Catalytic performance and hydrothermal durability of CeO₂–V₂O₅–ZrO₂/WO₃–TiO₂ based NH₃-SCR catalysts

Abstract

Ceria modified V₂O₅–ZrO₂/WO₃–TiO₂ catalysts with different Ce loading (0, 5, 10 wt%) have been evaluated in the NH₃-SCR process before and after hydrothermal aging (750 °C, 10 wt% H₂O/air for 12 h). Compared with only Zr containing catalysts, addition of Ce greatly enhances the low temperature activity of the fresh catalyst, but it deactivates obviously after aging. The catalysts were characterized by XRD, XPS, H₂-TPR and DRIFTS. The results suggest that not only the enrichment of Ce³⁺ and the increased redox properties, but also the more active adsorbed nitrates on CeO₂ modified catalysts benefit the SCR reaction. The catalyst deactivation after aging is mainly due to the sintering and segregation of CeO₂ on the catalysts surface, suggesting the poor hydrothermal stability of the Ce component. However, additional provided NO₂ will compensate for the activity loss due to hydrothermal aging and significantly improve the low temperature SCR activity, suggesting a high NO₂ sensitivity of the Ce component. Moreover, it was found that a ceria, zirconia containing catalyst exhibits superior SCR activities, with both the fresh and aged 1%V₂O₅–10%CeO₂–10%ZrO₂/WO₃–TiO₂ catalysts showing high NO_x conversions (>95%) and selectivity to N₂ (>98%) in a wide temperature range of 200–500 °C, at a space velocity of 120 000 h⁻¹ in a simulated exhaust containing 500 ppm NO_x (NO₂/NO = 1) and 5% H₂O.

---

1. Introduction

The selective catalytic reduction (SCR) of NO_x with NH₃ or urea has been proven to be the most chemically efficient and cost-effective technology to reduce NO_x emission from stationary and lean burn mobile sources.^1,2 One commercial catalyst technology is based on the V₂O₅–WO₃–TiO₂ system. It is highly efficient in the temperature range of 300–450 °C and has been widely used for industrial applications.³ However, the main drawbacks of these catalysts for mobile application are the insufficient low temperature activity, the poor thermal durability and the potential toxicity that might arise from vanadium during operations.⁴ In view of these defects, extensive research efforts have been made to modify V₂O₅ based catalysts with various components, such as SiO₂, SO₄²⁻, BaO, MnO₂ and rare earths.^5–13 Recently, Chapman D. M. has found that the loss of vanadia can be significantly reduced by reducing the loss of the surface area of the anatase-TiO₂ support during exposure and by minimizing the surface vanadia loading.⁴ In this aspect, zirconia modified TiO₂ based catalysts are found to exhibit superior thermal stability as NH₃-SCR catalysts.^14,15 This is due to the stabilized anatase phase of the TiO₂ support by the formation of Zr_xTi_1−xO₂ solid solutions.¹⁵ However, the insufficient low temperatures (<300 °C) activity still exists on these catalysts.

Due to the high oxygen storage capacity and excellent redox properties, ceria has been attracting much attention in the NH₃-SCR process.^13,16–20 Qi and Yang reported a MnO_x–CeO₂ catalyst that was highly active for the SCR of NO by NH₃ at low temperature.¹⁶ Xu et al.¹⁷ developed the CeO₂/TiO₂ catalyst, which has high activity in the temperature range of 275–400 °C. Chen et al.¹³ also found that the doping of ceria could greatly enhance the low temperature NH₃-SCR performance of V₂O₅ based catalysts for diesel application. However, limited information is available in the literature regarding the effect of hydrothermal aging on the reactivity of CeO₂ modified catalysts. Thermal durability of SCR catalysts has been much emphasized in vehicle applications,^4,11 because the exhaust temperature may reach above 700 °C during the regeneration of the diesel particulate filter (DPF) process.

In this work, we report the effect of harsh hydrothermal aging treatment (750 °C in 10 wt% H₂O/air for 12 h) on the activity of ceria modified 1 wt%V₂O₅–10 wt%ZrO₂/WO₃–TiO₂ catalyst. The performance of NH₃-SCR and NH₃ oxidation for both the fresh and aged catalysts was evaluated. The effect of H₂O and NO₂ on the activity of catalysts was also studied to simulate the practical conditions. Moreover, the structural effect of ceria modification on NH₃-SCR activity was investigated. DRIFT studies were also performed to investigate the effect of ceria on adsorbed species under reaction conditions. The aim is to probe how ceria would influence the NH₃-SCR performance of catalysts before and after aging.

2. Experimental

2.1. Catalysts preparation

The CeO₂ (0, 5, 10 wt%) modified V₂O₅–ZrO₂/WO₃–TiO₂ catalysts with 1 wt% V₂O₅, 10 wt% ZrO₂ loading were prepared by an impregnation method. The catalyst support, 92%TiO₂–8%WO₃ (WT) with 0.8 wt% of sulfate (from TG analysis), was commercially obtained from Millennium Inorganic Chemicals Inc. The complex of VO(CO₂)₂ was prepared by reacting V₂O₅ powder with oxalic acid (1 M) with continuous stirring at 70 °C for 30 min. Calculated amounts of zirconia acetate and cerium nitrate were then added into the solutions. Subsequently, the desired quantity of WO₃–TiO₂ powder was added into the mixed solutions and stirred for 1 h. The as-synthesized material was dried overnight at 110 °C and then calcined at 500 °C for 3 h to obtain the fresh catalyst. A further hydrothermal aging at 750 °C in 10 wt% H₂O/air for 12 h was carried out to obtain the aged catalysts. The fresh catalysts are denoted as V1CexZWT, where x denotes weight percentage of the ceria with respect to the TiO₂ support, and the reference catalyst is denoted as V1ZWT. The aged samples are denoted as V1CexZWT-A and V1ZWT-A, respectively.

2.2. Catalysts characterization

The powder X-ray diffraction (XRD) experiments were performed on an X’Pert Pro diffractometer employing Cu Kα radiation (λ = 0.15418 nm). The X-ray tube was operated at 40 kV and 40 mA. The X-ray powder diffractogram was recorded at a 0.02° interval in the range of 20° < 2θ < 80°. The identification of the phases was made with the help of JCPDS cards (Joint Committee on Powder Diffraction Standards). The mean crystallite sizes of titania were calculated using the Scherer equation.

The surface areas of the samples were measured by N₂ physisorption at −196 °C using the Brunauer–Emmett–Teller (BET) method with F-Sorb 3400 volumetric adsorption/desorption apparatus. The samples were degassed in flowing N₂ at 150 °C for 3 h before measurement.

The X-ray photoelectron spectroscopy (XPS) experiments were carried out on a PHI-1600 ESCA system. The binding energy was calibrated internally by the carbon deposit C1s binding energy (BE) at 284.8 eV.

H₂-TPR experiments were conducted using 100 mg of samples. Samples were pretreated at 500 °C for 1 h in O₂ (30 mL min⁻¹) flow and then cooled down to room temperature in N₂ flow. The samples were heated in 5% H₂/He flow (30 mL min⁻¹) from 50 to 800 °C by 10 °C min⁻¹ while H₂ consumption was recorded continuously.

In situ DRIFT spectra of adsorption species were collected on a Nicolet 6700 FTIR spectrometer equipped with a commercial high temperature chamber (Thermofisher) fitted with a ZnSe window. Prior to reactant gas (NH₃ or NO) chemisorption, the sample was heated up to 500 °C in a 20% O₂/N₂ flow (50 mL min⁻¹) in the chamber. After the pre-oxidation treatment at 500 °C for 30 min, the sample was cooled down to the target temperature. The sample was subsequently flushed by 50 mL min⁻¹ N₂ for 30 min to remove any physisorbed molecules for background collection. Then, the gas containing 3000 ppm NH₃ in N₂ (50 mL min⁻¹) passed through the sample at the target temperature for 30 min. The FTIR spectra were collected after purging the weakly adsorbed gas molecules by flowing N₂ for 15 min.

In situ FTIR spectra for NO, NH₃ and O₂ co-adsorption were collected at the similar conditions as the NH₃ chemisorptions, where 3000 ppm NO and 3000 ppm NH₃ in 4% O₂ + N₂ were introduced to the system.

2.3. Activity measurement

The catalytic activity measurement for the reduction of NO by NH₃ (NH₃-SCR) was carried out in a fixed bed reactor made of stainless steel with 0.5 mL catalysts (diluted by silica). The reaction gas mixture consists of 500 ppm NO, 500 ppm NH₃, 5% O₂ and N₂ in balance. The NO conversion was measured from 150 °C to 550 °C at a ramp rate of 5 °C min⁻¹. The total flow of the gas mixture was 1 L min⁻¹ at a gas hourly space velocity (GHSV) of 120 000 h⁻¹. The concentrations of NO, NO₂, N₂O and NH₃ were measured by a Thermofisher IS10 FTIR spectrometer equipped with a 2 m path length gas cell.

The N₂ concentrations were determined based on nitrogen balance from eqn (1).

N_2out = (1/2) ([NO]_in − [NO]_out) + (1/2) ([NH₃]_in − [NH₃]_out) − (1/2) NO_2out − N₂O_out (1)

And the N₂ selectivity for the SCR reaction was calculated from eqn (2).

N₂ selectivity (%) = 2N_2out/(2N_2out + 2N₂O_out + NO_2out) (2)

NH₃ oxidation activities were measured using the same reactor, with 500 ppm NH₃ and 5% O₂ in N₂ as a feed, from 250 to 550 °C. All catalysts were kept at each temperature for 10 min to obtain a stable conversion. N₂ selectivity was calculated based on the following equation:

N₂ selectivity (%) = (([NH₃]_in − [NH₃]_out) − NO_out − NO_2out − 2N₂O_out)/([NH₃]_in − [NH₃]_out) (3)

3. Results

3.1. Catalytic performance

3.1.1. NH₃-SCR performance. Fig. 1a illustrates the NH₃-SCR performance of various catalysts as a function of temperature. V1ZWT exhibits poor activity at low temperatures, with its NO conversions exceeding 80% only above 300 °C. Increasing Ce loadings greatly enhances the catalytic efficiencies below 350 °C, with V1Ce10ZWT exhibiting the optimum NH₃-SCR performance in the temperature range 200–500 °C. This strongly suggests the role of CeO₂ as active sites in the SCR reaction, as reported in other literature.^13,17 The sharp deactivation of V1Ce10ZWT above 500 °C is mainly due to the ammonia oxidation at high temperatures.²¹ After severe hydrothermal aging at 750 °C for 12 h, V1CexZWT-A (x = 5, 10) deactivates obviously below 300 °C. The promotional effect of ceria on low temperature SCR activity has been offset by aging treatment.

Fig. 1 NH₃-SCR performance on V1CexZWT and V1CexZWT-A catalysts as a function of temperature. (a) NO conversion. (b) N₂O generation and N₂ selectivity. (c) NO₂ generation.

N₂O formation and N₂ selectivity during the NH₃-SCR reaction are shown in Fig. 1b. On the V1ZWT catalyst, undesired N₂O is formed above 500 °C, which is a main side product on vanadia based catalysts.¹ Ceria addition strongly inhibits N₂O formation on fresh catalysts, while a significant amount of N₂O is found on aged ceria modified catalysts, which contributes to the decrease in N₂ selectivity of the aged catalysts.

Fig. 1c shows NO₂ concentration during the NH₃-SCR process. A little NO₂ is observed on the V1ZWT sample at 550 °C, while ceria greatly enhances NO₂ formation at high temperatures. This is mainly due to the direct NO oxidation by ceria at high temperatures.¹⁹ However, NO₂ concentration decreases on ceria modified catalysts after aging treatment. The increase of the NO₂ product on a fresh catalyst and the decrease on an aged one indicates the decreased oxidation ability of ceria modified catalysts.

3.1.2. NH₃ oxidation. NH₃ oxidation reaction was conducted to assess the reactivity of NH₃ to different products. Fig. 2a shows NH₃ conversion of all the catalysts. Compared to the V1ZWT catalyst, V1CexZWT (x = 5, 10) have much higher NH₃ conversions, shifting the NH₃ conversion profiles to lower temperature by about 80 °C. This clearly demonstrates the role of ceria as the main active sites for ammonia oxidation. However, after hydrothermal aging, NH₃ oxidation activity is obviously decreased on ceria modified catalysts, with the NH₃ conversion profile shifting to higher temperature by about 50 °C on V1Ce10ZWT-A. This suggests the decreased oxidation ability of the aged ceria modified catalysts, which may account for the decrease in SCR activity at low temperatures.

Fig. 2 NH₃ oxidation on V1CexZWT and V1CexZWT-A catalysts as a function of temperature. (a) NH₃ conversion. (b) NO and NO₂ generation. (c) N₂O generation and N₂ selectivity.

Fig. 2b and c show the NO, NO₂, N₂O formation and N₂ selectivity as a function of temperature during the NH₃ oxidation process. A small amount of N₂O is observed on the V1ZWT catalyst. With the addition of ceria, a significant amount of NO is observed while N₂O is inhibited during NH₃ oxidation at high temperatures. This high amount of NO side product also explains the drastic decrease in NH₃-SCR activity on V1Ce10ZWT at 550 °C. Moreover, NO₂ increases with the sharp increase of the NO product on ceria modified catalysts, indicating the strong oxidation capacity of CeO₂. After aging, both NO and NO₂ products decrease obviously, while N₂O increases on the V1Ce10ZWT-A catalyst, which results in the N₂ selectivity decrease. The change of side products from NO, NO₂ to N₂O after aging suggests hydrothermal deactivation of the CeO₂ component on the catalyst.

3.1.3. Effect of H₂O and NO₂. To evaluate the catalyst performance under practical conditions, the effect of H₂O and NO₂ on SCR activity was studied. Fig. 3 illustrates the influence of H₂O on NO conversions of catalysts. When H₂O (5 wt%) is added into the reaction gas, the activity of the V1ZWT catalyst decreases obviously in the temperature range 150–550 °C. While for the V1Ce10ZWT catalyst, the addition of H₂O has only a little inhibition effect in the temperature range 150–250 °C. Moreover, NO conversion above 500 °C is clearly enhanced. This result indicates that ceria modified catalysts have better resistance against water vapor.

Fig. 3 Effect of H₂O on NH₃ SCR activity of V1ZWT and V1Ce10ZWT catalysts. Reaction conditions: 500 ppm NO, 500 ppm NH₃, 5% O₂, 5% H₂O, N₂ balance.

Fig. 4a shows the activity of V1ZWT and V1Ce10ZWT in a NO_x feed containing various NO₂ concentrations (NO_x = 500 ppm, NO₂ = 0, 150, 250 ppm) with 500 ppm NH₃, 5% O₂ and 5% H₂O. For the V1ZWT catalyst, the promotional effect of NO₂ is limited, with only 35% of NO conversions obtained at 200 °C. Moreover, increasing the NO₂ concentration has little effect on low temperature (<300 °C) SCR activity. Comparatively, a clear positive effect of NO₂ on V1Ce10ZWT is observed, with the NO_x conversion enhanced from 35% (NO alone) to 95% (NO₂, NO = 250 ppm) at 200 °C, the overall NO_x conversion maintains above 95% from 200 to 500 °C, and the N₂O does not exceed 2 ppm at 550 °C. The positive role of NO₂ was demonstratively due to the occurrence of fast SCR reactions (2NH₃ + NO + NO₂ → 2N₂ + 3H₂O).^22–24

Fig. 4 Effect of NO₂ on NH₃ SCR activity of V1ZWT and V1Ce10ZWT catalysts. (a) NO_x conversion. (b) NO₂ concentration. Reaction conditions: 500 ppm NO_x (NO₂ = 0, 150, 250 ppm), 500 ppm NH₃, 5% O₂, 5% H₂O, N₂ balance.

In order to explore the difference in NH₃-SCR activities between V1ZWT and V1Ce10ZWT at low temperatures, the outlet NO₂ concentrations of the catalysts below 300 °C are shown in Fig. 4b. A large amount of NO₂ is observed on the V1ZWT catalyst, and the level of NO₂ is almost unchanged (in a feed of 150 ppm NO₂) or even increased (in a feed of 250 ppm NO₂) from 150 to 250 °C, indicating that less NO₂ is participating in the fast SCR reactions. And the increase of NO₂ under the conditions of feeding 250 ppm of NO₂ could be due to NO_x desorption on the V1ZWT catalyst. These results strongly suggest that V1ZWT has rather low sensitivity to NO₂ feed, increasing the NO₂ concentration or increasing the temperature has little promotional effect on SCR activity. Comparatively, only a little NO₂ is observed at 150 °C on the V1Ce10ZWT catalyst, and the level of NO₂ is below 2 ppm above 200 °C, indicating that NO₂ is more reactive on ceria modified catalysts. The performance of the V1Ce10ZWT-A catalyst with different NO₂/NO_x ratios is shown in Fig. 5a. The reaction conditions are 500 ppm NO_x (NO₂ = 0, 150, 250 ppm), 500 ppm NH₃, 5% O₂ in the absence or presence of 5% H₂O. Relatively low activity is observed below 300 °C with only NO added in the absence or presence of 5% H₂O. However, the NO_x removal efficiency is strikingly improved when NO₂ is introduced, with 90% of NO_x conversion shifting to low temperature by about 150 °C. Moreover, the presence of both NO₂ and H₂O can inhibit N₂O production, with N₂ selectivity significantly improved (Fig. 5b). It is found that the aged V1Ce10ZWT catalyst can still exhibit superior NH₃-SCR activity and N₂ selectivity in the temperature range 200–550 °C with NO₂/NO = 1 in the presence of H₂O.

Fig. 5 Effect of H₂O and NO₂ on NH₃ SCR activity of V1Ce10ZWT-A catalysts. (a) NO_x conversion. (b) N₂O generation and N₂ selectivity. Reaction conditions: 500 ppm NO_x (NO₂ = 0, 150, 250 ppm), 500 ppm NH₃, 5% O₂, N₂ balance, in the absence or presence of 5% H₂O.

3.2. BET & XRD

The BET surface areas are summarized in Table 1. It is found that CeO₂ does not have much effect on the BET surface area of fresh samples but slightly decreases it on aged samples. The X-ray powder diffraction patterns are shown in Fig. 6, and the calculated crystallite sizes of catalysts are summarized in Table 1. On fresh catalysts, all XRD peaks are from the anatase phase of TiO₂ and no ceria or zirconia diffractions can be detected, suggesting that CeO₂ and ZrO₂ are present as amorphous materials. Upon hydrothermal aging, the peaks assigned to anatase-TiO₂ on all catalysts become more intense. While the average crystallite sizes of aged catalysts are at the same level, ranging from 28.9 nm to 30.6 nm. This clearly indicates that CeO₂ modification does not influence much the structure stability of the TiO₂ support. However, WO₃ segregation on the V1ZWT-A catalyst is strongly inhibited by CeO₂ addition, which can be due to the Ce–W interactions on the catalyst surface.¹³ Moreover, the peak attributed to tetragonal ZrO₂ observed on V1ZWT-A slightly decreases on Ce–Zr modified aged catalysts, with Ce_0.75Zr_0.25O₂ solid solutions forming on the surface.²⁵ The active Ce component would segregate after aging, while the Zr component inclines to incorporate into the CeO₂ matrix to form Ce–Zr solid solutions, which is beneficial for the thermal stability of CeO₂.

Fig. 6 XRD patterns of the catalysts.

Table 1 Textural and structural properties of the catalysts

Sample	BET surface area/m^2 g^−1	TiO2 crystalline size/nm
V1ZWT	82.9	16.8
V1Ce5WT	82.1	16.5
V1Ce10WT	78.5	16.4
V1ZWT-A	21.6	28.9
V1Ce5ZWT-A	17.2	30.6
V1Ce10ZWT-A	16.1	29.2

---

3.3. XPS

The XPS spectra of chemical states of Ce over V1Ce10ZWT and V1Ce10ZWT-A catalysts are shown in Fig. 7. The bands labeled u′ and v′ represent the 3d¹⁰4f¹ initial electronic state corresponding to Ce³⁺, while the peaks labeled u, u′′, u′′′, v, v′′ and v′′′ represent the 3d¹⁰4f⁰ state of Ce⁴⁺ ions.²⁶ For V1Ce10ZWT, v and u′′′ peaks for Ce⁴⁺ were much weaker than v′ and u′ peaks for the Ce³⁺ state. This result suggests that the Ce component mainly stays in the state of 3+ on the catalyst surface. According to the literature,^18,27,28 the presence of the Ce³⁺ species could create a charge imbalance, and more oxygen vacancies can be generated, which is beneficial for the activation of surface oxygen species in the SCR reaction. This is critical for the improved low temperature activity of ceria modified catalysts. However, after severe hydrothermal aging, the characteristic peaks assigned to Ce³⁺ and Ce⁴⁺ almost disappear. This is probably due to the segregation of CeO₂ on the catalyst surface, which has little chance to be accessed due to its large diameters above the detection limit of XPS. And the disappearance of Ce³⁺ and segregation of the active CeO₂ component (as shown in Fig. 6) could contribute to the deactivation of V1Ce10ZWT-A catalysts at low temperature.

Fig. 7 XPS spectra of Ce 3d on the catalysts.

3.4. H₂-TPR test

H₂-TPR is usually carried out to study the redox property of various catalysts. As shown in Fig. 8, all the catalysts show a broad peak for the temperature range of 400–800 °C. The reduction profile of a commercial 8%WO₃–92%TiO₂ support (reference) is also shown for comparison. It is found that the TiO₂ support exhibits a major peak at around 621 °C. As the reduction of WO_x and TiO_x is usually at high temperatures above 800 °C,^29,30 this peak could be due to reduction of some sulfate species that existed on the commercial TiO₂ support, as reported in other literature.^8,31 The presence of sulfate is also confirmed by our DRIFT results (not shown here). For the V1ZWT catalyst, a strong reduction peak centered at 557 °C is observed. As pure ZrO₂ does not have redox properties,³² this peak on the V1ZWT sample could be due to the reduction of surface V⁵⁺ to V³⁺ and sulfate species. When Ce is introduced into the catalysts, a sharp peak at 608 °C and a novel shoulder peak around 306 °C are found. According to the literature on ceria based catalysts,³³ this shoulder peak is attributed to the reduction of surface oxygen of nonstoichiometric ceria of type Ce³⁺–O–Ce⁴⁺.

Fig. 8 H₂-TPR profiles on the catalysts.

On the other hand, the strong peak at 608 °C lines between the reduction peaks of TiO₂ and V1ZWT catalysts, but with higher intensity. This is probably due to the interactions between CeO₂, V₂O₅ and sulfates on the TiO₂ surface. These results strongly suggest the increased redox properties on CeO₂ modified catalysts. However, after hydrothermal aging, the intensity of these peaks decrease, which indicates the decreased redox properties on the V1Ce10ZWT-A catalyst.

3.5. DRIFT

3.5.1. NH₃ adsorption. Fig. 9 shows the DRIFTS spectra of V1CexZWT (x = 0, 10) during NH₃ adsorption between 50 and 300 °C. After exposing V1ZWT to NH₃ adsorption, sharp bands appear at 1453 and 1680 cm⁻¹, corresponding to asymmetric and symmetric vibrations of NH₄⁺ ions on Brønsted acid sites. The bands at 1242 and 1600 cm⁻¹ with low intensities are assigned to symmetric and asymmetric vibrations of coordinated NH₃ on Lewis acid sites.³⁴ A negative band is observed at 1370 cm⁻¹, which is assigned to the S=O stretching mode of sulfate on the catalyst surface.³⁵ These sulfate species are found to be derived from the commercial TiO₂ support, confirmed by our FTIR results (not shown here), and contribute to Brønsted acid sites on the surface.³⁶ With an increase in temperature, the intensity of the 1453 cm⁻¹ band decreases noticeably at high temperature, while the 1242 cm⁻¹ band still remains and shifts to a higher wave band at 1255 cm⁻¹. This indicates that NH₃ bonded to Lewis acid sites is more stable on the catalyst surface. With CeO₂ addition, the peaks assigned to Brønsted acid sites and Lewis acid sites decrease obviously. Moreover, the band assigned to Brønsted acid sites decreases more rapidly with increasing temperature on V1Ce10ZWT. These results suggest that the addition of CeO₂ decreases both the amount and strength of Brønsted acid sites on the catalyst surface. Although many research studies have proved the critical role of surface acidity in the SCR reaction,^37,38 the promotional effect of CeO₂ on NH₃-SCR activity could not be explained by the acidity factor in this catalyst system.

Fig. 9 DRIFT spectra of V1ZWT and V1Ce10ZWT catalysts arising from NH₃ adsorption at various temperatures.

3.5.2. NO + O₂ adsorption. The DRIFT spectra of NO + O₂ on V1CexZWT (x = 0, 10) at different temperatures are shown in Fig. 10. After exposing V1ZWT to NO + O₂ adsorption at various temperatures, several bands at 1238, 1290, 1581, 1627 cm⁻¹ are observed. The bands at 1238 and 1627 cm⁻¹ can be assigned to bridged nitrate.³⁹ The band at 1290 cm⁻¹ is attributed to monodentate nitrate,⁴⁰ while the band at 1581 cm⁻¹ is ascribed to bidentate nitrate.⁴¹ According to the literature,^42,43 nitrate could only weakly adsorb on the V₂O₅/TiO₂ catalyst. The noticeable nitrate adsorption on V1ZWT could be due to the presence of zirconia, which could enhance the NO_x adsorption process.¹⁵ With the addition of CeO₂, the adsorption of nitrate species is greatly enhanced, with a new band appearing at 1554 cm⁻¹, which overlaps those of the monodentate nitrate (1530 cm⁻¹) and bidentate nitrate species (1580 cm⁻¹).³⁹ Increasing the temperature decreases the amount of all the adsorbed nitrates on the catalysts. However, the bidentate nitrate (1581 cm⁻¹) on ceria modified catalysts is more stable than on the V1ZWT catalyst. These results indicate that CeO₂ improves the amount and strength of adsorbed nitrate species on the catalyst surface.

Fig. 10 DRIFT spectra of V1ZWT and V1Ce10ZWT catalysts arising from NO + O₂ adsorption at various temperatures.

3.5.3. In situ DRIFTS of the reaction between NO + O₂ and adsorbed NH₃ species. Fig. 11a shows the in situ DRIFTS spectra of the reaction between NO + O₂ and pre-adsorbed NH₃ species on the V1ZWT catalyst at 200 °C. After NH₃ pre-adsorption and N₂ purge, the catalyst surface was mainly covered by ionic NH₄⁺ (1440 cm⁻¹) and coordinated NH₃ (1327 cm⁻¹ and 1255 cm⁻¹). When NO + O₂ was introduced, the intensity of the bands due to NH₃ species was unchanged with the appearance of the bands attributed to nitrate species (bridging nitrate at 1625 cm⁻¹ and bidentate nitrate at 1577 cm⁻¹) in the first 3 min. As the nitrate further accumulated on the catalyst surface, the bands assigned to the adsorbed NH₃ slightly decreased at 5 min, and totally disappeared after 10 min. This result suggests that the adsorbed nitrate species on the V1ZWT catalyst are less active in NH₃-SCR reactions, which is also in accordance with the results of Section 3.1.3. For the V1Ce10ZWT catalyst (Fig. 11b), similar bands due to NH₃ adsorption were observed. After the catalyst was purged with NO + O₂, the adsorbed NH₃ species decreased quickly accompanied by the appearance of nitrate species (1619 cm⁻¹ and 1581 cm⁻¹) at 2 min. And surface NH₃ species almost vanished at 3 min. This result indicates that more active nitrate species are formed and participate in the NH₃-SCR reaction on ceria modified catalysts.

Fig. 11 DRIFT spectra of catalysts pretreated by exposure to 3000 ppm NH₃, followed by exposure to 3000 ppm NO + 4% O₂ at 200 °C for various times. (a) V1ZWT, (b) V1Ce10ZWT.

3.5.4. DRIFT spectra in a flow of NO + NH₃ + O₂. To investigate the surface species under SCR reaction conditions, DRIFT spectra were collected in a flow of NO + NH₃ + O₂ at various temperatures. On the V1ZWT catalyst (Fig. 12a), the bands attributed to ionic NH₄⁺ (1436 cm⁻¹), coordinated NH₃ (1335 cm⁻¹ and 1268 cm⁻¹) and bidentate nitrate (1580 cm⁻¹) appeared at 150 °C. Increasing the temperature resulted in a decrease in the band intensity ascribed to NH₃ species. However, the band ascribed to bidentate nitrate vanished at 200 °C, with a new band forming at 1598 cm⁻¹, which was assigned to coordinated NH₃ species.³⁴ Although noticeable NO_x adsorption was observed on the V1ZWT catalyst, the surface nitrate could not form with NH₃ supplied successively in the simulating gas above 200 °C. This suggests the less importance of nitrate species during the SCR reaction on the V1ZWT sample. For the V1Ce10ZWT catalyst (Fig. 12b), similar bands attributed to adsorbed NH₃ (1436 cm⁻¹, 1327 cm⁻¹ and 1263 cm⁻¹) and bidentate nitrate species (1575 cm⁻¹) were observed at 150 °C. However, the band ascribed to nitrate species could still exist even when the temperature is increased from 150 to 250 °C. The competition of NO_x adsorption with adsorbed NH₃ suggests that the surface nitrate can play an important role in the SCR reaction on ceria modified catalysts, which is beneficial for the enhanced NH₃-SCR activity of the catalyst.

Fig. 12 DRIFT spectra of catalysts arising from NH₃, NO and O₂ co-adsorption at 150, 200 and 250 °C. (a) V1ZWT, (b) V1Ce10ZWT.

4. Discussion

CeO₂ is found to enhance the low temperature NH₃-SCR activity of the V₂O₅–ZrO₂/WO₃–TiO₂ catalyst in the fresh state. NH₃ oxidation results suggest that the enhanced activity is directly correlated with the increased oxidation property of the catalysts. According to XRD and XPS results, ceria is highly dispersed on the catalyst surface (Fig. 6), and mainly exists in the form of Ce³⁺ oxide (Fig. 7). The enrichment of Ce³⁺ could create a charge imbalance, and more oxygen vacancies are formed. This could facilitate the activation and transportation of active oxygen species, and enhance the NO oxidation to NO₂ process, which is beneficial for the SCR reaction.^18,44 Furthermore, the H₂-TPR result reveals a new active oxygen species (Ce³⁺–O–Ce⁴⁺) with high reducibility by ceria addition (Fig. 8), indicating the enhanced redox property of the catalysts, which also contributes to the increase in SCR activity at low temperatures.

In addition, surface acidity of the catalyst is an important factor for vanadia based SCR reactions,^37,38 although some different mechanisms were proposed based on Brønsted or Lewis acid sites.^43,45–47 Chen L. et al.¹³ proposed that the promotional effect of Ce on V_0.1W₆Ti was due to the fact that the Ce additive could provide stronger and more active Brønsted acid sites. However, based on our NH₃-DRIFT results shown in Fig. 9, both the Brønsted acid sites (associated with V⁵⁺–OH, W⁶⁺–OH, S–OH) and Lewis acid sites (associated with Ti⁴⁺, Zr⁴⁺) decrease obviously on CeO₂ modified catalysts.^1,15,35,48 It could be suggested that although the presence of ceria could provide additional Brønsted acid sites,¹³ the coverage effect of CeO₂ on these NH₃ adsorption sites could be the overwhelming factor for the decrease in acidity in Brønsted and Lewis acid sites. However, the SCR activity is clearly improved by CeO₂ addition (Fig. 1). This result indicates that the surface acidity (Brønsted acid sites) is not the determining factor for low temperature activity enhancement on our ceria modified catalysts, while it is the enhancement of redox properties by ceria. In our previous work we have confirmed that the amount of Brønsted acid sites could not predict the SCR reaction rate.¹⁵

On the other hand, based on our DRIFT results, noticeable NO_x adsorption on V1ZWT and ceria modified catalysts was observed (Fig. 10). The nitrate species adsorbed on zirconia containing catalysts is inactive in reacting with adsorbed NH₃ species (Fig. 11a), and even could not be present under simulated SCR reaction conditions above 200 °C (Fig. 12a). Nevertheless, the amount and strength of nitrate species were both increased by CeO₂ addition (Fig. 10). And the nitrate on ceria modified catalysts is more active with adsorbed NH₃ (Fig. 11b) and could still exist under SCR reaction conditions at 250 °C (Fig. 12b). Compared with only ZrO₂ containing catalysts, the adsorbed nitrate plays a more important role in NH₃-SCR reaction on ceria modified catalysts.

The NO₂/NO_x ratio test from Section 3.1.3 indicates higher NO₂ sensitivity of ceria modified catalysts. According to the literature,^22–24 a VWTi catalyst is sensitive to fast SCR reactions. And the surface nitrates play a key role in reoxidizing the V-related redox sites during the reaction.⁴⁹ However, in the ZrO₂ containing VWTi catalysts, fast SCR reaction is less effective and NO₂ can hardly participate in the reaction. ZrO₂ is found to be inactive for SCR reactions.^50,51 The addition of zirconia may decrease the active V=O sites, which inhibits fast SCR reactions. The inhibition effect of ZrO₂ on the activity of the VWTi catalyst has been discussed in our previous work.¹⁵ With the addition of CeO₂, additional redox sites are provided. Moreover, the adsorbed surface nitrates on CeO₂ modified catalysts are stronger and more active in SCR reactions. These are beneficial for the enhanced fast SCR reactivity of the ceria modified catalysts.

After hydrothermal aging treatment, a clear decrease in SCR activity was observed on ceria modified catalysts (Fig. 1). This is mainly due to the sintering and segregation of the Ce component (Fig. 6), resulting in the decreased redox property and almost disappearance of Ce³⁺, and the process of NO oxidation to NO₂ is inhibited (as indicated by NO₂ formation from Fig. 1 and Fig. 2). CeO₂ is found to have poor hydrothermal stability in SCR reactions. However, additional provided NO₂ drastically improves the low temperature SCR activity of the V1Ce10ZWT-A catalyst (Fig. 5), indicating the high NO₂ sensitivity of ceria modified catalysts. It can be found that hydrothermal aging deactivates the capability of ceria for the NO oxidation process, while additional introduced NO₂ will compensate for the activity loss. It is worth noting that both the fresh and aged ceria modified catalysts can exhibit superior NH₃-SCR activity and N₂ selectivity in the temperature range 200–550 °C with NO₂/NO = 1 in the presence of 5 wt% H₂O, which is a good candidate catalyst for diesel applications.

5. Conclusions

The CeO₂ and ZrO₂ co-modified catalysts prepared by an impregnation method were highly active for NH₃-SCR reactions in the temperature range 220–500 °C. The higher Ce loading leads to higher NO conversion. The promotional effect of CeO₂ is attributed to the increased redox properties and the enrichment of Ce³⁺ on the surface. Moreover, DRIFTS results indicate that the adsorbed nitrates can be more active on CeO₂ modified catalysts, which is beneficial for SCR reactions. However, CeO₂ modified catalysts deactivate obviously at low temperature after severe hydrothermal aging at 750 °C, 10 wt% H₂O/air for 12 h. This is mainly due to the sintering and segregation of CeO₂ on the catalyst surface, suggesting the poor hydrothermal stability of the Ce component. Additional provided NO₂ will compensate for the activity loss due to hydrothermal aging and significantly improve the low temperature SCR activity, suggesting a high NO₂ sensitivity of the ceria component. It is worth noting that both the fresh and aged 1%V₂O₅–10%CeO₂–10%ZrO₂/WO₃–TiO₂ catalysts exhibit high NO_x conversions (>95%) and superior selectivity to N₂ (>98%) during 200–500 °C, at a space velocity of 120 000 h⁻¹ under conditions of NO₂/NO = 1 and 5% H₂O, which suggest that it is a good catalyst candidate for diesel applications.

Acknowledgements

The authors are grateful to the financial support from the National High-Tech Research and Development Program of China (No. 2011AA03A405), the Program of the Natural Science Foundation of China (No. 50972104) and the Research Fund for the Doctoral Program of Higher Education of China (No. 20090032110020).

Notes and references

1. G. Busca, L. Lietti, G. Ramis and F. Berti, Appl. Catal., B, 1998, 18, 1–36.
2. P. Forzatti, Catal. Today, 2000, 62, 51–65.
3. P. Forzatti, Appl. Catal., A, 2001, 222, 221–236.
4. D. M. Chapman, Appl. Catal., A, 2011, 392, 143–150.
5. M. Kobayashi, R. Kuma, S. Masaki and N. Sugishima, Appl. Catal., B, 2005, 60, 173–179.
6. M. Kobayashi and M. Hagi, Appl. Catal., B, 2006, 63, 104–113.
7. S. T. Choo, Y. G. Lee, I. S. Nam, S. W. Ham and J. B. Lee, Appl. Catal., A, 2000, 200, 177–188.
8. L. Baraket, A. Ghorbel and P. Grange, Appl. Catal., B, 2007, 72, 37–43.
9. S. T. Choo, S. D. Yim, I. S. Nam, S. W. Ham and J. B. Lee, Appl. Catal., B, 2003, 44, 237–252.
10. J. H. Goo, M. F. Irfan, S. D. Kim and S. C. Hong, Chemosphere, 2007, 67, 718–723.
11. M. A. L. Vargas, M. Casanova, A. Trovarelli and G. Busca, Appl. Catal., B, 2007, 75, 303–311.
12. M. Casanova, E. Rocchini, A. Trovarelli, K. Schermanz and I. Begsteiger, J. Alloys Compd., 2006, 408–412, 1108–1112.
13. L. Chen, J. H. Li and M. F. Ge, J. Phys. Chem. C, 2009, 113, 21177–21184.
14. N. Marcotte, B. Coq, C. Savill-Jovitt, P. Bichon, R. Cavalier and R. Durand, Appl. Catal., B, 2011, 105, 373–376.
15. A. J. Shi, X. Q. Wang, T. Yu and M. Q. Shen, Appl. Catal., B, 2011, 106, 359–369.
16. G. Qi and R. T. Yang, J. Catal., 2003, 217, 434–441.
17. W. Q. Xu, Y. B. Yu, C. B. Zhang and H. He, Catal. Commun., 2008, 9, 1453–1457.
18. T. Gu, Y. Liu, X. L. Weng, H. Q. Wang and Z. B. Wu, Catal. Commun., 2010, 12, 310–313.
19. X. Gao, Y. Jiang, Y. Zhong, Z. Y. Luo and K. F. Cen, J. Hazard. Mater., 2010, 174, 734–739.
20. S. S. R. Putluru, A. Riisager and R. Fehrmann, Catal. Lett., 2009, 133, 370–375.
21. G. Ramis, L. Yi, G. Busca, M. Turco, E. Kotur and R. J. Willey, J. Catal., 1995, 157, 523–535.
22. M. Koebel, G. Madia and M. Elsener, Catal. Today, 2002, 73, 239–247.
23. C. Ciardellia, I. Nova, E. Tronconi, D. Chatterjee, T. Burkhardt and M. Weibel, Chem. Eng. Sci., 2007, 62, 5001–5006.
24. E. Tronconi, I. Nova, C. Ciardelli, D. Chatterjee and M. Weibel, J. Catal., 2007, 245, 1–10.
25. B. M. Reddy, P. Lakshmanan and A. Khan, J. Phys. Chem. B, 2004, 108, 16855–16863.
26. F. B. Noronha, E. C. Fendley, R. R. Soares, W. E. Alvarez and D. E. Resasco, Chem. Eng. J., 2001, 82, 21–31.
27. X. W. Liu, K. B. Zhou, L. Wang, B. Y. Wang and Y. D. Li, J. Am. Chem. Soc., 2009, 131, 3140–3141.
28. S. X. Yang, W. P. Zhu, Z. P. Jiang, Z. X. Chen and J. B. Wang, Appl. Surf. Sci., 2006, 252, 8499–8505.
29. M. A. Reiche, E. Ortelli and A. Baiker, Appl. Catal., B, 1999, 23, 187–203.
30. M. A. Reiche, M. Maciejewski and A. Baiker, Catal. Today, 2000, 56, 347–355.
31. S. M. Jung and P. Grange, Appl. Catal., B, 2001, 32, 123–131.
32. D. Pietrogiacomi, A. Magliano, P. Ciambelli, D. Sannino, M. C. Campa and V. Indovina, Appl. Catal., B, 2009, 89, 33–40.
33. X. Gao, Y. Jiang, Y. C. Fu, Y. Zhong, Z. Y. Luo and K. F. Cen, Catal. Commun., 2010, 11, 465–469.
34. R. Q. Long and R. T. Yang, J. Catal., 2000, 196, 73–85.
35. R. T. Yang, W. B. Li and N. Chen, Appl. Catal., A, 1998, 169, 215–225.
36. J. P. Chen and R. T. Yang, J. Catal., 1993, 139, 277–288.
37. J. P. Chen and R. T. Yang, Appl. Catal., A, 1992, 80, 135–148.
38. L. Lietti, I. Nova, G. Ramis, L. Dall’Acqua, G. Busca, E. Giamello, P. Forzatti and F. Bregani, J. Catal., 1999, 187, 419–435.
39. F. D. Liu and H. He, Catal. Today, 2010, 153, 70–76.
40. G. M. Underwood, T. M. Miller and V. H. Grassian, J. Phys. Chem. A, 1999, 103, 6184–6190.
41. A. Trovarelli, Catal. Rev., 1996, 38, 439–520.
42. L. Lietti, I. Nova, E. Tronconi and P. Forzatti, Catal. Today, 1998, 45, 85–92.
43. G. Ramis, G. Busca, F. Bregani and P. Forzatti, Appl. Catal., 1990, 64, 259–278.
44. L. Chen, J. H. Li, M. F. Ge and R. H. Zhu, Catal. Today, 2010, 153, 77–83.
45. G. Ramis, G. Busca, V. Lorenzelli and P. Forzatti, Appl. Catal., 1990, 64, 243–257.
46. N. Y. Topsøe, H. Topsøe and J. A. Dumesic, J. Catal., 1995, 151, 226–240.
47. N. Y. Topsøe, H. Topsøe and J. A. Dumesic, J. Catal., 1995, 151, 241–252.
48. X. Guo, C. Bartholomew, W. Hecker and L. L. Baxter, Appl. Catal., B, 2009, 92, 30–40.
49. E. Tronconi, I. Nova and M. Colombo, Ind. Eng. Chem. Res., 2010, 49, 10374–10385.
50. R. T. Yang, J. P. Chen, E. S. Kikkinides and L. S. Cheng, Ind. Eng. Chem. Res., 1992, 31, 1440–1445.
51. D. Pietrogiacomi, A. Magliano, P. Ciambelli, D. Sannino, M. C. Campa and V. Indovina, Appl. Catal., B, 2009, 89, 33–40.

---