Study on Deactivation Mechanism of Catalytic Oxidation of Formaldehyde Over Ce_xZr_1-xO_y

Formaldehyde (HCHO) pollution was eliminated by catalyst effectively. The Ce_xZr_1-xO_y catalyst was prepared by coprecipitate method, and its performance of HCHO removal was investigated. The results showed that the HCHO removal rate decreased from 100% to 60.88% when the reaction time was from 24h to 48h.In order to study the deactivation mechanism of the Ce_xZr_1-xO_y catalyst,N₂ adsorption and desorption, SEM, EDS, FT-IR, XRD and XPS were used to characterize Ce_xZr_1-xO_y after different reaction times. Characterization results showed that the pore structure of Ce_xZr_1-xO_y did not change significantly with the extension of reaction time, and C₆H₉CeO₆ and Ce(CO₃)₂ increased gradually, while the active component Ce4+ decreased sharply. Therefore, the production and deposition of products are the main reasons for the deactivation of the Ce_xZr_1-xO_ycatalyst. The catalytic performance of regenerated Ce_xZr_1-xO_y catalyst was tested for 24h, and the results showed that it almost recovered to fresh Ce_xZr_1-xO_y catalyst. This provides a reference for the preparation of efficient catalyst for formaldehyde removal.

Keywords: Formaldehyde; Ce_xZr_1-xO_y; Catalytic Oxidation; Deactivation Mechanism

Highlights: (1) The Ce_xZr_1-xO_y catalyst had strong stability for formaldehyde. (2) The catalytic oxidation of formaldehyde produced cerium acetate and cerium carbonate, which deactivated the Ce_xZr_1-xO_y catalyst. (3) The deactivated Ce_xZr_1-xO_y catalyst had excellent catalytic performance after regeneration.

Formaldehyde (HCHO) pollution is caused by the release of building materials, household and daily chemicals, and the combustion of cigarettes and fuels [1]. The long-term exposure in formaldehyde environment could cause chronic poisoning, and even lead to leukemia, cancer and other serious diseases [2]. For formaldehyde pollution, researchers at home and abroad have developed formaldehyde purification methods to eliminate indoor formaldehyde pollution, which is divided into physical method [3], chemical method [4-6] and biological method [7,8]. Due to physical and biological methods have problems such as limited absorption capacity and time-consuming biological growth, chemical method has become a research focus of formaldehyde removal. For example, Ma et al. [9] improved the catalytic activity of Ag/CeO2 for HCHO by doping Na. The experimental result showed that the HCHO conversion rate was 30% at room temperature. Rong et al. [10] prepared a mesh structure 3D-MnO2 to remove HCHO with a conversion rate of 45%. Fang et al. [11] prepared MnOx/AC methanol catalyst for catalytic oxidation of HCHO. The study showed that the HCHO removal rate was basically maintained at 100% within 1000 min. Although these catalysts have achieved good results in removing HCHO, they all faced the problem that the activity of the catalyst decreased with the progress of the reaction. Therefore, it is an urgent scientific research project to find a highly stable catalyst for formaldehyde removal.

Ce, as a cheap and widely used lanthanide element, has special oxygen storage and release functions. ZrO2 has abundant surface oxygen vacancy and strong ion exchange ability, which can show unique catalytic effect when interacting with some active components in the system. Using Ce load in ZrO2 will have better oxygen storage capacity and redox capacity. For example, Fu QJ [12] prepared Pt/ Ce0.5Zr0.5O2 catalyst and investigated its catalytic combustion performance on C2H6. The results showed that Pt/ Ce0.5Zr0.5O2 had excellent activity on C2H6 combustion. CexZr1-xOy prepared by Ding YQ [13] showed high activity for CO oxidation and CH4 combustion. Gao X [14] prepared Ce0.7Zr0.3O2 for selective catalytic reduction of NO, and the conversion rate reached 100%. It can be seen from the above that Ce/Zr is a catalyst with excellent performance, but its application in catalytic oxidation of formaldehyde is rarely reported.

The Ce_xZr_1-xO_y catalyst was prepared by co-precipitation method, and its catalytic performance for removing HCHO was investigated. Through the characterization of Ce_xZr_1-xO_y at different reaction times, the reasons for deactivation were analyzed, which provided a reference for the development of the preparation of highly stable catalysts.

Cerium nitrate (Ce(NO3)3·6H2O) and zirconium nitrate (Zr(NO3)4·5H2O) were purchased from Macklin.Hydrazine hydrate (N2H4·H2O) was purchased from Tianjin Fengchuan Chemical Reagent Co., LTD.

The CexZr1-xOy catalyst was prepared by co-precipitation method [15,16]. First, Ce(NO3)3·6H2O and Zr(NO3)4·5H2O were dissolved with an appropriate amount of distilled water in a beaker, and got a mixed solution (the molar ratio of Ce to Zr was 4). Then, added slowly N2H4·H2O solution and adjusted the pH value of the mixed solution, prepared the CexZr1-xOy precipitate. After aging for 4h, the precipitate was filtered, washed and dried into the CexZr1-xOy crystallization [17]. Finally, the crystallization was calcined at 500 ⁰C for 4h to obtain the CexZr1-xOy catalyst.

The crystal structure of samples was detected by X-ray powder diffractometer (XRD, Bruker D8 Advance, Germany), using CuKα radiation, The intensity data were collected in a 2θ from 10° to 80°.

N2 adsorption was determined by the analyzer (Michael 2460, USA). The operating condition was controlled as follows: the samples were purified and degassed at 200 ⁰C for 3h and analyzed by static adsorption method under N2 atmosphere at 77K (liquid nitrogen).Specific surface area was calculated by BET equation.

The chemical state of element was determined by X-ray photoelectron spectroscopy (XPS,Thermo Scientific K-Alpha,USA). The radiation source was AlKα, the operating voltage was 12kV, and the binding energy was calibrated with internal standard carbon 1s peak (Eb= 284.80eV) with an accuracy of ±0.2eV.

The surface morphology of the catalyst was observed using a scanning electron microscope (SEM, Gemini300, Germany). Spectrometer (EDS, Oxford X-MAX,UK) was used for energy spectrum analysis. The acceleration voltage was 30kV, and the samples were dispersed with ethanol, dried, and sprayed with platinum.

The product functional groups on the surface of the catalyst were detected by the infrared spectrometer (FT-IR,Thermerfeld Nicolet iS20,USA).

The experimental process is shown in Figure 1. The HCHO catalytic oxidation reaction was operated in a fixed bed reactor under atmospheric pressure, 0.2g CexZr1-xOy catalyst and 50g SiO2 were loaded in the reactor. The volume fraction of each gas was 20% O2, 60% HCHO, and N2 as the equilibrium gas. The concentration of HCHO was determined by spectrophotometry [18].

where [HCHO]i (mg·L−1) is the initial concentration of HCHO before the test started, and [HCHO]f (mg·L−1) is the final concentration of HCHO at the end of the test.

To investigate the stability of the CexZr1-xOy catalyst, the catalytic oxidation reaction lasted for 48h, and the results are shown in Figure 2. When the catalytic oxidation experiment was carried out for 12 hours, the HCHO removal rate remained at 100%. Even after 24h reaction, formaldehyde removal rate could reach 99.67%, closed to complete degradation. The reason why the Ce_xZr_1-xO_y catalyst had such a high catalytic effect is that it has loose and porous morphology and abundant reaction sites. When the reaction lasted for 36h, the catalyst deactivation led to a significant decrease in HCHO degradation rate, but it still remained above 90%. With the progress of the reaction, the HCHO removal rate decreased to 78.29% at 42h. Until the end of catalytic oxidation, the HCHO removal rate was 60.88%. In order to improve the accuracy of the data, the experiment was repeated 5 times under the same conditions, and the results showed a high recurrence rate.

N2 adsorption and desorption were tested on the deactivated Ce_xZr_1-xO_y to explore the reasons for the decrease of catalyst activity. Table 1 shows the physical properties of Ce_xZr_1-xO_y after reaction at 0h, 12h, 24h and 48h. It can be seen that with the extension of reaction time, the specific surface area, pore volume and average pore size of Ce_xZr_1-xO_y tended to decrease, but the decrease rate tended to small. This phenomenon could be attributed to the formation of some solid species with higher SSA into the oxides [19]. In conclusion, the deactivation of Ce_xZr_1-xO_y was not caused by changes in specific surface area and pore structure, but was probably related to the blockage of pores by solid species SEM results of deactivated CexZr1-xOy and fresh CexZr1-xOy are shown in Figure 3. Figure 3(a) shows the morphology of fresh CexZr1-xOy, and its surface was smooth and flat without covering of particles. However, a large number of particles were deposited on the surface of deactivated CexZr1-xOy, as shown in Figure 3(b),Figure 3(c) and Figure 3(d). With the extension of reaction time, more and more particles were deposited, and even dense structures formed by particle agglomeration appeared. Figure 4 shows the EDS test results of CexZr1-xOy after the reaction. The atomic proportion of oxygen (O), carbon (C) and cerium (Ce) were much higher than that of other elements, among which carbon element might be brought by product CO2. Thus speculate that surface particle deposition is one of the causes of CexZr1-xOy deactivation, and the particles may be mainly composed of O, C and Ce elements.

Ce_xZr_1-xO_y was determined by Fourier transform infrared spectrometer. Firstly, as can be seen from Figure 5, absorption peaks appeared at 3381.71cm-1, 3377.42cm-1, 3378.97cm-1 and 3377.90cm-1, which neared 3300cm-1,indicating OH group [20,21]. Fresh Ce_xZr_1-xO_y peak had sharp shape without interference and it can infer to be OH group of unbound water. It may be that the low coordination O2- anions present on the basic support promoted the dissociation of water and produced OH group [22-24]. It can be used to supplement the OH group consumed in the decomposition process of HCHO [25-27]. After the reaction, the OH group absorption peaks of Ce_xZr_1-xO_y were wide and scattered, and a series of small peaks appeared in the range of 2700-2500cm-1, which are judged as characteristic peaks of carboxylic acid [28]. The absorption peaks of Ce_xZr_1-xO_y after 42-48h reaction were observed at 1564.28cm-1 and 1567.96cm-1, which are C=O antisymmetric stretching vibration of carboxylate, while the symmetric stretching vibration of 1440-1360cm-1 is weak [29-31]. It can be speculated that HCHO reacted with adsorbed oxygen species to produce formate [32], and then formed a cerium salt of organic acid with cerium. Secondly, the absorption peaks of Ce_xZr_1-xO_y after the reaction appeared at 1073.51cm-1 and 1080.64cm-1, which are C-O absorption peaks [33]. It is speculated that formaldehyde was catalyzed to generate CO2 products, which further reacted to form carbonate substances. In conclusion, the catalytic oxidation of formaldehyde by Ce_xZr_1-xO_y generates organic acid cerium salts and carbonate particles. This conclusion is consistent with EDS analysis.

Figure 6 shows the CexZr1-xOy XRD pattern of different reaction times. For Ce_xZr_1-xO_y before the reaction, the diffraction peaks at 2θ of 28.87 °, 33.47 °, 48.05°, 57.01°, 59.85°, 70.36° were attributed to the diffraction peaks of Ce_xZr_1-xO_y(JCPDS 28-0271). At the same time, the characteristic peaks of CeO2 and ZrO2 did not appear in the XRD pattern, indicating that the active component were evenly distributed on the surface of Ce_xZr_1-xO_y [34]. The diffraction peak intensity of Ce_xZr_1-xO_y (24h) increased, which may be due to the overlapping effect of particles on crystals. With the progress of the reaction (42-48h), the diffraction peak of Ce_xZr_1-xO_y faded rapidly and partially disappeared, indicating that Ce was further oxidized to Ce(CO3)2 and C6H9CeO6 in the reaction process, which is consistent with the infrared analysis results.At the same time, the diffraction peak intensity of Ce(CO3)2 and C6H9CeO6 slightly increased, indicating that their grain size gradually increased with the progress of the reaction, that is, Ce(CO3)2 and C6H9CeO6 on the surface of Ce_xZr_1-xO_yappeared agglomeration phenomenon. It is consistent with the SEM results. In conclusion, cerium acetate and cerium carbonate accumulated in the Ce_xZr_1-xO_y channel, hindering the contact between HCHO and the active component and weakening the activity of Ce_xZr_1-xO_y.

XPS was used to analyze the content and morphology of CexZr1-xOy element. In Figure 7,U and V correspond to the spin splitting orbits of Ce 3d3/2 and Ce 3d5/2 respectively. The peaks located at 897.82eV (V4), 888.28eV (V3),881.69eV (V1),916.08eV (U4),906.58eV (U3),900.28eV (U1) belong to Ce4+.The peaks located at 883.9eV (V2) and 902.34eV (U2) belong to Ce3+ [35]. It created a charge imbalance, forming some oxygen vacancies and unsaturated chemical bonds [36]. The existence of Ce3+ is due to the tiny particle size of cerium oxide, changes in the coordination of the Ce atoms, or changes in net charge caused by its shared anion with ZrO2 [37]. It can be obtained from Figure that the peak located at 534.1 and 529.5 eV, the former belong to the adsorbed oxygen (Oa) or surface hydroxyl oxygen of the catalyst, and the latter belong to the lattice oxygen (Ob) in the catalyst [38]. Surface hydroxyl oxygen can not only provide reaction sites for hydrogen bond adsorption of formaldehyde molecules, but also accelerate the catalytic oxidation of HCHO by using hydroxyl oxidation properties [39].

Through semi-quantitative calculation of peak area, the molar ratio of Ce4+ and Oa can be obtained, and the results are listed in Table 2. The content of Ce4+ decreased with the extension of reaction time, indicating that Ce4+ was involved in the catalytic oxidation reaction as the active component. At the same time, the Oa also decreased with Ce4+ concentration, indicating that Oa was consumed in the catalytic reaction. It can be concluded that Ce4+, as the active component, oxidizes HCHO to CO2 and H2O, and turn itself into Ce3+, while Ce3+ is oxidized to Ce4+ by Oa to supplement the active component and maintain high catalytic activity of CexZr1-xOy.

Combined with SEM, EDS, FT-IR and XRD characterization analysis, it is concluded that the formate reacted with Ce to produce C6H9CeO6,CO2 reacted with Ce to transform into Ce(CO3)2 in the catalytic oxidation of HCHO by CexZr1-xOy.These production not only occupied the active site, but also consumed Ce4+,which broke the benign cycle of "supplement-consumption-supplement" of active components. The above reasons results in the decreased activity of CexZr1-xOy.

To confirm that products deposition is the main cause of CexZr1-xOy deactivation, the deactivated CexZr1-xOy was collected and calcined at 700 ⁰C for 4h.The regenerated CexZr1-xOy was characterized by IR, XRD and SEM.

As shown in Figure 8, C-O absorption peaks of carbonate were not observed in infrared spectrum of the regenerated CexZr1-xOy. The carboxylic acid characteristic peak of the regenerated CexZr1-xOy was also not as obvious as that after the reaction. These indicating that Ce(CO3)2 and C6H9CeO6 had been decomposed after calcining, which is also confirmed in the XRD pattern (Figure 9).As can be seen from Figure 10,only a few particles remained on the surface of the regenerated CexZr1-xOy.According to the XRD pattern, these particles might be SiO2 that was not separated during the collection of deactivated CexZr1-xOy.From the characterization results, there is no significant difference between the calcined regenerated CexZr1-xOy and the fresh CexZr1-xOy.

For further verify that the catalytic activity was recovered after the decomposition of the product. The catalytic performance of the regenerated CexZr1-xOy was tested for 24h and the results are shown in Figure 11. Within 12h of the reaction, the HCHO removal rate of 100% was basically maintained. With the progress of the reaction, the HCHO removal rate was slightly reduced to 98.67% at 24h. After testing, the HCHO removal rate of regenerated CexZr1-xOy can basically reach the fresh CexZr1-xOy. These results confirm that the deposition of Ce(CO3)2 and C6H9CeO6 is the main cause of CexZr1-xOy deactivation.

The CexZr1-xOy catalyst had a good catalytic oxidation effect on HCHO removal, and HCHO removal rate was close to 100% within 24h of reaction. Characterization results of deactivated CexZr1-xOy catalyst showed that during the HCHO removal, the products of Ce(CO3)2 and C6H9CeO6 blocked the pore, occupied the reaction site, consumed the active component Ce4+, and led to the deactivation of the CexZr1-xOy catalyst.

Financial support from the National Natural Science Foundation of China (51568068), young and middle-aged academic and technical leaders reserve talent project of yunnan (cn) (202105AC160054) are gratefully acknowledged.

The authors declare no conflicts of interest.