Catalytic oxidation of NO by O₂ over CeO₂–MnO_x: SO₂ poisoning mechanism

Abstract

The catalytic oxidation of NO by O₂ was performed over a series of CeO₂–MnO_x catalysts with different molar ratios of Mn/Ce, which were prepared by the sol–gel method. The highest NO conversion efficiency reached 96% over the catalyst with a 0.4 Mn doping value at 238 °C. The possible reaction pathways of the catalytic oxidation process were proposed according to several characterization measurements. NO was adsorbed on the catalyst surface to form nitrates and then decomposed into NO₂. However, the catalyst was completely deactivated under an atmosphere of SO₂. NO conversion efficiency dramatically declined from 92% to 22% within only 400 min. Comparing the BET, TPR, TPD, XRD, XPS, FTIR, and TGA results of fresh and poisoned catalysts, the catalyst deactivation could be mainly attributed to manganese sulfate formation on the catalyst surface, which could only slightly decompose. The active sites for NO adsorption were occupied. Finally, the oxidation of NO to NO₂ was terminated due to lack of nitrates, which are intermediates for NO oxidation.

---

1. Introduction

Nitrogen oxides, particularly NO, are major pollutants that exist in coal-fired flue gas; they are mainly removed by direct reduction technology, including selective catalytic reduction (SCR)¹ and selective non-catalytic reduction (SNCR).² Recently, the oxidation method,^3–7 which converts the insoluble NO into highly soluble NO₂, NO₃, and N₂O₅ which then can be eliminated by wet flue gas desulfurization (WFGD)^3,8,9 together with sulfur dioxide, has been proposed. For the purpose of oxidation, plasma treatment,^3,4 injection of oxidants,^6–8 and catalytic oxidation by excessive oxygen^10,11 are often adopted. Among these methods, catalytic oxidation, without injection of the other oxidants, is considered to be an effective and economical approach.

Noble metals, typically platinum based catalysts, are known to demonstrate excellent NO oxidation efficiency.^12,13 The strong Pt–O bond causes the NO oxidation reaction to be thermodynamically endothermic on the catalyst surface.¹⁴ In order to weaken the Pt–O bond and enhance the adsorption ability of nitrogen oxides, doping with transition metals is often required.¹⁵ Meanwhile, the usage of noble metals is restricted by their high cost. However, transition metals present excellent redox abilities themselves and are more economical. Cerium and manganese oxides have shown especially good activity for SCR at low temperature.^16,17 Ceria (CeO₂), an abundant and inexpensive metal oxide, has been applied extensively as a catalyst carrier for its capacity to enhance and stabilize the dispersion of other metal oxides.^18,19 It also has excellent oxygen storage capacity and redox properties. Ceria is an oxygen reservoir, which can form large amounts of labile oxygen vacancies and oxygen species during the redox shift between Ce⁴⁺ and Ce³⁺ under oxidative and reductive conditions.^20,21 Manganese oxides have strong dynamic oxygen storage capacity and excellent redox properties; they are active and stable catalysts for catalytic oxidation.^22,23 Above all, ceria doped with manganese oxides was carefully selected as the catalyst for its potential catalytic oxidation ability for multi-pollutant removal.

Sulfur deactivation is one of the most significant factors restricting the application of the catalytic oxidation method. The deactivation mechanism of low temperature SCR has been investigated in recent studies.^24–26 The transition metal oxide deactivation process was mainly attributed to three aspects: (1) sulfates formed on the catalyst surface occupy the active sites; (2) reduction of metal oxides inhibits the redox shift; (3) the reduced acid sites prevent NH₃ adsorption. However, few studies have focused on the catalytic oxidation of NO. SCR occurs in a reductive atmosphere, while catalytic oxidation occurs in an oxidative atmosphere, so the deactivation mechanisms are expected to be different. Therefore, it is essential to investigate the mechanism of the effect of SO₂ on catalytic oxidation activity separately, which will be helpful to understand the deactivation mechanism and thus improve the resistance to SO₂. Therefore, a series of CeO₂–MnO_x catalysts with different Mn/Ce molar ratios were prepared and the optimal catalyst was used to carry out SO₂ poisoning tests. Accordingly, several characterization methods, including BET, X-ray diffraction (XRD), H₂-temperature programmed reduction (H₂-TPR), NO-temperature programmed desorption (NO-TPD), X-ray photoelectron spectroscopy (XPS), Fourier transform infrared spectroscopy (FT-IR), and thermogravimetric analysis (TGA), were carried out to study the sulfated catalyst.

2. Experimental section

2.1 Catalyst preparation

Catalysts were prepared through a sol–gel method.²⁷ Nitrate salts of Ce and Mn were dissolved in deionized water. The mole ratios of Mn/Ce used were 0, 0.2, 0.4, 0.6, and 0.8. Citrate was added to the solution to promote gel formation. The mole ratio of citrate to the nitrate salts was 2 : 1. All reagents were of analytical grade and were purchased from Sinopharm Chemical Reagent Co., Ltd. The mixture was stirred at 80 °C until a gel formed and then dried in an oven at 110 °C for 12 h. The dried gel was calcined at 500 °C for 3 h under air atmosphere in a muffle furnace. The obtained solid product was sieved in a 50-mesh sieve.

2.2 Catalyst characterization

For each catalyst sample, detailed characterization was performed through specific surface area and pore structure measurement, H₂-TPR, XRD, NO-TPD, XPS, FT-IR, and TGA.

Porous structure parameters were measured through N₂ adsorption at liquid N₂ temperature (77 K) using an automatic surface area and pore size analyzer (Micromeritics ASAP 2010). Prior to the measurements, each sample was degassed at 473 K for 4 h. The specific surface area and average pore diameter were determined by the BET method. The total pore volume was calculated by the single point adsorption of pores less than 180.2005 nm in diameter at P/P₀ = 0.99.

H₂-TPR measurements using an automatic temperature programmed chemisorption analyzer (Micromeritics AutoChem II 2920) were used to evaluate the reducibility of the metallic ions and the redox ability of the metal oxide catalysts. Prior to the test, 75 mg of each catalyst sample was purged at 120 °C in Ar atmosphere with a flow rate of 50 mL min⁻¹ for 1 h. Finally, the H₂ consumption of each sample was obtained by increasing the temperature from 50 °C to 800 °C with a ramp of 10 °C min in 5% H₂/Ar atmosphere (20 mL min⁻¹).

XRD patterns of the catalyst samples were recorded on a Rigaku D/max 2550PC diffractometer operating at 40 kV and 40 mA using Cu Kα radiation with a scanning rate of 4° min⁻¹ in the range of 2θ = 20° to 80°.

NO-TPD experiments carried out on the same instrument with H₂-TPR were used to evaluate the NO adsorption ability of the catalyst. The NO desorption signal was detected by a Hiden QIC20 mass spectrum instrument. Prior to the test, each sample was exposed at 50 °C in 5 vol% NO/Ar atmosphere with a flow rate of 50 mL min⁻¹ for 1 h, followed by Ar purge for another 30 min. Finally, the NO desorption of each sample was obtained by increasing the temperature from 50 °C to 800 °C with a ramp of 10 °C min⁻¹.

XPS spectra were used to evaluate the element valence and semi-quantitative concentration distribution on the catalyst surface. The spectra were determined using a photoelectron spectrometer (Thermo Scientific Escalab 250Xi) with a standard Al Kα source (1486.6 eV) operating in constant analysis energy (100 eV) mode at a scanning step of 0.1 eV. All binding energies were referenced to the C 1s line at 284.5 eV.

FT-IR spectroscopy was used to detect species generated on the catalyst surface during the catalytic oxidation process with and without SO₂. A Nicolet 5700 FTIR spectrometer with 0.09 cm⁻¹ resolution was used to carry out these measurements.

TGA experiments were conducted on a thermogravimetric analyzer (TA-Q500 TGA). 2 mg of each catalyst sample was loaded into the TGA reactor. The temperature of the TGA furnace was increased to 100 °C and kept constant for 10 min to remove adsorbed water. After equilibration at 50 °C for 10 min, the temperature was increased to 850 °C at a rate of 10 °C min⁻¹ under N₂ atmosphere. The TGA and DTA (differential thermal analysis) curves were obtained.

2.3 Activity and poisoning tests

Catalytic activity and SO₂ poisoning tests were carried out in a quartz tube-bed reactor heated by an electric furnace (Yifeng Furnace Co., Ltd). The reaction temperature was measured by a K-type thermocouple deposed into the catalyst bed. The simulated flue gas, mixed from bottled N₂, O₂, NO and SO₂, was supplied by Jinggong Gas Co., Ltd. (N₂ – 99.999%, O₂ – 99.999%, NO – 5%/balance N₂, SO₂ – 5%/balance N₂). A gas N₂ mixture containing 300 ppm NO and 10% O₂ was injected into the reactor separately to prevent the oxidation reaction in the absence of catalyst. 100 ppm SO₂ was injected into the reactor at 210 °C with the same stream of N₂/NO during the poisoning tests. The total gas flow rate was 1 L min⁻¹ and was controlled by mass flow controllers (MFC, Alicat Scientific Inc., America); the powder catalyst dosage was 3 g. The corresponding gas hourly space velocity (GHSV) was 20 000 h⁻¹. The concentrations of NO, NO₂, and SO₂ in the initial and outlet gas at the steady-state condition were measured by a Fourier transform infrared gas analyzer (Gasmet FTIR DX4000, Finland). The NO conversion efficiency was calculated by eqn (1):

[Conv.]_NO = [NO₂]_out /([NO]_out + [NO₂]_out) × 100% (1)

where [Conv.]_NO is the NO conversion efficiency, [NO₂]_out is the outlet NO₂ concentration of the reactor, and [NO]_out is the outlet NO concentration of the reactor.

3. Results and discussion

3.1 Catalytic activity of CeO₂–MnO_x

The catalytic activities of catalysts with different molar ratios of Mn/Ce are presented in Fig. 1. Only about 60% NO conversion efficiency is achieved with pure CeO₂ (Mn/Ce = 0) at 330 °C. When Mn is doped, the catalytic activities are distinctly improved. The NO conversion efficiency initially increases and then decreases with increasing Mn doping level. NO conversion efficiency increases with reaction temperature and then decreases with the thermodynamic equilibrium curve, given as a dotted line. It is obvious that the optimal molar ratio of Mn/Ce is 0.4, which displays excellent low temperature catalytic oxidation activity. The highest NO conversion efficiency is nearly 96% at 238 °C, and more than 60% can be achieved at 160 °C.

Fig. 1 NO conversion efficiencies of catalysts calcined at 500 °C with different molar ratios of Mn/Ce.

The BET surface areas, total pore volumes, and average pore diameters of catalysts with different molar ratios of Mn/Ce are summarized in Table 1. It can be seen that the BET surface area decreases from 48.6 m² g⁻¹ to 29.9 m² g⁻¹ and the average pore diameter increases from 4.1 nm to 12.0 nm with increasing Mn doping level. Meanwhile, the total pore volume changes very little. This illustrates that the doping process results in a decline in the number of pores and then in a decrease of the surface area.

Table 1 Porous structure parameters and lattice constants of catalysts^a

Sample	BET surface area (m^2 g^−1)	Total pore volume (mL g^−1)	Average pore diameter (nm)	Lattice constants (Å)
a The lattice constants are calculated by the Jade program based on the CeO2 (111) crystal plane according to the XRD data.
CeO2	48.6	0.05	4.1	5.4202
0.2Mn–Ce	46.7	0.08	6.9	5.4091
0.4Mn–Ce	39.8	0.07	7.1	5.4052
0.6Mn–Ce	32.8	0.09	10.9	5.3970
0.8Mn–Ce	29.9	0.09	12.0	5.3904
0.4Mn–Ce-sulfated	16.1	0.06	14.9	5.4031

---

Fig. 2 shows the H₂-TPR profiles for all the catalysts. The reduction peaks of pure CeO₂ (Mn/Ce = 0) at 450 °C and 750 °C are associated with the reduction of superficial Ce⁴⁺ ions and bulk lattice oxygen, respectively.^28,29 Meanwhile, these two peaks are very weak compared with those of the other Mn doped samples. The reduction peaks of the Mn doped samples at lower temperature can be mainly attributed to the two step reduction, including Mn₂O₃ to Mn₃O₄ at 200 to 220 °C and Mn₃O₄ to MnO at 310 to 342 °C.^30,31 Of course, the reduction of highly dispersed manganese oxide species strongly interacting with the ceria surface also makes contributions to these two reduction peaks. While the peak at higher temperature, similar to the peak of ceria, can be attributed to the reduction of MnO_x species with larger particles or intermediates as well as the reduction of the bulk lattice oxygen of CeO₂.³² The excellent low temperature catalytic activity of the Mn doped samples is demonstrated by the higher peak height and larger peak area compared to pure ceria. By comparison, the peak height and area increase significantly when the molar ratio of Mn/Ce increases from 0.2 to 0.4, and they are similar for the molar ratios of 0.4, 0.6, and 0.8. However, the reduction peak shifts to a higher temperature when the molar ratio increases from 0.6 to 0.8. As a result, the samples with 0.4 and 0.6 molar ratios displayed the best redox ability, which is in agreement with the activity test results shown in Fig. 1.

Fig. 2 H₂-TPR profiles for catalysts with different Mn doping proportions.

The XRD patterns of the catalysts with various molar ratios of Mn/Ce are shown in Fig. 3. Five diffraction peaks appear on all five catalyst samples, with diffraction angles located at 28.5°, 33.1°, 47.4°, 56.4°, and 59.2°. The diffraction angles are consistent with the fluorite structure of cubic CeO₂ (111), (200), (220), (311), and (222). This reveals that the framework of CeO₂ has not been changed by the doping of manganese oxides,^17,32 and the manganese oxides have been highly dispersed on the CeO₂ substrate. Meanwhile, Machida et al.³³ considered that Ce⁴⁺ might be substituted by Mn³⁺ in the fluorite structure due to their structural similarity. The smaller ionic radius of Mn³⁺ ion (0.066 nm) compared to Ce⁴⁺ ion (0.1098 nm) leads to the incorporation of Mn³⁺ into the fluorite lattice of CeO₂,³⁴ followed by the shrinkage of the crystal cell of CeO₂. This gives rise to a decrease in the lattice constant of CeO₂ when Mn is doped and a slight decline with increasing Mn doping level, which is shown in Table 1. This is evidenced by the fact that the main diffraction peaks of CeO₂ shift to slightly higher Bragg angle values and become shorter and broader with increasing Mn doping level.

Fig. 3 XRD characterization for catalysts with different Mn doping proportions: (a) whole patterns and (b) enlarged-zone patterns.

3.2 SO₂ poisoning process

In order to investigate the SO₂ poisoning process during catalytic oxidation of NO, the catalyst with a 0.4 molar ratio of Mn/Ce was tested, and the results are shown in Fig. 4. The NO conversion efficiency remains stable at about 92% for 60 min before SO₂ is injected. When SO₂ is added to the simulated gas, the NO conversion efficiency decreases gradually with reaction time. After 400 min injection, the SO₂ gas is stopped, and the conversion efficiency decreases from 92% to 22%. SO₂ leads to severe deactivation of the catalyst for NO oxidation. The SO₂ concentration in the outlet gas is also shown. No increase can be observed in the first 100 min, which can be attributed to the adsorption on the catalyst. After that, SO₂ concentration increases with reaction time. When SO₂ is stopped, the conversion efficiency remains stable and no recovery can be observed. This illustrates that the catalyst deactivation is an irreversible process. After this test, the sulfated catalyst that was exposed to the atmosphere of SO₂, NO, and O₂ for 400 min at 210 °C was measured by a series of characterization measurements to reveal the poisoning mechanism.

Fig. 4 NO conversion efficiency on CeO₂–MnO_x catalyst under 100 ppm SO₂ atmosphere (molar ratio of Mn/Ce: 0.4, reaction temperature: 210 °C).

3.3 Characterization of poisoning catalyst

3.3.1 BET. As is shown in Table 1, the surface area decreases from 39.8 m² g⁻¹ to 16.1 m² g⁻¹, and the average pore diameter increases from 7.1 nm to 14.9 nm when the catalyst is sulfated. Meanwhile, there is no obvious change for the total pore volume. This illustrates that the pore numbers on the catalyst surface decrease during the poisoning process, which then results in a decrease in the catalyst activity.

3.3.2 TPR. Compared with fresh catalyst in Fig. 5, the reduction peaks at 209 °C, 310 °C, and 750 °C cannot be observed for the sulfated catalyst. Meanwhile, a strong reduction peak appears at 480 °C, which can be attributed to the reduction of SO₄²⁻.³⁵ Combined with the above TPR analysis, the disappearance of these three peaks demonstrates that the manganese oxides on catalyst surface are replaced by manganese sulfate; the bulk lattice oxygen is also substituted by other species, which will be revealed by other characterization measurements.

Fig. 5 H₂-TPR profiles of fresh and sulfated catalyst: the sulfated catalyst was measured after the poisoning test, where it was exposed to SO₂ + NO + O₂ for 400 min at 210 °C. The following figures are the same.

3.3.3 NO-TPD. The intensity of the NO desorption on the sulfated catalyst was much weaker than that on fresh catalyst, as shown in Fig. 6. This illustrates that the NO adsorption ability was largely restrained after the poisoning process. The NO adsorbed on the catalyst surface exists in the form of nitrate species, which are coordinated by its oxygen atoms. The oxygen atoms used to bridge NO are covered by SO₄²⁻ after sulfuration. This can also be related to the decrease of NO adsorption ability over the sulfated catalyst.

Fig. 6 NO-TPD profiles of fresh and sulfated catalyst.

3.3.4 XRD. As is shown in Fig. 7, the XRD patterns after sulfuration show few changes compared with fresh catalyst, which suggests that sulfuration does not destroy the fluorite structure of CeO₂–MnO_x. According to the above results, SO₄²⁻ is indeed formed on the sulfated catalyst. It is reasonable to conclude that manganese sulphates are preferentially formed instead of cerium sulphates. Previous studies have also indicated that manganese oxides can protect ceria from sulfur poisoning.^36,37

Fig. 7 XRD patterns of fresh and sulfated catalyst.

3.3.5 XPS. The XPS spectra of the catalysts were investigated and are shown in Fig. 8. Detailed spectra were recorded for the Ce 3d, Mn 2p, and O 1s regions. The characteristic peaks of Ce 3d shown in Fig. 8(a) are composed of two types, named as v, corresponding to the 3d_5/2 states, and u, corresponding to the 3d_3/2 states, respectively. As is listed in Table 2, the v, v′′, and v′′′ peaks and u, u′′, and u′′′ peaks are attributed to the Ce⁴⁺ species.³⁸ Moreover, the v′ and u′ peaks actually represent Ce³⁺ components.^39,40 This suggests that Ce⁴⁺ and Ce³⁺ coexist on the catalyst surface. The mole ratios of Ce⁴⁺/(Ce⁴⁺ + Ce³⁺) for fresh and sulfated catalyst calculated from the XPS spectra are 0.79 and 0.75, respectively. The weak decrease in the Ce⁴⁺ proportion can be ascribed to the reduction of Ce⁴⁺ to Ce³⁺ as a result of exposure to SO₂.

Fig. 8 XPS spectra of fresh and sulfated catalyst over the spectral regions of Ce 3d, Mn 2p, and O 1s.

Table 2 The Ce 3d binding energies (eV) and valence compositions on fresh and sulfated catalyst

Name	v	v′	v′′	v′′′	u	u′	u′′	u′′′	Ce^4+/Ce
Fresh	882.3	883.9	888.7	898.3	900.8	902.2	907.5	916.8	0.79
Sulfated	882.4	883.6	888.9	898.4	900.9	902.4	907.5	916.9	0.75

---

Fig. 8(b) shows the photoelectron spectra of the Mn 2p region of the catalysts. The region is composed of a spin orbit doublet with Mn 2p_3/2 with a binding energy of 642 eV and Mn 2p_1/2 observed at 653 eV, corresponding to a coexisting valence state of Mn⁴⁺ and Mn³⁺. The Mn 2p_3/2 region can be decomposed into several components by peak-fit processing. Two peaks are assigned to Mn³⁺ (ref. 41 and 42) and Mn⁴⁺ (ref. 41 and 43) for the fresh catalyst, whereas a new peak corresponding to Mn²⁺ (ref. 44) appears for the sulfated catalyst. The binding energies are all listed in Table 3. The strong interaction between manganese and cerium oxides leads to some differences of the binding energies existing between these catalysts and pure MnO, Mn₂O₃, and MnO₂.^44,45 After sulfuration, the proportion of Mn⁴⁺ and Mn³⁺ listed in Table 3 certainly decreases and is transformed into Mn²⁺ according to eqn (2) and (3). Furthermore, the binding energies of Mn⁴⁺ and Mn³⁺ shift to higher values after sulfuration, which may be related to the formation of sulfates on the catalyst surface.

MnO₂ + SO₂ → MnSO₄ (2)

2Mn₂O₃ + 4SO₂ + O₂ → 4MnSO₄ (3)

Table 3 The Mn 2p_3/2 binding energies (eV) and valence compositions on fresh and sulfated catalyst

Name	Mn^2+	Mn^3+	Mn^4+	Mn^2+/Mn	Mn^3+/Mn	Mn^4+/Mn
Fresh	—	641.6	643.8	0	0.67	0.33
Sulfated	640.8	641.9	644.5	0.10	0.60	0.30

---

The O 1s spectra of fresh and sulfated catalysts are illustrated in Fig. 8(c). The O 1s spectra are asymmetric and complicated due to the interaction between the oxygen species of manganese and cerium oxides. The spectra are decomposed to lattice oxygen O_α, chemisorbed oxygen O_β, and adsorbed OH groups O_γ.^31,46,47 The binding energies and distribution of oxygen species concentrations are all listed in Table 4. The weak increase of the adsorbed OH groups might be assigned to the hydrolyzation of sulfates formed on the catalyst surface.⁴⁸ It can be observed that the O_β proportion increases distinctly after sulfuration. This can be attributed to three aspects: (1) SO₂ consumes some highly ionized lattice oxygen of the metal oxides during the poisoning process, and sulfates accumulate on the catalyst surface. The oxygen species in the sulfates (S=O) can be assigned to chemisorbed oxygen. (2) The sulfate formation causes unbalanced electronic distribution on the catalyst surface and then generates more vacancies to adsorb the oxygen species.⁴⁸ (3) The Ce³⁺ ions are beneficial to create charge imbalance, labile oxygen vacancies, and unsaturated chemical bonds on the catalyst surface,^49,50 which generates more chemisorbed oxygen on the catalyst surface. Therefore, the slight increase of Ce³⁺ described above facilitates the increase of chemisorbed oxygen. Generally, the chemisorbed oxygen is often considered to be the most active oxygen species and to be beneficial in most of the catalytic oxidation reactions,^51,52 including the oxidation of NO to NO₂.^10,20 However, the biggest contribution to the increase of chemisorbed oxygen is sulfate formation, which cannot enhance the catalytic activity.

Table 4 The binding energies (eV) and distribution of oxygen species on fresh and sulfated catalyst

Name	Oα (lattice oxygen)	Oβ (chemisorbed oxygen)	Oγ (adsorbed OH groups)	Oα/O	Oβ/O	Oγ/O
Fresh	529.2	531.1	533.5	0.67	0.31	0.02
Sulfated	529.4	531.4	533.3	0.55	0.41	0.04

---

3.3.6 FT-IR. FT-IR analysis was carried out to investigate the diversification in the surface adsorbed species after sulfuration. Infrared spectra of the fresh, used, and sulfated catalyst covering the range of 875 to 2000 cm⁻¹ are recorded in Fig. 9. For the fresh catalyst, the bands at 1048, 1121, 1395, and 1614 cm⁻¹ can be assigned to incompletely decomposed nitrate species on manganese and cerium oxides.^17,53 The used catalyst was tested after exposure to an atmosphere of NO and O₂ for 400 min at 210 °C. Compared with the weak band at 1395 cm⁻¹ for the fresh catalyst, the used catalyst displays a stronger band at 1386 cm⁻¹, corresponding to free nitrates.^54,55 It can be speculated that these nitrates are the intermediates in NO oxidation to NO₂. Firstly, NO is adsorbed on surface active sites of the catalyst to form nitrosyls. Then the unstable nitrosyls are oxidized into nitrates by lattice oxygen and finally decomposed into NO₂. After sulfuration, the band that represents the free nitrates shifts to 1417 cm⁻¹, and the absorbance decreases. This illustrates that the exposure to SO₂ inhibits the formation of nitrates on the catalyst surface, and the catalytic reaction will be restrained due to a lack of intermediates. Furthermore, two stronger bands at 1045 and 1133 cm⁻¹ can be observed after sulfuration. These bands can be assigned to either surface or bulk sulfates.^56,57 These sulfates can be partly attributed to reactions (1) and (2), as shown before. Meanwhile, SO₂ oxidation to SO₃ according to eqn (4) and (5) might be promoted. The active sites that are used for adsorbing nitrogen oxides and intermediates of NO oxidation to NO₂ are occupied by sulfur oxides and sulfates. Therefore, the catalytic activity of NO oxidation decreases continuously after exposure to SO₂.

2SO₂ + O₂ ↔ 2SO₃ (4)

SO₂ + NO₂ ↔ SO₃ + NO (5)

Fig. 9 FT-IR spectra of fresh, used and sulfated catalyst: the used catalyst was measured after exposure to NO + O₂ for 400 min at 210 °C.

3.3.7 TGA. In order to evaluate the thermal stability of the fresh catalyst and investigate the sulfate decomposition temperature of the sulfated catalyst, thermogravimetric analysis (TGA) was performed, and the results are shown in Fig. 10. The weight loss of fresh catalyst is only less than 3.0% after thermal treatment, suggesting good thermal stability. The derivative DTA curve displays two peaks at 200 °C and 720 °C, respectively. Because water adsorbed on the sample catalyst was removed, the first weight loss peak should be attributed to the decomposition of residual nitrates on the metal oxides. The second peak can be assigned to the phase transitions of metal oxides.⁵⁸ For the sulfated catalyst, the weight loss is slightly larger than that of the fresh catalyst after thermal treatment. However, the weight loss at a higher temperature region, corresponding to the sulfate decomposition,^59,60 occupies about 73% of the total weight loss. It can be concluded that regenerating the catalyst by the decomposition of the sulfates on the catalyst surface requires a temperature higher than 700 °C; this process requires further investigation.

Fig. 10 TGA profiles of fresh and sulfated catalyst.

3.4 Poisoning mechanism over CeO₂–MnO_x

Unlike the reductive atmosphere for SCR,^24–26 the SO₂ deactivation mechanism for catalytic oxidation of NO occurs in an oxidative atmosphere, and thus is somewhat different. Various kinds of characterization measurements were carried out to obtain the possible reaction pathways of the catalytic oxidation and poisoning process, which has not been studied in previous works. Nitric oxides (NO) are adsorbed on the catalyst surface to form nitrosyls and then oxidize into nitrates by the lattice oxygen, with evidence of the nitrate peak shown in the FTIR spectra for the catalyst after the oxidation reaction. Meanwhile, Ce⁴⁺ and Mn⁴⁺ are reduced to Ce³⁺ and Mn³⁺, respectively. Then nitrates are decomposed into nitrogen dioxides (NO₂) and one lattice oxygen is restored. After the nitrogen dioxides dissociate from the catalyst surface, Ce and Mn are returned to their previous valence states by adsorbed oxygen and the original lattice oxygen. The active sites are regained and a new cycle begins. However, when the catalyst is exposed to sulfur dioxide (SO₂) atmosphere, the sulfur dioxides occupy the active sites to form sulphites. Sulfates are formed on the catalyst surface, oxidized by the lattice oxygen. Hereafter, some sulfates decompose into sulfur trioxides (SO₃) that will desorb from the catalyst surface. The reduced Ce³⁺ can be oxidized to Ce⁴⁺ again by adsorbed oxygen and nitrogen dioxide. As a result, only a slight decrease was observed in the Ce⁴⁺ proportion after sulfuration. The undecomposed sulfates are formed on Mn sites based on the XRD and XPS (Mn 2p) results presented. Due to the higher temperature required for the manganese sulfate decomposition shown in the TGA results, the active sites are only slightly regenerated. The active sites are occupied by sulfates so that the catalytic reaction cannot continue. This leads to a lack of nitrates, which are a decisive intermediate for NO oxidation to NO₂. The main difference between catalytic oxidation of NO and SCR can be observed: in terms of SCR, the reductive atmosphere could inhibit sulfate formation to some extent, whereas the oxidative atmosphere promotes sulfate formation for catalytic oxidation.

4. Conclusions

CeO₂–MnO_x mixed oxide catalysts with different molar ratios of Mn/Ce prepared by the sol–gel method were conducted to evaluate the catalytic activity of NO oxidation. The highest NO conversion efficiency of nearly 96% at 238 °C was reached by the catalyst with 0.4 Mn doping value. Physico-chemical characterizations of the catalysts were carried out, coupled with BET, TPR, and XRD.

The SO₂ poisoning process was tested with the optimal catalyst at 210 °C. Exposure to SO₂ leads to severe catalyst deactivation for NO oxidation. When SO₂ is added, NO conversion efficiency decreases from 92% to 22% after 400 min. A series of characterization measurements over the sulfated catalyst were performed to reveal the poisoning mechanism with the help of BET, TPR, TPD, XRD, XPS, FTIR, and TGA. According to the comparison between fresh and sulfated catalyst, the catalyst deactivation could be mainly attributed to manganese sulfate formed on the catalyst surface, which occupies the active sites for NO adsorption. Finally, the catalytic oxidation is inhibited due to the lack of nitrates. Furthermore, the oxidative atmosphere promotes sulfuration of the catalyst.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (51422605), the National Basic Research Program of China (2012CB214906) and the Zhejiang Provincial Natural Science Foundation (LR16E060001).

References

1. B. Li, Z. Ren, Z. Ma, X. Huang, F. Liu, X. Zhang and H. Yang, Catal. Sci. Technol., 2016 10.1039/c5cy01430a.
2. D. Q. Dao, L. Gasnot, K. Marschallek, A. El Bakali and J. F. Pauwels, Energy Fuels, 2010, 24, 1696–1703.
3. S. L. Ding, Q. Yu, Y. Z. Zhang, Y. Liu, C. X. Xie and G. Yu, J. Adv. Oxid. Technol., 2015, 18, 114–122.
4. I. Jogi, E. Stamate, C. Irimiea, M. Schmidt, R. Brandenburg, M. Holub, M. Bonislawski, T. Jakubowski, M. L. Kaariainen and D. C. Cameron, Fuel, 2015, 144, 137–144.
5. K. Li, X. L. Tang, H. H. Yi, P. Ning, D. J. Kang and C. Wang, Chem. Eng. J., 2012, 192, 99–104.
6. K. Skalska, J. S. Miller and S. Ledakowicz, Chem. Eng. Sci., 2011, 66, 3386–3391.
7. Z. H. Wang, J. H. Zhou, Y. Q. Zhu, Z. C. Wen, J. Z. Liu and K. Cen, Fuel Process. Technol., 2007, 88, 817–823.
8. Y. S. Mok and H. J. Lee, Fuel Process. Technol., 2006, 87, 591–597.
9. Z. H. Wang, X. Zhang, Z. J. Zhou, W. Y. Chen, J. H. Zhou and K. F. Cen, Energy Fuels, 2012, 26, 5583–5589.
10. M. Y. Zhang, C. T. Li, L. Qu, M. F. Fu, G. M. Zeng, C. Z. Fan, J. F. Ma and F. M. Zhan, Appl. Surf. Sci., 2014, 300, 58–65.
11. Z. Guo, Z.-H. Huang, M. Wang and F. Kang, Catal. Sci. Technol., 2015, 5, 827–829.
12. R. L. Arevalo, K. Oka, H. Nakanishi, H. Kasai, H. Maekawa, K. Osumi and N. Shimazaki, Catal. Sci. Technol., 2015, 5, 882–886.
13. Y. Li, M. Q. Shen, J. Q. Wang, T. M. Wan and J. Wang, Catal. Sci. Technol., 2015, 5, 1731–1740.
14. S. Ovesson, B. I. Lundqvist, W. F. Schneider and A. Bogicevic, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 71, 1–5.
15. M. C. Escano, T. Q. Nguyen, H. Nakanishi and H. Kasai, J. Phys.: Condens. Matter, 2009, 21, 1–6.
16. S. Andreoli, F. A. Deorsola and R. Pirone, Catal. Today, 2015, 253, 199–206.
17. G. S. Qi, R. T. Yang and R. Chang, Appl. Catal., B, 2004, 51, 93–106.
18. S. Pengpanich, V. Meeyoo, T. Rirksomboon and K. Bunyakiat, Appl. Catal., A, 2002, 234, 221–233.
19. A. M. Silva, L. O. O. Costa, A. P. M. G. Barandas, L. E. P. Borges, L. V. Mattos and F. B. Noronha, Catal. Today, 2008, 133, 755–761.
20. X. Li, S. Zhang, Y. Jia, X. Liu and Q. Zhong, J. Nat. Gas Chem., 2012, 21, 17–24.
21. G. S. Qi and R. T. Yang, J. Catal., 2003, 217, 434–441.
22. V. S. Escribano, E. F. Lopez, J. M. Gallardo-Amores, C. D. Martinez, C. Pistarino, M. Panizza, C. Resini and G. Busca, Combust. Flame, 2008, 153, 97–104.
23. L. W. Jia, M. Q. Shen, J. J. Hao, T. Rao and J. Wang, J. Alloys Compd., 2008, 454, 321–326.
24. B. Q. Jiang, Z. B. Wu, Y. Liu, S. C. Lee and W. K. Ho, J. Phys. Chem. C, 2010, 114, 4961–4965.
25. S. W. Pan, H. C. Luo, L. Li, Z. L. Wei and B. C. Huang, J. Mol. Catal. A: Chem., 2013, 377, 154–161.
26. Z. Y. Sheng, Y. F. Hu, J. M. Xue, X. M. Wang and W. P. Liao, J. Rare Earths, 2012, 30, 676–682.
27. Z. Wang, F. Lin, S. Jiang, K. Qiu, M. Kuang, R. Whiddon and K. Cen, Fuel, 2016, 166, 352–360.
28. B. de Rivas, R. Lopez-Fonseca, M. A. Gutierrez-Ortiz and J. I. Gutierrez-Ortiz, Appl. Catal., B, 2011, 101, 317–325.
29. S. Ricote, G. Jacobs, M. Milling, Y. Y. Ji, P. M. Patterson and B. H. Davis, Appl. Catal., A, 2006, 303, 35–47.
30. J. Trawczynski, B. Bielak and W. Mista, Appl. Catal., B, 2005, 55, 277–285.
31. P. Venkataswamy, K. N. Rao, D. Jampaiah and B. M. Reddy, Appl. Catal., B, 2015, 162, 122–132.
32. P. Yang, S. S. Yang, Z. N. Shi, Z. H. Meng and R. X. Zhou, Appl. Catal., B, 2015, 162, 227–235.
33. M. Machida, M. Uto, D. Kurogi and T. Kijima, Chem. Mater., 2000, 12, 3158–3164.
34. Z. Wang, G. L. Shen, J. Q. Li, H. D. Liu, Q. Wang and Y. F. Chen, Appl. Catal., B, 2013, 138, 253–259.
35. S. J. Yang, Y. F. Guo, H. Z. Chang, L. Ma, Y. Peng, Z. Qu, N. Q. Yan, C. Z. Wang and J. H. Li, Appl. Catal., B, 2013, 136, 19–28.
36. M. Casapu, O. Krocher and M. Elsener, Appl. Catal., B, 2009, 88, 413–419.
37. K. Tikhomirov, O. Krocher, M. Elsener, M. Widmer and A. Wokaun, Appl. Catal., B, 2006, 67, 160–167.
38. J. Liu, Z. Zhao, J. Wang, C. Xu, A. Duan, G. Jiang and Q. Yang, Appl. Catal., B, 2008, 84, 185–195.
39. M. Y. Smirnov, A. V. Kalinkin, A. V. Pashis, A. M. Sorokin, A. S. Noskov, K. C. Kharas and V. I. Bukhtiyarov, J. Phys. Chem. B, 2005, 109, 11712–11719.
40. X. Y. Wang, Q. Kang and D. Li, Appl. Catal., B, 2009, 86, 166–175.
41. M. Kang, E. D. Park, J. M. Kim and J. E. Yie, Appl. Catal., A, 2007, 327, 261–269.
42. M. R. Morales, B. P. Barbero and L. E. Cadus, Appl. Catal., B, 2006, 67, 229–236.
43. D. Delimaris and T. Ioannides, Appl. Catal., B, 2008, 84, 303–312.
44. L. J. Liu, Q. A. Yu, J. Zhu, H. Q. Wan, K. Q. Sun, B. Liu, H. Y. Zhu, F. Gao, L. Dong and Y. Chen, J. Colloid Interface Sci., 2010, 349, 246–255.
45. X. F. Tang, Y. G. Li, X. M. Huang, Y. D. Xu, H. Q. Zhu, J. G. Wang and W. J. Shen, Appl. Catal., B, 2006, 62, 265–273.
46. V. P. Santos, M. F. R. Pereira, J. J. M. Orfao and J. L. Figueiredo, Appl. Catal., B, 2010, 99, 353–363.
47. C. H. Zhang, C. Wang, W. C. Zhan, Y. L. Guo, Y. Guo, G. Z. Lu, A. Baylet and A. Giroir-Fendler, Appl. Catal., B, 2013, 129, 509–516.
48. X. S. Du, X. Gao, L. W. Cui, Y. C. Fu, Z. Y. Luo and K. F. Cen, Fuel, 2012, 92, 49–55.
49. B. Guan, H. Lin, L. Zhu and Z. Huang, J. Phys. Chem. C, 2011, 115, 12850–12863.
50. X. Gao, Y. Jiang, Y. Zhong, Z. Y. Luo and K. F. Cen, J. Hazard. Mater., 2010, 174, 734–739.
51. F. Larachi, J. Pierre, A. Adnot and A. Bernis, Appl. Surf. Sci., 2002, 195, 236–250.
52. H. L. Li, C. Y. Wu, Y. Li and J. Y. Zhang, Environ. Sci. Technol., 2011, 45, 7394–7400.
53. K. I. Hadjiivanov, Catal. Rev.: Sci. Eng., 2000, 42, 71–144.
54. A. L. Goodman, E. T. Bernard and V. H. Grassian, J. Phys. Chem. A, 2001, 105, 6443–6457.
55. M. Kantcheva, J. Catal., 2001, 204, 479–494.
56. H. Abdulhamid, E. Fridell, J. Dawody and M. Skoglundh, J. Catal., 2006, 241, 200–210.
57. T. Yamaguchi, T. Jin and K. Tanabe, J. Phys. Chem., 1986, 90, 3148–3152.
58. J. Cheng, J. Yu, X. Wang, L. Li, J. Li and Z. Hao, Energy Fuels, 2008, 22, 2131–2137.
59. D. W. Kwon, K. B. Nam and S. C. Hong, Appl. Catal., B, 2015, 166, 37–44.
60. W. S. Kijlstra, M. Biervliet, E. K. Poels and A. Bliek, Appl. Catal., B, 1998, 16, 327–337.

---