Preparation of novel CeMo(x) hollow microspheres for low-temperature SCR removal of NO_x with NH₃

Abstract

A series of novel Mo doped CeO₂ hollow microspheres have been successfully synthesized using carbon microspheres as templates, which were characterized by XRD, SEM, TEM, BET and used to selective catalytic reduction (SCR) of NO_x with ammonia at lower temperature. Compared with pure CeO₂ hollow microspheres, Mo doped CeO₂ hollow microspheres showed a better catalytic performance from 200–400 °C. The optimal molar ratios of Mo/Ce and surface active sites to remove NO_x were also investigated, which can significantly affect the catalytic performances. Among all the hollow microspheres with different molybdenum doping ratios, the obtained CeMo(0.3) hollow microspheres showed not only the best SCR performance at lower temperature but also excellent stability and H₂O resistance, which was attributed to the higher Brønsted acid sites and reasonable Ce³⁺ content and chemisorbed oxygen species, and the synergic effect between CeO₂ and MoO₃ based on the XPS, NH₃-TPD, H₂-TPR and in situ DRIFTS results. Additionally, according to the reactivity of adsorbed NH₃ and NO_x species based on in situ DRIFTS, the excellent low-temperature NH₃-SCR activity of CeMo(0.3) hollow microspheres was resulted from the reactivity of the more active NH₄⁺ ions with gaseous NO₂ molecules (i.e., the “fast SCR” reaction).

---

Introduction

Nitrogen oxide (NO_x) is a serious threat to the environment, and has resulted in many environmental issues, such as acid rain, photochemical smog, ozone depletion, and greenhouse effects. Selective catalytic reduction with ammonia (NH₃-SCR) is a widely used and effective deNOx technology.^1–3 According to the reaction window, the SCR process contains three types, they are high – (>450 °C), medium – (320 to 450 °C) and low-temperature (120 to 320 °C) SCR technology.⁴ At present, commercial V₂O₅–WO₃/TiO₂ and V₂O₅–MoO₃/TiO₂ catalysts show better deNOx performance from 300 to 400 °C, which belongs to medium temperature SCR catalyst.^5,6 Nevertheless, several inevitable drawbacks of the toxicity of vanadium species, higher operating temperatures and poor tolerance for dust and alkali metals still exist.⁷ Therefore, a great deal of efforts have been made to develop highly efficient lower-temperature SCR catalysts without vanadium element, which can be located downstream of the electrostatic precipitator and desulfurization.⁸

Recently, ceria oxides (CeO₂) nanomaterials have become an ideal SCR catalyst candidates owing to its high oxygen storage capacity, excellent redox property, non-toxicity and low cost.^9–11 To promote their catalytic performance, many CeO₂-based catalysts have been doped with special metal oxides, such as SnO_x–CeO₂, TiO₂–CeO₂, Nb₂O₅–CeO₂ and so on.^12–14 Nevertheless, these catalysts still have several defects, such as the narrow temperature window, weak thermal stability at higher temperature and poor SO₂ resistance. As “chemical” and “structural” promoters, MoO₃ was a common activator in the SCR reaction. Several reports have proved that Mo species could promote the catalytic activity, thermal stability and SO₂/H₂O durability of the catalysts.^15,16 It was also investigated that MoO₃ can promote the adsorption and activation of NH₃ thus improve the catalytic activity.¹⁷ Therefore, the combination of CeO₂ and MoO₃ was prepared and applied for eliminating NO_x by NH₃-SCR just as reported by Peng et al.¹⁸

Furthermore, “structure and morphology effect” of nanomaterials have attracted a lot of attentions recently.^19–21 Among numerous of structures (nanospheres, nanowire, nanosheet, mesoporous and hollow spheres), the hollow microspheres have attracted much attention particularly, which can offer short diffusion length of reaction gases and efficient channels for mass transport.^22,23 Zhang et al. reported that the hollow Mn_xCo_3−xO₄ nanocages exhibited better low-temperature catalytic activity, higher stability, and SO₂ tolerance than conventional Mn_xCo_3−xO₄ nanoparticles in deNOx process attributed to the hollow and porous architectures.²⁰ Moreover, Ca-doped FeO_x hollow microspheres have been synthesized for catalytic oxidation of 1,2-dichlorobenzene (o-DCB) and exhibited excellent activity.²¹ To the best of our knowledge, there were few studies about the catalytic properties of Mo doped CeO₂ hollow microspheres in the NH₃-SCR reaction until now.

In our present study, a series of CeMo(x) (x = 0, 0.1, 0.2, 0.3, 0.4, 0.5) hollow microspheres have been synthesized successfully by using carbon microspheres as template to remove NO_x. The as-prepared hollow microspheres have been tested systematically, including XRD, TEM, BET, XPS, TPR, TPD and XPS. With these CeMo(x) hollow microspheres, the NO_x conversion ratio can be improved obviously from 200 to 400 °C with a gas hourly space velocity (GHSV) of 30 000 h⁻¹. Furthermore, several key performance indicators were also investigated, such as the stability, H₂O resistance and SO₂ tolerance. Finally, the reaction mechanism of Mo-doped CeO₂ hollow microspheres was proposed based on the results of in situ DRIFTS.

Experimental

Preparation of materials

All chemicals used here were analytical grade without further purification. In order to prepare the novel CeMo(x) hollow microspheres, the carbon microsphere was firstly prepared as template. In brief, 0.6 g cetyltrimethylammonium bromide (CTAB) and 9 g glucose were dissolved in 90 mL distilled water. After being stirred for 30 min, the solution was transferred into a Teflon-lined autoclave and maintained at 180 °C for 9 h. After cooling to room temperature, the black precipitates were collected and washed with distilled water for five times. Finally, the obtained carbon microspheres were dried in air at 80 °C.

To prepare Mo doped CeO₂ hollow microspheres, 0.2 g carbon microspheres were dispersed in 50 mL N,N-dimethylformamide (DMF) homogeneously under ultrasonication for 1 h. At the same time, Ce(NO₃)₃·6H₂O and H₃[P(Mo₃O₁₀)₄]·xH₂O with required ratio were dissolved in 50 mL DMF solution and stirred for 1 h. Then the above mixture was added drop by drop to the above carbon solution with magnetic stirring, then 3 mL distilled water was added dropwise. After that, the mixture was treated under ultrasonication for another 1 h and then aged at room temperature for 24 h. The next step was to collect the sample by centrifugation and wash it with distilled water and ethanol for five times until the supernatant matter was limpid, followed by dried at 80 °C in air. Finally, the obtained product was calcined in air at 400 °C for 10 h to produce hollow microspheres. The nominal molar ratio of Mo/Ce was 0, 0.1, 0.2, 0.3, 0.4 and 0.5 and the corresponding samples were denoted as CeO₂, CeMo(0.1), CeMo(0.2), CeMo(0.3), CeMo(0.4) and CeMo(0.5). As a comparison, pure MoO₃ hollow microspheres were also prepared under the same condition. What's more, the bulk CeMo(0.3) samples were prepared by conventional coprecipitation method using NH₄HCO₃ as precipitant.

Characterization

The obtained materials were measured by powder X-ray diffraction (XRD) using a Rigaku D/MAX 2500 v/PC instrument with a monochromatized Cu Kα radiation (λ = 1.5418 Å) operated at 40 kV and 40 mA. Field-emission scanning electron microscopy with an accelerating voltage of 3 kV (FESEM, S-4800) was carried out to observe morphologies of the samples. Transmission electron microscopy (TEM) images were examined on a JEOL JEM 2100F microscope with an accelerating voltage of 200 kV. The nitrogen adsorption–desorption isotherms were characterized using a Quantachrome Autosorb IQ instrument at −196 °C. The specific surface areas were calculated using the Brunauer–Emmett–Teller (BET) method and the pore size distributions of the samples were calculated by a cylindrical pore model (BJH method). By referencing the C 1s peak at 284.6 eV to correct the binding energy, the X-ray photoelectron spectroscopy (XPS) spectra was carried out using a surface analysis system (Thermal ESCALAB 250) with Al Kα radiation. Temperature programmed reduction by hydrogen (H₂-TPR) was examined on a Micromeritics Autochem 2920 II instrument, the sample was heated to 900 °C at a rate of 10 °C min⁻¹ and H₂ consumption was monitored by TCD. Temperature programmed desorption of ammonia (NH₃-TPD) was obtained on a Quantachrome Autosorb iQ-C-TCD-MS from 50 to 800 °C at a heating rate of 10 °C min⁻¹. The in situ DRIFTS were recorded by Thermo Nicolet iS50 FTIR spectrometer, the spectra was collected 64 scans at a spectral resolution of 4 cm⁻¹ in Kubelka–Munk format. Prior to each experiment, catalysts were pretreated at 300 °C in pure N₂ for 1 h and then cooled to 50 °C to perform the test.

Catalytic performance tests

The NH₃-SCR measurement was carried out in a fixed-bed quartz reactor and the inner diameter of the reactor is 10 mm. The typical gas mixture used here was made up of 500 ppm NO, 500 ppm NH₃, 5 vol% O₂, 10 vol% H₂O (while necessary), 100 or 200 ppm SO₂ (while necessary) and N₂ acted as balance gas, and the total flow rate of feed gas was 200 mL min⁻¹. During every SCR activity test, 500 mg 40–60 mesh sample was used with a gas hourly space velocity (GHSV) of 30 000 h⁻¹. The concentration variation of NO, NO₂, and SO₂ were detected by a KM-940 flue gas analyzer (Kane International Limited, UK) and N₂O and NH₃ were measured by Thermo Nicolet iS50 FTIR spectrometer. NO_x conversion ratio and N₂ selectivity were calculated in a steady state according to the following equations:

where [NO_x] = [NO] + [NO₂] and the [NO_x]_in and [NO_x]_out indicated the inlet and outlet concentration at steady-state, respectively.

Results and discussion

NH₃-SCR activity

The hollow microspheres with different Mo/Ce molar ratios (0, 0.1, 0.2, 0.3, 0.4 and 0.5) were used as NH₃-SCR catalyst to remove NO_x. As shown in Fig. 1a, pure CeO₂ hollow microspheres achieved a relatively low NO_x conversion ratio (less than 80%) and pure MoO₃ presented a much lower NO_x conversion (less than 70%). After doping molybdenum, the SCR activity was notably enhanced to almost 100% from 250 to 400 °C, indicating that the synergetic effects between Ce and Mo could lead to better SCR performance.²⁴ It is clear that 88% NO_x conversion can be achieved over CeMo(0.3) at 200 °C, much higher than the other materials (less than 53%). And from 250 to 400 °C CeMo(0.3) hollow microspheres exhibited almost 100% NO_x conversion ratio, much better than the commercial V₂O₅–WO₃/TiO₂ catalysts when below 300 °C and superior to bulk CeMo(0.3) at higher temperature (Fig. 1b). Besides, the N₂ selectivity of CeMo(0.3) hollow microspheres was more than 80% (Fig. S1, ESI†) and CeMo(0.3) hollow microspheres seemed to be more resistant to different GHSVs (Fig. S2, ESI†).

Fig. 1 (a) NO_x conversion ratio of CeMo(x) hollow microspheres. (b) NO_x conversion ratio of different kind of catalysts. Reaction conditions: [NO] = [NH₃] = 500 ppm, [O₂] = 5%, balanced in N₂, GHSV = 30 000 h⁻¹.

To further study the catalytic activity of CeMo(0.3) hollow microspheres, the stability, H₂O resistance and SO₂ tolerance were also investigated. As shown in Fig. 2a, the NO_x conversion ratio of CeMo(0.3) maintained steady at either 200 or 250 °C for 40 h. Fig. 2b showed that after the addition of 10 vol% H₂O, the NO_x conversion ratio merely reduced from 100% to 97.6%. And when H₂O was shut off, the NO_x conversion ratio recovered quickly. In addition, the influence of different SO₂ concentrations on CeMo(0.3) hollow microspheres was also discussed at 250 °C. Under 100 ppm SO₂ the NO_x conversion ratio decreased from 100% to 88.6%, while with 200 ppm SO₂ the NO_x conversion ratio reduced from 100% to 75.7% after 18 h. And after stopping SO₂, NO_x conversion ratio nearly regenerated to the previous level. Moreover, the co-effect of H₂O and SO₂ were also performed. The SCR activity was down to 84.5% one hour later and gradually decreased to 69.8% after 18 h when introducing 10 vol% H₂O and 100 ppm SO₂. When it came to 10 vol% H₂O and 200 ppm SO₂, the NO_x conversion ratio decreased to 65.1% after 1 h and a tiny rebound to 70.6% was found after 2 h, and then the conversion ratio was on a gentle downward slide to 58.6%. And for comparison, bulk CeMo(0.3) was also tested the stability, H₂O resistance and SO₂ tolerance at 250 °C. It is shown in Fig. S3† that bulk CeMo(0.3) sample showed a good stability but poor H₂O resistance and SO₂ tolerance. Moreover, CeMo(0.3) hollow microspheres possessed a better H₂O resistance and SO₂ tolerance than the commercial WO₃–V₂O₅/TiO₂ catalysts at 300 °C (Fig. S4, ESI†).

Fig. 2 (a) The study of thermal stability and (b) H₂O resistance and SO₂ tolerance at 250 °C on CeMo(0.3) hollow microspheres. Reaction conditions: [NO] = [NH₃] = 500 ppm, [O₂] = 5%, balanced in N₂, GHSV = 30 000 h⁻¹.

Structural property

The main mineral phases of the prepared materials were characterized by X-ray diffraction (XRD). As shown in Fig. 3a, the XRD pattern of pure CeO₂ hollow microspheres showed distinct diffraction peaks at 28.5, 33.1, 47.5, 56.3, 59.1, 69.4, 76.7 and 79.1°, which could be indexed as a cubic fluorite structure (JCPDS 81-0792) with face-centered structure of (111), (200), (220), (311), (222), (400), (331) and (420), typically.²⁵ However, no diffraction peaks related to oxidized molybdenum species were founded, indicating that Mo species were finely dispersed on the surface of cubic CeO₂ in the form of relatively small crystallite size or existed as amorphous species.¹⁵ And the peak intensity decreased gradually with increased Mo/Ce ratio from 0.1 to 0.5, indicating that the Mo species can inhibit the particle growth of CeO₂.²⁶ Calculated by the Scherrer equation from the XRD results, the average particle sizes of CeMo(x) hollow microspheres decreased from 7.2 to 4.5 nm (listed on Table 1) with increasing Mo content. It was shown in Fig. 3b that bulk CeMo(0.3) sample showed distinct diffraction peaks related to cubic CeO₂ with higher intensity than CeMo(0.3) hollow microspheres. By the way, pure MoO₃ sample presented much higher XRD intensity just as Fig. S5† showed, resulting a crystallite size of 28.2 nm.

Fig. 3 (a) XRD patterns of CeMo(x) hollow microspheres: CeO₂ (1), CeMo(0.1) (2), CeMo(0.2) (3), CeMo(0.3) (4), CeMo(0.4) (5) and CeMo(0.5) (6). (b) The comparison on XRD patterns of CeMo(0.3) hollow microspheres and bulk CeMo(0.3).

Table 1 BET specific surface area, pore volume, pore diameter and crystallite size of different materials^e

Materials	SBET^a (m^2 g^−1)	Pore volume^b (cm^3 g^−1)	Pore diameter^c (nm)	Crystallite size^d (nm)
a BET specific surface areas.b BJH desorption pore volume.c Average pore diameters.d CeO2 crystallite size calculated by the Scherrer equation from the XRD results.e “h”-represented “hollow microspheres”, “b”-represented “bulk”.
h-CeO2	58.1	0.19	14.8	7.2
h-CeMo(0.1)	41.7	0.13	12.8	6.2
h-CeMo(0.2)	24.3	0.10	17.4	5.9
h-CeMo(0.3)	20.2	0.11	26.8	5.2
h-CeMo(0.4)	21.8	0.12	26.2	4.9
h-CeMo(0.5)	23.5	0.11	20.8	4.5
h-MoO3	2.8	0.012	20.5	28.2
b-CeMo(0.3)	44.9	0.09	11.4	6.1

---

The morphology and microstructure of the CeMo(x) hollow microspheres were investigated by FESEM. From Fig. S3a,† the particle size of CeO₂ hollow microspheres was about 200–300 nm. While diameter of different CeMo(x) hollow microspheres are of 100–200 nm after doping Mo (Fig. S6b–4f, ESI†). As is well known, the hollow structure can provide more active sites to adsorb and activate reaction gases, which could benefit for the catalytic activity.²⁷ Obviously it is easy to verify its hollow structure from the broken part on SEM images. In addition, the CeMo(0.3) hollow microspheres was further verified by TEM images. As shown in Fig. 4b, the wall thickness of CeMo(0.3) was approximate 15 nm. On the HRTEM image, several lattice fringes of 0.27 nm can be observed clearly, which was in good agreement with the crystal plane (200) of CeO₂. And the lattice fringes of 0.21 nm were attributed to the crystal plane (141) of MoO₃, confirming the existence of Mo species. Meanwhile, it is easy to find that MoO₃ is coupled with CeO₂ from the energy-dispersive X-ray spectroscopy (EDS) elemental mapping of CeMo(0.3) hollow microspheres (Fig. 2d–f), indicating that MoO₃ is highly dispersed on the surface of CeMo(0.3) hollow microspheres.²⁸ Moreover, the energy-dispersive X-ray spectroscopy inside Fig. 2a confirmed the coexistence of Ce and Mo.

Fig. 4 Morphology and microstructure of CeMo(0.3) hollow microsphere by (a) FESEM, (b) TEM, (c) HRTEM, energy-dispersive X-ray spectroscopy (EDS) elemental mapping for (d) Ce and Mo, (e) Ce, and (f) Mo.

Moreover, N₂ adsorption–desorption performance of CeMo(x) hollow microspheres were carried out to observe its textural properties, and all the isotherms can be classified as type IV (Fig. 5).^29–31 The BET specific surface area, BJH desorption pore volume and average pore diameter of the prepared materials are tabulated in Table 1. Because of the MoO₃’s smaller specific surface area (2.8 m² g⁻¹), their specific surface area decreased from 58 to 20 m² g⁻¹ with the increase of Mo/Ce ratio.^15,32 In addition, the isotherm of bulk CeMo(0.3) sample was also classified to type IV as Fig. S8† depicted.

Fig. 5 N₂ adsorption–desorption isotherm of novel hollow CeMo(x) microspheres.

It is noted that the element valence state is very important during NH₃-SCR progress. To take a look at the metal oxidation states and the content in the surface layer, XPS measurements of hollow microspheres were performed and the results were shown in Fig. 6a and b. The sub-bands of Ce 3d peak labeled as u, u₂, u₃, v, v₂ and v₃ were ascribed to the 3d¹⁰4f⁰ state of Ce⁴⁺, and u₁ and v₁ represented the 3d¹⁰4f¹ initial electronic state corresponding to Ce³⁺. The ratios of Ce³⁺/(Ce³⁺ + Ce⁴⁺) and SCR activities (250 and 200 °C) of CeMo(x) hollow microspheres were shown in Table 2. As we can see, the Ce³⁺ ratio increased to 19.34% when the least Mo content was doped to pure CeO₂ and decreased gradually with increasing Mo loading. It is reported that Ce³⁺ species could make the surface chemisorbed oxygen increased by inducing a charge imbalance, the vacancies and unsaturated chemical bonds on the surface of catalysts.⁹ And Kwon et al. proved that Ce⁴⁺ could facilitate the adsorption of NH₃, and the adsorbed NH₃ can switch to NH₂, which can react with gaseous NO to generate N₂ and H₂O.^33,34 By compared with bulk CeMo(0.3) sample, the CeMo(0.3) hollow microspheres possessed more Ce³⁺ (see Fig. S9†).

Fig. 6 XPS spectra of the prepared CeMo(x) hollow microspheres, (a) Ce 3d and (b) O 1s [(1) CeO₂, (2) CeMo(0.1), (3) CeMo(0.2), (4) CeMo(0.3), (5) CeMo(0.4) and (6) CeMo(0.5)].

Table 2 XPS results of the prepared hollow microspheres with different Mo loading

Materials	Mo/Ce	O 1S (eV)		Oα ratio (%)	Ce^3+ ratio (%)	NOx conversion at 250 °C (%)	NOx conversion at 200 °C (%)
		Oα	Oβ
h-CeO2	0	531.18	529.39	40.40	12.08	44.2	27.9
h-CeMo(0.1)	0.28	531.17	529.98	42.42	19.34	88.9	41.4
h-CeMo(0.2)	0.44	531.12	529.91	33.51	17.75	96.1	53.4
h-CeMo(0.3)	0.56	531.50	530.19	27.41	16.72	99.0	88.4
h-CeMo(0.4)	0.63	531.96	530.27	22.47	14.27	93.4	47.9
h-CeMo(0.5)	0.65	532.35	530.43	19.17	12.88	93.9	36.2
h-MoO3	—	531.82	530.96	25.28	—	13.0	14.0
b-CeMo(0.3)	0.33	531.60	529.90	21.10	12.92	97.6	83.2

---

The O 1s XPS of the pure and Mo doped CeO₂ hollow samples were shown in Fig. 6b. The peak at 531.18 eV corresponds to chemisorbed oxygen (O_α) and the peak of pure CeO₂ at 529.439 eV sample can be assigned to lattice oxygen (O_β). After doping Mo element, both of the peaks shifted to lower binding energy side, and the O_α content decreased.³⁵ As we all know, active surface oxygen (O_α) is beneficial to the activation of adsorbed NH₃ molecules on catalyst surfaces.³⁶ And O_α also plays a significant role in oxidation reactions, which can not only oxidize NO into NO₂ to facilitate the “fast SCR” process, but excessive O_α can also consume NH₃ to produce nitrogen oxide byproducts to narrow the operating window, which was verified by NO or NH₃ oxidation reaction (Fig. S10†).³⁷ Thus it is very important to find the optimal doping ratio to generate favorable O_α concentration. According to the SCR activity, CeMo(0.3) hollow microspheres possess the suitable O_α ratio (27.41%). From the analysis of O 1s XPS of bulk CeMo(0.3) sample, there was much more O_α on CeMo(0.3) hollow sample (Fig. S12†).

The redox behaviors of catalysts were characterized by H₂-TPR (Fig. 7a). The H₂-TPR profile of pure CeO₂ exhibits two small broad peaks at 475 and 750 °C. As is known to all, the low temperature peak between 400 and 600 °C can be ascribed to the reduction of surface oxygen and high temperature peak between 600 and 800 °C is assigned to the reduction from Ce⁴⁺ to Ce³⁺.^38,39 When doping less Mo to CeO₂ (Mo/Ce < 0.3), there is a smaller shift to lower temperature for the reduction peak, and when the ratio increased to 0.3, the interaction between Ce and Mo species was so strong that there was only one reduction peak at 580 °C. With increasing Mo/Ce molar ratios from 0.3 to 0.5, the reduction peaks shifted to higher temperatures, which might be caused by the coverage of Mo species or the surface aggregation, which can be confirmed by Peng et al.^15,18 Transition metals (such as Mo) doped on CeO₂ could result in low reducibility, because of the decrease of labile surface oxygen and defect sites. It was shown in Fig. S14† that MoO₃ can be reduced at high temperature, from MoO₃ to MoO₂ at 770 °C and further reduced to Mo at 1000 °C, which was in accordance with the literature.²⁶ Although bulk CeMo(0.3) sample showed lower reduction temperature (Fig. S15†), it seems that the stronger interaction between Ce and Mo species came into being compared to CeMo(0.3) hollow microspheres.

Fig. 7 (a) H₂-TPR and (b) NH₃-TPD analysis of CeMo(x) hollow microspheres.

The surface acid strength and amount were measured by NH₃-TPD and the calculation results of the obtained profiles (Fig. S16†) are shown in Fig. 7b. The TPD profile of CeO₂ shows a major peak centered at 110 °C and a minor peak centered at 338 °C. As is well known that the coordinated NH₃ molecular bound to the Lewis acid sites is more thermally stable than the NH₄⁺ ions fixed on the Brønsted acid sites, it could be speculate that the lower peaks are due to NH₄⁺ ions originating from the Brønsted acid sites. In addition, the desorption peak area increased with acid amount.²⁹ Generally, with enhancing the acid sites, more NH₃ species could be absorbed on the surface of hollow microspheres, then their NH₃-SCR performance was promoted.⁴⁰ It is quite clear that at lowest Mo content, a remarkable decline of total acid sites was founded, and the total acid sites rised up from CeMo(0.1) to CeMo(0.3) and then decreased in CeMo(0.4) and CeMo(0.5). Among the hollow materials containing Mo, CeMo(0.3) hollow microspheres possess the most Brønsted acid sites, which is responsible for the best SCR activities below 250 °C. Furthermore, the more acid sites over the CeMo(0.3) hollow microspheres led to the better H₂O resistance ability, because it is easy absorb more NH₃ than H₂O. It has been reported that the adsorbed NH₃ is very important for not only Eley–Rideal but also Langmuir–Hinshelwood mechanisms in NH₃-SCR reaction.^41,42 Compared to bulk CeMo(0.3) sample as shown Fig. S17,† CeMo(0.3) hollow microspheres possessed more Brønsted acid sites, which was the main reason for increased deNOx activity.

In situ DRIFTS

The adsorption behavior of NH₃, NO and O₂ on the surface of different samples was investigated by in situ DRIFTS. Prior to each adsorption experiment, the sample was flushed with N₂ at 300 °C for 1 h to eliminate the external adsorbed materials. Furthermore, NH₃ or NO + O₂ would be inputted after cooling samples to 100 °C, then DRIFTS data will be recorded with increasing temperature.

The NH₃ or NO + O₂ adsorption performance of CeO₂ and CeMo(0.3) hollow microspheres were presented in Fig. 8. At the high wavenumber range of Fig. 8a and b, the band at 3685 cm⁻¹ on CeO₂ and the 3642 cm⁻¹ band on CeMo(0.3) were due to the adsorption of NH₄⁺ onto –OH. A small band at 3542 cm⁻¹ appeared on CeO₂, while on CeMo(0.3) a broad band from 3542 cm⁻¹ to 3178 cm⁻¹ came into being, which were ascribed to NH₃ bonded on Lewis acid.^33,42 Within the range of 1800–1000 cm⁻¹, more bands were obtained on CeO₂, the 1621, 1566, 1307, 1174 cm⁻¹ bands were attributed to coordinatively adsorbed NH₃ on Lewis acid sites,^2,26,39 the band at 1470 cm⁻¹ was assigned to NH₄⁺ bonded on Brønsted acid site and a small shoulder peak at 1528 cm⁻¹ was attributed to amide species (–NH₂) with Brønsted acid sites.⁴³ However, for CeMo(0.3) hollow microspheres, small peaks vanished and three main peaks can be ascribed to symmetric deformation of coordinated ammonia on Lewis acid sites (1605 and 1180 cm⁻¹) and NH₄⁺ peak at 1412 cm⁻¹ can be coordinated to Brønsted acid sites in the range of 1800–1000 cm⁻¹.^32,39,43 It was shown in Fig. S18† that CeMo(0.3) hollow microspheres gave rise to much more Brønsted acid sites than pure CeO₂ hollow microspheres for the peak at 1412 cm⁻¹ on CeMo(0.3) hollow microspheres was much higher than that at 1470 cm⁻¹ on pure CeO₂. With increasing temperature, the intensities of all of the bands decreased, nevertheless the bands on CeMo(0.3) were stronger than CeO₂ hollow microspheres, indicating that the doping Mo could enhance the adsorption of NH₃ on both Lewis acid sites and Brønsted acid sites, which is consistent with NH₃-TPD result. Moreover, the bands of NH₄⁺ ions (1412 and 3642 cm⁻¹) almost faded, while the coordinated NH₃ (1605 cm⁻¹) still can be observed at 300 °C. Therefore, the coordinated NH₃ on Lewis acidic sites are much more stable than NH₄⁺ species on the Brønsted acidic sites at high temperatures.⁴⁴

Fig. 8 In situ DRIFTS of NH₃ adsorption on (a) CeO₂ and (b) CeMo(0.3) hollow microspheres; NO + O₂ adsorption on (c) CeO₂ and (d) CeMo(0.3) hollow microspheres.

Similar with NH₃ adsorption, more peaks were detected on CeO₂. And the peak at 1611 cm⁻¹ was ascribed to asymmetric frequency of gaseous NO₂ molecules, while the corresponding peak of CeMo(0.3) hollow spheres shifted to 1607 cm⁻¹. For CeO₂ hollow microspheres, three bands (1543, 1526, 1305 cm⁻¹) can be attributed to monodentate nitrate, the bands of 1543 and 1526 cm⁻¹ strengthened with an increase in temperature while the 1305 cm⁻¹ band became weaker and overlapped with the 1255 cm⁻¹ band (corresponding to NO_x species).^38,42,44 And the bands at 1472 and 1222 cm⁻¹ on CeO₂ sample were assigned to the NO₂⁻ and nitrate species.⁴¹ For CeMo(0.3) hollow microspheres, the peak at 1888 cm⁻¹ was assigned to weakly adsorbed mononitrosyls (–NO) on the surface of the sample and a shoulder peak (1509 cm⁻¹) was assigned to nitrates. Two close peaks at 1182 and 1178 cm⁻¹ on CeO₂ and CeMo(0.3) hollow microspheres respectively were attributed to NO⁻ ad-species.⁴⁴ From comparison of the intensity of all peaks on CeO₂ and CeMo(0.3) hollow microspheres, it is indicated that much more gaseous NO₂ molecules can be adsorbed on the surface of CeMo(0.3) hollow microspheres, which was beneficial to “fast SCR” reaction to enhance the low-temperature SCR reaction.

Reaction mechanism analysis

The reactivity of surface-adsorbed NH₃ and NO_x species were characterized by in situ DRIFTS to investigate the possible reaction pathway at different temperatures (Fig. 9). After being exposed to NH₃ for 30 min, CeMo(0.3) hollow microspheres was mainly covered by coordinated NH₃ (1605, 1180 cm⁻¹) and NH₄⁺ (1412 cm⁻¹). After NO + O₂ was introduced, the bands ascribed to ionic NH₄⁺ vanished quickly and the bands of coordinated NH₃ decreased gradually at 150 °C (Fig. 9a). However, all bands of ionic NH₄⁺ and coordinated NH₃ disappeared after 2 min at 350 °C (Fig. 5b). As time goes on, bands of gaseous NO₂ molecules (1607 cm⁻¹) and nitrates (1509 cm⁻¹) appeared gradually and the intensity of NO_x bands enhanced. The results indicated that the NH₄⁺ ions on Brønsted acidic sites can react with NO_x species at a lower temperature, and at high temperature both NH₄⁺ ions on Brønsted acidic sites and coordinated NH₃ on Lewis acidic sites can participate in the SCR reaction.

Fig. 9 In situ DRIFTS of CeMo(0.3) hollow microspheres under 500 ppm NO at different times after pre-adsorption 500 ppm NH₃ at (a) 150 °C and at (b) 350 °C, and CeMo(0.3) hollow microspheres under 500 ppm NH₃ at different times after pre-adsorption 500 ppm NO at (c) 150 °C and (d) 350 °C.

After CeMo(0.3) hollow microspheres was pretreated by NO + O₂ for 30 min followed by N₂ purging for 20 min, several bands related to NO_x species (1607, 1509, 1267, and 1030 cm⁻¹) was detected. Then the gas was changed to NH₃. At 150 °C (Fig. 9c), gaseous NO₂ molecules (1607 cm⁻¹) on the surface of CeMo(0.3) were reduced by NH₃ gradually, while NO₂⁻ (1267 cm⁻¹) and nitrate (1509 cm⁻¹) species still existed after 30 min. Nevertheless, when the temperature increased to 350 °C all the bands assigned to NO_x species decreased sharply (Fig. 9d). The above results indicated that gaseous NO₂ molecules can react with NH₃ at low temperature and all the NO_x species (including gaseous NO₂ molecules, nitrite and nitrate) can be reduced by NH₃ at high temperature.

In summary, the proposed reaction mechanism of NH₃-SCR reaction to remove NO_x was exhibited in Fig. 10. The NH₃ adsorbed on the surface of CeMo(0.3) hollow microspheres to form NH₄⁺ ions bonded on Brønsted acidic sites and coordinated NH₃ related to Lewis acidic sites, and NO_x adsorbed mainly in the form of gaseous NO₂ molecules, nitrite and nitrate. At a lower temperature, it is mainly the NH₄⁺ ions on Brønsted acidic sites react with gaseous NO₂ molecules (i.e. the so-called “fast SCR” reaction).⁴⁵ Nevertheless, all the gaseous NO₂, nitrite and nitrate can participate in the reaction with both ionic NH₄⁺ and coordinated NH₃ to generate N₂ at high temperature.^46,47 Furthermore, the reactions between NH₃ and preadsorbed NO + O₂ species over CeMo(0.3) hollow microsphere was completed almost in 2 minutes, which was much shorter than the reactions between NO + O₂ species and preadsorbed NH₃. Therefore, the adsorbed behaviour of NH₃ played an important role in the SCR reactions.

Fig. 10 The proposed mechanism of NH₃-SCR on CeMo(0.3) hollow microspheres.

Conclusions

In this study, a series of novel Mo doped CeO₂ hollow microspheres were prepared successfully to be utilized for low-temperature NH₃-SCR reaction. Systematical characterizations, including XRD, TEM, BET, XPS, TPR, TPD and in situ DRIFTS were performed on the prepared samples, aiming at a better understanding of the SCR behavior and reaction mechanism. The results indicated that CeMo(0.3) hollow microspheres possessed a higher Brønsted acid sites and favourable content of Ce³⁺ and chemisorbed oxygen species. Moreover, there was strong interaction between Ce and Mo species. In summary, CeMo(0.3) hollow microspheres exhibits a much better NH₃-SCR activity, stability and H₂O resistance at low-temperature region. Moreover, the reactivity of the different kind of adsorbed NH₃ and NO_x species was also studied. It is indicated that the better low-temperature NH₃-SCR performance was due to the reaction pathway of the more active NH₄⁺ ions and gaseous NO₂ molecules.

Acknowledgements

This study was financially supported by Natural Science Foundation of China (21377061 and 81270041), Independent Innovation fund of Tianjin University (2015XRG0020), Key Laboratory of Colloid and Interface Chemistry (Shandong University, Ministry of Education) (201401), Key Technologies R&D Program of Tianjin (13ZCZDSF00300), and by Natural Science Foundation of Tianjin (Grant No. 15JCYBJC48400).

Notes and references

1. E. Park, M. Kim, H. Jung, S. Chin and J. Jurng, ACS Catal., 2013, 3, 1518–1525.
2. Y. Liu, T. T. Gu, X. L. Weng, Y. Wang, Z. B. Wu and H. Q. Wang, J. Phys. Chem. C, 2012, 116, 16582–16592.
3. N. Z. Yang, R. T. Guo, Q. S. Wang, W. G. Pan, Q. L. Chen, C. Z. Lu and S. X. Wang, RSC Adv., 2016, 6, 11226–11232.
4. J. H. Li, H. Z. Chang, L. Ma, J. M. Hao and R. T. Yang, Catal. Today, 2011, 175, 147–156.
5. Y. L. Fu, Y. F. Zhang, G. Q. Li, J. Zhang and F. Tian, RSC Adv., 2016, 6, 8539–8548.
6. Y. Peng, J. H. Li, L. Chen, J. H. Chen, J. Han, H. Zhang and W. Han, Environ. Sci. Technol., 2012, 46, 2864–2869.
7. M. A. Zamudio, N. Russo and D. Fino, Ind. Eng. Chem. Res., 2011, 50, 6668–6672.
8. Z. B. Wu, B. Q. Jiang, Y. Liu, H. Q. Wang and R. B. Jin, Environ. Sci. Technol., 2007, 41, 5812–5817.
9. C. X. Liu, L. Chen, J. H. Li, L. Ma, H. Arandiyan, Y. Du, J. Y. Xu and J. M. Hao, Environ. Sci. Technol., 2012, 46, 6182–6189.
10. W. C. Zhan, X. Y. Zhang, Y. L. Guo, L. Wang, Y. Guo and G. Z. Lu, J. Rare Earths, 2014, 32, 146–152.
11. X. L. Li, Y. H. Li, S. S. Deng and T. A. Rong, Catal. Commun., 2013, 40, 47–50.
12. X. Gao, Y. Jiang, Y. Zhong, Z. Y. Luo and K. F. Cen, J. Hazard. Mater., 2010, 174, 734–739.
13. R. Y. Qu, X. Gao, K. F. Cen and J. H. Li, Appl. Catal., B, 2013, 142–143, 290–297.
14. G. S. Qi and R. T. Yang, J. Catal., 2003, 217, 434–441.
15. S. P. Ding, F. D. Liu, X. Y. Shi, K. Liu, Z. H. Lian, L. J. Xie and H. He, ACS Appl. Mater. Interfaces, 2015, 7, 9497–9506.
16. H. L. Koh and H. K. Park, J. Ind. Eng. Chem., 2013, 19, 73–79.
17. Z. M. Liu, S. X. Zhang, J. H. Li and L. L. Ma, Appl. Catal., B, 2014, 144, 90–95.
18. Y. Peng, R. Y. Qu, X. Y. Zhang and J. H. Li, Chem. Commun., 2013, 49, 6215–6217.
19. D. S. Zhang, L. Zhang, C. Fang, R. H. Gao, Y. L. Qian, L. Y. Shi and J. P. Zhang, RSC Adv., 2013, 3, 8811–8819.
20. J. F. Li, J. Z. Wang, X. Liang, Z. J. Zhang, H. K. Liu, Y. T. Qian and S. L. Xiong, ACS Appl. Mater. Interfaces, 2014, 6, 24–30.
21. X. D. Ma, X. Feng, J. Guo, H. Q. Cao, X. Y. Suo, H. W. Sun and M. H. Zheng, Appl. Catal., B, 2014, 147, 666–676.
22. S. C. Ma, L. Q. Sun, L. N. Cong, X. G. Gao, Y. Cao, X. Guo, L. H. Tai, P. Mei, Y. P. Zeng, H. M. Xie and R. S. Wang, J. Phys. Chem. C, 2013, 117, 25890–25897.
23. L. Zhang, L. Y. Shi, L. Huang, J. P. Zhang, R. H. Gao and D. S. Zhang, ACS Catal., 2014, 4, 1753–1763.
24. Y. Y. Wang, B. X. Shen, C. He, S. J. Yue and F. M. Wang, Environ. Sci. Technol., 2015, 49, 9355–9363.
25. P. Maitarad, J. Han, D. S. Zhang, L. Y. Shi, S. Namuangruk and T. Rungrotmongkol, J. Phys. Chem. C, 2014, 118, 9612–9620.
26. H. Z. Chang, M. T. Jong, C. Z. Wang, R. Y. Qu, Y. Du, J. H. Li and J. M. Hao, Environ. Sci. Technol., 2013, 47, 11692–11699.
27. C. M. A. Parlett, K. Wilson and A. F. Lee, Chem. Soc. Rev., 2013, 42, 3876–3893.
28. L. Zhang, D. S. Zhang, J. P. Zhang, S. X. Cai, C. Fang, L. Huang, H. R. Li, R. H. Gao and L. Y. Shi, Nanoscale, 2013, 5, 9821–9829.
29. D. S. Zhang, L. Zhang, L. Y. Shi, C. Fang, H. R. Li, R. H. Gao, L. Huang and J. P. Zhang, Nanoscale, 2013, 5, 1127–1136.
30. S. H. Zhan, M. Y. Qiu, S. S. Yang, D. D. Zhu, H. B. Yu and Y. Li, J. Mater. Chem. A, 2015, 2, 20486–20493.
31. P. F. Ji, J. L. Zhang, F. Chen and M. Anpo, J. Phys. Chem. C, 2008, 112, 17809–17813.
32. L. Lietti, I. Nova, G. Ramis, L. D. Acqua, G. Busca, E. Giamello, P. Forzatti and F. Bregani, J. Catal., 1999, 187, 419–435.
33. D. W. Kwon, K. B. Nam and S. C. Hong, Appl. Catal., B, 2015, 166–167, 37–44.
34. 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, 136–137, 19–28.
35. X. N. Lu, C. Y. Song, S. H. Jia, Z. S. Tong, X. L. Tang and Y. X. Teng, Chem. Eng. J., 2015, 260, 776–784.
36. Y. Peng, W. W. Yu, W. K. Su, X. Huang and J. H. Li, Catal. Today, 2015, 242, 300–307.
37. H. Hu, S. X. Cai, H. R. Li, L. Huang, L. Y. Shi and D. S. Zhang, ACS Catal., 2015, 5, 6069–6077.
38. Z. H. Lian, F. D. Liu and H. He, Ind. Eng. Chem. Res., 2014, 53, 19506–19511.
39. Z. M. Liu, S. X. Zhang, J. H. Li, J. Z. Zhu and L. L. Ma, Appl. Catal., B, 2014, 158–159, 11–19.
40. S. X. Cai, D. S. Zhang, L. Y. Shi, J. Xu, L. Zhang, L. Huang, H. R. Li and J. P. Zhang, Nanoscale, 2014, 6, 7346–7353.
41. L. Chen, Z. C. Si, X. D. Wu and D. Weng, ACS Appl. Mater. Interfaces, 2014, 6, 8134–8145.
42. J. L. Zuo, Z. H. Chen, F. R. Wang, Y. H. Yu, L. F. Wang and X. H. Li, Ind. Eng. Chem. Res., 2014, 53, 2647–2655.
43. L. Zhang, W. X. Zou, K. L. Ma, Y. Cao, Y. Xiong, S. G. Wu, C. J. Tang, F. Gao and L. Dong, J. Phys. Chem. C, 2015, 119, 1155–1163.
44. L. Zhang, J. Pierce, V. L. Leung, D. Wang and W. S. Epling, J. Phys. Chem. C, 2013, 117, 8282–8289.
45. S. J. Yang, S. C. Xiong, Y. Liao, X. Xiao, F. H. Qi, Y. Peng, Y. W. Fu, W. P. Shan and J. H. Li, Environ. Sci. Technol., 2014, 48, 10354–10362.
46. M. E. Yu, C. T. Li, G. M. Zeng, Y. Zhou, X. N. Zhang and Y. E. Xie, Appl. Surf. Sci., 2015, 342, 174–182.
47. T. Zhang, R. Y. Qu, W. K. Su and J. H. Li, Appl. Catal., B, 2015, 176–177, 338–346.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra10720f

---