English

• 0.   Introduction

  Manganese oxide possesses copious Lewis acidic sites^[1], reactive oxygen species, and multiple valence states^[2-3], which results in its excellent denitration (DeNO_x) activity for NH₃-selective catalytic reduction (NH₃-SCR) at low temperatures. However, it is susceptible to inactivation due to sulfur poisoning at low temperatures, which results in its inability for industrial application. Therefore, how to improve sulfur tolerance of manganese catalysts has become the focus of research in recent years. Previous research has disclosed that sulfur tolerance of manganese catalysts can be promoted by doping elements, changing the synthesis route, and adjusting the morphology of the catalyst^[4]. The mobility of oxygen in manganese catalysts can be enhanced by cerium doping, which results in the improvement of low-temperature activity. In addition, SO₂ tolerance of manganese catalysts is also promoted after modification of cerium as CeO₂ restrains the sulfation of MnO_x^[5]. Nb doping leads to the rise of the Mn⁴⁺ ratio and adsorption of oxygen in manganese catalysts, which is beneficial for the generation of NO₂. Therefore, the activity and sulfur tolerance of the catalyst is enhanced^[6]. Ho doping improves the ratio of chemisorption oxygen, Mn⁴⁺, and surface acidic properties of manganese catalysts, which is a benefit for the removal of NO_x and increase of sulfur tolerance of the catalysts^[7]. Modification of Sm not only leads to the rise of weakly adsorbed Lewis acid sites and redox properties of manganese catalysts but also prevents the active sites from being sulfated, which results in the improvement of NO_x purification and SO₂ and water tolerance at low temperatures^[8]. It was disclosed that the opened pore structure of flower-like Mn-Co catalyst brings about its high superficial area, which is beneficial for the diffusion, adsorption, and reaction of reactants. Mn-Co catalyst with a flower-like structure also possesses more Lewis acidic sites, which promotes the removal of NO_x and the resistance to water at low temperatures^[9]. CeO₂-MnO_x catalyst with core-shell structure leads to uniform distribution of CeO₂ on the outer surface of MnO_x nanoparticles. The CeO₂ shell facilitates the diffusion of reactant molecules and prevents MnO_x from poisoning and deactivation, thus improving the tolerance to sulfur^[10]. The study of Chen et al.^[3] found that Mn-Ce/TiO₂ prepared by reverse co-precipitation method behaves with relatively low crystallinity and high surface area, which is beneficial for the dispersion of active ingredients. Moreover, the catalyst contains a high ratio of Ce³⁺, Mn⁴⁺, adsorption oxygen, and acid sites, which results in its excellent DeNO_x activity and resistance to sulfur and water. According to the research of Zhang, the in-situ precipitation method not only promotes the electron transfer between MnO_x and CeO_x, but also enhances the specific surface and surface acidity of Ce-Mn/TiO₂ catalyst, which results in its excellent activity and sulfur tolerance at low temperatures^[4]. Our previous research discovered that modification of Ho significantly enhanced the NO_x conversion, sulfur and water tolerance of Mn-Ce/TiO₂ at low temperatures. However, considering its application in practical engineering, sulfur and water tolerance of Mn-Ce/TiO₂ should be further enhanced. Based on this, HoCeMn/TiO₂ was synthesized by dipping and co-precipitation methods respectively. The effect of the preparation method on structure, NO_x conversion performance, tolerance to sulfur and water of catalysts was investigated. The results obtained are instrumental in the preparation and modification of catalyst for NO_x conversion at low temperatures.

  1.   Experimental

  1.1   Preparation of catalyst

  Catalyst with a mass ratio of 13.20∶12.28∶5∶100 of MnO_x, CeO₂, Ho₂O₃, and TiO₂ was synthesized by impregnation and co-precipitation methods respectively. Ce(NO₃)₃·6H₂O (Kelong Chemical), Ho(NO₃)₃·5H₂O (Kelong Chemical), 50% Mn(NO₃)₂ (Kelong Chemical), and TiO₂ (TiD Chemical) were used as raw materials. In the impregnation method, Ce(NO₃)₃·6H₂O, Ho(NO₃)₃·5H₂O, and 50% Mn(NO₃)₂ solution were dissolved in ultrapure water to form mixed solutions first. Then, TiO₂ powder was dipped in the mixed solutions accompanied by strong stirring. The mixture was dried and then calcined. The calcination temperature and time were 500 ℃ and 3 h respectively. The obtained catalyst was named HoCeMnTi-I. As for the co- precipitation method, TiO₂ was added into mixed solutions containing Ce(NO₃)₃, Ho(NO₃)₃, and Mn(NO₃)₂ to form a suspension. Then 1 mol·L^-1 ammonium carbonate was added into the suspension to form a precipitate and subsequently stirred for 24 h. After this, the precipitate was filtered, washed, dried, and calcined. The drying and roasting conditions were the same as the impregnation method. The obtained catalyst by this method was named HoCeMnTi-C. After reacting in flue gas containing H₂O and SO₂, the catalysts were labeled as HoCeMnTi-Ia and HoCeMnTi-Ca respectively.

  1.2   Characterization

  The Brunauer-Emmett-Teller (BET) surface area of the catalysts was carried out on a Quadrasorb SI analyzer at -196 ℃ after pretreatment at 300 ℃ for 3 h. The crystal structures of the catalysts were finished on an X-ray diffractometer (XRD, PANalytical X′Pert PRO). Cu Kα (λ=0.154 18 nm) was used as radiation source. The operating current and voltage were 40 kV and 100 mA respectively. The scan range was 10°-80°. The composition of surface elements was analyzed by X-ray photoelectron spectroscopy (XPS, Kratos Analytcal Inc, Al Kα excitation source, hν=1 486 eV). The electron binding energy of the element was corrected by C1s (284.8 eV).

  Temperature-programmed desorption of ammonia (NH₃-TPD) was performed at the chemisorption instrument (VDSorb 91x). The catalysts were heat treated at 350 ℃ for 60 min at pure He flow to remove the surface adsorbate, then cooled to 50 ℃. After this, the NH₃/He (5∶95, V/V) mixture was injected until saturated. Subsequently, He was injected to blow off physically adsorbed ammonia. Lastly, the catalysts were heated at 10 ℃·min^-1 to reach 850 ℃ in He flow to desorb the NH₃ adsorbed on acidic sites.

  Temperature-programmed reduction of hydrogen (H₂-TPR) was also performed on the VDSorb 91x chemisorption instrument. In this experiment, the catalysts were heat treated at 350 ℃ for 60 min in pure Ar flow to get rid of the surface adsorbate and then dropped to 50 ℃. After this, they were heated to 950 ℃ in 10% H₂/Ar (10∶90, V/V) mixture at a rate of 8 ℃·min^-1 to experiment.

  The in situ diffuse reflectance infrared Fourier transform spectra (In situ DRIFTS) were carried out on Nicolet iS50 spectrometer. The reaction gas included NH₃, NO_x, SO₂, and O₂, their concentrations were 340, 600, 340 mg·m^-3, and volume fraction of 6% respectively. N₂ was carrier gas.

  1.3   Activity test

  Catalysts with sizes of 20-40 mesh were used for the activity test. The composition of the simulated exhaust gas was 600 mg·m^-3 NO_x, 340 mg·m^-3 NH₃, 340 mg·m^-3 SO₂, volume fraction of 6% O₂, and volume fraction of 5% H₂O. The space velocity on catalysts was 10 000 h^-1. The concentration of NO_x, NH₃, and N₂O was detected by a 3012H flue gas analyzer and Nicolet iS50 FTIR spectrometer respectively. NO_x conversion rate (α_NOx) and N₂ selectivity (S_N2) were determined by Formula 1 and 2 respectively:

  $ \alpha_{\mathrm{NO}_x}=\left(1-\frac{\rho_{\mathrm{NO}_x, \text { out }}}{\rho_{\mathrm{NO}_x \text {, in }}}\right) \times 100 \% $

  (1)

  where ρ_NOx is the inlet concentration of NO_x, mg·m^-3; ρ_{NOx, out} is the outlet concentration of NO_x, mg·m^-3.

  $ S_{\mathrm{N}_2}=\left(1-\frac{2 \rho_{\mathrm{N}_2 \mathrm{O}, \text { out }}}{\rho_{\mathrm{NO}_x, \text { in }}+\rho_{\mathrm{NH}_3, \text { in }}-\rho_{\mathrm{NO}_x, \text { out }}-\rho_{\mathrm{NH}_3, \text { out }}}\right) \times 100 \% $

  (2)

  Where ρ_{NH3, in} is the inlet concentration of NH₃, mg·m^-3; ρ_{NH3, out} is the outlet concentration of NH₃, mg·m^-3, and ρ_{N2O, out} is the outlet concentration of N₂O, mg·m^-3.

  2.   Results and discussion

  2.1   Texture properties

  The XRD patterns of fresh and used catalysts are shown in Fig. 1. It is clear that the catalysts displayed obvious diffraction peaks at 25.3°, 36.9°, 37.8°, 38.8°, 48.0°, 53.9°, 55.1°, 62.7°, 68.8°, 70.3°, and 75.1°, which are ascribed to anatase^[8]. Besides, two diffraction peaks at 28.9° and 33.1° were also observed, which are attributed to CeO₂ and MnO_x respectively^[11-12]. As shown in the inset of Fig. 1, the peak intensity of CeO₂ (28.9°) and MnO_x (33.1°) in HoCeMnTi-I was slightly stronger than that in HoCeMnTi-C, suggesting that the co-precipitation method slightly improved the dispersity of active component. After reacting in water and sulfur-containing exhaust gas, the intensity of the diffraction peaks of the two catalysts did not change obviously. In addition, the diffraction peaks ascribed to sulfates were not observed, demonstrating that few sulfates are generated or the sulfates on the surface of the catalysts are amorphous.

  Figure 1

  

  Figure 1.  XRD patterns of fresh and used catalysts

  Inset: the corresponding enlarged XRD patterns

  下载: 全尺寸图片 幻灯片

  The texture parameters of catalysts are listed in Table 1. It was apparent that the catalysts prepared by the two methods had mesoporous structures. Moreover, the preparation method has an apparent impact on the specific surface of catalysts. The BET surface area of HoCeMnTi-C was higher than that of HoCeMnTi-I, demonstrating that the co-precipitation method improves the dispersion of components and alleviates the aggregation of particles, which is conducive to the increase of active sites on catalysts, and then improves the conversion of NO_x. The surface area of catalysts decreased after the reaction due to the formation of sulfates, which covered the surface of catalysts. HoCeMnTi-C exhibited a lower decrease in BET surface area than HoCeMnTi-I, demonstrating its better sulfur resistance performance.

  Table 1

  

  Table 1.  Texture parameters of the catalysts

  下载: 导出CSV

  Catalyst	Surface area/(m2·g-1)	Pore volume/(cm3·g-1)	Pore size/nm
  HoCeMnTi-I	80	0.31	13.1
  HoCeMnTi-C	96	0.36	13.6
  HoCeMnTi-Ia	75	0.28	14.5
  HoCeMnTi-Ca	93	0.33	13.2

  2.2   XPS results

  Fig. 2a-2f display XPS spectra of Ce3d, Mn2p, O1s, Ho4d, Ti2p, and S2p respectively. As shown in Fig. 2a, Ce3d spectra can be divided into eight peaks. The peaks signed as "v₁" and "u₁" are attributed to Ce³⁺, while those signed as "v₀", "v₂", "v₃" and "u₀", "u₂", "u₃" are attributed to Ce^{4+ [13]}. The XPS spectra of Mn2p_3/2 (Fig. 2b) were fitted into three peaks at a binding energy of 643.3-644.0, 639.5-641.1, and 638.0-638.8 eV, which correspond to Mn⁴⁺, Mn³⁺, and Mn²⁺ respectively^[14-15]. The binding energy of Mn2p_3/2 of HoCeMnTi-C was higher than that of HoCeMnTi-I, demonstrating different chemical environments of active components in the two catalysts. Ho4d XPS spectra exhibited two peaks at a binding energy of 165.5-165.9 and 158.8-159.4 eV, which correspond to the feature of Ho³⁺ and Ho^{2+ [16]}. The result indicates the existence of Ho₂O₃. It has been disclosed that the existence of Ho₂O₃ is favorable to the enhancement of Lewis acid sites in catalysts, which is conducive to SCR reaction^[17]. The XPS spectra of O1s were fitted into two peaks. The peak located at 527.7-528.2 eV is attributed to the surface adsorbed oxygen (O_α), while that located at 526.7-526.2 eV arises from the lattice oxygen (O_β). Ti2p spectra also displayed two peaks (Fig. 2e) at a binding energy of 455.0-455.5 and 460.8-461.4 eV, which originate from Ti2p_3/2 and Ti2p_1/2 respectively^[18]. In contrast to HoCeMnTi-I, the O1s and Ti2p spectra of HoCeMnTi-C moved to lower binding energy, demonstrating the enhanced interaction between the active species and support in the co- precipitation method. The possible reason is as follows. In the co-precipitation method, the active component and support react in an ionic state, while in the impregnation method, the active component is adsorbed on support in the form of ions. So, the interaction between the active component and support in the co-precipitation method is stronger than that in the impregnation method.

  Figure 2

  

  Figure 2.  XPS spectra of (a) Ce3d, (b) Mn2p, (c) O1s, (d) Ho4d, (e) Ti2p, and (f) S2p of fresh and used catalysts

  下载: 全尺寸图片 幻灯片

  Table 2 lists the surface element composition calculated based on XPS spectra of the catalysts. Compared with HoCeMnTi-I, HoCeMnTi-C displayed a higher proportion of Ce³⁺, Mn⁴⁺, and O_α. Previous research showed that Ce³⁺ is instrumental in the production of oxygen vacancies through the reaction of 2CeO₂ → Ce₂O₃+O*^[4]. So, a high ratio of O_α is along with a high proportion of Ce³⁺. In addition, the redox cycles of 2MnO₂ → Mn₂O₃+O, Mn₂O₃+2CeO₂ → 2MnO₂+Ce₂O₃ and Ce₂O₃+1/2O₂ → 2CeO₂ in MnO_x-CeO_x catalysts are profitable for the production of active oxygen species^[19]. The high proportion of Mn⁴⁺ and Ce³⁺ in HoCeMnTi-C demonstrates that the electron transport between MnO_x and CeO_x is improved by the co-precipitation method. The coexistence of Ce⁴⁺ and Ce³⁺ is beneficial for the proceeding of SCR reaction^[20]. In contrast, Mn⁴⁺ is profitable for the oxidation of NO and promotes the fast SCR reaction^[21]. The active oxygen is instrumental in improving the oxidative activity of the catalyst^[22]. So, the high proportion of Ce³⁺, Mn⁴⁺, and O_α impart HoCeMnTi-C excellent denitration performance, which will be confirmed in the activity result.

  Table 2

  

  Table 2.  Atomic fractions of elements on the catalyst surface

  下载: 导出CSV

  Catalyst	Atomic fraction/%							nCe3+/nCe	nOα/nO	nMn4+/nMn3+	nMn4+/(nMn2++nMn3++nMn4+)
  	Ti	Mn	Ce	Ho	O	S	N
  HoCeMnTi-I	22.59	1.21	0.62	0.79	74.79	—	—	24.24	35.03	0.43	15.06
  HoCeMnTi-C	21.97	1.51	0.56	0.55	75.42	—	—	27.49	42.37	0.47	16.02
  HoCeMnTi-Ia	19.83	2.94	0.26	0.18	76.79	1.25	1.13	32.71	44.37	0.37	10.18
  HoCeMnTi-Ca	19.56	3.84	0.26	0.10	75.50	0.53	0.62	30.49	45.24	0.38	12.94

  After the reaction, the XPS spectra of Mn2p and Ce3d of the two catalysts both shifted to higher binding energy. The result indicates that Mn and Ce were sulfated. Ce3d XPS spectra of the two catalysts both shifted about 0.3 eV. Mn2p_3/2 spectra of HoCeMnTi-Ia shifted about 0.9 eV, while that of HoCeMnTi-Ca shifted about 0.4 eV. The result demonstrates more serious sulfation of Mn than Ce, and HoCeMnTi-I suffers from more severe damage of SO₂. The Ti2p spectra of HoCeMnTi-Ia also moved about 0.3 eV to higher binding energy, indicating that TiO₂ in HoCeMnTi-I also suffers from sulfation after reaction and forms Ti(SO₄)₂. After being used, the content of Mn⁴⁺ decreased owing to the formation of MnSO₄. In this course, electrons transfer from SO₂ to Mn⁴⁺ to form SO₃, and then SO₃ is combined with Mn²⁺ to form MnSO₄^[23]. In addition, the proportion of Ce³⁺ and O_α increased after reaction. The increase of Ce³⁺ is caused by the formation of Ce₂(SO₄)₃ through the reaction between CeO₂ and SO₂^[24]. In contrast, the increase of O_α is owing to the following two reasons. One is the conversion of SO₂ to SO₃^[25]. The other is the accumulation of excessive adsorption of oxygen caused by the blockage of the oxygen transfer route (O^2- → O₂^2- → O^2-)^[26]. The atomic ratio of Ce³⁺ and O_α in HoCeMnTi-Ca increased by 10.9% and 6.8% respectively. While the Mn⁴⁺ ratio (atomic fraction) of HoCeMnTi-Ca decreased by 19.2%. The variation of Ce³⁺, O_α, and Mn⁴⁺ in HoCeMnTi-C was relatively small, demonstrating its outstanding sulfur resistance. Fig. 2f shows the S2p XPS spectra of the catalysts after being used. The peaks located at 166.8 and 165.8 eV belonged to HSO₄^- and SO₄^2- respectively^[9]. The area of HSO₄^- was bigger than that of SO₄^2-, demonstrating that bisulfates are predominant on the two catalysts. HoCeMnTi-Ca displayed lower content of S and N, indicating that few sulfates and bisulfates form on it. The result further confirms better sulfur resistance of HoCeMnTiC.

  2.3   H₂-TPR and NH₃-TPD results

  H₂-TPR spectra of the catalysts are shown in Fig. 3. It was clear that the spectra of the fresh HoCeMnTi-C and HoCeMnTi-I catalysts were composed of four H₂ consumption peaks. For the HoCeMnTi-C catalyst, the peak located at about 290 ℃ originates from the reduction of Mn⁴⁺ to Mn³⁺ and surface chemisorbed oxygen. The peaks located at 315, 420, and 525 ℃ are correspond to the redox of Mn³⁺ to Mn^8/3+, Mn^8/3+ to Mn²⁺, and Ce⁴⁺ to Ce³⁺ respectively^[3]. Compared to HoCeMnTi-C, the reduction peaks of HoCeMnTi-I shifted to higher temperatures. The result indicates that HoCeMnTi-C is more likely to be reduced, which is ascribed to its higher ratio of O_α. It has been disclosed that the high ratio of O_α makes catalysts more reducible because of their stronger mobility from interior to surface^[2-3]. Moreover, HoCeMnTi-C also displayed a bigger reductive peak area below 300 ℃ and in the whole test temperatures (Table 3), demonstrating its higher amount of H₂ consumption. The high amount of H₂ consumption below 300 ℃ is attributed to the higher ratio of Mn⁴⁺ and O_α as proved by XPS results. The result shown above demonstrates better low-temperature redox performance of HoCeMnTi-C. That is to say, the co-precipitation method is beneficial in improving the redox properties of catalysts. Prior research has disclosed that the activity of manganese declines in the sequence of MnO₂ > Mn₅O₈ > Mn₂O₃ > Mn₃O₄ > MnO^[26]. Consequently, a high ratio of Mn⁴⁺ imparts HoCeMnTi-C outstanding NO_x conversion performance at low temperatures, which will be testified in the activity result. After the reaction, the reduction peaks of the two catalysts moved to higher temperatures, indicating the decline in the redox performance of the catalysts. The result is consistent with the decrease of O_α, as it reacts with SO₂ first in the NH₃-SCR reaction^[27]. In addition, the number of the redox peaks of the catalysts decreased, but the peak area increased. The decrease in the number of peaks originates from the overlap of the redox peaks of sulfates and active oxides^[28]. The increase in peak area is due to the decomposition of (NH₄)₂SO₄ and NH₄HSO₄, which results in the increment of H₂ depletion^[27]. The peak area of HoCeMnTi-Ca rose by 28%, while that of HoCeMnTi-Ia rose by 47%. The rise of the peak area of HoCeMnTi-Ca was lower than that of HoCeMnTi-Ia, indicating the production of less (NH₄)₂SO₄ and NH₄HSO₄ on HoCeMnTi-Ca. The result demonstrates better sulfur tolerance of HoCeMnTi-C.

  Figure 3

  

  Figure 3.  H₂-TPR spectra of the catalysts

  下载: 全尺寸图片 幻灯片

  Table 3

  

  Table 3.  Amounts of H₂ consumption and NH₃ desorption

  下载: 导出CSV

  Catalyst	Peak area
  	H2 consumption peak below 300 ℃	Total H2 consumption	Weak acid	Total acid
  HoCeMnTi-C	16 689	41 491	175	243
  HoCeMnTi-I	3 814	39 854	84	178
  HoCeMnTi-Ca	—	53 248	157	286
  HoCeMnTi-Ia	—	58 583	138	247

  NH₃-TPD spectra of catalysts are exhibited in Fig. 4. It was apparent that the fresh HoCeMnTi-C and HoCeMnTi-I catalysts displayed three NH₃ desorption peaks, demonstrating the existence of different acidic sites on the catalyst surface. The peaks at 140-240 ℃ stem from the NH₃ desorbed from weak acidic sites and NH₄⁺ adsorbed on weak Brønsted acidic sites, while those located at 280-360 ℃ are due to the desorption of NH₃ strongly adsorbed on Brønsted acidic sites and Lewis acidic sites^[29]. The results indicate the presence of both Bønsted acidic and Lewis acidic sites on the catalyst surface. The peak area of NH₃ desorption on HoCeMnTi-C was greater than that on HoCeMnTi-I (Table 3), demonstrating that HoCeMnTi-C possessed more amount of surface acidic sites. A large amount of NH₃ adsorption on HoCeMnTi-C is attributed to its higher ratio of Mn⁴⁺ as NH₃ is more likely to adsorb on Mn⁴⁺ to form Mn⁴⁺-NH₃^[30]. The results demonstrate that co-precipitation is beneficial in improving the amount of surface acidic sites of catalysts. A previous study has disclosed that a higher amount of NH₃ adsorption on Brønsted acidic sites is beneficial for NH₃-SCR reaction at low temperatures^[31]. In contrast with Brønsted acidic sites, Lewis acidic sites can keep a high conversion rate of NO and low yields of N₂O in the same conditions^[23]. Thus, it can be deduced that HoCeMnTi-C shows much higher NO conversion and N₂ selectivity, which will be proved in activity and N₂ selectivity results. Besides, the increase of acidic sites is also beneficial for preventing the adsorption of SO₂ on the catalyst surface, and consequently enhances the ability for sulfur tolerance of the catalysts^[23], which will be confirmed in the result of the sulfur resistance test. After being used in H₂O and SO₂-containing gas, HoCeMnTi-C still maintained three peaks, but the area of peaks below 240 ℃ (α and β peaks) decreased, demonstrating the decrease of NH₃ adsorption. The area of peak at 310 ℃ (γ peak) rose obviously, which may be attributed to the decomposition of (NH₄)₂SO₄/NH₄HSO₄ on the catalyst surface^[27]. For HoCeMnTi-I, the NH₃ desorption of α (160 ℃), β (230 ℃), and γ (280 ℃) peaks all increased. Another peak located at 330 ℃ appeared after the reaction. The increase of NH₃ desorption below 280 ℃ may be attributed to the decomposition of (NH₄)₂SO₄, whereas that located at 330 ℃ originates from the decomposition of NH₄HSO₄ because of its higher decomposition temperatures^[32]. The increase in the peak area of HoCeMnTi-Ca and HoCeMnTi-Ia was 15% and 38% respectively. The rise of the NH₃ desorption peak area in HoCeMnTi-Ca was lower than that in HoCeMnTi-Ia, suggesting that fewer sulfates are formed on HoCeMnTi-Ca. The result demonstrates better SO₂ resistance of HoCeMnTi-C. The existence of sulfates has a great impact on the surface acidity of the catalyst, which affects the adsorption and activation of NH₃.

  Figure 4

  

  Figure 4.  NH₃-TPD spectra of the catalysts

  下载: 全尺寸图片 幻灯片

  2.4   In-situ DRIFTS analysis

  2.4.1   NH₃ adsorption

  DRIFTS of the NH₃ adsorption are displayed in Fig. 5. As shown in Fig. 5, the bands at 3 357, 3 258, 1 620, 1 430, and 1 237 cm^-1 were observed on HoCeMnTi-I after injection of NH₃ at 50 ℃. The bands at 3 357, 3 258, 1 620, and 1 237 cm^-1 originated from NH₃ on Lewis acidic sites^[33]. The band at 1 430 cm^-1 is attributed to NH₄⁺ on Brønsted acidic sites^[18]. With the rise of temperatures, the bands at 3 357, 3 258, and 1 237 cm^-1 gradually became weak and then disappeared at 250 ℃. But the bands at 1 620 and 1 430 cm^-1 increased and shifted to high wavenumber with the increase of temperatures. The result may be attributed to the change in the surface state of the catalyst after the rise of temperatures, which leads to the increase of NH₃ adsorption sites. A new peak ascribed to NO_x appeared at 1 357 cm^-1 over 350 ℃, demonstrating the reaction of the adsorbed NH₃ and chemisorbed oxygen^[34], which results in insufficiency of reductant and declining of NH₃-SCR performance^[35]. As for HoCeMnTi-C, the bands at 3 366, 3 257, 1 602, 1 549, 1 291, 1 225, and 1 154 cm^-1 occurred, which stemmed from NH₃ on Lewis acid. The bands at 1 653 and 1 435 cm^-1 originate from NH₄⁺ on Brønsted acidic sites^[36]. With the increase of the temperatures, all the bands became weak clearly. In contrast to HoCeMnTi-I, HoCeMnTi-C showed stronger bands of the NH₃ adsorption. In addition, the NH₃ adsorption bands on HoCeMnTi-C disappeared at higher temperatures, which was consistent with the NH₃-TPD result. The results discussed above demonstrate that the co- precipitation method not only improves the NH₃ adsorption of HoCeMnTi-C but also prevents its oxidation, which promotes the SCR reaction obviously^[34].

  Figure 5

  

  Figure 5.  DRIFTS of the catalysts after NH₃ adsorption

  Lewis acid; B: Brønsted acid.

  下载: 全尺寸图片 幻灯片

  2.4.2   NO+O₂ adsorption

  Fig. 6 displays DRIFT spectra of NO+O₂ adsorption. The introduction of NO+O₂ led to the appearance of three peaks at 1 604, 1 557, and 1 297 cm^-1 on HoCeMnTi-I, which originate from bridging nitrate (1 604 cm^-1), bidentate nitrate (1 557 cm^-1), and monodentate nitrate (1 297 cm^-1) respectively^[37-38]. The rise of temperatures led to the decline and disappearance of bridging nitrate and bidentate nitrate, but monodentate nitrate was maintained on the surface of the catalyst. A new band (1 344 cm^-1) originated from nitro compound appeared at 250 ℃ and increased with the increase of temperatures, which is consistent with Li′s study^[38]. As for HoCeMnTi-C, the bands attributed to bridging nitrate (1 647 cm^-1), monodentate nitrate (1 521 cm^-1), linear nitrite (1 470, 1 418 cm^-1) bridging nitrite (1 215 cm^-1), and bidentate nitrate (1 100 cm^-1)^[39] occurred after adsorption of NO+O₂ at 50 ℃. With the increase in the temperatures, bridging nitrate, monodentate nitrate, linear nitrite, and bidentate nitrate declined gradually and then disappeared. The bridged nitrite (1 215 cm^-1) converted to more stable bridging nitrate at 150 ℃ and then declined and disappeared with the rise of temperatures. Nitro compound (1 340 cm^-1) on HoCeMnTi-C occurred at 450 ℃. The band intensity on HoCeMnTi-C was stronger than that on HoCeMnTi-I, demonstrating its stronger capacity for the adsorption of NO_x species, which facilitates the conversion of NO in SCR reaction.

  Figure 6

  

  Figure 6.  DRIFTS of the catalysts after NO+O₂ adsorption

  下载: 全尺寸图片 幻灯片

  2.4.3   Reaction between adsorbed NH₃ or NO+O₂ and NO+O₂ or NH₃

  The DRIFTS of catalysts in NO+O₂ flow after NH₃ adsorption are shown in Fig. 7. For HoCeMnTi-I, the injection of NH₃ caused the appearance of the bands at 3 360, 3 262, 1 630, 1 074, 1 678, 1 479, and 1 323 cm^-1. The bands at 3 360, 3 262, 1 630, and 1 074 cm^-1 are related to Lewis acidic sites. The bands at 1 678 and 1 479 cm^-1 are related to Brønsted acidic sites, and that at 1 323 cm^-1 is due to the formation of monodentate nitrate as a result of the reaction between the absorbed NH₃ and active oxygen^[21]. The bands at 3 360, 3 262 cm^-1 declined gradually and then remained unchanged after injection of NO+O₂ for 15 min. While the bands at 1 479 and 1 074 cm^-1 did not change significantly. The bands at 1 678, 1 630, 1 354, and 1 246 cm^-1 increased gradually with the injection of NO+O₂ due to the formation of nitrate species which overlapped with the bands of NH₃ species^[37]. The result indicates that most of the NH₃ species on HoCeMnTi-I were stable, and the formation of nitrate species was not obvious. As for HoCeMnTi-C, the bands at 3 384, 3 624, 1 301, and 1 169 cm^-1 declined gradually and then diminished after injection of NO+O₂ for 15 min. Meanwhile, the bands at 1 601, 1 558, 1 269, and 1 111 cm^-1 increased due to the generation of nitrate species. The results suggest that most of the NH₃ adsorbed on the HoCeMnTi-C surface can be activated and react with NO. The results shown above indicate that the adsorbed NH₃ on HoCeMnTi-C was more reactive than that on HoCeMnTi-I, which is attributed to its large amounts of acidic sites. It has been disclosed that large amounts of acidic sites are beneficial for both adsorption and activation of NH₃.

  Figure 7

  

  Figure 7.  DRIFTS of catalysts in NO+O₂ flow after adsorption of NH₃ at 150 ℃

  下载: 全尺寸图片 幻灯片

  The DRIFTS of the catalysts in NH₃ flow after adsorption of NO+O₂ are shown in Fig. 8. It was obvious that the surface of HoCeMnTi-I was covered by nitro compound (1 349 cm^-1), bidentate nitrates (1 558 and 1 018 cm^-1), monodentate nitrate (1 288 cm^-1), and chelate nitrite (1 137 cm^-1) after exposing in NO+O₂. With the introduction of NH₃, the bands attributed to NO_x species remained stable, but new bands originated from adsorbed NH₃ species (3 364, 3 264 cm^-1) appeared and increased, and then remained stable after 15 min. As for HoCeMnTi-C, the bands attributed to bridging nitrates (1 635 and 1 259 cm^-1), monodentate nitrate (1 521 cm^-1), nitro compound (1 361 cm^-1), and bidentate nitrate (1 107 cm^-1) were observed after injection of NO+O₂. When the flow switched to NH₃, the band at 1 259 cm^-1 increased and shifted gradually to high wavenumber, which may be attributed to the overlap of coordinated NH₃ and bridging nitrates. Other bands attributed to NO_x species remained stable. In addition, new bands at 3 365, 3 260, and 1 169 cm^-1 attributed to coordinated NH₃ species appeared after exposure to NH₃ flow for 5 min and increased with time. The results discussed above indicate that the adsorbed NO_x species on the two catalysts can′t react with NH₃.

  Figure 8

  

  Figure 8.  DRIFTS of the catalysts in NH₃ flow after adsorption of NO+O₂ at 150 ℃

  下载: 全尺寸图片 幻灯片

  The in-situ DRIFTS results suggest that the adsorbed NH₃ species on the two catalysts can participate in the SCR reaction, but the adsorbed NO_x species do not take part in this process. Therefore, the SCR reaction on the two catalysts both follows the Eley-Rideal (E-R) mechanism. Additionally, the reaction between adsorbed NH₃ and NO on HoCeMnTi-C is more quickly, indicating its better activity.

  2.4.4   Influence of SO₂ on adsorption of NH₃ or NO+O₂

  Fig. 9 displays the adsorption of NH₃ on the catalysts in the gas containing 340 mg·m^-3 SO₂. It was clear that the intensity of the bands ascribed to NH₃ species on HoCeMnTi-I increased after exposure to NH₃+SO₂. The result demonstrates the enhancement of NH₃ adsorption in the presence of SO₂, which is ascribed to the increased surface acidic sites as a result of the formation of sulfates^[40-41]. Besides, new bands attributed to the surface and bulk sulfates appeared at 1 443, 1 226, and 1 059 cm^{-1 [42]}. For HoCeMnTi-C, the intensity of the bands attributed to adsorbed NH₃ species (3 361, 3 262, 1 653-1 597, and 1 300-1 219 cm^-1) remained almost unchanged after injection of SO₂. Additionally, new bands originated from the NH₃ bounded to the Brønsted acidic site (1 472 cm^-1), surface sulfate (1 398 cm^-1), and bulk sulfate (1 019 cm^-1)^[43] appeared. The increased NH₃ adsorption and appearance of sulfates confirmed sulfur poisoning of the catalysts. The increase of the band intensity on HoCeMnTi-C was lower than that on HoCeMnTi-I, suggesting that SO₂ had less impact on the NH₃ adsorption of HoCeMnTi-C.

  Figure 9

  

  Figure 9.  In-situ DRIFTS of the catalysts in NH₃ or NH₃+SO₂

  下载: 全尺寸图片 幻灯片

  Fig. 10 displays the adsorption of NO+O₂ in the presence of SO₂. After injection of SO₂, the bands related to NO_x species (1 558, 1 288, 1 142 cm^-1) on HoCeMnTi-I disappeared or decreased, and the bands at 1 326 cm^-1 originated from surface sulfates occurred^[6]. For HoCeMnTi-C, only the bands of NO_x species at 1 250 and 1 101 cm^-1 disappeared, other bands of NO_x species (1 635, 1 521 cm^-1) remained almost unchanged. Moreover, a new band attributed to surface sulfate appeared at 1 327 cm^-1 after injection of SO₂. The results discussed above indicate the competition adsorption between SO₂ and NO, which inhibits the formation of the nitrate species and then leads to the decline of activity. The variation of the band intensity on HoCeMnTi-C was less than that on HoCeMnTi-I, demonstrating that SO₂ had less impact on NO adsorption of HoCeMnTi-C.

  Figure 10

  

  Figure 10.  In-situ DRIFTS of the catalysts in NO+O₂ or NO+O₂+SO₂

  下载: 全尺寸图片 幻灯片

  2.5   Catalytic properties

  Fig. 11 and 12 display the NO_x conversion and N₂ selectively of catalysts respectively. As displayed in Fig. 11, the conversion of NO_x on HoCeMnTi-C was 53% at 50 ℃. That was to say the T₅₀ (the temperatures at which the NO_x conversion reached 50%) of HoCeMnTi-C was approximately 50 ℃. The NO_x conversion on HoCeMnTi-C reached 100% at 130 ℃, and then it remained stable until 290 ℃. When the temperature exceeded 290 ℃, the NO_x conversion declined due to the excessive oxidation of NH₃. But for HoCeMnTi-I, the NO_x conversion was only 32% at 50 ℃ and arrived at 50% at 65 ℃. At 150 ℃, the NO_x conversion reached 100% and remained stable until 250 ℃, then declined with the rise of the temperatures. The ΔT₈₀ (the region of temperatures that the NO_x conversion exceeded 80%) of HoCeMnTi-C and HoCeMnTi-I were 262 and 220 ℃ respectively. Thus, HoCeMnTi-C showed wider ΔT₈₀ and higher NO_x conversion, suggesting its better activity. It was clear from Fig. 12 that the N₂ selectivity of the two catalysts was nearly 100% below 230 ℃. When the temperature exceeded 230 ℃, the selectivity of N₂ declined with the increase of the temperatures. It has been disclosed that the excellent oxidation performance of Mn-based catalysts results in the generation of large amounts of N₂O which suppresses the selectivity of N₂^[44]. The image inset in Fig. 12 shows the concertation of N₂O in the activity tests. It was obvious that the N₂O concentration on both of the two catalysts was low below 230 ℃. When the temperature exceeded 230 ℃, the concentration of N₂O increased with the rise of the temperatures. HoCeMnTi-C displayed lower concertation of N₂O than that of HoCeMnTi-I, which results in its higher N₂ selectivity. From the above analysis, it is easy to know that HoCeMnTi-C showed better activity and N₂ selectivity than HoCeMnTi-I. Based on the characterization results mentioned above, the following reasons may account for this. (ⅰ) HoCeMnTi-C has a higher ratio of Mn⁴⁺, Ce³⁺, and O_α. A large amount of Mn⁴⁺ promotes the oxidation of NO due to its comparatively high oxidation state^[35]. The high content of Ce³⁺ is beneficial for the generation of oxygen vacancies. The existence of oxygen vacancies enhances the mobility of oxygen and the redox properties of the catalyst, which is conducive to the oxidation of NO^[18]. Due to the generation of NO₂, the conversion of NO proceeds by the means of a fast SCR reaction (i.e. 2NH₃+NO+NO₂ → 2N₂+3H₂O)^{[21, 45]}, which significantly enhances the activity of HoCeMnTi-C. (ⅱ) HoCeMnTi-C possesses excellent reducibility properties, which are beneficial for the enhancement of the redox cycle and promoting the SCR reaction^[42]. (ⅲ) HoCeMnTi-C shows stronger surface acidity and possesses more amount of surface acidic sites. The increase of surface acidic sites is beneficial for the NH₃ adsorption. NH₃ linked to Lewis acidic sites can keep high conversion of NO and low production of N₂O^[39]. (ⅳ) HoCeMnTi-C displays better adsorption capacity for NH₃ and NO_x species. More adsorbed NH₃ can be activated and react with NO in a relatively short time. That is, the reaction between the adsorbed NH₃ and NO on HoCeMnTi-C is more easily, demonstrating its better activity.

  Figure 11

  

  Figure 11.  NO_x conversions on the catalysts

  下载: 全尺寸图片 幻灯片

  Figure 12

  

  Figure 12.  N₂ selectivity and N₂O concentration (Inset) of the catalysts

  下载: 全尺寸图片 幻灯片

  2.6   H₂O and SO₂ tolerance of catalysts

  Fig. 13 displays the conversion of NO_x over the catalysts in the presence of H₂O, SO₂, and SO₂+H₂O. The NO_x conversion on the catalysts declined after the injection of 5% H₂O. The decrease of the NO_x conversion is due to the competitive adsorption between NH₃ and H₂O which results in the decrease of NH₃ adsorption^[46-47]. The NO_x conversion on HoCeMnTi-C decreased to 98% while that on HoCeMnTi-I reduced to 96% after the introduction of H₂O. So, the influence of H₂O on the two catalysts was weak. That is to say, both of the two catalysts possess excellent water resistance, which may be attributed to the extensive surface acidic site of the catalysts. It has been disclosed that Brønsted acidic sites are beneficial for promoting water resistance of the catalysts^[48]. Besides, Lewis acidic sites are more favorable to enhance the water resistance of catalysts than Brønsted acidic sites due to the weak interaction between Lewis acidic sites and water^[49]. So, HoCeMnTi-C behaves with better resistance to water than HoCeMnTi-I due to its more acidic sites. After the water was cut off, the conversion of NO_x recovered gradually owing to the disappearance of competitive adsorption between NH₃ and H₂O.

  Figure 13

  

  Figure 13.  NO_x conversion in H₂O, SO₂, and H₂O+SO₂ containing atmosphere at 160 ℃

  下载: 全尺寸图片 幻灯片

  The presence of 340 mg·m^-3 SO₂ brought about a significant decline of the NO_x conversion. After being exposed to the exhaust containing SO₂ for 540 min, the NO_x conversion on HoCeMnTi-I decreased to 84%, while that on HoCeMnTi-C decreased to 89%. The results indicate that HoCeMnTi-C had better SO₂ tolerance than HoCeMnTi-I. Simultaneous introduction of H₂O and SO₂ resulted in a rapid decrease in the NO_x conversion, suggesting the aggravation of damage of SO₂ to catalysts in the presence of water. After 540 min, the NO_x conversion on HoCeMnTi-I and HoCeMnTi-C decreased to 79% and 86% respectively. HoCeMnTi-C maintained higher conversion of NO_x than HoCeMnTi-I, indicating its better SO₂ resistance performance. After SO₂ or H₂O+SO₂ was shut off, the conversion of NO_x gradually increased to a certain level, but couldn′t be recovered. The result demonstrates irreversible damage to catalysts. According to NH₃-TPD and in-situ DRIFTS results, HoCeMnTi-C possessed more acidic sites. The increased surface acidic sites are beneficial to suppress the adsorption of SO₂ on the surface of HoCeMnTi-C and prevent the formation of sulfates. As it has been proved by XPS results, few sulfates were formed on HoCeMnTi-C. The damage to the catalysts arises from the formation of surface and bulk sulfates.

  3.   Conclusions

  HoCeMnTi-C prepared by the co-precipitation method possesses more Ce³⁺, Mn⁴⁺, and O_α because of the strong interaction between TiO₂ and the active component, which leads to its outstanding redox properties at low temperatures. In addition, the co-precipitation method results in the increase of surface acidic sites of HoCeMnTi-C, which promotes the adsorption of NH₃. The increased surface acidic sites and outstanding redox properties impart HoCeMnTi-C high conversion of NO_x at low temperatures. The SCR reaction on catalysts complies with the E-R mechanism. The increase of surface acidic sites is also beneficial to prevent the adsorption of H₂O and SO₂ on HoCeMnTi-C, which significantly promotes water and sulfur tolerance of HoCeMnTi-C. Sulfur damage of catalysts originates from the generation of surface and bulk sulfates which causes the coverage and destruction of active ingredients and then results in the decline of NO_x conversion.

  
  
• 1. [1]

     Liu H H, Gao F Y, Ko S J, Luo N, Tang X L, Duan E H, Yi H H, Zhou Y S. Low-temperature NH₃-SCR performance of a novel @Mn composite denitrification catalyst[J]. J. Environ. Sci., 2024, 137:  271-286. doi: 10.1016/j.jes.2022.12.010

  2. [2]

     Ye L M, Lu P, Chen D S, Chen D Y, Wu H W, Dai W J, Gan Y L, Xiao J Y, Xie Z W, Li Z W, Huang H B. Activity enhancement of acetate precursor prepared on MnO_x-CeO₂ catalyst for low-temperature NH₃-SCR: Effect of gaseous acetone addition[J]. Chin. Chem. Lett., 2021, 32:  2509-2512. doi: 10.1016/j.cclet.2020.12.040

  3. [3]

     Chen C, Xie H D, He P W, Liu X, Yang C, Wang N, Ge C M. Comparison of low-temperature catalytic activity and H₂O/SO₂ resistance of the Ce-Mn/TiO₂ NH₃-SCR catalysts prepared by the reverse co-precipitation, co-precipitation and impregnation method[J]. Appl. Surf. Sci., 2022, 571:  151285. doi: 10.1016/j.apsusc.2021.151285

  4. [4]

     Zhang X L, Zhang X C, Yang X J, Chen Y Z, Hu X R, Wu X P. CeMn/TiO₂ catalysts prepared by different methods for enhanced low-temperature NH₃-SCR catalytic performance[J]. Chem. Eng. Sci., 2021, 238:  116588. doi: 10.1016/j.ces.2021.116588

  5. [5]

     Wei L, Cui S P, Guo H X, Ma X Y, Zhang L J. DRIFT and DFT study of cerium addition on SO₂ of manganese-based catalysts for low temperature SCR[J]. J. Mol. Catal. A-Chem., 2016, 421:  102-108. doi: 10.1016/j.molcata.2016.05.013

  6. [6]

     Sun P, Huang S X, Guo R T, Li M Y, Liu S M, Pan W G, Fu Z G, Liu S W, Sun X, Liu J. The enhanced SCR performance and SO₂ resistance of Mn/TiO₂ catalyst by the modification with Nb: A mechanistic study[J]. Appl. Surf. Sci., 2018, 447:  479-488. doi: 10.1016/j.apsusc.2018.03.245

  7. [7]

     Li W, Zhang C, Li X, Tan P, Zhou A L, Fang Q Y, Chen G. Ho-modified Mn-Ce/TiO₂ for low-temperature SCR of NO_x with NH₃: Evaluation and characterization[J]. Chin. J. Catal., 2018, 39:  1653-1663. doi: 10.1016/S1872-2067(18)63099-2

  8. [8]

     Liu L J, Kai Xu K, Su S, He L M, Qing M X, Chi H Y, Liu T, Hu S, Wang Y, Xiang J. Efficient Sm modified Mn/TiO₂ catalysts for selective catalytic reduction of NO with NH₃ at low temperature[J]. Appl. Catal. A-Gen., 2020, 592:  117413. doi: 10.1016/j.apcata.2020.117413

  9. [9]

     Zhu Y J, Xiao X X, Wang J T, Cheng Ma C, Jia X F, Qiao W M, Ling L C. Enhanced activity and water resistance of hierarchical flower-like Mn-Co binary oxides for ammonia-SCR reaction at low temperature[J]. Appl. Surf. Sci., 2021, 569:  150989. doi: 10.1016/j.apsusc.2021.150989

  10. [10]

      Li S H, Huang B C, Yu C L. A CeO₂-MnO_x core-shell catalyst for low-temperature NH₃-SCR of NO[J]. Catal. Commun., 2017, 98:  47-51. doi: 10.1016/j.catcom.2017.04.046

  11. [11]

      Wu X M, Yu X L, He X Y, Jing G H. Insight into low-temperature catalytic NO reduction with NH₃ on Ce-doped manganese oxide octahedral molecular sieves[J]. J. Phys. Chem. C, 2019, 123(17):  10981-10990. doi: 10.1021/acs.jpcc.9b01048

  12. [12]

      Meng D M, Zhan W C, Guo Y, Guo Y L, Wang Y S, Wang L, Lu G Z. A highly effective catalyst of Sm-Mn mixed oxide for the selective catalytic reduction of NO_x with ammonia: Effect of the calcination temperature[J]. J. Mol. Catal. A-Chem., 2016, 420:  272-281. doi: 10.1016/j.molcata.2016.04.028

  13. [13]

      Qiu L, Meng J J, Pang D D, Zhang C L, Ouyang F. Reaction and characterization of Co and Ce doped Mn/TiO₂ catalysts for low-temperature SCR of NO with NH₃[J]. Catal. Lett., 2015, 145:  1500-1509. doi: 10.1007/s10562-015-1556-x

  14. [14]

      Chuang C, Yan Z D, Zhang C L, Zhang Y S, Jiang M, Ruan L N, Xiao M, Yu Y B, He H. Design of Ca-type todorokite catalysts with highly active for the selective reduction of NO_x by NH₃ at low temperatures[J]. J. Environ. Sci., 2024, 138:  697-708. doi: 10.1016/j.jes.2023.04.025

  15. [15]

      Thirupathi B, Smirniotis P G. Nickel-doped Mn/TiO₂ as an efficient catalyst for the low-temperature SCR of NO with NH₃: Catalytic evaluation and characterizations[J]. J. Catal., 2012, 288:  74-83. doi: 10.1016/j.jcat.2012.01.003

  16. [16]

      Secu M, Cernea M, Secu E C, Vasile B S. Structural characterization and photoluminescence of nanocrystalline Ho-doped BaTiO₃ derived from sol-gel method[J]. J. Nanopart. Res., 2011, 13:  3123-3128. doi: 10.1007/s11051-011-0224-3

  17. [17]

      Hang T J, Zhang Y P, Zhuang K, Lu B, Zhu Y W, Shen K. Preparation of honeycombed holmium-modified Fe-Mn/TiO₂ catalyst and its performance in the low temperature selective catalytic reduction of NO_x[J]. J. Fuel Chem. Technol., 2018, 46(3):  319-327. doi: 10.1016/S1872-5813(18)30015-X

  18. [18]

      Zhang Z P, Li R M, Wang M J, Li Y S, Tong Y M, Yang P P, Zhu Y J. Two steps synthesis of CeTiO_x oxides nanotube catalyst: Enhanced activity, resistance of SO₂ and H₂O for low temperature NH₃-SCR of NO_x[J]. Appl. Catal. B-Environ., 2021, 282:  119542. doi: 10.1016/j.apcatb.2020.119542

  19. [19]

      Mu W T, Zhu J, Zhang S, Guo Y Y, Su L Q, Li X Y, Li Z. Novel proposition on mechanism aspects over Fe-Mn/ZSM-5 catalyst for NH₃-SCR of NO_x at low temperature: Rate and direction of multifunctional electron-transfer-bridge and in-situ DRIFTS analysis[J]. Catal. Sci. Technol., 2016, 6(20):  7532-7548. doi: 10.1039/C6CY01510G

  20. [20]

      Chen L, Ren S, Liu L, Su B X, Yang J, Chen Z C, Wang M M, Liu Q C. Catalytic performance over Mn-Ce catalysts for NH₃-SCR of NO at low temperature: Different zeolite supports[J]. J. Environ. Chem. Eng., 2022, 10:  107167. doi: 10.1016/j.jece.2022.107167

  21. [21]

      Wang F M, Shen B X, Zhu S W, Wang Z. Promotion of Fe and Co doped Mn-Ce/TiO₂ catalysts for low temperature NH₃-SCR with SO₂ tolerance[J]. Fuel, 2019, 249:  54-60. doi: 10.1016/j.fuel.2019.02.113

  22. [22]

      Huang X S, Dong F, Zhang G D, Tang Z C. Design and identify the confinement effect of active site position on catalytic performance for selective catalytic reduction of NO with NH₃ at low temperature[J]. J. Catal., 2023, 420:  134-150. doi: 10.1016/j.jcat.2023.02.020

  23. [23]

      Jiang L J, Liu Q C, Ran G J, Kong M, Ren S, Yang J, Li J L. V₂O₅-modifified Mn-Ce/AC catalyst with high SO₂ tolerance for low temperature NH₃-SCR of NO[J]. Chem. Eng. J., 2019, 370:  810-821. doi: 10.1016/j.cej.2019.03.225

  24. [24]

      Wang Y L, Li X X, Zhan L, Li C, Qiao W M, Ling L C. Effect of SO₂ on activated carbon honeycomb supported CeO₂-MnO_x catalyst for NO removal at low temperature[J]. Ind. Eng. Chem. Res., 2015, 54(8):  2274-2278. doi: 10.1021/ie504074h

  25. [25]

      France L J, Yang Q, Li W, Chen Z H, Guang J Y, Guo D W, Wang L F, Li X H. Ceria modified FeMnO_x-enhanced performance and sulphur resistance for low-temperature SCR of NO_x[J]. Appl. Catal. B-Environ., 2017, 206:  203-215. doi: 10.1016/j.apcatb.2017.01.019

  26. [26]

      Fang X, Liu Y J, Chen L Z, Cheng Y. Influence of surface active groups on SO₂ resistance of birnessite for low temperature NH₃-SCR[J]. Chem. Eng. J., 2020, 399:  125798. doi: 10.1016/j.cej.2020.125798

  27. [27]

      Fan A D, Jing Y, Guo J X, Shi X K, Yuan S D, Li J J. Investigation of Mn doped perovskite La-Mn oxides for NH₃-SCR activity and SO₂/H₂O resistance[J]. Fuel, 2022, 310:  122237. doi: 10.1016/j.fuel.2021.122237

  28. [28]

      Jiang L J, Liang Y, Liu W Z, Wu H L, Aldahri T, Carrero D S, Liu Q C. Synergistic effect and mechanism of FeO_x and CeO_x co-doping on the superior catalytic performance and SO₂ tolerance of Mn-Fe-Ce/ACN catalyst in low-temperature NH₃-SCR of NO_x[J]. J. Environ. Chem. Eng., 2021, 9(6):  106360. doi: 10.1016/j.jece.2021.106360

  29. [29]

      Fang N J, Guo J X, Shu S, Luo H D, Li J J, Chu Y H. Effect of calcination temperature on low temperature NH₃-SCR activity and the resistance of SO₂ with or without H₂O over Fe-Mn-Zr catalyst[J]. J. Taiwan Inst. Chem. Eng., 2018, 93:  277-288. doi: 10.1016/j.jtice.2018.07.027

  30. [30]

      Jia B H, Guo J X, Shu S, Fang N J, Li J J, Chu YH. Effects of different Zr/Ti ratios on NH₃-SCR over MnO_x/Zr_yTi_1-yO₂: Characterization and reaction mechanism[J]. Mol. Catal., 2017, 443:  25-37. doi: 10.1016/j.mcat.2017.09.019

  31. [31]

      Fan Y X, Zhang J, Yang L X, Lu M X, Ying T T, Deng B H, Dai W L, Luo X B, Zou J P, Luo S L. Enhancing SO₂-shielding effect and Lewis acid sites for high efficiency in low-temperature SCR of NO with NH₃: Reinforced electron-deficient extent of Fe³⁺ enabled by Ti⁴⁺ in Fe₂O₃[J]. Sep. Purif. Technol., 2023, 311:  123272. doi: 10.1016/j.seppur.2023.123272

  32. [32]

      Shi X K, Guo J X, Shen T, Fan A D, Liu Y J, Yuan S D. Improvement of NH₃-SCR activity and resistance to SO₂ and H₂O by Ce modified La-Mn perovskite catalyst[J]. J. Taiwan Inst. Chem. Eng., 2021, 126:  102-111. doi: 10.1016/j.jtice.2021.06.056

  33. [33]

      Guo K, Ji J W, Osuga R, Zhu Y X, Sun J F, Tang C J, Kondo J N, Dong L. Construction of Fe₂O₃ loaded and mesopore-confined thin-layer titania catalyst for efficient NH₃-SCR of NO_x with enhanced H₂O/SO₂ tolerance[J]. Appl. Catal. B-Environ., 2021, 287:  119982. doi: 10.1016/j.apcatb.2021.119982

  34. [34]

      Yang C X, Zhang K X, Zhang Y K, Peng G J, Yang M, Wen J J, Xie Y, Xia F T, Jia L J, Zhang Q L. An environmental and highly active Ce/Fe-Zr-SO₄^2- catalyst for selective catalytic reduction of NO with NH₃: The improving effects of CeO₂ and SO₄^2-[J]. J. Environ. Chem. Eng., 2021, 9:  106799. doi: 10.1016/j.jece.2021.106799

  35. [35]

      Ma Y P, Li W, Wang H M, Chen J J, Wen J J, Xu S Y, Tian X Y, Gao L Y, Hou Z C, Zhang Q L Yang H. Enhanced performance of iron-cerium NO_x reduction catalysts by sulphuric acid treatment: The synergistic effect of surface acidity and redox capacity[J]. Appl. Catal. A-Gen., 2021, 621:  118200. doi: 10.1016/j.apcata.2021.118200

  36. [36]

      Qin B, Guo R T, Wei L G, Yin X F, Yin T Y, Zhou J, Qiu Z Z. A highly effective NbMnCeO_x catalyst for NH₃-SCR and in situ DRIFTS for investigating the reaction mechanism[J]. J. Environ. Chem. Eng., 2022, 10:  108564. doi: 10.1016/j.jece.2022.108564

  37. [37]

      Zhang Q L, Fan J, Ning P, Song Z X, Liu X, Wang L Y, Wang J, Wang H M, Long K X. In situ DRIFTS investigation of NH₃-SCR reaction over CeO₂/zirconium phosphate catalyst[J]. Appl. Surf. Sci., 2018, 435:  1037-1045. doi: 10.1016/j.apsusc.2017.11.180

  38. [38]

      Liu Z M, Zhu J Z, Li J H, Ma L L, Woo S I. Novel Mn-Ce-Ti mixed-oxide catalyst for selective catalytic reduction of NO_x with NH₃[J]. ACS Appl. Mater. Interfaces, 2014, 6:  14500-14508. doi: 10.1021/am5038164

  39. [39]

      Qin Q J, Chen K, Xie S Z, Li L L, Ou X M, Wei X L, Luo X T, Dong L H, Li B. Enhanced SO₂ and H₂O resistance of MnTiSnO_y composite oxide for NH₃-SCR through Sm modification[J]. Appl. Surf. Sci., 2022, 583:  152478. doi: 10.1016/j.apsusc.2022.152478

  40. [40]

      Cao F, Su S, Xiang J, Wang P Y, Hu S, Sun L S, Zhang A C. The activity and mechanism study of Fe-Mn-Ce/γ-Al₂O₃ catalyst for low temperature selective catalytic reduction of NO with NH₃[J]. Fuel, 2015, 139:  232-239. doi: 10.1016/j.fuel.2014.08.060

  41. [41]

      Liu L J, Su S, Chen D Z, Shu T, Zheng X T, Yu J Y, Feng Y, Wang Y, Hu S, Xiang J. Highly efficient NH₃-SCR of NO_x over MnFeW/Ti catalyst at low temperature: SO₂ tolerance and reaction mechanism[J]. Fuel, 2022, 307:  121805. doi: 10.1016/j.fuel.2021.121805

  42. [42]

      Wang C Z, Gao F Y, Ko S J, Liu H H, Yi H H, Tang X L. Structural control for inhibiting SO₂ adsorption in porous MnCe nanowire aerogel catalysts for low-temperature NH₃-SCR[J]. Chem. Eng. J., 2022, 434:  134729. doi: 10.1016/j.cej.2022.134729

  43. [43]

      Liu J, Guo R T, Li M Y, Sun P, Liu S M, Pan W G, Liu S W, Sun X. Enhancement of the SO₂ resistance of Mn/TiO₂ SCR catalyst by Eu modification: A mechanism study[J]. Fuel, 2018, 223:  385-393. doi: 10.1016/j.fuel.2018.03.062

  44. [44]

      Zhao B H, Ran R, Guo X G, Cao L, Xu T F, Chen Z, Wu X D, Si Z C, Weng D. Nb-modified Mn/Ce/Ti catalyst for the selective catalytic reduction of NO with NH₃ at low temperature[J]. Appl. Catal. A-Gen., 2017, 545:  64-71. doi: 10.1016/j.apcata.2017.07.024

  45. [45]

      Zhou J, Guo R T, Zhang X F, Liu Y Z, Duan C P, Wu G L, Pan W G. Cerium oxide-based catalysts for low-temperature selective catalytic reduction of NO_x with NH₃: A review[J]. Energy Fuel, 2021, 35:  2981-2998. doi: 10.1021/acs.energyfuels.0c04231

  46. [46]

      Liu H Z, Chen Z, Wang H M, You C F. Active centers response to SO₂ and H₂O poisoning over Fe-W-Ni exchanged zeolite for high-temperature NH₃-SCR: Experimental and DFT studies[J]. Appl. Surf. Sci., 2021, 570:  151105. doi: 10.1016/j.apsusc.2021.151105

  47. [47]

      Lai J K, Wachs I E. A perspective on the selective catalytic reduction (SCR) of NO with NH₃ by supported V₂O₅-WO₃/TiO₂ catalysts[J]. ACS Catal., 2018, 8:  6537-6551. doi: 10.1021/acscatal.8b01357

  48. [48]

      Ma S B, Zhao X Y, Li Y S, Zhang T R, Yuan F L, Niu X Y, Zhu Y J. Effect of W on the acidity and redox performance of the Cu_0.02Fe_0.2W_aTiO_x (a=0.01, 0.02, 0.03) catalysts for NH₃-SCR of NO[J]. Appl. Catal. B-Environ., 2019, 248:  226-238. doi: 10.1016/j.apcatb.2019.02.015

  49. [49]

      Ma S B, Tan H S, Li Y S, Wang P Q, Zhao C, Niu X Y, Zhu Y J. Excellent low-temperature NH₃-SCR NO removal performance and enhanced H₂O resistance by Ce addition over the Cu_0.02Fe_0.2Ce_yTi_1-yO_x (y=0.1, 0.2, 0.3) catalysts[J]. Chemosphere, 2020, 243:  125309. doi: 10.1016/j.chemosphere.2019.125309

• 

  Figure 1  XRD patterns of fresh and used catalysts

  Inset: the corresponding enlarged XRD patterns

  下载: 全尺寸图片 幻灯片
  

  Figure 2  XPS spectra of (a) Ce3d, (b) Mn2p, (c) O1s, (d) Ho4d, (e) Ti2p, and (f) S2p of fresh and used catalysts

  下载: 全尺寸图片 幻灯片
  

  Figure 3  H₂-TPR spectra of the catalysts

  下载: 全尺寸图片 幻灯片
  

  Figure 4  NH₃-TPD spectra of the catalysts

  下载: 全尺寸图片 幻灯片
  

  Figure 5  DRIFTS of the catalysts after NH₃ adsorption

  Lewis acid; B: Brønsted acid.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  DRIFTS of the catalysts after NO+O₂ adsorption

  下载: 全尺寸图片 幻灯片
  

  Figure 7  DRIFTS of catalysts in NO+O₂ flow after adsorption of NH₃ at 150 ℃

  下载: 全尺寸图片 幻灯片
  

  Figure 8  DRIFTS of the catalysts in NH₃ flow after adsorption of NO+O₂ at 150 ℃

  下载: 全尺寸图片 幻灯片
  

  Figure 9  In-situ DRIFTS of the catalysts in NH₃ or NH₃+SO₂

  下载: 全尺寸图片 幻灯片
  

  Figure 10  In-situ DRIFTS of the catalysts in NO+O₂ or NO+O₂+SO₂

  下载: 全尺寸图片 幻灯片
  

  Figure 11  NO_x conversions on the catalysts

  下载: 全尺寸图片 幻灯片
  

  Figure 12  N₂ selectivity and N₂O concentration (Inset) of the catalysts

  下载: 全尺寸图片 幻灯片
  

  Figure 13  NO_x conversion in H₂O, SO₂, and H₂O+SO₂ containing atmosphere at 160 ℃

  下载: 全尺寸图片 幻灯片

  Table 1.  Texture parameters of the catalysts

  Catalyst	Surface area/(m2·g-1)	Pore volume/(cm3·g-1)	Pore size/nm
  HoCeMnTi-I	80	0.31	13.1
  HoCeMnTi-C	96	0.36	13.6
  HoCeMnTi-Ia	75	0.28	14.5
  HoCeMnTi-Ca	93	0.33	13.2

  下载: 导出CSV

  Table 2.  Atomic fractions of elements on the catalyst surface

  Catalyst	Atomic fraction/%							nCe3+/nCe	nOα/nO	nMn4+/nMn3+	nMn4+/(nMn2++nMn3++nMn4+)
  	Ti	Mn	Ce	Ho	O	S	N
  HoCeMnTi-I	22.59	1.21	0.62	0.79	74.79	—	—	24.24	35.03	0.43	15.06
  HoCeMnTi-C	21.97	1.51	0.56	0.55	75.42	—	—	27.49	42.37	0.47	16.02
  HoCeMnTi-Ia	19.83	2.94	0.26	0.18	76.79	1.25	1.13	32.71	44.37	0.37	10.18
  HoCeMnTi-Ca	19.56	3.84	0.26	0.10	75.50	0.53	0.62	30.49	45.24	0.38	12.94

  下载: 导出CSV

  Table 3.  Amounts of H₂ consumption and NH₃ desorption

  Catalyst	Peak area
  	H2 consumption peak below 300 ℃	Total H2 consumption	Weak acid	Total acid
  HoCeMnTi-C	16 689	41 491	175	243
  HoCeMnTi-I	3 814	39 854	84	178
  HoCeMnTi-Ca	—	53 248	157	286
  HoCeMnTi-Ia	—	58 583	138	247

  下载: 导出CSV
•