Preparation of Ce x Zr1-x O2 by Different Methods and Its Catalytic Oxidation Activity for Diesel Soot.

Novel Ce x Zr1-x O2 (x = 0.67, 0.8, 0.9, 1.0) catalysts were designed and synthesized by solvothermal, calcination, and sol-gel methods and were used to catalyze oxidation of soot from diesel vehicle exhaust. The influence of catalysts synthesized by different methods and Ce/Zr molar ratios on the performance was investigated. These catalysts were characterized by XRD, N2 adsorption-desorption, FT-IR, TEM, XPS, H2-temperature programmed reduction (TPR), and O2-temperature programmed desorption (TPD) techniques. The results indicated that Ce0.8Zr0.2O2 prepared by the calcination method has excellent activity and stability at low temperature. The soot ignition point is 322 °C, and the ratio of soot conversion reaches 90% at 497 °C, which is lower than that from the solvothermal and sol-gel methods. The XRD, Raman, SEM, XPS and H2-TPR results reveal that the structure and oxygen adsorption properties are crucial to soot oxidation activity, and Zr4+ is successfully doped into the CeO2 lattice and forms a homogeneous solid solution. Nanostructured Ce0.8Zr0.2O2 with 110.2 m2/g surface areas is produced. The proportion of chemical oxygen and surface adsorbed oxygen in the catalyst prepared from the calcination method is the highest at 23.18%. The structure may lead to charge imbalance, unsaturated bonds, and oxygen vacancies, thus increasing the adsorption of oxygen on the catalyst surface.

Introduction

Recently, diesel engines have been widely utilized for heavy tools and trucks due to their lower fuel consumption and higher thermal efficiency than gasoline engines. In addition, the number of diesel vehicles has also been increasing.[1] However, diesel engine emissions, such as gas phase nitrogen oxides (NO), particulate matter (PM), unburned hydrocarbons (HCs), and carbon monoxide (CO), are considered a major source of air pollution.[2] Until now, diesel particulate filters (DPFs) combined with oxidized catalysts have been the most promising after-treatment technology, removing soot particulates in the range of 200–500 °C.

Many studies have been devoted to developing effective soot oxidation catalysts, such as noble metals, perovskites, spinel-type oxides, alkaline metal oxides, rare earth metal oxides, and transition metal oxides.[3−6] Recently, Ce-based catalysts have become increasingly notable because of their excellent redox properties and remarkable oxygen storage capacity. However, pure CeO2 has several shortcomings, for instance, its poor activity and easy sintering. Many attempts have been made to combine pure CeO2 with other transition metals.[7−12] Wang et al.[13] used the impregnation method to dope Cs into 3%Co/Ce0.5Sn0.5O2 to obtain a 10%Cs/3%Co/Ce0.5Sn0.5O2 catalyst with excellent soot oxidation properties. Experiments show that Cs doping not only promotes the generation of more surface oxygen defects and the activation of surface chemisorbed oxygen but also greatly improves the mobility of surface active oxygen species. Zuo et al.[14] prepared Co–Ce composite oxide catalysts by citric acid complexation. Due to the interaction between Co3O4 and CeO2, the oxygen adsorption and desorption capacity and the redox capacity were enhanced, and the effects of the redox mechanism and the oxygen spillover mechanism were enhanced at the same time. Reactive oxygen species (superoxide and peroxide) and carbon–oxygen intermediates (carbonyl and formate species) have also been found in soot combustion, and the reaction to generate formate species is an important part of the soot–O2 reaction. Liu et al.[15] prepared Ce0.84Zr0.16O2 catalyst with a cubic fluorite structure by the sol–gel method. They reported that amorphous Ce0.84Zr0.16O2 catalysts possess a higher contact area between the catalyst and soot and have more reactive oxygen species on the catalyst surface. Zhang et al.[16] prepared CeZr1–O2 (x = 0.27, 0.50, and 0.73) by the coprecipitation method and then aged them at 700–1000 °C to study the changes in reducibility and soot oxidation activity during the aging process. It was found that the H2 consumption in the Ce0.27Zr0.73O2 catalyst increased with increasing aging temperature, and the catalyst showed a strong reduction ability even at 1000 °C. Zhang et al.[17] doped Al2O3, ZrO2, La2O3, and Y2O3 into Pt/CeO2–MnO catalysts by a coprecipitation method to improve the thermal stability of the catalyst and improve the soot thermal stability and catalytic activity. However, little research has been conducted on the comparison of CeZr1–O2 catalysts prepared by hydrothermal, sol–gel, and direct calcination methods for catalyzing soot oxidation. Therefore, we report the incorporation of Zr into a CeO2 lattice formed a CeZr1–O2 solid solution by hydrothermal, sol–gel, and direct calcination to oxidize soot. The effects of different preparation methods and molar ratios of Ce/Zr in CeZr1–O2 solid solutions were investigated to improve soot oxidation activity.

Experimental Section

Catalyst Preparation

Hydrothermal Method

Ce(NO3)2·6H2O and ZrOCl2·8H2O were used as the precursors. These precursors were dissolved in 60 mL of deionized water, and the obtained solutions were mixed in a reaction vessel. The pH level of the solution was adjusted to 9 by using a 14 mol/L NaOH solution. The reaction solution was transferred to a Teflon-lined stainless-steel autoclave, which was sealed and placed in an oven preheated to 180 °C for 24 h. The reaction products were centrifuged and then washed with water and ethanol several times. The powder thus obtained was then pulverized using a mortar and pestle and calcined at 500 °C for 4 h. Finally, CeZr1–O2 with different Ce/Zr molar ratios was obtained, named CeZr1–O2-H.

Sol–Gel Method

Ce(NO3)2·6H2O and ZrOCl2·8H2O were added to a 15 mL mixed solution (9 mL of ethylene glycol and 6 mL of methyl alcohol) under ultrasound. Then, the mixed solution was constantly stirred for 6 h. The reaction products were collected by centrifugation and then washed with water and ethanol several times. The powder thus obtained was then pulverized using a mortar and pestle and calcined at 500 °C for 4 h. Finally, CeZr1–O2 with different Ce/Zr molar ratios was obtained, named CeZr1–O2-S.

Calcination Method

Ce(NO3)2·6H2O and ZrOCl2·8H2O were pulverized using a mortar and pestle and calcined at 500 °C for 4 h. Finally, CeZr1–O2 with different Ce/Zr molar ratios was obtained, named CeZr1–O2-C.

Physical and Chemical Characterization

The structural features of the catalysts were determined with X-ray diffraction (XRD) using a D8 Advance X-ray polycrystal diffractometer (Bruker, Germany) with Cu Kα radiation (λ = 1.5406 Å). The measured voltage was 40 kV and 40 mA, and the detection range was 10–80° (2θ). A Renishaw Invia instrument was used for Raman spectrum analysis. With 532 nm as the excitation source, the Raman spectrum ranges from 200 to 1000 cm–1. The morphology of the catalyst was observed by field emission scanning electron microscopy (FE-SEM, Zeiss merLIN). A Mack ASAP2460 instrument was used to measure the BET specific surface area, pore volume, and pore size distribution. The sample was degassed at 300 °C for 4 h to remove residual moisture before analysis. The specific surface area of the catalyst was calculated by the Brunauer–Emmett–Teller (BET) method, while the pore size and pore volume were calculated by the Barrett–Joyner–Halenda (BJH) method.

A Thermo Scientific K-Alpha X-ray photoelectron spectrometer (XPS) from Thermo Fisher Scientific was used for the X-ray photoelectron spectroscopy (XPS). The binding energy was corrected according to the standard C 1s = 284.80 eV. The test parameters were Al/Mg target, high pressure, 14 kV, power 250 W, and vacuum at 10–8 Torr. The redox performance of the samples in this experiment was determined by the automatic chemisorption apparatus (TPR) of MAC Auto Chem II 2920. The 100 mg sample was weighed and placed in a quartz tube, activated at 450 °C for 1 h in an Ar atmosphere of 40 mL·min–1, and cooled to room temperature. After that, the gas was switched to a H2–Ar mixture, 10.0% by volume fraction (40 mL·min–1), and the reaction temperature was increased from room temperature to 800 °C (10 °C·min–1). The thermal conductivity detector (TCD) monitored the hydrogen consumption online.

Activity Measurements

The prepared catalysts were assessed by temperature-programmed oxidation reactions (TPO) in a fixed-bed tubular quartz reactor (ø = 8 mm), each of which ranged from 250 to 650 °C at a rate of 3 °C min–1. The soot model involved is Printex-U (diameter 25 nm, purchased from Degussa). Then, 100 mg of catalyst and 10 mg of soot particles were added to a mortar, ground for 10 min to make close contact, and placed in a fixed-bed tubular quartz reactor. The oxidation reaction was carried out in a mixture of 10% O2 and 0.5% NO or 10% O2 at a flow rate of 500 mL/min, with N2 as the equilibrium gas. The flue gas analyzer was used to analyze the composition of the outlet gas. The catalytic activity was evaluated by Ti (soot ignition point), Tm (maximum soot combustion rate), and T90 (90% soot conversion temperature).

Results and Discussion

Catalytic Performance for Soot Combustion

Table shows the soot combustion performance of different catalysts without NO. As shown in Table , before the addition of catalyst, the Ti and Tm of soot are 445 and 575 °C, respectively, and the soot conversion rate reaches 90% at 627 °C. After the addition of catalyst, the ignition temperature of soot is below 380 °C, Tm is also within 500 °C, and the temperature decreases 40–130 °C when the soot conversion rate reaches 90%. Among these catalysts, Ce0.8Zr0.2O2-C has the best catalytic performance, with Ti, Tm, and T90 being 322, 415, and 497 °C, respectively. Figure shows the CO2 concentration and soot conversion of the catalyst with TPO and no NO. The catalytic performance of Ce0.8Zr0.2O2-C is obviously better than that of other catalysts. Figure a,c illustrates that the catalytic oxidation performance of the catalyst prepared by the calcination method is better than that of catalysts prepared by the other methods.

Table 1

Carbon Smoke Combustion Performance of Different Catalysts without NO

sample	Ti (°C)	Tm (°C)	T90 (°C)
none	445	575	627
Ce0.9Zr0.1O2-H	378	486	579
Ce0.8Zr0.2O2-H	356	492	587
Ce0.9Zr0.1O2-S	361	470	565
Ce0.8Zr0.2O2-S	372	482	573
CeO2-C	357	433	547
Ce0.9Zr0.1O2-C	334	412	514
Ce0.8Zr0.2O2-C	322	415	497
Ce0.67Zr0.33O2-C	365	501	535

Figure 1

Carbon dioxide concentration (a, b) and soot conversion (c, d) of catalyst under TPO: (a, c) different proportions of Ce/Zr by calcination; (b, d) catalysts prepared by different methods in the same proportion.

Table shows the soot combustion performance of different catalysts with NO. It can be seen from Table that Ce0.8Zr0.2O2-C has the best catalytic performance, with Ti, Tm, and T90 being 325, 407, and 478 °C, respectively. After the addition of NO, the Ti and Tm of the catalysts are slightly different from those without addition, but the temperature of T90 is greatly reduced. NO still has a certain catalytic effect on soot combustion. Figure demonstrates that the catalytic activity of Ce0.8Zr0.2O2-C is still optimal, with Ti, Tm, and T90 at 325, 407, and 458 °C, respectively. In particular, the fraction of T90 is greatly improved, which is obviously better than that of catalysts prepared by other methods.

Table 2

Soot Combustion Performance of Different Catalysts with NO

catalysts	Ti (°C)	Tm (°C)	T90 (°C)
none	448	548	590
Ce0.8Zr0.2O2-H	350	482	519
Ce0.8Zr0.2O2-S	367	472	539
Ce0.8Zr0.2O2-C	325	407	458

Figure 2

Carbon dioxide concentration (a) and soot conversion (b) of catalysts with TPO and NO.

Structural Features of As-Prepared Catalysts

XRD and Raman Results of As-Prepared Catalysts

Figure shows the XRD pattern of the CeZr1–O2 catalyst. Figure a shows that the diffraction crystal planes of the CeO2-C catalyst synthesized by the calcination method correspond to (111), (220), (220), and (311) at 2θ values of 28.63°, 33.18°, 47.71° and 56.56°, which are characteristic diffraction peaks when CeO2 is a cubic fluorite structure.[18] The other catalysts synthesized under the calcination method did not show the characteristic diffraction peak of ZrO2, but the peak moved more to the right with increasing Zr content, indicating that Zr4+ was doped into the CeO2 lattice and formed a homogeneous solid solution.[19]Figure b shows that the characteristic diffraction peaks of Ce0.8Zr0.2O2 catalysts synthesized by different methods are consistent with the diffraction peaks that would appear when CeO2 has a cubic fluorite structure. There is no characteristic diffraction peak of ZrO2, indicating that each synthesis method can make Zr4+ enter the CeO2 lattice and form a solid solution. When Zr4+ is doped into CeO2 to form a homogeneous solid solution, the crystallinity of CeO2 is reduced, more defects and vacancies are formed, and the activity of CeO2 is finally improved.[20]

Figure 3

XRD patterns of the CeZr1–O2 catalyst: CeZr1–O2-C (a) and Ce0.8Zr0.2O2 synthesized by different methods (b)

Figure shows the Raman spectrum of the catalyst. Figure a shows that the Ce0.8Zr0.2O2 catalyst prepared by all methods has a characteristic peak at 461 cm–1, which is because CeO2 of the cubic fluorite structure has the F 2g mode of Ce–O–Ce symmetric stretching vibration.[21] This shows that the catalysts prepared by all methods have cubic phase structures, which is consistent with the above XRD results. As seen from Figure b, shoulder peaks of the Ce0.8Zr0.2O2 catalyst prepared by all methods appeared at 270 and 601 cm–1, which were associated with tetragonal zirconia phase.[22] Raman peaks at 601 and 1178 cm–1 can be attributed to oxygen vacancies in the CeO2 lattice.[23,24] The oxygen vacancy is the adsorption–desorption center of gaseous oxygen, which can enhance the mobility of oxygen, thus facilitating the catalytic oxidation reaction and improving the catalytic oxidation activity of the catalyst toward soot particles.

Figure 4

Raman spectrum of the catalysts: CeZr1–O2-C (a) and CeO2 and Ce0.8Zr0.2O2 synthesized by different methods (b).

SEM Results of As-Prepared Catalysts

Figure shows SEM images of catalysts prepared by various methods. It is obvious that the catalysts prepared by the calcination method have irregular spherical granular structures. The catalyst prepared by the solvothermal method accumulates irregular large spherical particles.[16,25] It may be that the irregular spherical particles of the catalyst prepared by calcination lead to an increase in the specific surface area, which makes the contact surface area of soot particles larger, thus improving the performance of the catalyst.

Figure 5

SEM of the CeZr1–O2.

N2 Adsorption–Desorption Results of the As-Prepared Catalysts

Figure a demonstrates the nitrogen adsorption–desorption isotherm of the catalyst. N2 adsorption and desorption isotherms of catalysts end up with the final reversal of isotherms in the region with a higher P/P0 value, and the adsorption branch of isotherms is inconsistent with the desorption branch of isotherms; thus, a hysteresis loop can be observed.[26] Compared with the classification diagram of physical adsorption isotherms proposed by IUPAC, it can be seen that the adsorption isotherms of the catalysts prepared in this work are Langmuir IV physical adsorption isotherms. According to IUPAC’s classification of hysteresis loops, the catalysts prepared demonstrate H3 hysteresis loops, which also indicates that the catalysts have mesoporous structure.[27]Figure b shows the pore size distribution of different catalysts, from which it can be seen that the pore diameter of catalysts is distributed in the range of 6.6–10.2 nm, and the pore size of the Ce0.8Zr0.2O2 catalyst moves toward low pore size, indicating that doping can effectively reduce the mesoporous diameter of catalysts, thereby improving the specific surface area of catalysts, and ultimately improving the catalytic performance of catalysts. As seen from Table S1, the maximum BET specific surface area of Ce0.8Zr0.2O2 catalysts synthesized by different methods is 110.2 m2/g, and the pore size is 10.2 nm. The BET specific surface area of the Ce0.8Zr0.2O2-C catalyst is larger than that of Ce0.8Zr0.2O2-H and Ce0.8Zr0.2O2-S. However, the specific surface area of CeO2 prepared by the same preparation method is less than that of Ce0.8Zr0.2O2-C, indicating that the doping with Zr4+ can improve the specific surface area of the catalyst.[28] Therefore, increasing the specific surface area can improve the catalytic performance of the catalyst.

Figure 6

Nitrogen adsorption–desorption isotherm (a) and pore size distribution (b) of the catalysts.

XPS Results of As-Prepared Catalysts

Figure a,b illustrates the XPS Ce 3d spectra of the catalyst. Ce 3d peaks are divided into Ce 3d3/2 and Ce 3d5/2 due to spin orbital splitting. The u line in Figure a,b corresponds to the Ce 3d3/2 spin orbital component of the catalyst, and the v line corresponds to the Ce 3d5/2 spin orbital component of the catalyst.[29] These three pairs of peaks (u (901.7 eV), u″ (908.9 eV), u‴ (917.0 eV) and v (882.0 eV), v″ (889.7 eV), v‴ (899.1 eV)) are attributed to characteristic peaks of Ce4+,[25] whereas v′ (884.5 eV) and u′ (903.7 eV) are attributed to characteristic peaks of Ce3+.[30] Therefore, both Ce4+ and Ce3+ exist in the catalysts prepared in this work. Table S2 lists the proportion of catalyst Ce3+, in which the highest proportion in Ce0.8Zr0.2O2-C catalyst reaches 23.21%. The transition from Ce4+ to Ce3+ will cause lattice oxygen migration of the catalyst, resulting in oxygen vacancies, which are conducive to better catalytic oxidation activity toward soot particles.[31] Ce0.8Zr0.2O2-C has the best catalytic activity for soot because it has a higher proportion of Ce3+.

Figure 7

XPS spectra: Ce 3d of catalysts with different Ce/Zr ratios (a); Ce 3d of catalysts with different preparation methods (b); O 1s of catalysts with different Ce/Zr ratios (c); O 1s of catalysts with different preparation methods (d).

Figure c,d shows the XPS O 1s spectra of the catalyst. In this figure, 528.0–529.7 eV can be attributed to lattice oxygen (Oα), 530.8–531.6 eV to chemical oxygen (Oβ), and 531.7–532.6 eV to surface adsorbed oxygen (Oδ).[32] Because both Oβ and Oδ are on the surface of the catalyst, it is easier for them to contact directly with soot particles, thus facilitating the catalytic purification of carbon particles during catalytic oxidation.[33] The proportions of Oβ + Oδ in various catalysts can be obtained from Table S2, and the proportions of Oβ + Oδ in Ce0.8Zr0.2O2-C are the highest, reaching 23.18%. Therefore, the XPS O 1s spectra are consistent with the results of Ce0.8Zr0.2O2-C’s catalytic oxidation activity for soot particles.

H2-TPR Results of As-Prepared Catalysts

H2-TPR was used to verify the reducibility of the catalyst because the reducibility of the catalyst is of great significance to the soot combustion reaction. The results are shown in Figure . The multiple reduction peaks at 340–550 °C can be seen as a result of the step-by-step reduction of lattice oxygen on the surface of the catalyst,[34] while the reduction peak at approximately 770 °C can be attributed to the reduction of lattice oxygen in the bulk phase of the catalyst. When the temperature is above 600 °C, soot can be actively burned under aerobic conditions. Therefore, the reduction peak at 770 °C is not useful for the low-temperature catalytic purification of carbon particles. As shown in Figure , under the low temperature reduction peak of 340–550 °C, the reduction peak temperature of the Ce0.8Zr0.2O2-C catalyst prepared by calcination is obviously lower than that of other catalysts.[16] H2 consumption is the quantitative analysis listed in Table . It can be seen from Table that the hydrogen consumption of Ce0.8Zr0.2O2-S is the highest (0.78 mmol/g), which is higher than that of Ce0.8Zr0.2O2-C (0.74 mmol/g). However, the reduction peak temperature of Ce0.8Zr0.2O2-S is higher than that of Ce0.8Zr0.2O2-C; therefore the oxidation activity of Ce0.8Zr0.2O2-C catalyst is better at low temperature.[35] In conclusion, the Ce0.8Zr0.2O2-C catalyst can provide more oxygen species to participate in the catalytic oxidation reaction at lower temperatures, which may show a better catalytic purification effect in the carbon particle catalytic purification reaction.[36] This result is also consistent with the results of catalyst activity tests.

Figure 8

H2-TPR profiles of the CeZr1–O2.

Table 3

Hydrogen Consumption of the Catalysts

catalyst	CeO2-C	Ce0.8Zr0.2O2-C	Ce0.8Zr0.2O2-H	Ce0.8Zr0.2O2-S
H2 consumption (mmol/g)	0.68	0.74	0.23	0.78

Stability Analysis of Catalyst

The stability of the catalysts under the two atmospheres was also tested, and the results are shown in Figure . After 3 cycles in 10% O2 atmosphere, the Tm of Ce0.8Zr0.2O2-C catalyst increased by 36 °C, and the Tm of Ce0.8Zr0.2O2-H and Ce0.8Zr0.2O2-S catalysts increased by 35 and 40 °C, respectively. After 3 cycles in 10% O2 and 0.5% NO atmosphere, the Tm of Ce0.8Zr0.2O2-C catalyst increased by 32 °C, and the Tm of Ce0.8Zr0.2O2-H and Ce0.8Zr0.2O2-S catalysts increased by 34 °C. However, in the two atmospheres, the Tm of the three catalysts only increased slightly after one cycle (ΔTm ≤ 5 °C). The increase of Tm after three cycles of the catalysts was similar, indicating that the catalysts had similar stability and were relatively stable.

Figure 9

Catalytic soot oxidation activities of the catalyst (three cycles): (a) 10% O2; (b) 10% O2 + 0.5% NO).

Probable Reaction Mechanism for Soot Combustion

XRD and Raman characterization results show that Zr4+ was successfully doped into the CeO2 lattice to form a uniform solid solution. BET characterization showed that the introduction of Zr was beneficial to increase the specific surface area of the catalyst, which was conducive to better contact between the catalyst and soot particles. The incorporation of Zr into CeO2 promotes the reduction of Ce4+ to Ce3+, which will generate a large number of oxygen vacancies to balance the charge. XPS results also showed that % (Ce3+) and %(Oβ + Oδ) of Ce0.8Zr0.2O2-C were higher than those of CeO2-C. This is one of the reasons why the oxidation activity of Ce0.8Zr0.2O2-C is better than that of CeO2-C.[37] At the same time, it can be seen from the results of H2-TPR that after the incorporation of Zr into CeO2, both the reduction peak temperature and hydrogen consumption are superior to pure CeO2. This can also improve the catalytic activity for soot oxidation.[15] Under the conditions of 2000 ppm NO atmosphere, there may be two or more oxidation paths of soot to CO2 (see Figure ).[38,39] For example, NO is oxidized to NO2, which has a strong oxidation capacity and can oxidize soot to CO2. Reactive oxygen species on the surface of the catalyst can react directly with soot to produce CO2. Oxygen in the atmosphere is more likely to react with oxygen vacancies to generate O2–, which can oxidize soot to CO2 more quickly than reactive oxygen species on the surface of the catalyst.[40,41]

Figure 10

Catalytic mechanism diagram of soot combustion.

Conclusions

In summary, a series of CeZr1–O2 catalysts were prepared by hydrothermal, calcination, and sol–gel methods in this work. The catalytic oxidation activity of the catalysts in soot combustion was studied. The doping of Zr4+ into the CeO2 lattice improved the shortcomings of CeO2, such as poor thermal stability and great difference in catalytic activity. It was found that the catalyst prepared by calcination had the best catalytic activity. Under the conditions with no NO, the ignition point (Ti), fastest combustion rate temperature (Tm), and 90% soot conversion temperature (T90) of the Ce0.8Zr0.2O2-C catalyst were 322, 415, and 497 °C, respectively. The Ti, Tm, and T90 of soot catalyzed by the Ce0.8Zr0.2O2-C catalyst were 325, 407, and 458 °C, respectively, after 2000 ppm of NO was injected. Therefore, according to the experimental results and oxygen overflow mechanism, it can be inferred that NO can promote the combustion of soot and the effect is relatively significant. The reduction capacity and oxygen storage capacity of each catalyst were consistent with the change in catalytic activity for soot combustion, and the order from large to small is Ce0.8Zr0.2O2-C > Ce0.8Zr0.2O2-H > Ce0.8Zr0.2O2-S. In this work, CeZr1–O2 catalysts prepared by three methods were compared, which also provided a better synthesis method for further optimization of CeO2-based catalysts.