Evaluation of cerium modification over Cr/Ti-PILC for NO catalytic oxidation and their mechanism study

Abstract

A series of cerium modified Cr/Ti-PILC catalysts were evaluated, which showed a remarkable increase in the activity of NO oxidation. The aim of this novel design was to investigate the mechanism of cerium modification over the Cr/Ti-PILC catalyst. Physicochemical characteristics were investigated in detail by various techniques such as BET, TPD (NO and O₂), XPS, PL, EPR and DRIFTS. The analysis results demonstrated that cerium modification could facilitate the generation of oxygen vacancy via charge transfer, promote the formation of surface superoxide ions (O₂⁻), and increase the amount of surface nitrates. Furthermore, the original oxidation pathway of Cr/Ti-PILC was maintained by cerium modification. The experimental results showed that the NO conversion of CrCe(0.25)/Ti-PILC catalyst was increased to nearly 66.9% at 300 °C.

---

1. Introduction

Nitrogen oxides (NOx) originating from the combustion of fossil fuels are hazardous to the environment and human health, and cause ozone depletion, photochemical smog and acid deposition.¹ Because of very strict regulations on NOx emissions, nitric oxide (NO), the main source of NOx, must be carefully controlled prior to its release in the atmosphere. The most common technology currently used for the removal of NO is the well-known selective catalytic reduction with ammonia (NH₃-SCR).^2,3 However, considering the limitations of high cost and possible corrosion in the NH₃-SCR reaction, the method of selective catalytic oxidation (SCO) might be identified a promising way.^4–6 The most distinguishing feature is dry catalytic oxidation, in which outlet gas mixtures (NO and NO₂) are adequately absorbed by an alkaline solution if the ratio of NO/NOx is about 60%.⁷ Hence, many efforts have been focused on developing suitable catalysts for NO oxidation.

It is noteworthy that Ti-PILC has been investigated as one kind of attractive supports, which was applied to investigate the catalytic reaction.^8–11 In addition, it is necessary to point out that chromium oxides with the Cr³⁺/Cr⁶⁺ redox couple can absorb the acid gas (such as NO) and facilitate complete oxidation;^12,13 therefore, we preliminarily focused on chromium oxides supported on TiO₂-pillared clay to catalytically oxidize NO. The good oxidation performance of CrOx catalyst was verified in our previous work.¹⁴ In addition, cerium oxides, with strong oxygen mobility and excellent properties of oxygen storage and release, are beneficial for activating oxygen via oxygen vacancies and improving the redox properties of the catalyst.^15,16 Although we achieved NO conversion of 69% at 350 °C by the conventional impregnation method and determined the promotion effect of ceria in NO oxidation,¹⁷ the adsorption properties and the functional mechanisms of cerium modification over Cr/Ti-PILC via oxygen vacancies are still unclear. Furthermore, a lower reaction temperature window could be considered according to the thermodynamic equilibrium limitation.

Inspired by the abovementioned consideration, the preparation method was optimized via ultrasonic irradiation to widen the activity temperature window. Moreover, both textural property and adsorption property were taken into account for SCO reaction. Furthermore, the role of oxygen vacancies was considered to be the focus of cerium modified Cr/Ti-PILC in catalysis. Finally, the evolution of surface-bound species was investigated by DRIFTS to prove the effect of oxygen vacancies. Further, the mechanism of cerium modification over Cr/Ti-PILC for NO oxidation was proposed reasonably.

2. Experimental

2.1 Materials

Butyl titanate, nitric acid and acetone were provided by Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Ethanol, chromium nitrate and cerium nitrate were purchased from Shanghai Chemical Reagent Co., Ltd. (Shanghai, China). All the chemicals were of analytical grade. In addition, the original material was purified Na-montmorillonite (Na-MMT) and its cation exchange capacity (CEC) was 102 meq. per 100 g, which was purchased from Zhejiang Institute of Geology & Mineral Resources.

2.2 Preparation of the catalysts

The support was prepared by ion-exchange method and the catalysts were synthesized by wet impregnation method via ultrasonic irradiation. A stoichiometric amount of tetra-n-butyl titanate was added dropwise to nitric acid solution under vigorous stirring and aged at room temperature, which then formed titanium pillaring agents. Then, the pillaring solution, with the Ti/clay ratio of 18 mmol g⁻¹, was added dropwise to an aqueous clay suspension. The collected mixture was kept under vigorous stirring for 12 h at 60 °C. Finally, the modified mixture was washed repeatedly, dried, grinded and calcined for 3 h at 500 °C. It was referred to as TiO₂-pillared clay (support, abbreviated as Ti-PILC).

In the next step of catalyst preparation, metal oxides were introduced into the dried Ti-PILC by wet impregnation under ultrasonic irradiation. About 1 g Ti-PILC was added to Cr (NO₃)₃·9H₂O (and) Ce (NO₃)₄·6H₂O solution. After stirring at 60 °C for 6 h, the suspension was agitated ultrasonically for 30 min. The impregnated products were collected by filtration, drying and calcination. The total loading of active components was 10 wt%. For simplification, these samples were denoted as Cr/Ti-PILC and CrCe(x)/Ti-PILC. For example, CrCe(0.25)/Ti-PILC implied the catalyst with a molar ratio of Ce/Cr = 0.25.

2.3 Catalytic activity test

The catalytic performance of NO oxidation was carried out in a fixed-bed flow micro reactor at atmospheric pressure. Typically, 300 mg sample (sieve fraction of 60–80 mesh) was placed in a quartz reactor (i.d. 6.8 mm). The reactant mixture (400 ppm NO, 8% O₂ and N₂ balance) was fed to the reactor with a total gas flow rate of 100 mL min⁻¹. Prior to the measurement of catalytic activity, each sample was pretreated for 2 h to avoid errors caused by NO adsorption. The concentrations of inlet and outlet mixture, including NO, NO₂, O₂, were analyzed by the Ecom-JZKN 12 flue gas analyzer (made in Germany). The exit gas from the micro reactor passed through a trap containing the concentrated alkaline solution and then vented out. The NO conversion to NO₂ was defined as follows:

NO Conversion = (NO_inlet − NO_outlet)/NO_inlet × 100% (1)

2.4 Characterizations of the catalysts

Specific surface areas of different catalysts were determined by N₂ adsorption–desorption measurements at 77 K by employing the Brunauer–Emmett–Teller (BET) method (Gold App V-sorb 2800), and the total pore volume and pore size of the samples were calculated by the Barrett–Joyner–Halenda (BJH) method.

Temperature-programmed desorption (TPD) was carried out on automated chemisorption analyzer (Quantachrome Instruments). About 200 mg of sample was used. After NO (or O₂) saturation in 1 h, the gas was switched to He for 0.5 h. Subsequently, TPD was performed by ramping the temperature from 10 °C min⁻¹ to 700 °C in He (70 mL min⁻¹). The desorption of NO (or O₂) was detected by a thermal conductivity detector (TCD).

X-ray photoelectron spectra (XPS) were performed on a Thermo ESCALAB 250 (USA) apparatus with Al Kα X-rays (hv = 1486.6 eV) as radiation source operated at 150 W. The samples were compensated for charging with low-energy electron beam, and the peak of C1s (binding energy = 284.4 eV) was used to correct for sample charging.

The electron transformation of the catalyst was characterized by photoluminescence (PL) spectra at room temperature using a LabRAM-HR800-type spectrophotometer (Jobin Yvon Co., France) with a He–Cd laser (λ = 325 nm) as the excitation light source.

The electron paramagnetic resonance (EPR) measurements were made at room temperature using a Bruker EMX-10/12-type spectrometer (∼9.7 GHz) in the X-band.

Diffused reflectance infrared Fourier transform spectroscopy (DRIFTS) spectra were obtained on a Nicolet IZ10 FTIR spectrometer, equipped with a liquid-nitrogen-cooled MCT detector. 32 scans were averaged for each spectrum, which were recorded at a resolution of 4 cm⁻¹.

3. Results and discussion

3.1 Structural characterization: BET

Some researchers have reported that the increase in specific surface areas is conducive for improving catalytic activity.^18,19 Hence, nitrogen adsorption–desorption isotherms were employed to determine the surface area and pore size distributions of the samples, and the results are displayed in Fig. 1. As can be observed, all the isotherms were of type IV according to the IUPAC classification, with a hysteresis loop type H3, indicating that these samples exhibited the characteristic behavior of layered materials with slit-like pores.²⁰ The corresponding pore size distributions, calculated by the BJH method, are presented in Fig. 1b, which further indicated the presence of mesopores (2–50 nm). Mesoporous structure would make a significant contribution toward generation of high specific surface area. Moreover, it could be seen from the pore size distributions that the average particle pores of Cr/Ti-PILC and CrCe(0.25)/Ti-PILC were smaller than that of Ti-PILC.

Fig. 1 N₂ adsorption–desorption isotherms (a) and pore size distributions (b) of the samples.

The structural parameters calculated from nitrogen adsorption–desorption isotherms are summarized in Table 1. Obviously, the introduction of metal oxides to TiO₂-pillared clays caused a significant decrease in surface area, which indicated that metal oxides partly occupied the pores of TiO₂-pillared clays. Moreover, it was observed that the introduction of cerium influenced the surface area of the catalysts. The surface area decreased with the variances of particle sizes. From our results, it could be seen that the CrCe(0.25)/Ti-PILC catalyst with a pore size of 7 nm exhibited the smallest specific surface area of 138 m² g⁻¹. Although a relatively large surface area could provide more active sites for the adsorption of reactant molecules for promoting catalytic conversion, the specific surface area was not a crucial factor for comparing catalytic activity in NO oxidation. The slight difference in the textural properties between Cr/Ti-PILC and CrCe(0.25)/Ti-PILC may be connected with the oxidation and emancipation of gaseous nitrogen oxides during the thermal decomposition of the precursors, which will be discussed in Section 3.6.

Table 1 Surface area, pore volume measurements and pore size of the samples

Samples	Surface area (m^2 g^−1)	Pore volume (10^−2 cm^3 g^−1)	Pore size (nm)
Ti-PILC	207	58	13
Cr/Ti-PILC	160	33	9
CrCe(0.25)/Ti-PILC	138	24	7

---

3.2 Catalytic activity test

Fig. 2 displays the catalytic performance of the samples for NO oxidation as a function of temperature from 200 to 380 °C. The gradual increase in NO conversion with temperature was observed for all the catalysts in the medium temperature range (200–320 °C); however, the catalytic activity slightly decreased at 320 °C. The catalytic activity decreased in the following order: CrCe(0.25)/Ti-PILC > CrCe(0.13)/Ti-PILC > CrCe(0.5)/Ti-PILC > Cr/Ti-PILC. The results demonstrated that the activity of the Cr/Ti-PILC catalyst was only 55.1% when reaction temperature reached up to 350 °C. Compared to the Cr/Ti-PILC sample, a clearly promotional impact of cerium modification was obtained and the optimal molar ratio of Ce/Cr reached to 0.25. Especially for the CrCe(0.25)/Ti-PILC catalyst, its activity temperature window ranged from 290 to 350 °C and NO conversion was increased to nearly 66.9% at 300 °C, while NO conversion of 69% at 350 °C was achieved by Zhang.¹⁷ Compared to our previous study,¹⁷ it is necessary to point out that more superior catalytic performance and wider temperature window might be attributed to the effect of ultrasonic irradiation during preparation. Therefore, it could be deduced that cerium modification improved NO conversion and expanded the temperature window of Cr/Ti-PILC.

Fig. 2 Catalytic performance of the samples for NO oxidation at various temperatures. Reaction conditions: 400 ppm NO, 8% O₂, N₂ balance, GHSV = 35 400 h⁻¹.

The same shape of catalytic activity curve, despite very different activity, may support the suggestion that the introduced cerium species act as additional active sites for NO oxidation. Meng²¹ et al. found that the improvement in the activity for cerium-containing material was related to the effect of oxygen vacancies in the redox reaction. In other words, there may be a new reaction pathway on the site of ceria, which helped the generation of more NO₂ to improve the catalytic performance. Thereby, XPS, PL and EPR were applied to analyze the electronic states and to further demonstrate the oxygen vacancies of the samples. In particular, the evolution of surface-bound species was essential, which was recorded by in situ DRIFTS spectra to reveal the oxidation reaction. For simplicity, we evaluated the performance of cerium modification with Cr/Ti-PILC and CrCe(0.25)/Ti-PILC, respectively.

3.3 Adsorption properties (NO-TPD, O₂-TPD)

To evaluate the reversibility of gas adsorption (NO or O₂) in different samples, the TPD tests were performed. The TPD profiles of NO over the different catalysts are shown in Fig. 3a. As the temperature increased from 50 to 550 °C, the desorption spectra of typical catalysts were dominated by two main temperature ranges for NO desorption, suggesting that NO adsorbed on two different sites.²² According to the literature,^23,24 the first desorption peak (α) from 100 to 250 °C was attributed to bridged nitrate and the second one (β) at around 450–700 °C was assigned to bidentate nitrate. Moreover, compared to the NO desorption area, the ability of surface adsorption–desorption of NO over CrCe(0.25)/Ti-PILC catalyst was stronger than that of Cr/Ti-PILC catalyst. This was in agreement with the catalytic activity showed in Fig. 2. This result implied that the addition of ceria could improve the amount of adsorbed NO over the Cr/Ti-PILC catalyst, especially the bidentate nitrate. In particular, the similar curve shape of NO desorption showed that NO adsorption site occurred at the site of chromium oxides. Furthermore, the adsorbed NO could be easily desorbed from 100 to 400 °C, which was consistent with the test temperature region. This result suggested that the adsorption NO was unstable.²⁵

Fig. 3 NO-TPD profile (a) and O₂-TPD profile (b) of Cr/Ti-PILC and CrCe(0.25)/Ti-PILC.

It was an effective method to determine the mobility of oxygen species via O₂-TPD, and the spectra are displayed in Fig. 3b. Compared to the Cr/Ti-PILC catalyst, oxygen desorption area of the CrCe(0.25)/Ti-PILC catalyst increased and its temperature decreased dramatically, which suggested that the capability of oxygen activation was advanced with cerium modification. Particularly, the different curve shape of O₂ desorption showed that O₂ adsorption site acted on ceria. In general, the O₂ desorption amount was consistent with oxygen consumption. Synthetically considering the desorption temperature and oxygen consumption, the CrCe(0.25)/Ti-PILC catalyst possessed the higher ability to activate oxygen. From this, it could be seen that low desorption temperature and large oxygen consumption were favorable for significantly increasing the catalytic oxidation activity, which was consistent with the findings by Meng et al.²¹ In addition, the desorption peak of CrCe(0.25)/Ti-PILC catalyst remained unchanged at high temperature (600 °C), which further confirmed that CeOx could resist the sintering of the catalysts. Interestingly, Ce 3d XPS (Fig. 4a) results revealed that oxygen vacancies were always present in the presence of cerium, thus it is believable that active oxygen species were formed via oxygen vacancies on the surface of CrCe(0.25)/Ti-PILC catalyst. Considering the relatively low desorption temperature, such oxygen species could be O₂⁻. These similar results have been noted previously.²⁵

Fig. 4 XPS spectra of Cr/Ti-PILC and CrCe(0.25)/Ti-PILC: (a) Cr 2p; (b) Ce 3d.

3.4 Surface composition: X-ray photoelectron spectroscopy (XPS)

For further investigating the impact of the surface chemical state of various elements, the surface atomic concentrations of Cr 2p and Ce 3d are summarized in Table 2, and their high-resolution XPS spectra are presented in Fig. 4.

Table 2 XPS elements and cation surface concentration of catalysts^a

Samples		Cr/Ti-PILC	CrCe(0.25)/Ti-PILC
a Ratios were calculated from bulk composition.
Atomic Concentration	Cr (at%)	9.6	8.7
	Ce (at%)		2
	O (at%)	66	67
Cation ratios	Cr^6+/(Cr^6+ + Cr^3+)	44%	47%
	Ce^3+/(Ce^3+ + Ce^4+)		16%
Oxygen ratios	Oα/(Oα + Oβ)	63%	64%

---

The Cr 2p spectra of the samples are exhibited in Fig. 4a. It was clear that two peaks at about 587 and 576 eV were assigned to 2p_1/2 and 2p_3/2 of Cr³⁺ species, respectively. The peaks at about 579 and 578 eV were attributed to the 2p_3/2 of Cr⁶⁺ species.²⁶ It was obvious that the binding energy of the CrCe(0.25)/Ti-PILC catalyst was higher than that of the Cr/Ti-PILC. According to the research of Li. et al.,²⁷ it was inferred as the result of charge imbalance caused by the addition of cerium. And the reason might be that electrons around the chromium species (such as Cr³⁺) transferred to ceria, which facilitated the formation of oxygen vacancies. Furthermore, the area ratio of Cr⁶⁺/(Cr³⁺ + Cr⁶⁺) also confirmed this suggestion. As shown in Table 2, note that the area ratio of Cr⁶⁺/(Cr³⁺ + Cr⁶⁺) slightly increased when cerium oxides were introduced. It further confirmed the fact that the formation of oxidized Cr⁶⁺ species was conducive to promote the catalytic performance of NO oxidation.

The chemical states of Ce 3d were investigated, and the results are shown in Fig. 4b. The peaks labeled v are attributed to Ce 3d_5/2, and the peaks labeled u represent Ce 3d_3/2.^28,29 The dominant peaks denoted by v, v′′, v′′′ (882.7, 888.9, 898.3 eV) and u, u′′, u′′′ (901.0, 908.3 ± 1.6, 916.8 eV) represented the 3d¹⁰4f⁰ state of Ce⁴⁺ ions, whereas the peaks marked v′ and u′(885.8, 904.0 ± 0.4 eV) represented the 3d¹⁰4f¹ initial electronic state, corresponding to Ce³⁺ ions.^30,31 Migani et al.³² reported that an oxygen vacancy was associated with the formation of two Ce³⁺ 4f¹ ions. Expectedly, a small quantity of Ce³⁺ (v′ and u′) created the opportunity of the formation of oxygen vacancies on the surface of the catalyst. This suggested that the ceria modified sample generated the reduced ceria (Ce₂O₃) forming oxygen vacancies, which is consistent with our PL results. Therefore, the richness of surface Cr⁶⁺ could promote the adsorption and activation of NO, and the increase of Ce³⁺ might be beneficial to form oxygen vacancies.

The O1s core level spectra of two representative samples are displayed in Fig. S1.† Two types of surface oxygen species were distinguished. The lower binding energy of 530.1 eV was attributed to the lattice oxygen (denoted as O_β) and the other strong peak with higher binding energy (around 532.1 eV) was assigned to chemisorbed oxygen (denoted as O_α).³³ As shown in Fig. 3b and Table 2, the ratio of chemisorbed oxygen to the entire oxygen in the CrCe(0.25)/Ti-PILC catalyst was slightly higher than that of Cr/Ti-PILC, which indicated that the cerium modified sample was beneficial for the formation of oxygen vacancy on the sample surface. It was reported that³⁴ surface chemisorbed oxygen was the most active and played a critical role in oxidation reaction via oxygen vacancies. In fact, the results of the activity test are in agreement with this conclusion. Moreover, the binding energy of O1s shifted toward higher position after ceria modification. It could be because of charge imbalance leading to increased electron density of O in the presence of cerium.

3.5 Characterization of surface superoxide ions

For SCO reaction, the adsorption of NO and the formation and desorption of NO₂ is the key procedure. According to the finding of Wang et al.,³⁵ NO molecule could react with superoxide ions to generate the NOx group, including NO₂. Subsequently, the crucial role of superoxide ions has been confirmed by other studies.³⁶ Considering the fact that cerium oxides activated the oxygen for generating superoxide ion via oxygen vacancies, it was necessary to confirm the formation of superoxide ions on the catalyst surface in the presence of ceria.

Superoxide ions could be formed by oxygen adsorption onto the surface oxygen vacancies.³⁶ The effect of oxygen vacancies via charge transfer was determined by PL spectra, as shown in Fig. 5. The catalysts displayed obvious PL signals with a similar curve shape. Accordingly, the noticeable peak at 533 nm was attributed to the vacancy with one trapped electron, i.e., F⁺ center.³⁴ It was accompanied by the formation of superoxide radicals (O₂⁻) on O₂ over ceria and the generation of NO⁻ on NO via chromium. Obviously, the peak intensity at 533 nm of cerium modified catalyst was lower than that of the unmodified one. It indicated that CeO₂, as trapping site, could capture charge of oxygen vacancies to separate electron-hole pairs and then inhibit the oxygen vacancies diffuse. It caused luminescence quenching and the decrease in PL intensity. In other words, the higher charge transfer resulted in a weaker PL intensity. This result is consistent with the conclusion of Xu et al.³⁷ that the lower the intensity of PL spectra, the higher the catalytic activity. Viagin et al.³⁸ also proposed that cerium ions in the 4+ valence state with 4f⁰ electronic configuration did not possess any luminescent properties. That is, the difference of the intensity of PL spectra could be attributed to oxygen vacancies on the ceria sites. Therefore, it was reasonable to believe that the reduced ceria (Ce₂O₃) was beneficial to form oxygen vacancies.

Fig. 5 PL spectra of Cr/Ti-PILC and CrCe(0.25)/Ti-PILC.

In fact, this viewpoint was further confirmed by the EPR spectra of superoxide ions displayed in Fig. 6. Accordingly, the g-value (g = 2.003) was characteristic of paramagnetic materials containing superoxide ions,³⁹ and we detected the signals at g = 2.0028. Moreover, two superimposed signals denoted by A and B were observed on the samples. The A signal was attributed to Ce³⁺ species with removable ligands and the B signal was due to Ce³⁺ ions stabilized by some lattice defects.⁴⁰ Overall, only Ce³⁺ (3d¹⁰4f¹ state) could be detected by EPR due to its unpaired electron⁴¹ and there was no signal for Cr(VI). The stronger intensity and larger peak area in CrCe(0.25)/Ti-PILC demonstrated that more superoxide ions were generated in CrCe(0.25)/Ti-PILC than in Cr/Ti-PILC, which implied that more oxygen vacancies were produced after cerium modification. Such observations were consistent with the results of our XPS results. Considering the fact that the oxygen vacancies in the metal oxide are the centers of positive charges, O₂ bounded electrons could easily form superoxide ions (O₂⁻) in the presence of oxygen vacancies, which were proven to be important intermediates in the catalytic oxidation of NO.

Fig. 6 EPR spectra of superoxide ions over the samples.

3.6 Diffused reflectance infrared fourier transform spectroscopy (DRIFTS)

To further clarify the evolution of surface-bound species under different conditions, in situ DRIFTS spectra of NO adsorption at 300 °C and the subsequent introduction of O₂ were recorded, which are displayed in Fig. 7a and b. This was a time-dependent process. For the Cr/Ti-PILC catalyst, after the feeding of NO, several peaks were detected at 1860, 1760, 1620, 1520 and 1430 cm⁻¹. According to the literature,^42,43 the strong peak at 1430 cm⁻¹ was due to monodentate nitrite, while the weak band at about 1520 cm⁻¹ was assigned to bidentate nitrate on the surface of the sample, which was consistent with the result of NO-TPD analysis. Considering the fact that the transformation of NO to nitrates required oxygen species, the formation of bidentate nitrate in the initial stage of NO adsorption without gaseous O₂ could be attributed to lattice oxygen in the Cr/Ti-PILC catalyst. These results indicated that nitrite was the dominant intermediate and less nitrate was generated in the absence of O₂. In particular, a set of significant bands were detected at 2132, 1972 and 1860 cm⁻¹. These bands are associated with three redox assignments of nitrosyl ligands with the entire NO⁺, NO˙, and NO⁻ series.^44–48 Accordingly, the formation of NO⁺ might confirm that chromium oxide with high valence state, such as Cr(VI), could transfer NO to NO⁺. Tang et al.⁴⁹ found that NO⁺ transformed NO into nitrates and readily adsorbed NO₂. In addition, Wu et al.⁵⁰ also proved that NO⁺ could be oxidized preferentially compared with other nitrogen-containing species. Meanwhile, NO⁻ might be generated by the interaction between NO gas and the vacancy of chromium species. Moreover, in the OH stretching region, a new band at 3680 cm⁻¹ and a broad absorption peak between 3600 and 3000 cm⁻¹ were observed, which were attributed to the vibrations of surface hydroxyl (OH) species.⁴⁹ Furthermore, a remarkable peak at 1620 cm⁻¹ was detected, which was attributed to the asymmetric vibration of gaseous NO₂ molecule.⁴⁹ This observation was in accordance with catalytic oxidation reaction. Once the O₂ supply was switched on, a new peak at 1380 cm⁻¹, assigned to free nitrate with asymmetric stretching, and the weak band at 1430 cm⁻¹, assigned to monodentate nitrite, were detected.^23,42,51 It suggested that nitrite was gradually inhibited and nitrates were formed dominantly in the selective oxidation of NO in the presence of O₂. Nevertheless, the bands at 1860, 1760 and 1520 cm⁻¹ weakened gradually after O₂ was introduced. In fact, these bands were attributed to NO⁺ and bidentate nitrate, respectively. This result implied that the interaction between NO⁺ and nitrates occurred easily and then transferred to a new species in the presence of O₂. Combined with the enhanced band at 1620 cm⁻¹, the new species might be identified as gaseous NO₂ molecule.

Fig. 7 In situ DRIFT spectra of Cr/Ti-PILC (a) and CrCe(0.25)/Ti-PILC (b) exposed to 4000 ppm NO for 60 min (spectrum was recorded at 10, 20 and 60 min) and followed by introduction of O₂ for 60 min at 300 °C.

However, it is worth noting that there were distinct differences in the details between two samples, which are clearly visible in Fig. 7c and d. First, a broad absorption peak between 3600 and 3000 cm⁻¹ was observed in the OH stretching region through cerium modification. It suggested that ceria were active and conducive to react with O₂ molecule to form new species, which led to the vibrations of surface hydroxyl species.⁵² Second, by comparing Cr/Ti-PILC with CrCe(0.25)/Ti-PILC in Fig. 7d, it was observed that the growth of the bands at 1620, 1430 and 1360 cm⁻¹ was significant, which was attributed to gaseous NO₂ molecule, monodentate nitrite and free nitrate, respectively. Hence, it was reasonable to deduce that cerium modification over Cr/Ti-PILC could not only form more nitrite to maintain the original oxidation pathway but could also produce more free nitrate and gaseous NO₂ molecule. This observation was consistent with the activity results. In fact, O₂ played a crucial role in the oxidation of NO. On one hand, O₂ could partly oxidize nitrite to nitrate. On the other hand, O₂ could react with oxygen vacancy to form superoxide radical (O₂⁻), which was conducive to further oxidize NO to nitrates. Particularly, the experimental results of DRIFTS revealed that O₂ adsorption was considered to be a rate-determining step, which obeyed the Langmuir–Hinshelwood mechanism (L–H model).

3.7 Catalytic mechanism

According to the abovementioned observations, a possible mechanism of cerium modification over Cr/Ti-PILC could be proposed to explain the phenomenon of catalytic oxidation NO. It is illustrated in Fig. 8 and described as follows: The dispersive CeOx and CrOx, especially CrO₃ and Ce₂O₃, might play a different role in catalytic oxidation because of their different active sites. For the Cr/Ti-PILC catalyst, NO was adsorbed on Cr⁶⁺ sites to form NO⁺ while NO was also adsorbed on vacancies based on chromium species to generate NO⁻, which were accompanied by partial charge transfer. Both NO⁺ and NO⁻ could be preferentially transferred to nitrites/nitrates by active oxygen species. Indeed, the unstable nitrites were oxidized to nitrate by O₂. Finally, the ion pair NO⁺NO₃⁻ was decomposed to two NO₂ molecules.³⁵ For cerium modified sample, considering that oxygen vacancy on ceria captures the gas-phase oxygen to form superoxide radical (O₂⁻), a new reaction pathway on the site of ceria was investigated. That is, the strong interaction between superoxide ions (O₂⁻) and NO resulted in the formation of free nitrates. After the release of NO₂, the oxygen vacancy was recovered reversibly. In a word, for the CrCe(0.25)/Ti-PILC catalyst, the original oxidation pathway of Cr/Ti-PILC was maintained and the formation of nitrates was clearly improved via superoxide ions. In summary, the entire catalytic mechanism obeyed the Langmuir–Hinshelwood mechanism (L–H model).

Fig. 8 Catalytic mechanism of NO oxidation over cerium modified Cr/Ti-PILC catalyst.

4. Conclusion

The catalytic activity of NO removal was clearly increased and the activity temperature window was advanced for the CrCe(0.25)/Ti-PILC catalyst in the selective catalytic oxidation of NO. The formation of oxygen vacancies and superoxide ions were confirmed by PL and EPR, respectively. Moreover, both the reversibility of gas (NO or O₂) adsorption and adsorption sites were evaluated by TPD profiles. The investigation of DRIFTS verified that the evolution of surface-bound species such as NO⁺, nitrites and nitrates. The mechanism of the CrCe(0.25)/Ti-PILC catalyst was attributed to the fact that cerium modification could facilitate the generation of oxygen vacancies via charge transfer, which maintained the original oxidation pathway of Cr/Ti-PILC. This aspect resulted in the increasing amount of superoxide ions (O₂⁻) and the generation of nitrates, which were conducive to the catalytic oxidation of NO.

Acknowledgements

This work was financially supported by the Assembly Foundation of The Industry and Information Ministry of the People's Republic of China 2012 (543), the National Natural Science Foundation of China (U1162119), Scientific Research Project of Environmental Protection Department of Jiangsu Province (2013003) and (201112), Research Fund for the Doctoral Program of Higher Education of China (20113219110009) and Industry-Academia Cooperation Innovation Fund Projects of Jiangsu Province (BY2012025).

References

1. Z. Y. Sheng, Y. F. Hu, J. M. Xue and X. M. Wang, J. Rare Earths, 2012, 30, 676.
2. M. F. Irfan, J. H. Goo and S. D. Kim, Appl. Catal., B, 2008, 78, 267.
3. M. P. Ruggeri, A. Grossale and I. Nova, Catal. Today, 2012, 184, 107.
4. A. Setiabudi, J. L. Chen and G. Mul, Appl. Catal., B, 2004, 51, 9.
5. K. Krishna, A. Bueno-López and M. Makkee, Appl. Catal., B, 2007, 75, 201.
6. K. Tikhomirov, O. Kröcher and M. Elsener, Appl. Catal., B, 2006, 64, 72.
7. K. Li, X. Tang and H. H. Yi, Chem. Eng. J., 2012, 192, 99.
8. K. V. Bineesh, D. R. Cho, S. Y. Kim, B. R. Jermy and D. W. Park, Catal. Commun., 2008, 9, 2040.
9. K. V. Bineesh, S. Y. Kim, B. R. Jermy and D. W. Park, J. Ind. Eng. Chem., 2009, 15, 207.
10. N. Jagtap and V. Ramaswamy, Appl. Clay Sci., 2006, 33, 89.
11. Q. Q. Huang, S. F. Zuo and R. X. Zhou, Appl. Catal., B, 2010, 95, 327.
12. C. M. Pradier, F. Rodrigues and P. Marcus, Appl. Catal., B, 2000, 27, 73.
13. H. D. Liu, L. Q. Wei and R. L. Yue, Catal. Commun., 2010, 11, 829.
14. W. Cai, Q. Zhong, S. L. Zhang and J. X. Zhang, RSC Adv., 2013, 3, 7009.
15. A. Masalov, O. Viagin and P. Mak simchuk, Langmuir, 2014, 145, 61.
16. S. Y. Christou and M. C. Alvarez-Galvan, Appl. Catal., B, 2011, 106, 103.
17. J. X. Zhang, S. L. Zhang, W. Cai and Q. Zhong, Appl. Surf. Sci., 2013, 268, 535.
18. R. O. Fuentes and R. T. Baker, J. Phys. Chem. C, 2009, 113, 914.
19. C. S. Wright, R. I. Walton, D. Thompsett, J. Fisher and S. E. Ashbrook, Adv. Mater., 2007, 19, 4500.
20. J. G. Yu, J. C. Yu, B. Cheng, S. K. Hark and K. Iu, J. Solid State Chem., 2003, 174, 372.
21. Z. Q. Zou, M. Meng and Y. Q. Zha, J. Phys. Chem. C, 2010, 114, 468.
22. H. Y. Huang and R. T. Yang, Langmuir, 2001, 17, 4997.
23. S. Kameoka, Y. Ukisu and T. Miyadera, Phys. Chem. Chem. Phys., 2000, 2, 367.
24. W. S. Kijlstra, D. S. Brands, E. K. Poels and A. Bliek, J. Catal., 1997, 171, 208.
25. W. Cai, Q. Zhong, S. L. Zhang and W. Zhao, Chem. Eng. J., 2014, 236, 223.
26. S. J. Zhang, Ultrason. Sonochem., 2012, 19, 767.
27. P. Li, Y. Xin and Q. Li, Environ. Sci. Technol., 2012, 46, 9600.
28. F. Larachi, J. Pierre and A. Adnot, Appl. Surf. Sci., 2002, 195, 236.
29. C. X. Liu, L. Chen and J. H. Li, Environ. Sci. Technol., 2012, 46, 6182.
30. X. Li, S. J. Wei and Z. L. Zhang, Catal. Today, 2011, 175, 112.
31. L. Chen, J. H. Li and M. F. Ge, J. Phys. Chem. C, 2009, 113, 21177.
32. G. Preda, A. Migani, K. M. Neyman and S. T. Bromley, J. Phys. Chem. C, 2011, 115, 5817.
33. X. F. Tang, Y. G. Li, X. Huang and Y. D. Xu, Appl. Catal., B, 2006, 62, 265.
34. S. L. Zhang, H. Y. Li and Q. Zhong, Appl. Catal., A, 2012, 435–436, 156.
35. X. Wang, H. Y. Chen and W. M. H. Sachtler, J. Catal., 2001, 197, 281.
36. Y. T. Li and Q. Zhong, J. Hazard. Mater., 2009, 172, 635.
37. J. J. Xu, Y. H. Ao, D. G. Fu and C. W. Yuan, Appl. Surf. Sci., 2008, 254, 3033.
38. A. Masalov, O. Viagin, P. Maksimchuk, V. Seminko and I. Bespalova, Langmuir, 2014, 145, 61.
39. M. Anpo, M. Che, B. Fubini, E. Garrone and E. Giamello, Top. Catal., 1999, 8, 189.
40. A. Martinez-Arias, M. Fernandez-Garcia, C. Belver, J. C. Conesa and J. Soria, Catal. Lett., 2000, 65, 197.
41. J. Matta, D. Courcot, E. Abi-Aad and A. Aboukaïs, Chem. Mater., 2002, 14, 4118.
42. W. S. Kijlstra, D. S. Brands, E. K. Poels and A. Bliek, J. Catal., 1997, 171, 208.
43. B. Q. Jiang, Z. B. Wu and Y. Liu, et al., J. Phys. Chem. C, 2010, 114, 4961.
44. Y. Chen, F. Lin and R. E. Shepherd, Inorg. Chem., 1999, 38, 973.
45. B. Azambre, L. Zenboury, F. Delacroix and J. V. Weber, Catal. Today, 2008, 137, 278.
46. K. Hadjiivanov, V. Avreyska, D. Klissurski and T. s. Marinova, Langmuir, 2002, 18, 1619.
47. E. Ivanova, K. Hadjiivanov, D. Klissurski, M. Bevilacqua, T. Armaroli and G. Busca, Microporous Mesoporous Mater., 2001, 46, 299.
48. K. Hadjiivanov, B. Tsyntsarski and T. Nikolova, Phys. Chem. Chem. Phys., 1999, 1, 4521.
49. N. Tang, Y. Liu, H. Q. Wang and Z. B. Wu, J. Phys. Chem. C, 2011, 115, 8214.
50. J. C. S. Wu and Y. T. Cheng, J. Catal., 2006, 237, 393.
51. P. Sazama, L. Čapek, H. Drobná and Z. Sobalík, J. Catal., 2005, 232, 302.
52. B. Azambre, I. Atribak and A. Bueno-Lopez, J. Phys. Chem. C, 2010, 114, 13300.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra05471g

---