Selective Catalytic Reduction of NO with NH 3 over Ce-W-Ti Oxide Catalysts Prepared by Solvent Combustion Method

: In this work, a series of Ce-W-Ti catalysts were synthesized using a solution combustion method for the selective catalytic reduction (SCR) of NO with NH 3 at low temperatures. The reaction performance of NH 3 -SCR of NO was signiﬁcantly improved over the Ce-W-Ti catalysts compared to Ce 0.4 Ti and W 0.4 Ti catalysts, while Ce 0.2 W 0.2 Ti showed the best activity among all the samples. The Ce 0.2 W 0.2 Ti catalyst exhibited over 90% removal of NO and 100% N 2 selectivity in the temperature range of 250–400 ◦ C at a gas hourly space velocity (GHSV) of 120,000 mL · g − 1 · h − 1 . The Ce-W-Ti catalysts were characterized using N 2 adsorption-desorption, X-ray diffraction, X-ray photoelectron spectrometry and temperature programmed desorption of NH 3 to establish the structure-activity relationships of the Ce-W-Ti catalysts. The excellent catalytic performance of the Ce 0.2 W 0.2 Ti catalyst could be associated with the larger speciﬁc surface area, highly dispersed Ce and W species, increased amount of surface adsorbed oxygen (O ads ) and enhanced total acidity on the catalyst surfaces.


Introduction
The emission of nitrogen oxides (NO x ) from stationary sources and lean-burn engines remains one of the major challenges in environmental pollution control due to the negative effect of NO x on air quality and human health. Selective catalytic reduction (SCR) of NO x with NH 3 (NH 3 -SCR) over TiO 2 supported V 2 O 5 -WO 3 or V 2 O 5 -MoO 3 catalysts have been extensively investigated due to the high activity of these catalysts in NO reduction [1][2][3][4]. However, the toxicity of active vanadium species and the narrow operation temperature window (300-420 • C) limit the use of these catalysts on a commercial scale to meet the stringent NO x emission legislations, especially when the boilers are operated at low loads and the temperatures are far below the operation temperature window of the vanadium-based catalysts [5,6].
In recent years, significant efforts have been devoted to the development of an environmentally-benign catalyst with a wide operation temperature window and high deNO x activity [7,8]. Various types of transition and rare earth metals based catalysts have shown an excellent reactivity in NH 3 -SCR reaction over a wide temperature range [9][10][11]. Among the non-vanadium NH 3 -SCR catalysts, cerium-based catalysts have been intensively investigated considering their unique properties of high oxygen-storage capacity and facilitated redox cycles between Ce 4+ and Ce 3+ , e.g., Ce-Ti, Ce-Zr, Ce-Cu-Ti and Ce-Mn, etc. [12][13][14][15][16][17].
Recently, tungsten, a typical promoter and stabilizer in vanadium-based catalysts, has been used as a dopant to improve the performance of Ce-based catalysts [18][19][20]. Chen et al. reported that the doping of W into CeO 2 /TiO 2 catalysts significantly enhanced the SCR of NH 3 in the temperature range of 150-300 • C, compared to the catalysts without W doping. The catalyst sample with 6 wt.% W exhibited the best performance [21]. Jiang et al. found Ce 20 W 10 Ti catalysts prepared by a single step sol-gel method exhibited a NO conversion of 90% at 275 • C at a gas hourly space velocity (GHSV) of 90,000 h −1 [22]. They also developed a Ce-W-Ti catalyst prepared by sol-gel method, which exhibited 95% NO conversion rate (250 • C) at the same GHSV [23]. Shan et al. reported that Ce 0.2 W 0.1 Ti catalysts synthesized by a facile homogeneous precipitation method showed 100% NO conversion and N 2 selectivity in a wide temperature range of 250 • C to 400 • C at a high gas hourly space velocity (GHSV) of 250,000 h −1 , which was significantly higher than that using Ce-Ti catalysts without W doping [24,25]. Currently, Ce-W-Ti catalysts were assumed to be one of the potential candidates for the commercial vanadium-based catalysts [26]. The performance of the Ce-W-Ti catalysts were closely related to the redox properties between the active metals.
To further improve the activity of the Ce-W-Ti catalysts in low temperature NH 3 -SCR, a novel solution combustion method has been used to prepare a series of Ce-W-Ti catalysts with different Ce/W ratios and tested over a wide temperature range. The effect of these catalysts on the conversion of NO and N 2 selectivity has been investigated. A range of catalyst characterization techniques including N 2 adsorption-desorption, X-ray diffraction (XRD), X-ray photoelectron spectrometry (XPS), temperature programmed desorption of NH 3 (NH 3 -TPD) has been used to better understand the properties of the Ce-W-Ti catalysts and their effects on the SCR of NO with NH 3 .

Catalyst Preparation
A series of Ce-W-Ti catalysts with different Ce/W ratios were prepared by a solution combustion method. Tetrabutyl titanate, cerium nitrates and ammonium metatungstate were used as the precursors together with ethanol, nitric acid, ethanedioic acid and glycine for the synthesis of the catalysts. All chemicals were reagent grade from Aladdin Co. Ltd. (Shanghai, China). Typically, tetrabutyl titanate (8.508 g) was firstly dissolved in a 60 mL water-ethanol (v/v = 1/1) solution and stirred for 1 h. Stoichiometric amounts of cerium nitrate and ammonium metatungstate were dissolved in deionized water under vigorous stirring. The molar ratio of metal ions to titanium species was kept at 0.4. After mixing the two solutions, glycine was added dropwisely as a fuel and complexant. Note, the molar ratio of glycine to nitrates was 1:1. The obtained mixture was heated in a water bath at 50 • C for 3 h, followed by evaporation at 300 • C in a muffle furnace. The solution was boiled and ignited during the evaporation. In this way, a homogeneous brown powder was obtained. The samples were pelleted and sieved to 40-60 meshes for the test of catalytic activity. In this work, the Ce-W-Ti catalysts were denoted as Ce x W 0.4−x Ti where x was the molar ratio of Ce species to Ti species in the catalysts.

Catalyst Characterizations
N 2 adsorption-desorption isotherms were obtained using a Quantachrome (Boynton Beach, FL, USA) Autosorb-1 apparatus at −196 • C. Prior to N 2 adsorption, the catalysts were degassed under vacuum at 300 • C for 4 h. The specific surface area of the catalysts was determined by the Brunauer-Emmet-Teller (BET) method. Total pore volume and average pore size of the catalysts were evaluated using the Barrett-Joyner-Halenda (BJH) method.
Powder X-ray diffraction (XRD) patterns of the catalysts were recorded using a Rigaku D/Max-2550 instrument (Tokyo, Japan) equipped with a Cu Kα radiation source in the 2Θ range of 10-80 • .
X-ray photoelectron spectroscopy spectra were measured with a Thermo ESCALAB 250 (Waltham, MA, USA) using Al Kα radiation source (hυ = 1486.6 eV). The spectra were calibrated with the C 1s binding energy (B. E.) value of 284.6 eV to eliminate the sample charging effects.
Temperature programmed desorption of ammonia (NH 3 -TPD) was performed using a Micrometrics AutoChem II 2920 chemisorption analyzer (Ottawa, ON, Canada). Before each test, the samples (100 mg) were purged in a He gas flow at 350 • C for 1 h and then cooled down to 50 • C. Then, the samples were saturated with 2 vol.% NH 3 /He gas stream until equilibrium was reached, followed by flushing with He at 100 • C to avoid NH 3 physisorption. Finally, the samples were heated from 100-500 • C at the heating rate of 10 • C·min −1 . The signal of NH 3 desorption was recorded using a gas chromatography equipped with a thermal conductivity detector (TCD).

Catalytic Activity Measurements
The catalytic activity test for all the catalysts was carried out in a fixed-bed reactor packed with 0.4 g catalyst (40-60 meshes). The feed gas contained 500 ppm NO, 500 ppm NH 3 , 5% O 2 and balanced N 2 . The total flow rate of the feed gas was fixed at 800 mL·min −1 , corresponding to a gas hourly space velocity (GHSV) of 120,000 mL·g −1 ·h −1 . The inlet and outlet gas compositions were monitored on-line by a Gasmet Dx-4000 Fourier transform infrared (FTIR) gas analyzer (Helsinki, Finland). The gas analyzer was calibrated with standard gas cylinders of NO, NO 2 , N 2 O, SO 2 and NH 3 before the experiments. All the measurements were recorded when the reaction reached a steady-state at each temperature. NO conversion and N 2 selectivity was defined as follows: where [NO] Figure 1 shows the influence of different Ce/W molar ratios of the Ce-W-Ti catalysts on the conversion of NO in the temperature range of 150-450 • C. Almost no catalytic activity was found for W 0.4 Ti at a temperature below 250 • C, while the conversion of NO increased quasi-linearly from 4.6% to 64.7% when increasing the temperature from 250 to 450 • C. As for the Ce 0.4 Ti catalyst, the NO conversion increased from 5.8% to 85.0% in the tested temperature range (150-450 • C). The addition of Ce into the W-Ti catalyst dramatically improved NO conversion and shifted the operation temperature of the catalysts to a lower temperature. A maximum NO conversion of 90.1% to 100% was obtained between 250 • C and 450 • C over the Ce 0.2 W 0.2 Ti catalysts. However, further increasing the doping amount of Ce or W decreased the NO conversion as the value was in the range of 74.8% to 98.8% for Ce 0.3 W 0.1 Ti and 57.5% to 93.2% for Ce 0.1 W 0.3 Ti, respectively. The selectivity of N 2 showed a similar trend. As for W 0.4 Ti, the N 2 selectivity decreased from 98.0% at 300 • C to 83.8% at 450 • C, while this value decreased from 99.7% to 90.4% for Ce 0.4 Ti. The N 2 selectivity was significantly enhanced when using the Ce-W-Ti catalysts, 100% N 2 selectivity was achieved over Ce 0.2 W 0.2 Ti catalyst between 150 • C and 350 • C and further increasing the temperature to 450 • C slightly decreased the N 2 selectivity to 98.5%. The results of NO conversion and N 2 selectivity indicate that strong interactions exist between Ce and W species in the Ce-W-Ti catalysts, which could contribute to the reaction performance of NH 3 -SCR process. A series of characterizations were carried out to analyze the physico-chemical properties of the Ce-W-Ti catalysts and the relationships between the catalyst structure and the reaction performance.  Figure 2 shows the effect of SO2 on the NH3-SCR activity over the best-performed Ce0.2W0.2Ti catalyst at 300 °C with and without H2O over 10 h. Before introducing SO2 and H2O into the catalyst layer, the NO conversion rate is 100% at a steady state. When 500 ppm SO2 was added, the catalytic activity of Ce0.2W0.2Ti dramatically decreases to around 60% in 8 h [22]. Previous work confirmed that the SO2 could react with NH3 and form the NH4HSO4, which may block the active sites on the surface of Ce0.2W0.2Ti catalyst 26 . On the other hand, the formation of highly thermal stable sulfate species of Ce(SO4)2 and Ce2(SO4)3 was also expected on the catalyst surfaces, which would reduce the relative concentration of Ce 3+ /Ce 4+ redox pair on the catalyst surface, and consequently impose a negative effect on the catalytic activity [27,28]. The co-introduction of 500 ppm SO2 and 5% H2O significantly affects the catalytic activity of Ce0.2W0.2Ti. The catalytic activity is reduced more quickly compared to the case of introducing SO2 alone with only 65% after 8 h. After cutting off SO2 and H2O from the feeding gas, the activity slightly recovers to around 70%. It is obvious that the addition of W species into the original Ce-Ti catalyst remarkably enhances the resistance of catalysts towards SO2 and H2O. Similar observations were also reported in previous work of cerium-based catalysts [27,29].

N2 Adsorption and Desorption
N2 adsorption and desorption experiments were performed to analyze the structure of the CexW0.4−xTi catalysts. All catalysts exhibited typical type III isotherms, indicating the formation of multilayers in the catalysts and relative weak interactions between N2 and the catalysts [30]. The  Figure 2 shows the effect of SO 2 on the NH 3 -SCR activity over the best-performed Ce 0.2 W 0.2 Ti catalyst at 300 • C with and without H 2 O over 10 h. Before introducing SO 2 and H 2 O into the catalyst layer, the NO conversion rate is 100% at a steady state. When 500 ppm SO 2 was added, the catalytic activity of Ce 0.2 W 0.2 Ti dramatically decreases to around 60% in 8 h [22]. Previous work confirmed that the SO 2 could react with NH 3 and form the NH 4 HSO 4 , which may block the active sites on the surface of Ce 0.2 W 0.2 Ti catalyst 26 . On the other hand, the formation of highly thermal stable sulfate species of Ce(SO 4 ) 2 and Ce 2 (SO 4 ) 3 was also expected on the catalyst surfaces, which would reduce the relative concentration of Ce 3+ /Ce 4+ redox pair on the catalyst surface, and consequently impose a negative effect on the catalytic activity [27,28]. The co-introduction of 500 ppm SO 2 and 5% H 2 O significantly affects the catalytic activity of Ce 0.2 W 0.2 Ti. The catalytic activity is reduced more quickly compared to the case of introducing SO 2 alone with only 65% after 8 h. After cutting off SO 2 and H 2 O from the feeding gas, the activity slightly recovers to around 70%. It is obvious that the addition of W species into the original Ce-Ti catalyst remarkably enhances the resistance of catalysts towards SO 2 and H 2 O. Similar observations were also reported in previous work of cerium-based catalysts [27,29].  Figure 2 shows the effect of SO2 on the NH3-SCR activity over the best-performed Ce0.2W0.2Ti catalyst at 300 °C with and without H2O over 10 h. Before introducing SO2 and H2O into the catalyst layer, the NO conversion rate is 100% at a steady state. When 500 ppm SO2 was added, the catalytic activity of Ce0.2W0.2Ti dramatically decreases to around 60% in 8 h [22]. Previous work confirmed that the SO2 could react with NH3 and form the NH4HSO4, which may block the active sites on the surface of Ce0.2W0.2Ti catalyst 26 . On the other hand, the formation of highly thermal stable sulfate species of Ce(SO4)2 and Ce2(SO4)3 was also expected on the catalyst surfaces, which would reduce the relative concentration of Ce 3+ /Ce 4+ redox pair on the catalyst surface, and consequently impose a negative effect on the catalytic activity [27,28]. The co-introduction of 500 ppm SO2 and 5% H2O significantly affects the catalytic activity of Ce0.2W0.2Ti. The catalytic activity is reduced more quickly compared to the case of introducing SO2 alone with only 65% after 8 h. After cutting off SO2 and H2O from the feeding gas, the activity slightly recovers to around 70%. It is obvious that the addition of W species into the original Ce-Ti catalyst remarkably enhances the resistance of catalysts towards SO2 and H2O. Similar observations were also reported in previous work of cerium-based catalysts [27,29].

N2 Adsorption and Desorption
N2 adsorption and desorption experiments were performed to analyze the structure of the CexW0.4−xTi catalysts. All catalysts exhibited typical type III isotherms, indicating the formation of multilayers in the catalysts and relative weak interactions between N2 and the catalysts [30]. The

N 2 Adsorption and Desorption
N 2 adsorption and desorption experiments were performed to analyze the structure of the Ce x W 0.4−x Ti catalysts. All catalysts exhibited typical type III isotherms, indicating the formation of multilayers in the catalysts and relative weak interactions between N 2 and the catalysts [30]. The hysteresis loop was demonstrated to be H3 type for the catalysts, which is associated with the existence of slit-shaped pores formed from the aggregation of plate-like particles [31]. Table 1 gives the information of the specific surface area (S BET ), the pore volume and the average pore diameter of the catalysts based on the calculations from the isotherms. Compared with Ce 0.4 Ti and W 0.4 Ti, the Ce x W 0.4−x Ti catalysts possessed larger S BET and pore volume, indicating more surface active sites formed on the Ce x W 0.4−x Ti catalysts for the SCR reaction. The highest S BET (108.1 m 2 ·g −1 ) and pore volume (0.121 cm 3 ·g −1 ) was observed for the Ce 0.2 W 0.2 Ti catalyst, while further increasing of the loading amount of Ce or W resulted in decreased S BET and pore volume. It is well recognized that larger specific surface area was beneficial for offering more active sites for the NH 3 -SCR reaction [23,25]. Moreover, Ce 0.2 W 0.2 Ti showed the smallest average pore diameter of 3.9 nm compared to the other samples. Zhao et al. also reported that the decrease of average pore diameter and pore volume could improve the performance of the NH 3 -SCR process [32].

XRD Analysis
The XRD patterns of the catalysts are presented in Figure 3. All the catalysts presented the typical diffraction peaks of anatase TiO 2 (JCPDS 21-1272). The characteristic diffraction peaks of Ce and W species were not observed for all catalysts even the loading amount of Ce and W may exceed the limit of monolayer dispersion. This indicated that the active Ce and W species were in amorphous phase or well dispersed on the TiO 2 surfaces. Similar observations were reported in a previous study [22]. The crystallization degrees of the TiO 2 support were smaller in the Ce x W 0.4−x Ti catalysts compared to the Ce 0.4 Ti and W 0.4 Ti indicating the smaller crystalline sizes of the Ce x W 0.4−x Ti catalysts, considering the lower peak intensities of the major typical diffraction peaks of TiO 2 (1 0 1) at 2θ = 25.3 • . This phenomenon indicated a smaller crystalline size of the Ce-W-Ti catalysts compared to the Ce 0.4 Ti and W 0.4 Ti catalysts, which was beneficial for the NH 3 -SCR of NO and could be confirmed by the broadening of the characteristic peaks in the Ce-W-Ti catalysts [33,34]. hysteresis loop was demonstrated to be H3 type for the catalysts, which is associated with the existence of slit-shaped pores formed from the aggregation of plate-like particles [31]. Table 1 gives the information of the specific surface area (SBET), the pore volume and the average pore diameter of the catalysts based on the calculations from the isotherms. Compared with Ce0.4Ti and W0.4Ti, the CexW0.4−xTi catalysts possessed larger SBET and pore volume, indicating more surface active sites formed on the CexW0.4−xTi catalysts for the SCR reaction. The highest SBET (108.1 m 2 ·g −1 ) and pore volume (0.121 cm 3 ·g −1 ) was observed for the Ce0.2W0.2Ti catalyst, while further increasing of the loading amount of Ce or W resulted in decreased SBET and pore volume. It is well recognized that larger specific surface area was beneficial for offering more active sites for the NH3-SCR reaction [23,25]. Moreover, Ce0.2W0.2Ti showed the smallest average pore diameter of 3.9 nm compared to the other samples. Zhao et al. also reported that the decrease of average pore diameter and pore volume could improve the performance of the NH3-SCR process [32].

XRD Analysis
The XRD patterns of the catalysts are presented in Figure 3. All the catalysts presented the typical diffraction peaks of anatase TiO2 (JCPDS 21-1272). The characteristic diffraction peaks of Ce and W species were not observed for all catalysts even the loading amount of Ce and W may exceed the limit of monolayer dispersion. This indicated that the active Ce and W species were in amorphous phase or well dispersed on the TiO2 surfaces. Similar observations were reported in a previous study [22]. The crystallization degrees of the TiO2 support were smaller in the CexW0.4−xTi catalysts compared to the Ce0.4Ti and W0.4Ti indicating the smaller crystalline sizes of the CexW0.4−xTi catalysts, considering the lower peak intensities of the major typical diffraction peaks of TiO2 (1 0 1) at 2θ = 25.3°. This phenomenon indicated a smaller crystalline size of the Ce-W-Ti catalysts compared to the Ce0.4Ti and W0.4Ti catalysts, which was beneficial for the NH3-SCR of NO and could be confirmed by the broadening of the characteristic peaks in the Ce-W-Ti catalysts [33,34].   Figure 4 shows the XPS spectra of O 1s for the Ce-W-Ti catalysts. Two major peaks of different surface oxygen species could be obtained by fitting the curves. The peaks located between 529.6 eV and 530.1 eV could be ascribed to the lattice oxygen (O lat ), while the peaks at 531.2 eV to 531.6 eV belonged to the surface-adsorbed oxygen (O ads ) [22]. The relative concentrations of O ads /(O ads + O lat ) are calculated and presented in Table 1. The Ce-W-Ti catalysts possessed higher relative concentrations of O ads /(O ads + O lat ) in the range of 21.0% to 33.8% compared to the Ce 0.4 Ti (17.4%) and W 0.4 Ti (16.0%) catalysts, while the highest O ads ratio was observed over the Ce 0.2 W 0.2 Ti catalyst. The conventional SCR reaction cycles indicated that redox properties were crucial for the reaction performance at low temperatures. Shan et al. reported that the interactions between Ce and W species could enrich the amount of Ce 3+ and surface oxygen vacancies, and that these two species were directly correlated with the surface oxygen species [35]. Surface-adsorbed oxygen (O ads ) is more chemically reactive due to their higher mobility on the catalyst surfaces compared to the lattice oxygen (O lat ) species, which was beneficial for the oxidation of NO to NO 2 , one of the initial steps of NH 3 -SCR [24,36]. The order of O ads was in good agreement with the sequence of the NH 3 -SCR performance.

XPS Analysis
Appl. Sci. 2018, 8, x FOR PEER REVIEW 6 of 9 Figure 4 shows the XPS spectra of O 1s for the Ce-W-Ti catalysts. Two major peaks of different surface oxygen species could be obtained by fitting the curves. The peaks located between 529.6 eV and 530.1 eV could be ascribed to the lattice oxygen (Olat), while the peaks at 531.2 eV to 531.6 eV belonged to the surface-adsorbed oxygen (Oads) [22]. The relative concentrations of Oads/(Oads + Olat) are calculated and presented in Table 1. The Ce-W-Ti catalysts possessed higher relative concentrations of Oads/(Oads + Olat) in the range of 21.0% to 33.8% compared to the Ce0.4Ti (17.4%) and W0.4Ti (16.0%) catalysts, while the highest Oads ratio was observed over the Ce0.2W0.2Ti catalyst. The conventional SCR reaction cycles indicated that redox properties were crucial for the reaction performance at low temperatures. Shan et al. reported that the interactions between Ce and W species could enrich the amount of Ce 3+ and surface oxygen vacancies, and that these two species were directly correlated with the surface oxygen species [35]. Surface-adsorbed oxygen (Oads) is more chemically reactive due to their higher mobility on the catalyst surfaces compared to the lattice oxygen (Olat) species, which was beneficial for the oxidation of NO to NO2, one of the initial steps of NH3-SCR [24,36]. The order of Oads was in good agreement with the sequence of the NH3-SCR performance.

NH3-TPD Analysis
Acidity of the catalysts is crucial for efficiency NH3 adsorption during the SCR process. The acidity and the relative strength of all the acid sites on the surfaces of the Ce-W-Ti catalysts were analyzed using NH3-TPD measurement. A broad desorption peak spanned between 100 °C and 400 °C for all samples in Figure 5, which could be attributed to the desorption of NH3 molecules on the weak and medium acid sites [3,37]. It is reported that the thermal stability of restrained NH4 + ions on the Lewis acid sites were stronger than those on the Brønsted acid sites. The NH3 desportion peaks below 200 °C belonged to the Brønsted acid sites, while the desorption peaks at higher temperatures were attributed to the Lewis acid sites [35]. For Ce0.4Ti and W0.4Ti, the strong desorption peaks at 150 °C and 181 °C belonged to the NH4 + ions adsorbed on the Brønsted acid sites, while the peaks located at 295 °C and 311 °C could be ascribed to the desorption of NH3 species adsorbed on the

NH 3 -TPD Analysis
Acidity of the catalysts is crucial for efficiency NH 3 adsorption during the SCR process. The acidity and the relative strength of all the acid sites on the surfaces of the Ce-W-Ti catalysts were analyzed using NH 3 -TPD measurement. A broad desorption peak spanned between 100 • C and 400 • C for all samples in Figure 5, which could be attributed to the desorption of NH 3 molecules on the weak and medium acid sites [3,37]. It is reported that the thermal stability of restrained NH 4 + ions on the Lewis acid sites were stronger than those on the Brønsted acid sites. The NH 3 desportion peaks below 200 • C belonged to the Brønsted acid sites, while the desorption peaks at higher temperatures were attributed to the Lewis acid sites [35]. For Ce 0.4 Ti and W 0.4 Ti, the strong desorption peaks at 150 • C and 181 • C belonged to the NH 4 + ions adsorbed on the Brønsted acid sites, while the peaks located at 295 • C and 311 • C could be ascribed to the desorption of NH 3 species adsorbed on the Lewis acid sites. For the Ce-W-Ti catalysts, the NH 3 desorption peaks shifted to higher temperatures (between 158 • C and 167 • C) compared to the Ce 0.4 Ti catalyst, while the intensities of the desorption peaks were significantly increased. Moreover, the intensities of the desorption peaks above 200 • C were also enhanced over the Ce-W-Ti catalysts, indicating the existence of stronger Lewis acid sites [38]. This phenomenon could be attributed to the synergistic effect between the cerium and tungsten species. The amounts of NH 3 uptake of all the catalysts were given in Table 1  Lewis acid sites. For the Ce-W-Ti catalysts, the NH3 desorption peaks shifted to higher temperatures (between 158 °C and 167 °C) compared to the Ce0.4Ti catalyst, while the intensities of the desorption peaks were significantly increased. Moreover, the intensities of the desorption peaks above 200 °C were also enhanced over the Ce-W-Ti catalysts, indicating the existence of stronger Lewis acid sites [38]. This phenomenon could be attributed to the synergistic effect between the cerium and tungsten species. The amounts of NH3 uptake of all the catalysts were given in Table 1 as evidence for the total acidity on the catalysts. The Ce-W-Ti catalysts show stronger acidity compared to the Ce0.4Ti and W0.4Ti catalysts, while the Ce0.2W0.2Ti catalyst possessed the largest NH3 uptake of 193.9 μmol·g −1 .
Further increases or decreases in the Ce/W molar ratio decreases the acidity on the catalysts. It was well acknowledged that the acidity of the catalysts played a significant role in the process of NH3-SCR, and more acid sites on the surface Ce0.2W0.2Ti catalysts could be favorable for the conversion of NO in the low temperature range [3].

Conclusions
In this work, a series of Ce-W-Ti catalysts were synthesized using a solution combustion method for selective catalytic reduction (SCR) of NO with NH3 at low temperatures. The removal of NO was significantly improved over the Ce-W-Ti catalysts compared to Ce0.4Ti and W0.4Ti catalyst. Among all the catalysts, Ce0.2W0.2Ti exhibited over 90% NO removal efficiency and 100% N2 selectivity in the temperature range of 250°C and 400 °C at the GHSV of 120,000 mL·g −1 ·h −1 . A series of catalyst characterizations including N2 adsorption-desorption, XRD, XPS and NH3-TPD were performed to establish the structure-activity relationships of the Ce-W-Ti catalysts. The interactions between Ce and W species significantly affected the physic-chemical properties of the catalysts. The Ce-W-Ti catalysts exhibited larger specific surface areas compared to the Ce0.4Ti and W0.4Ti catalysts, while the Ce and W species were highly dispersed on the surfaces of the catalysts. The interactions between Ce and W species also increased the amount of surface adsorbed oxygen (Oads) and the total acidity of the catalysts, which were beneficial for the adsorption of the reactants and oxidation of the intermediates during the NH3-SCR process. To sum up, the catalytic performance of the NH3-SCR was improved over the Ce0.2W0.2Ti catalysts.

Conclusions
In this work, a series of Ce-W-Ti catalysts were synthesized using a solution combustion method for selective catalytic reduction (SCR) of NO with NH 3 at low temperatures. The removal of NO was significantly improved over the Ce-W-Ti catalysts compared to Ce 0.4 Ti and W 0.4 Ti catalyst. Among all the catalysts, Ce 0.2 W 0.2 Ti exhibited over 90% NO removal efficiency and 100% N 2 selectivity in the temperature range of 250 • C and 400 • C at the GHSV of 120,000 mL·g −1 ·h −1 . A series of catalyst characterizations including N 2 adsorption-desorption, XRD, XPS and NH 3 -TPD were performed to establish the structure-activity relationships of the Ce-W-Ti catalysts. The interactions between Ce and W species significantly affected the physic-chemical properties of the catalysts. The Ce-W-Ti catalysts exhibited larger specific surface areas compared to the Ce 0.4 Ti and W 0.4 Ti catalysts, while the Ce and W species were highly dispersed on the surfaces of the catalysts. The interactions between Ce and W species also increased the amount of surface adsorbed oxygen (O ads ) and the total acidity of the catalysts, which were beneficial for the adsorption of the reactants and oxidation of the intermediates during the NH 3 -SCR process. To sum up, the catalytic performance of the NH 3 -SCR was improved over the Ce 0.2 W 0.2 Ti catalysts.