Study on the simultaneous reduction of diesel engine soot and NO with nano-CeO₂ catalysts

Abstract

Three nanometer cerium oxide (nano-CeO₂) catalysts were prepared by a precipitation method in order to reduce particulate matter (PM) and nitrogen oxide (NO_x) emissions from diesel engines through after-treatment technology. X-ray diffraction (XRD), Raman spectroscopy, scanning electron microscopy (SEM), and Brunauer–Emmett–Teller (BET) analysis were employed to characterize the catalysts. The catalytic activity was evaluated by analyzing the ignition temperature, peak temperature of soot combustion and conversion of nitric oxide (NO) to nitrogen (N₂). The results show that the average particle sizes of the three prepared CeO₂ catalysts are 7, 12 and 20 nm, respectively, which are much smaller than that of commercial CeO₂ (comm-CeO₂); also, the prepared CeO₂ catalysts have larger specific surface areas. The catalytic activity of CeO₂ increases as its specific surface area increases. The efficiencies of the catalysts for the oxidation of CeO₂ on soot are 20, 12 and 7 nm, in descending order. The ignition temperatures of soot combustion were reduced by 124 °C, 109 °C and 93 °C and the peak temperatures decreased by 185 °C, 104 °C and 102 °C when the three CeO₂ catalysts were employed. The conversion of NO is greater than 70% at 370 °C to 550 °C, which indicates that prepared CeO₂ has a wider temperature window. The presence of oxygen vacancies on the CeO₂ surface greatly increases the adsorption energy of NO and can promote the reduction reaction of NO.

---

1. Introduction

Diesel engines are widely used by virtue of their favourable power performance, economy and durability. However, due to the high temperature of diffusion combustion and the presence of local over-rich mixtures, more particulate matter (PM) and nitrogen oxides (NO_x) are generated by diesel engines. The main components of PM are soot, soluble organic matter (SOF), and sulphate and metal substances; the diameters of 90% of these particles are smaller than 1 µm,¹ which is severely detrimental to human health and the atmospheric environment. Nitrogen oxides (NO_x) are mainly composed of nitric oxide (NO) and nitrogen dioxide (NO₂); NO accounts for about 90% of NO_x.² NO_x and volatile organic compounds react with each other and form “photochemical smog” close to the surface of the earth, causing long-term effects on the climate, human health and ecological environments.³ A trade-off relationship exists between soot and NO_x emissions for diesel engines. It is very difficult to reduce these emissions by internal purification alone. Therefore, the development of new diesel engine emission control technologies to effectively reduce these two emissions has become a subject of concern to both domestic and foreign engine researchers.

Diesel particulate filters (DPFs) can effectively reduce diesel particulate emissions with a collection efficiency higher than 90%.⁴ However, most of the particles are deposited in the filter, which leads to increasing exhaust back pressure and deterioration of the economic and dynamic efficiency of the engine. In order to solve this problem, passive regeneration techniques such as chemical catalysis can be utilized to reduce the activation energy of particle oxidation and the ignition temperature. Thus, the particles can be burnt out by the energy from diesel exhaust under high gas temperatures to achieve the goal of DPF regeneration.^5,6 In addition, by coating a proper catalyst within the DPF, NO_x reduction from exhaust gas can be achieved, thus realizing the purpose of reducing both PM and NO_x emissions from diesel engines. However, it is difficult to select a highly active catalyst for particle oxidation.

Because of their rich electronic energy level structures, rare earth-based catalysts exhibit unique chemical and physical properties; therefore, these catalysts play important roles in the fields of industrial catalysis,^7,8 energy utilization⁹ and environmental governance.^10,11 Among the existing rare earth oxides, cerium dioxide has attracted much attention in the catalysis field due to its low cost, unique crystal structure and Ce³⁺/Ce⁴⁺ reversible conversion.^12–20 Its qualities are not limited to catalytic oxidation of soot; it can also be used in the transformation of hydrocarbons (HC), carbon monoxide (CO) and NO_x into H₂O, carbon dioxide (CO₂) and N₂. As the core of an oxygen storage/oxygen release material, CeO₂ can rapidly store oxygen when the oxygen content in exhaust is excessively high, then release the oxygen when the oxygen content is low; hence, it plays the role of an “oxygen buffer”. It is therefore essential to improve the catalytic activity and prolong the catalytic life of CeO₂.

The health risks of CeO₂ are related not only to its toxicity, but also to its exposure scale.^21,22 The aim of this study is to coat CeO₂ as a catalyst on a DPF carrier, at the current technological level, with minimal peeling. Even if a small amount of nano-CeO₂ detached from the carrier, it would accumulate with the soot particles generated in the engine combustion process rather than dispersing in water or soil. Therefore, its concentration would not exceed 1 mg L⁻¹, which is considered to be a safe ecological concentration.

The application of CeO₂ in after-treatment technology for diesel engines has become a prevalent research topic in recent years. With regards to experimental investigation, research by Ulrich et al.²³ showed that cerium-based fuel additives have only slight effects on engine performance and secondary environmental pollution. Teraoka²⁴ and Kagawa²⁵ adopted a temperature programmed reaction (TPR) technique and simulated the composition of diesel engine tail gas to study the reaction of mixtures with increasing temperature with close contact between a CeO₂ catalyst and dry soot; they investigated the ability of CeO₂ to reduce PM. Atribak et al.²⁶ found that the catalytic activity of CeO₂ on soot decreased with increasing temperature, and the oxidation reduction activity was higher when the catalyst was in close contact with soot. Daturi et al.²⁷ utilized H₂ to greatly restore the surface of CeO₂ because high concentrations of surface oxygen vacancies can degrade NO_x without reducing agents. Furthermore, through infrared (IR) and mass spectrometry (MS) analysis, it was confirmed that the NO_x degradation activity was directly related to the oxygen vacancy concentration on the CeO₂ surface and the quantity of oxygen exchange. Furthermore, with respect to theoretical calculations, Watson et al.²⁸ found that the formation energy of CeO₂ oxygen vacancies was directly related to the catalytic activity of CeO₂. Deng²⁹ and Pawan³⁰ showed that oxygen vacancies were more likely to form on the surface of CeO₂ by molecular mechanics studies. Jiang et al.³¹ studied the relationship between the formation energy of CeO₂ bulk phase oxygen vacancies and external oxygen pressure (P_O2) as well as temperature; they pointed out that the formation energy of oxygen vacancies decreased with decreasing P_O2 and increasing temperature. Li et al.³² found that CeO₂ had a high adsorption capacity for NO at 303 K (4.9 molecules per nm²) and that 40% of the NO in CeO₂ was due to chemical adsorption in the form of NO₂⁻. Furthermore, due to a change in the IR adsorption frequency, it was assumed that NO also played a role in the conversion of Ce⁴⁺/Ce³⁺ on the surface of CeO₂.

Previous researchers have carried out constructive investigations on soot catalytic oxidation and the conversion of NO_x with CeO₂. However, basically, no research has combined both theoretical and experimental investigations of the control of PM and NO_x emissions from diesel engines. Therefore, this paper mainly studies the reduction of both soot and NO_x using nano-CeO₂ catalysts. Three groups of CeO₂ with different microstructures were first prepared, and then their microstructures were studied by characterization methods. Afterwards, the catalytic activities of the three CeO₂ catalysts were evaluated by examining their soot oxidation characteristics and NO conversion. Finally, the adsorption energy of NO on the surface of CeO₂ was calculated theoretically.

2. Experimental

2.1. Catalyst preparation

Nano-sized CeO₂ was prepared by a precipitation method. The raw material, Ce(NO₃)₃·6H₂O, was used to prepare salt solutions with certain concentrations, with ammonia as the precipitating agent. Different microstructures of CeO₂ were prepared by positive and negative precipitation methods. Absolute ethyl alcohol was added to the solution as a surface active agent to reduce the surface tension between water molecules, thus inhibiting aggregation. Ultrasonic wave oscillation was used to disperse precipitated particles during the precipitation process. Meanwhile, the reaction temperature was controlled by a thermostatic water bath. The samples were placed in an air-drying oven at 110 °C to dry overnight; then, they were heated at 10 °C min⁻¹ and roasted at 450 °C for 4 hours in a muffle furnace. The prepared samples were roughly ground in a ceramic crucible and finely ground in an agate crucible, then collected and set aside.

2.2. Catalyst characterization

The powder X-ray diffraction (XRD) patterns were obtained using a diffractometer (D8 ADVANCE, Bruker, Germany) adopting Cu Kα radiation (λ = 0.15418 nm). The X-ray tube was operated at 40 kV and 30 mA. The X-ray diffractograms were recorded at 0.02° intervals within the 20° ≤ 2θ ≤ 80° range at a scanning velocity of 4° per min.

The Raman spectra were recorded with a Renishaw inVia Reflex spectrometer at room temperature and atmospheric pressure. The emission line at 514.5 nm from an Ar⁺ ion laser was focused on the sample under a microscope; the analyzing spot was about 1 mm. The power of the incident beam on the sample was 3 mW. The wavenumber values obtained from the Raman spectra were accurate to 2 cm⁻¹.

Scanning electron microscopy (SEM) was carried out on a field emission scanning electron microscope (S-4800, Hitachi, Japan). Samples were coated with platinum to improve conductivity. The amplification range of the S-4800 was 20 times to 8 × 10⁵ times. The minimum resolution was 1.0 nm, and the electron acceleration voltage range was 0.5 to 30 kV.

N₂ adsorption–desorption isotherms of the samples were measured using an ASAP 2460 aperture analyzer from Micromeritics Instrument Company. The samples were degassed at 180 °C under vacuum for no less than 6 hours before the tests.

2.3. Catalytic testing

The evaluation indicators of catalyst activity mainly include soot ignition temperature (T_i), peak temperature (T_m) and the conversion of NO to N₂ (X_NO), where T_i is defined as the temperature when the total mass of the sample is reduced by 5% in the reaction and T_m is the temperature corresponding to the maximum weightlessness rate. In a catalytic activity evaluation system, lower T_i and T_m and higher X_NO represent better catalytic activity.

The test engine was a 4-cylinder, turbocharged, inter-cooled, electronic control common-rail diesel engine, and its main technical parameters are shown in Table 1. PM samples were collected under an ESC 13-mode test cycle using the AVL SPC472 partial flow particle acquisition system. Thermogravimetric analysis (TGA) of the particle samples was conducted on a METTLER TGA/DSC1 thermogravimetric analyzer, which has a built-in high precision electronic balance and temperature sensor. In the thermogravimetric tests, the sample weight was 3 mg, and the mass fraction of oxygen atmosphere was about 12%, which is close to the oxygen content in diesel exhaust. High-purity N₂ was used as the protective gas, and its flow velocity was constant at 100 mL min⁻¹. The programmed temperature range was 40 °C to 800 °C, and the heating rate was maintained at 15 °C min⁻¹.

Table 1 Specifications of the testing engine

Item	Specification
Type	4-Cylinder, in-line, turbocharged and intercooled
Bore × stroke (mm)	105 × 118
Combustion chamber type	Direct injection ω type
Compression ratio	17.5
Displacement (L)	4.09
Max. torque/speed (Nm/r/min)	400/1500
Rated power/speed (kW/r/min)	95/2600
Max. injection pressure (MPa)	160
Fuel injection system	Electronic controlled high pressure common-rail

---

Fig. 1 shows the catalytic activity determination reaction system, which is composed of a gas distribution system, a catalytic reaction section and a measurement section. Firstly, the collected diesel soot particles and CeO₂ were evenly mixed and placed in a fixed-bed quartz reactor. Secondly, the reactor was placed in an electric heating furnace with a programmable temperature controller. The programmed temperature was constantly increased until the soot completely underwent oxidation-combustion, and 5% NO + 10% O₂ + 85% He were used as reaction gases to simulate diesel exhaust gas atmosphere; He was the carrier gas. Thirdly, the NO conversion (X_NO) at each reaction temperature was calculated from the NO concentration in the simulated exhaust ([NO]_inlet) and the outlet N₂ concentration ([N₂]); these data were used as an indicator to measure the catalyst activity. The N₂ concentration was detected with a gas chromatograph. The specific calculation formula is shown in eqn (1).

X_NO = 2[N₂]/[NO]_inlet × 100% (1)

Fig. 1 Technical drawing of the catalytic activity evaluation system. (1)–(3): Cylinders ((1) He, (2) He + NO, (3) O₂). (4) Flowmeter. (5) Mixed gas cylinders. (6) Battery. (7) Fixed bed quartz reactor. (8) Reactor. (9) Program temperature controller. (10) Gas chromatograph. (11) Computer.

3. Results and discussion

3.1. Catalyst characterization

The XRD profiles of the CeO₂ catalysts are shown in Fig. 2. It can be seen that the diffractograms of CeO₂ show the diffraction peaks (28.5, 33.1, 47.5, 56.3 and 59.0°) of cerianite with a fluorite-like structure (JCPDS 34-0394). The crystal structures of the three CeO₂ catalysts did not change. Using the Scherrer formula and the half peak width of the CeO₂(111) diffraction peak, the average particle sizes of the three CeO₂ samples were found to be 7, 12 and 20 nm, which are significantly lower than the particle size (100 nm) of the commercial CeO₂ catalyst purchased from Sinopharm, marked as comm-CeO₂.

Fig. 2 XRD patterns of the CeO₂ samples.

Fig. 3 shows the Raman patterns of CeO₂ with different structures; there are two main peaks. The absorption peak near 450 cm⁻¹ is attributed to the typical F_2g vibrating belt of the CeO₂ cubic crystal structure, which is attributed to the absorption modes of oxygen ion vibration and the surrounding cation vibration, representing the vibration between O²⁻ and Ce⁴⁺. The relatively weak absorption peak of the Raman shift at 570 cm⁻¹ is attributed to lattice defects, which result from oxygen vacancies. This indicates that the different structures of nano-CeO₂ have lattice defects, and there are oxygen vacancies in their structures. The shift of the Raman peak may be due to CeO₂ crystal sintering caused by the loss of lattice oxygen, which is caused by thermoinduction. The intensities of the F_2g Raman vibration peaks of CeO₂ are different. A higher intensity of Raman vibration peaks indicates that the CeO₂ is more crystalline and that the long-range order is also superior. For the soot oxidation reaction, the oxygen vacancy position in the CeO₂ catalyst is the active site, which implies that greater oxygen vacancy activity will result in better catalytic oxidation of soot.

Fig. 3 Raman shifts of the CeO₂ samples.

The particle diameters in the unit areas were analyzed by Nano Measure software using the SEM images to calculate the average particle diameters. Fig. 4 shows the SEM images of the CeO₂ samples. From Fig. 4(a), the particle diameter range of the first nano-CeO₂ sample is 22.1 to 35.2 nm, with an average particle diameter of 29 nm. From Fig. 4(b), the particle diameter range of the second nano-CeO₂ sample is 18.9 to 25.1 nm, with an average particle diameter is 22 nm. From Fig. 4(c), the particle diameter range of the third nano-CeO₂ sample is 15.5 to 22.3 nm, with an average particle diameter of 19 nm. The appearances of the three groups of particles are relatively uniform; the particles are mostly spherical or near-spherical, with clear boundaries. Furthermore, they show the same trend of average particle diameter as obtained through the Scherrer formula. This proves that the particle aggregation is effectively controlled in the process of preparation, which improves dispersion.

Fig. 4 SEM images of CeO₂ samples: (a) 20 nm; (b) 12 nm; (c) 7 nm.

Table 2 shows the surface areas, pore volumes and average pore diameters of the three CeO₂ samples. The surface areas were determined using the linear portion of the Brunauer–Emmett–Teller (BET) model, and the average pore sizes were calculated using the Barrett–Joyner–Halenda (BJH) formula from the desorption branch of the N₂ adsorption isotherm prior to these measurements. As shown in Table 2, the surface areas of the prepared CeO₂ samples are significantly greater than that of comm-CeO₂. A greater surface area indicates that there are more active sites on the surface of CeO₂ per unit mass. Furthermore, a greater surface area provides more opportunities for the catalyst to be in close contact with the reactants. These are all favorable characteristics for the adsorption and activation of reactant molecules.

Table 2 Surface areas, pore volumes and pore diameters of the CeO₂ samples

Sample	Surface (m^2 g^−1)	Pore volume (cm^3 g^−1)	Average pore diameter (nm)
7 nm-CeO2	59	0.04	2.9
12 nm-CeO2	68	0.09	5.7
20 nm-CeO2	89	0.19	8.5
Comm-CeO2	8.1	0.04	2.8

---

3.2. Activity tests

TGA is used to obtain the relationship (TG curve) between the mass and temperature of a sample under a temperature control program. The differential thermogravimetric curve (DTG) represents the relationship between the rate of mass change and the temperature of the sample, which is obtained from the first-order derivative of the TG curve. The particle oxidation characteristics of the diesel engine were analyzed with the thermogravimetric method, which effectively reflects the oxidation and combustion properties of PM in a specific atmosphere.

Fig. 5 shows the TG curves of PM in the simulated exhaust gas atmosphere, which are the TG curves and DTG curves of particle weightlessness. As shown in the figure, the two obvious weightlessness segments in the TG curve without CeO₂ correspond to the two peaks in the DTG curve. In the TG curve, the 23% weightlessness in the low temperature segment was due to SOF, the 74% weightlessness in the high temperature segment was due to dry soot, and the 3% residual particles were mainly metals and sulphates at the end stage of the oxidation.

Fig. 5 TG and DTG curves of diesel particles.

In order to explore the effects of CeO₂ on the catalytic oxidation properties of particles, the three CeO₂ catalysts and diesel particulate samples of equal quality were mixed separately for the TG tests. The weightlessness of the particles was measured in percentage, and the three different TG and DTG curves are shown in Fig. 5. As shown in the DTG curves, there are three peak weightlessness rates of the particles with CeO₂, which can be attributed to the volatilization of HCs with low boiling points, the oxidation of HCs with high boiling points, and the oxidative combustion of dry soot. Compared with the DTG curves of the pure particles, the peak temperature of the particles with CeO₂ decreased; this proves that the addition of CeO₂ promoted the oxidative combustion of some components of SOF and dry soot.

Table 3 gives the weightlessness characteristic parameters of the pure particles and the particles with different CeO₂ microstructure parameters. T_s is defined as the temperature corresponding to the peak weightlessness rate of HCs with low boiling points in SOF. T_p is defined as the temperature corresponding to the peak weightlessness rate of HCs with high boiling points in SOF. T_i is the ignition temperature of dry soot, and T_m is the combustion peak temperature of dry soot. By comparing the TG curves, it is found that the weightlessness of SOF is basically coincident with the volatile phases. T_s, caused by the volatilization of HCs with low boiling points, is about 200 °C. The particles with the CeO₂ catalyst appear to enable oxidation combustion of high boiling point HCs, and T_p is about 300 °C. Thus, it can be seen that CeO₂ has little effect on low boiling point HCs; however, it promotes the oxidative combustion of high boiling point HCs. As for the catalytic oxidation of dry soot with CeO₂, both T_i and T_m decrease to different degrees. The corresponding DTG curve moves to the low temperature segment, and both T_i and T_m reach the lowest values with 20 nm CeO₂. This is consistent with the conclusions of the specific surface area analysis above. The catalytic oxidation efficiencies of the three CeO₂ catalysts for the soot particles are in the order of 20 nm > 12 nm > 7 nm. The T_i values decrease by 124 °C, 109 °C and 93 °C, respectively, and the T_m values also decrease by 185 °C, 104 °C and 102 °C, respectively.

Table 3 Characteristic parameters of the weightlessness of diesel particles

Sample	SOF weightlessness (°C)		Soot weightlessness (°C)
	T s	T p	T i	T m
None	191	—	472	598
7 nm-CeO2	200	300	379	496
12 nm-CeO2	199	295	363	494
20 nm-CeO2	207	302	348	413

---

Fig. 6 gives the NO conversion curves of the different CeO₂ samples. As the figure shows, NO conversion with the prepared CeO₂ samples is significantly higher than with comm-CeO₂. As the temperature increases, the NO conversion increases initially and then decreases. The NO conversion with 20 nm CeO₂ reaches the highest value of 70% at 350 °C. The NO conversions with the three CeO₂ samples are all higher than 70% at 370 °C to 550 °C, which proves that the prepared CeO₂ has a wide temperature window.

Fig. 6 Conversion curves of NO to N₂.

3.3. Surface DFT calculations

3.3.1. CeO₂ and CeO₂(110) models. Kohn–Sham density functional theory (DFT) calculations were performed with the generalized gradient approximation (GGA) using the Perdew–Burke–Ernzerhof (PBE) exchange–correlation functional.^33–35 Super cell models of CeO₂ and CeO₂(110) are shown in Fig. 7. Firstly, geometry optimization of the unit cell structure was conducted. The cut-off energy was set as 400 eV, and the self-consistent field energy convergence criterion was 1.0 × 10⁻⁶ eV. The optimal convergence energy was smaller than 1.0 × 10⁻³ eV, and the force threshold was set to 0.02 eV Å⁻¹. The Brillouin zone was sampled with a 2 × 2 × 1 k point grid according to the Monkhorst–Pack scheme. The vacuum gap was set to 10 Å.

Fig. 7 Schematic cell models of CeO₂ and CeO₂(110)-2 × 2.

The lattice constant of CeO₂ obtained from the optimization is 5.47 Å, which is in good agreement with the experimental value of 5.41 Å.³⁶ The reliability of the selected model was verified.

The adsorption energies of the NO molecules on the surface of CeO₂ were quantified by the adsorption energy (E_ad), which is defined in eqn (2):

E_ad = E(CeO₂) + E(NO) − E(NO/CeO₂) (2)

where E(NO/CeO₂) and E(CeO₂) are the total energies of the CeO₂ films with and without adsorbed NO, and E(NO) is the total energy of the NO molecules in the gas phase.

3.3.2. Analysis of NO on the CeO₂(110) surface. Fig. 8 shows the adsorption models of NO in a single oxygen vacancy and in an oxygen vacancy pair. The calculation results indicate that the adsorption energy of NO on the clean CeO₂(110) surface is 12.9 kJ mol⁻¹; this is due only to physical adsorption. This shows that the adsorption of NO on the surface of CeO₂(110) is caused by van der Waals forces, and there is no chemical reaction between the two species. When there is an oxygen vacancy on the surface of CeO₂(110), the adsorption energy of NO is 128.1 kJ mol⁻¹, which is due to chemical adsorption. This shows that the adsorption of NO on the CeO₂(110) surface is caused by a chemical bond generated from a chemical reaction. When there is an oxygen vacancy pair on the surface of CeO₂(110), the adsorption energy is 332.1 kJ mol⁻¹. It can be seen that the presence of oxygen vacancies greatly increases the adsorption energy of NO and can promote the reduction reaction of NO on the surface of CeO₂(110).

Fig. 8 Optimized structures of NO adsorbed on an oxygen vacancy.

3.4. Reaction mechanism

Soot is usually formed because of the thermal decomposition of fuel at a high temperature under oxygen deficit conditions. CeO₂ can undergo a transformation from the stoichiometric CeO₂ (+4) valance state to the Ce₂O₃ (+3) state via a relatively low-energy reaction. CeO₂ supplies the oxygen for the reduction of the soot and is converted to Ce₂O₃ according to chemical eqn (3).³⁷ CeO₂, as an oxidation catalyst, also lowers the carbon combustion activation temperature and thus enhances soot oxidation, promoting complete combustion.

4CeO₂ + C(soot) → 2Ce₂O₃ + CO₂ (3)

Because of its high thermal stability, Ce₂O₃ formed from the reaction of soot remains active after enhancing the initial combustion and is reoxidized to CeO₂ through the reduction of NO.³⁷ NO can be reduced to N₂ according to chemical eqn (4).³⁸ Thus, NO emissions can be inhibited to some extent; meanwhile, CeO₂ can be regenerated.

2Ce₂O₃ + 2NO → 4CeO₂ + N₂ (4)

In summary, the catalytic activity of CeO₂ toward soot oxidation suggests a new method to promote the internal online regeneration of DPF.

4. Conclusions

In this study, three different microstructures of nano-CeO₂ were prepared as test subjects, and the effects of the microstructures on the catalytic activity of CeO₂ were systemically studied with characterization methods. Furthermore, the activities of the catalysts were evaluated by analysis of the ignition temperature and peak temperature of soot combustion as well as the conversion of NO to N₂. Several conclusions can be drawn, as follows:

The particle sizes of the three CeO₂ catalysts prepared by a precipitation method are 7, 12 and 20 nm, respectively, which are significantly smaller than the particle size of comm-CeO₂. Moreover, compared with comm-CeO₂, the specific surface areas of the prepared CeO₂ samples are relatively large, which could effectively improve the oxidation characteristics of soot and promote the effective conversion of NO.

The CeO₂ catalytic oxidation efficiencies for soot decrease successively in the order of 20 nm, 12 nm and 7 nm. The ignition temperatures of soot combustion are reduced by 124 °C, 109 °C and 93 °C, respectively, and the peak temperatures are reduced by 185 °C, 104 °C and 102 °C, respectively, with the three CeO₂ catalysts.

As the particle sizes of the three prepared CeO₂ samples are relatively small, the effects of particle size on their catalytic activities can be negligible. The catalytic activity is closely related to the specific surface area; this activity increases with increasing specific surface area.

With increasing temperature, the conversion of NO first increases and then decreases. NO conversion with 20 nm-CeO₂ reaches the highest value of 70% at 350 °C. The conversions of the three CeO₂ catalysts are greater than 70% at 370 °C to 550 °C, which indicates that the prepared CeO₂ has a wide temperature window.

The presence of oxygen vacancies on the CeO₂ surface greatly increases the adsorption energy of NO and promotes the reduction reaction of NO on the CeO₂(110) surface.

Acknowledgements

We would like to acknowledge the financial support from the Natural Science Major Research Project of Jiangsu Province University (14KJA470001), the State Key Laboratory of Engines, Tianjin University (K2016-05), the Natural Science Foundation of Jiangsu Province, China (BK20160538) and the Graduate Student Scientific Research Innovation Project of Jiangsu Province University (KYLX_1038).

Notes and references

1. B. R. Stanmore, J. F. Brilhac and P. Gilot, Carbon, 2001, 39, 2247–2268.
2. A. Santhosham and P. Aghalayam, RSC Adv., 2016, 6, 59513–59526.
3. F. R. Cassee, E. C. van Balen, C. Singh, D. Green and H. Muijser, Crit. Rev. Toxicol., 2011, 41, 213–229.
4. G. E. Batley, B. Halliburton, J. K. Kirby, C. L. Doolette and D. Navarro, Environ. Toxicol. Chem., 2013, 32, 1896–1905.
5. S. Chatterjee, R. Conway, S. Viswanathan, M. Blomquist, B. Klueseuer, S. Anderson, SAE conference paper, 2003-01-0048, 2003, pp. 1–19.
6. M. Mohr, A. M. Forss and U. Lehmann, Environ. Sci. Technol., 2006, 40, 2375–2383.
7. M. Issa, C. Petit, A. Brillard and J. F. Brilhac, Fuel, 2008, 87, 740–750.
8. D. Fino and V. Specchia, Powder Technol., 2008, 18, 64–73.
9. Y. Guo, G. Lu, Z. Zhang, S. Zhang, Y. Qi and Y. Liu, Catal. Today, 2007, 126, 296–302.
10. R. J. Gorte, AIChE J., 2010, 56, 1126–1135.
11. G. A. Deluga, J. R. Salge, L. D. Schmidt and X. E. Verykios, Science, 2004, 303, 993–997.
12. V. D. A. Pierer and B. M. Weckhuysen, Angew. Chem., Int. Ed., 2002, 114, 4924–4926.
13. E. Aneggi, C. D. Leitenburg, J. Llorca and A. Trovarelli, Catal. Today, 2012, 197, 119–126.
14. E. P. Mahofa, T. B. Narsaiah, C. S. Chakra and P. Kumar, Materials Today Proceedings, 2015, 2, 4451–4456.
15. G. R. Li, D. L. Qu, Z. L. Wang, C. Y. Su, Y. X. Tong and L. Arurault, Chem. Commun., 2009, 48, 7557–7559.
16. M. Sugiura, M. Ozawa, A. Suda, T. Suzuki and T. Kanazawa, Bull. Chem. Soc. Jpn., 2005, 78, 752–767.
17. J. Yu, Z. C. Si, M. Zhu, X. D. Wu, L. Chen, D. Weng and J. S. Zou, RSC Adv., 2015, 102, 83594–83599.
18. M. Annamalai, B. Dhinesh, K. Nanthagopal, P. Sivaramakrishnan and J. R. Lalvani, Energy Convers. Manage., 2016, 123, 372–380.
19. E. Aneggi, M. Boaro, C. D. Litenburg, G. Dolcetti and A. Trovarelli, J. Alloys Compd., 2006, 37, 1096–1102.
20. G. Dutta, U. V. Waghm, T. Baidya, M. S. Hegde, K. R. Priolkar and P. R. Sarode, J. Am. Chem. Soc., 2006, 18, 3249–3256.
21. P. C. Shukla, T. Gupta, N. K. Labhsetwar and A. K. Agarwal, RSC Adv., 2016, 61, 55884–55893.
22. H. X. Mai, L. D. Sun, Y. W. Zhang, R. Si, W. Feng, H. P. Zhang, H. C. Liu and C. H. Yan, J. Phys. Chem., 2005, 109, 24380–24385.
23. A. Ulrich and A. Wichser, Anal. Bioanal. Chem., 2003, 377, 71–81.
24. Y. Teraoka, K. Nalano, S. Kagawa and W. F. Shangguan, Appl. Catal., B, 1995, 5, 181–185.
25. S. Kagawa, W. F. Shangguan and Y. Teraoka, Appl. Catal., B, 1996, 8, 217–227.
26. I. Atribak, A. Bueno-Lopez and A. Garcia-Garcia, J. Catal., 2008, 259, 123–132.
27. M. Daturi, N. Bion, J. Saussey, J. C. Lavalley, C. Hedouin, T. Sequelong and G. Blanchard, Phys. Chem. Chem. Phys., 2001, 5, 252–255.
28. M. Nolan, S. C. Parker and G. W. Watson, Surf. Sci., 2005, 595, 223–232.
29. D. Changshun, H. Qingqing, Z. Xiying, H. Qun, S. Wenli and L. Caiyuan, Appl. Surf. Sci., 2016, 389, 1033–1049.
30. K. Pawan, K. Parmod and R. C. Meena, J. Alloys Compd., 2016, 672, 543–548.
31. Y. Jiang, J. B. Adams, M. V. Schilfgaarde, R. Sharma and P. A. Crozier, Appl. Phys. Lett., 2005, 87, 141917–141920.
32. G. Li, A. K. Kaneko and S. Ozeki, Langmuir, 1997, 13, 5894–5899.
33. M. Gaus, Q. Cui and M. Elstner, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2014, 4, 49–61.
34. K. Lejaeghere, V. V. Speybroeck, G. V. Oost and S. Cottenier, Crit. Rev. Solid State Mater. Sci., 2012, 39, 1–24.
35. L. Juan, W. Chen, L. Tongxiang and L. Wensheng, Appl. Surf. Sci., 2016, 390, 273–282.
36. M. Fu, J. Rare Earths, 2014, 32, 153–158.
37. V. Sajith, C. B. Sobhan and G. P. Peterson, Adv. Mech. Eng., 2010, 2, 144–148.
38. F. Bozek, J. Dvorak and J. Mares, WSEAS Transactions Topics Environment, 2011, 7, 233–243.

---