPR, and TPD techniques, and the effects of the addition of CeO₂ on the structure and morphology, surface characteristics, acid site distribution, and redox properties of the catalyst were investigated. The addition of 6% CeO₂ promotes lattice shrinkage of TiO₂, improves the dispersion of TiO₂, and increases the specific surface area of the catalyst. The chemical adsorption of oxygen and low-valence metal ions (Fe²⁺/Ce³⁺) on the catalyst surface are enhanced, providing more oxygen vacancies and surface active sites for the NH₃-SCR reaction. At the same time, the redox capacity of the catalyst is enhanced, which facilitates the NH₃-SCR reaction. The addition of CeO₂ promotes the formation of weak and medium-strong acid sites on the catalyst surface, significantly improving the low-temperature denitration capacity of the Fe–Mn/TiO₂ catalyst. In situ DRIFTS results indicate that the NH₃-SCR reaction on the surface of the 8Fe₂O₃–6CeO₂–40MnO₂/TiO₂ catalyst follows both L–H and E–R mechanisms, with the L–H mechanism being dominant.

Conflicts of interest

There are no conflicts to declare.

Data availability

All relevant data are within the paper.

References

1. F. Zheng, C. Liu, X. Ma, Z. Zhou and J. Lu, Fuel Process. Technol., 2023, 247, 107773.
2. T. Boningari and P. G. Smirniotis, Curr. Opin. Chem. Eng., 2016, 13, 133–141.
3. E. D. Park, Molecules, 2024, 29, 4506.
4. Z. Fan, J.-W. Shi, C. Niu, B. Wang, C. He and Y. Cheng, Chem. Eng. J., 2020, 398, 125572.
5. J.-K. Lai and I. E. Wachs, ACS Catal., 2018, 8, 6537–6551.
6. L. Han, S. Cai, M. Gao, J. Hasegawa, P. Wang, J. Zhang, L. Shi and D. Zhang, Chem. Rev., 2019, 119, 10916–10976.
7. Q. Liu, J. Yang, M. Luo, Z. Yang and Q. Zhao, Catal. Lett., 2021, 151, 966–979.
8. J. Mu, X. Li, W. Sun, S. Fan, X. Wang, L. Wang, M. Qin, G. Gan, Z. Yin and D. Zhang, Ind. Eng. Chem. Res., 2018, 57, 10159–10169.
9. M. N. Khan, L. Han, P. Wang, J. He, B. Yang, T. Yan, L. Shi and D. Zhang, Chem. Eng. J., 2020, 397, 125535.
10. G. Zhai, Z. Han, H. Du, Y. Gao and X. Pan, Environ. Sci. Pollut. Res., 2022, 29, 17295–17308.
11. Z. Liu, C. Chen, J. Zhao, L. Yang, K. Sun, L. Zeng, Y. Pan, Y. Liu and C. Liu, Chem. Eng. J., 2020, 379, 122288.
12. C. Gao, J.-W. Shi, Z. Fan, B. Wang, Y. Wang, C. He, X. Wang, J. Li and C. Niu, Appl. Catal., A, 2018, 564, 102–112.
13. J. Huang, H. Huang, H. Jiang and L. Liu, Catal. Today, 2019, 332, 49–58.
14. L. Xu, X.-S. Li, M. Crocker, Z.-S. Zhang, A.-M. Zhu and C. Shi, J. Mol. Catal. A: Chem., 2013, 378, 82–90.
15. J. Rong, W. Zhao, W. Luo, K. Kang, L. Long, Y. Chen and X. Yao, J. Rare Earths, 2023, 41, 1323–1335.
16. B. Shen, T. Liu, N. Zhao, X. Yang and L. Deng, J. Environ. Sci., 2010, 22, 1447–1454.
17. T. Zhu, X. Zhang, W. Bian, Y. Han, T. Liu and H. Liu, Catalysts, 2020, 10, 135.
18. S. M. Lee, K. H. Park and S. C. Hong, Chem. Eng. J., 2012, 195–196, 323–331.
19. S. Ma, X. Zhao, Y. Li, T. Zhang, F. Yuan, X. Niu and Y. Zhu, Appl. Catal., B, 2019, 248, 226–238.
20. J. Xu, H. Yu, C. Zhang, F. Guo and J. Xie, New J. Chem., 2019, 43, 3996–4007.
21. H. Vidal, J. Kašpar, M. Pijolat, G. Colon, S. Bernal, A. Cordón, V. Perrichon and F. Fally, Appl. Catal., B, 2000, 27, 49–63.
22. D. H. Kim, Y. J. Park, K.-Y. Lee, H. P. Ha and D. W. Kwon, Appl. Catal., A, 2022, 643, 118770.
23. C. Yang, J. Yang, Q. Jiao, D. Zhao, Y. Zhang, L. Liu, G. Hu and J. Li, Ceram. Int., 2020, 46, 4394–4401.
24. T. Doan, P. Dam, K. Nguyen, T. H. Vuong, M. T. Le and T. H. Pham, Catalysts, 2020, 10, 321.
25. K. Cheng, J. Liu, T. Zhang, J. Li, Z. Zhao, Y. Wei, G. Jiang and A. Duan, J. Environ. Sci., 2014, 26, 2106–2113.
26. A. Stahl, Z. Wang, T. Schwämmle, J. Ke and X. Li, Catalysts, 2017, 7, 71.
27. C. Sun, H. Liu, W. Chen, D. Chen, S. Yu, A. Liu, L. Dong and S. Feng, Chem. Eng. J., 2018, 347, 27–40.
28. Z. Lian, F. Liu, H. He, X. Shi, J. Mo and Z. Wu, Chem. Eng. J., 2014, 250, 390–398.
29. Z. Liu, Y. Yi, S. Zhang, T. Zhu, J. Zhu and J. Wang, Catal. Today, 2013, 216, 76–81.
30. B. Qin, R. Guo, L. Wei, X. Yin, T. Yin, J. Zhou and Z. Qiu, J. Environ. Chem. Eng., 2022, 10, 108564.
31. R. Guo, B. Qin, L. Wei, T. Yin, J. Zhou and W. Pan, Phys. Chem. Chem. Phys., 2022, 24, 6363–6382.
32. K. A. Michalow-Mauke, Y. Lu and K. Kowalski, ACS Catal., 2015, 5, 5657–5672.
33. L. Yan, Y. Liu, H. Hu, H. Li, L. Shi and D. Zhang, ChemCatChem, 2016, 8, 2267–2278.
34. J.-W. Shi, Y. Wang, R. Duan, C. Gao, B. Wang, C. He and C. Niu, Catal. Sci. Technol., 2019, 9, 718–730.

---