Tailored Synthesis of Catalytically Active Cerium Oxide for N, N-Dimethylformamide Oxidation

Cerium oxide nanopowder (CeOx) was prepared using the sol–gel method for the catalytic oxidation of N, N-dimethylformamide (DMF). The phase, specific surface area, morphology, ionic states, and redox properties of the obtained nanocatalyst were systematically characterized using XRD, BET, TEM, EDS, XPS, H2-TPR, and O2-TPO techniques. The results showed that the catalyst had a good crystal structure and spherelike morphology with the aggregation of uniform small grain size. The catalyst showed the presence of more adsorbed oxygen on the catalyst surface. XPS and H2-TPR have confirmed the reduction of Ce4+ species to Ce3+ species. O2-TPR proved the reoxidability of CeOx, playing a key role during DMF oxidation. The catalyst had a reaction rate of 1.44 mol g−1cat s−1 and apparent activation energy of 33.30 ± 3 kJ mol−1. The catalytic performance showed ~82 ± 2% DMF oxidation at 400 °C. This work’s overall results demonstrated that reducing Ce4+ to Ce3+ and increasing the amount of adsorbed oxygen provided more suitable active sites for DMF oxidation. Additionally, the catalyst was thermally stable (~86%) after 100 h time-on-stream DMF conversion, which could be a potential catalyst for industrial applications.


Introduction
N, N-Dimethylformamide (DMF) is a colorless transparent liquid which has been widely used as a chemical raw material [1]. DMF is miscible with water and most organic solvents and is known as a solvent in the battery industry (12%), polyacrylic fibre industry (22%), pharmacy and cosmetic industry (39%), polyurethane lacquers (17%), and pesticide formulations (10%) [2][3][4][5]. It is stable and difficult to biodegrade and has certain biological toxicity [6][7][8][9]. It is present in industrial wastewater as a nitrogen-containing volatile organic compound (NVOC). DMF typically has pungent odors that can harm the human body when inhaled into the respiratory system or through direct contact with the skin, as are its derivatives, such as NOx, which is a pollutant to the environment [10,11]. For this reason and for economic reasons, the transformation of DMF into harmless compounds has garnered increasing attention in recent years.
The general conversion methods for the NVOCs treatment are as follows: biotreatment (biotechnological purification) [1], activated carbon adsorption method [12], pervaporation separation [13], microbial degradation [14], electrochemical oxidation [15], and low-pressure pyrolysis [16]. However, the above methods require expensive conversion costs or complex post-treatment processes, which greatly limit the application of these methods of DMF conversion. Researchers have recently developed a method of catalytic oxidation of DMF using metal oxides [11,17,18]. Compared to other methods, metal oxide catalytic oxidation of DMF has the advantages of relatively low cost, easy operation, and

Catalyst Preparation
The catalyst is prepared using the sol-gel method, and the preparation process is shown in Figure S1 in the Supplementary Material (SM) and the literature [26]. Cerium nitrate hexahydrate (Ce(NO 3 ) 3 .6H 2 O, 99.9%, Sinopharm Chemical Reagent Co, Ltd., Shanghai, China) was used as a precursor because it is easily soluble in water and ethanol. The preparation process was divided into six steps: (i) 6.0 g of cerium nitrate hexahydrate was dissolved in 60.0 mL distilled water at 60 • C and stirred in a beaker for 10 min; (ii) 1.0 g of sodium hydroxide was dissolved in 30.0 mL of distilled water at 60 • C and stirred for 10 min; (iii) the already prepared solutions were mixed and continuously stirred for 10 min. (iv) Furthermore, 100.0 mL of ethanol was added to the solution and stirred at 60 • C for 10 min to make it fully react. (v) Then, 15.0 mL of aqueous ammonia solution was added to the final mixture to adjust the pH value to 8. The mixture was then dried in an oven at 100 • C for 2 h to evaporate the solvent. (vi) Finally, the powdered sample was transferred to an electric furnace and calcined at 500 • C for 120 min at a rate of 5.6 • C min −1 .

Catalyst Characterization
A variety of techniques were used to systematically characterize the prepared nanocatalyst. The phase information of the nanoparticles was analyzed using RIGAKU ULTIMA IV X-ray diffraction (XRD) at the scanning angle range of 5.0 • −90.0 • , and the scanning step was in the range of 0.5 • using Cu Kα. The powder samples' surface areas and textural characteristics were measured using an Autosorb IQ instrument (Quantachrome) at 77.4 • C to obtain a BET (Brunauer-Emmett-Teller). The morphology of the ceria oxides was observed via transmission electron microscopy (TEM) using a NIPPON Electronics 2100 microscope operated at 200 kV. The surface chemical composition of the sample was obtained via X-ray photoelectron spectroscopy (XPS) (Thermo Fly's NEXSA) with an analytical voltage of 1486.8 eV and an analysis voltage of 15.0 kV. The redox properties were elaborated using temperature-programmed reduction (H 2 -TPR) and temperature-programmed oxidation ((O 2 -TPO) AutoChem II 2920, Micromeritics). The optical properties were evaluated with a SHIMADZU UV−2600i visible-near-infrared spectrophotometer.

Catalytic Test
The catalytic performance of prepared Ce 2 O 3 for DMF was analyzed in a fixed bed reactor under atmospheric pressure. The length of the reactor tube was 900 mm with an inner diameter of 8 mm. The device diagram of the catalytic test can be observed in Figure  S2 (see Supplementary Materials). An amount of 0.1 g of the CeO x catalyst sample was placed into the reactor. The reaction gases were DMF and O 2, with Ar as the equilibrium gas. Ar was used as the carrier gas to move DMF into the reactor. The concentration of DMF was 1000 ppm, 10.0% O 2 was diluted in Ar with a total flow rate of 50.0 mL min −1 , and the corresponding gas space velocity (GHSV) was 30,000 mL g −1 cat h −1 . A digital electric furnace controlled the temperature at increments of about 5 • C min −1 . The exhaust gases were detected and analyzed using an FTIR spectrometer within a spectral range of 400-4000 cm -1 . Before the catalytic test, the samples were pretreated in Ar (50.0 mL min −1 ) at 400 • C for 1 h to remove any surface-adsorbed substances. The conversion of the DMF is calculated using the following equation: where [X in ] and [X out ] represent the initial molar concentration of DMF at temperature T and the final molar concentration of DMF at the outlet, respectively.

Characterization Results and Analysis
3.1. Catalyst Structure and Texture Analysis Figure 1 shows the XRD patterns of the nanoparticles. The prepared sample has crystalline diffraction peaks. The as-prepared catalyst CeO x corresponds to the cubic phase cerium oxide (CeO 2 ) (JCPDS No.01-089-8436), which is in agreement with the literature [27,28]. The strong and sharp diffraction peak indicates the pure crystal structure of the sample. The crystallite sizes (D) and micro-strain (ε) on the (111), (220), and (311) planes with the highest intensity were calculated using the Debye-Scherrer Formulas (3) and (4).
where β is the diffraction broadening on the peak at half height for Bragg's angle θ and λ represents the wavelength of the X-ray radiation. For more details on the calculation, see Table 1. The average grain size is 20.07 ± 0.2 nm, and the average microstrain is 0.12 ± 0.02%. With the increase in the crystallite size, the microstrain has a downward trend. This is because inherent defects may cause the microstrain in the crystal lattice, such as vacancies, spatial barriers, and concentration changes during the preparation process.
The CeO x nanocatalyst in this study is fine crystals, which are lower than 100 nm. The BET results show that the specific surface area of the sample is 3.8 ± 0.2 m 2 g −1 , with a pore volume of 0.01 ± 0.001 cm 3 g −1 and a pore size of 25.7 ± 0.2 nm at the relative pressure of P/P 0 = 0.98. Figures S3 and S4 (see Supplementary Materials) depict type 3 nitrogen adsorption-desorption isotherm and the pore size distribution, respectively. It is worthy to note that the pore size obtained in this work is quite smaller than those of the previously reported CeO 2 (53.48 nm) [1] and CeO 2 -nanocube [23]. Small crystallite and pore size are advantageous for abundant oxygen species, playing a key role during DMF oxidation [20]. Therefore, the obtained CeO x is expected to offer a good DMF conversion.

Morphology
To obtain the structural changes of the CeO x catalyst due to calcination and reaction temperature, the surface morphology and bulk elemental constituent of the prepared nanopowder were characterized using TEM and EDS, as shown in Figure 2. The surface of the nanocatalyst has a spherelike morphology, which contrasts with the 3D ball-like shape of self-assembled M-CeO 2 [11,23] and the large particles of CeO 2 supported on multiwall carbon nanotube (CNTs) [1]. It can also be observed that the catalyst prepared has an aggregation of uniform small grain size. Looking at the EDS elemental distribution in Figure 2, it can be observed that the cerium and oxygen have similar distribution and local aggregation. The carbon arising from the carbon tape that was used as support during EDS analysis, as shown in Figure 2. It can be seen from the elemental distribution that cerium accounts for 53.60% of cerium, 24.20% carbon, and 22.20% oxygen. However, the 24% carbon is not considered because it arises from the carbon tape employed during EDS analysis. The observed spherelike morphology with small grain size agglomeration is in good line with XRD and BET results, which will positively affect the movement of absorbed oxygen and active sites advantageous for the oxidation reaction. gregation. The carbon arising from the carbon tape that was used as support during EDS analysis, as shown in Figure 2. It can be seen from the elemental distribution that cerium accounts for 53.60% of cerium, 24.20% carbon, and 22.20% oxygen. However, the 24% carbon is not considered because it arises from the carbon tape employed during EDS analysis. The observed spherelike morphology with small grain size agglomeration is in good line with XRD and BET results, which will positively affect the movement of absorbed oxygen and active sites advantageous for the oxidation reaction.

Elemental Composition and Ionic State
XPS was used to detect the surface chemical oxidation state of the Ce element. Figure  3 shows the deconvolution of the electron binding energy (BE) in the Ce 3d spectrum, while Table 2 shows the details of the distribution of Ce species. The XPS spectra of Ce 3d were divided into two spin-orbit components 3d3/2 and 3d5/2, using fityk 1.

Elemental Composition and Ionic State
XPS was used to detect the surface chemical oxidation state of the Ce element. Figure 3 shows the deconvolution of the electron binding energy (BE) in the Ce 3d spectrum, while Table 2 shows the details of the distribution of Ce species. The XPS spectra of Ce 3d were divided into two spin-orbit components 3d 3/2 and 3d 5/2 , using fityk 1.  [29,30]. Ce 3d mainly exists in the form of Ce 4+ because the cerium in the catalyst is in the form of oxides. The ratio of Ce 3d 5/2 is less than in Ce 3d 3/2 . The decrease in the binding energy causes the ratio of Ce 3+ /Ce 4+ to decrease, which could be due to Ce 3+ being converted to Ce 4+ .  [29,30]. Ce 3d mainly exists in the form of Ce 4+ because the cerium in the catalyst is in the form of oxides. The ratio of Ce 3d5/2 is less than in Ce 3d3/2. The decrease in the binding energy causes the ratio of Ce 3+ /Ce 4+ to decrease, which could be due to Ce 3+ being converted to Ce 4+ .    Note: BE refers to binding energy, and RA refers to the relative area of the peak. Figure 4 is an XPS spectrum of the valence state and distribution of elements before the catalytic test, and Table 3 gives the detailed distribution of elements in the XPS spectrum. The O 1s peak was fitted into two to estimate the chemical states of different oxygen species. As shown in Figure 4, the O 1s core-shell exhibits two prominent peaks. The lower binding energy peak corresponds to lattice oxygen O Lat (O 2− ) at 529.1 eV. The higher binding energy peak corresponds to the adsorbed oxygen (O Ads ), expressed as OH -, and H 2 O at 532.5 and 534.7, respectively. These results are in agreement with the literature [30]. Surface oxygen is mainly composed of O Lat and O Ads . O Lat exists in the form of oxygen combined with transition metal elements, and O Ads mainly exists in the form of hydroxide ions and oxygen ions. The proportion of O Ads on the sample surface is higher than that of O Lat . The high proportion of adsorbed oxygen could enhance the catalyst's ability to react with the surface products to form products such as H 2 O, NO 2 , and CO 2 , which improves the mass transfer driving force of product diffusion, thus improving the reaction rate.

Redox Properties
H2-TPR studies the reducibility of the prepared catalysts, and the results of the H2 consumption by CeOx as a function of temperature are shown in Figure 5a and Table 4. With the increase in reduction temperature, three hydrogen consumption peaks with different strengths were observed: a small peak at 575.6 °C, the second one at 589.9 °C, and a third wide peak appear at 641.8 °C. The temperature peak at <600 °C is due to the reduction of Ce 4+ to Ce 3+ on the cerium oxide surface with oxygen vacancy, while the one > 600 °C represents a reduction of Ce 4+ to Ce 3+ by H2 inside the bulk oxygen from cerium oxides species [1,31]. Therefore, the H2-TPR profiles show that the CeOx could result from the

Redox Properties
H 2 -TPR studies the reducibility of the prepared catalysts, and the results of the H 2 consumption by CeO x as a function of temperature are shown in Figure 5a and Table 4. With the increase in reduction temperature, three hydrogen consumption peaks with different strengths were observed: a small peak at 575.6 • C, the second one at 589.9 • C, and a third wide peak appear at 641.8 • C. The temperature peak at <600 • C is due to the reduction of Ce 4+ to Ce 3+ on the cerium oxide surface with oxygen vacancy, while the one > 600 • C represents a reduction of Ce 4+ to Ce 3+ by H 2 inside the bulk oxygen from cerium oxides species [1,31]. Therefore, the H 2 -TPR profiles show that the CeO x could result from the removal of bulk oxygen of cerium oxide. The oxygen mobility in the CeOx was studied using O2-TPO. Figure 5b illustrates the O2 consumption as a function of the temperature. The peak areas of the TPO profiles over the synthesized catalysts represent oxygen consumption at 693.4 °C. The calculated peak area is listed in Table 4. As shown in Figure 5b, the O2 consumption is high on CeOx. This indicates that oxygen is not consumed. The reoxidation initiates at ⁓300 °C, and the catalyst completely recovered at around ⁓693.4 °C. These findings are in good agreement with previous studies [32,33]. In summary, TPR and TPO results reveal that CeOx could be more active from the range of 300 to 700 °C for DMF oxidation.

Optical Properties
The adsorption spectrum of the cerium oxide catalyst, as well as the Tauc's plot displaying the estimated bandgap energy (Eg), were investigated in the region of 200-800 nm, and the results are illustrated in Figure 6. Figure 6a shows an adsorption's decrement as the wavelength extends to the visible region. The intense peak referring to the strong absorbance appeared at about 300-350 nm. According to the literature, this peak is related to the charge transfer from O 2p to Ce 4f, which overruns the well-known f-f spin-orbit splitting of the Ce 4f state [34,35]. The estimation of the optical bandgap energy Eg, can be evaluated by the following equation: where A is a refractive index constant and n is a constant related to the transition's nature (in this work, n = 2 considering the cerium oxide presented a direct bandgap).  The oxygen mobility in the CeO x was studied using O 2 -TPO. Figure 5b illustrates the O 2 consumption as a function of the temperature. The peak areas of the TPO profiles over the synthesized catalysts represent oxygen consumption at 693.4 • C. The calculated peak area is listed in Table 4. As shown in Figure 5b, the O 2 consumption is high on CeO x . This indicates that oxygen is not consumed. The reoxidation initiates at~300 • C, and the catalyst completely recovered at around~693.4 • C. These findings are in good agreement with previous studies [32,33]. In summary, TPR and TPO results reveal that CeO x could be more active from the range of 300 to 700 • C for DMF oxidation.

Optical Properties
The adsorption spectrum of the cerium oxide catalyst, as well as the Tauc's plot displaying the estimated bandgap energy (E g ), were investigated in the region of 200-800 nm, and the results are illustrated in Figure 6. Figure 6a shows an adsorption's decrement as the wavelength extends to the visible region. The intense peak referring to the strong absorbance appeared at about 300-350 nm. According to the literature, this peak is related to the charge transfer from O 2p to Ce 4f, which overruns the well-known f-f spin-orbit splitting of the Ce 4f state [34,35]. The estimation of the optical bandgap energy E g , can be evaluated by the following equation: where A is a refractive index constant and n is a constant related to the transition's nature (in this work, n = 2 considering the cerium oxide presented a direct bandgap). The corresponding Tauc plot of (αhν) 2 is presented in Figure 6b. The Eg for the catalyst was observed from the straight line's intersection with the photon energy hν axis. Eg was determined to be 2.93 ± 0.05 eV. This value is smaller compared to bandgap energies of cerium oxide nanoparticles prepared in the literature from other preparation methods, such as coprecipitation [36], microwave [35], and microwave-assisted hydrothermal [37]. As previously reported, small Eg facilitates the movement of OLat at the catalyst's surface, enhancing the catalytic activity [26]. Therefore, the catalyst in this work is expected to exhibit a good conversion during DMF oxidation.

Catalytic Test
The catalytic performance of CeOx for DMF oxidation was studied in the temperature range of 100-400 °C. Figure 7a shows the catalytic behavior of DMF on the prepared CeOx catalyst. As shown in Figure 7a, the catalyst had little effect on the conversion of DMF before 300 °C. The conversion of DMF gradually increased from 300 °C and reached 81.48 ± 2% at 400 °C. In addition, Figure 7b shows the selectivity of other products-CO2 and NO2-produced by the reaction at 400 °C. The main products are CO2 and NO2, with selectivities of 62.48 ± 2% and 37.52 ± 2%, respectively. As illustrated in Figure 8 and Figure  S5 (see Supplementary Materials), it is worth noting that the as-prepared catalyst retained its catalytic activity and reproducibility after three running tests, indicating that it is reusable for DMF oxidation. The DMF conversion in this study showed a higher conversion compared to those reported for Cu Ru/C (73.00%) [38], Pd/C (51.00%) [39], Ru/C (50.00%) [40], and Cu/ZnO/Al2O3 (40.60%) [41].
According to the XRD analysis, CeOx has a small crystallite size, which is beneficial for oxidation reactions [42]. The performance of CeOx is related to the spherical and smooth structure and the agglomeration of fine particles (see TEM results), which are conducive to enriching the movement of lattice oxygen. XPS results further confirmed the abundance of OLat. This catalytic behavior can be attributed to the good content of OLat and the small amount of loading detected on the catalyst surface. In addition, the catalyst presented a large number of Ce 4+ species, which was proven to have a positive effect on DMF oxidation [43]. In this work, CeOx nanopowder was active between 300 and 400 °C, confirming the H2-TPR and O2-TPO results. Therefore, the catalytic activity of CeOx can be attributed to the above properties. The corresponding Tauc plot of (αhν) 2 is presented in Figure 6b. The E g for the catalyst was observed from the straight line's intersection with the photon energy hν axis. E g was determined to be 2.93 ± 0.05 eV. This value is smaller compared to bandgap energies of cerium oxide nanoparticles prepared in the literature from other preparation methods, such as coprecipitation [36], microwave [35], and microwave-assisted hydrothermal [37]. As previously reported, small E g facilitates the movement of O Lat at the catalyst's surface, enhancing the catalytic activity [26]. Therefore, the catalyst in this work is expected to exhibit a good conversion during DMF oxidation.

Catalytic Test
The catalytic performance of CeO x for DMF oxidation was studied in the temperature range of 100-400 • C. Figure 7a shows the catalytic behavior of DMF on the prepared CeO x catalyst. As shown in Figure 7a, the catalyst had little effect on the conversion of DMF before 300 • C. The conversion of DMF gradually increased from 300 • C and reached 81.48 ± 2% at 400 • C. In addition, Figure 7b shows the selectivity of other products-CO 2 and NO 2 -produced by the reaction at 400 • C. The main products are CO 2 and NO 2 , with selectivities of 62.48 ± 2% and 37.52 ± 2%, respectively. As illustrated in Figure 8 and Figure S5 (see Supplementary Materials), it is worth noting that the as-prepared catalyst retained its catalytic activity and reproducibility after three running tests, indicating that it is reusable for DMF oxidation. The DMF conversion in this study showed a higher conversion compared to those reported for Cu Ru/C (73.00%) [38], Pd/C (51.00%) [39], Ru/C (50.00%) [40], and Cu/ZnO/Al 2 O 3 (40.60%) [41].
According to the XRD analysis, CeO x has a small crystallite size, which is beneficial for oxidation reactions [42]. The performance of CeO x is related to the spherical and smooth structure and the agglomeration of fine particles (see TEM results), which are conducive to enriching the movement of lattice oxygen. XPS results further confirmed the abundance of O Lat . This catalytic behavior can be attributed to the good content of O Lat and the small amount of loading detected on the catalyst surface. In addition, the catalyst presented a large number of Ce 4+ species, which was proven to have a positive effect on DMF oxidation [43]. In this work, CeO x nanopowder was active between 300 and 400 • C, confirming the H 2 -TPR and O 2 -TPO results. Therefore, the catalytic activity of CeO x can be attributed to the above properties.  Furthermore, the catalytic performance of the as-prepared catalysts was compared based on the specific reaction rate and the apparent activation energy, as illustrated in Figure 9, which were calculated using the Arrhenius equation in the DMF conversion range within 15% [44,45]. The following equations were used to calculate the reaction rate and the apparent activation energy: k = A exp (−Ea/RT) (6) r (mol g −1 s −1 ) = F0 X/Wcat (7) where k is the rate constant, Ea is the apparent activation energy, R is the gas constant, T is the temperature, A the pre-exponential factor, F0 refers to the molar flow rate, X is the conversion achieved at a certain temperature, and Wcat is the weight of the catalyst. DMF's oxidation started as the reaction rate increased with a gradual temperature increase, as shown in Figure 9a. The Ea in this work was found to be 33.30 ± 3 kJ mol -1 , which is smaller than those of CeO2-nanorods (37.48 kJ mol −1 ) [23], CeO2-nanocubes (47.32 kJ mol −1 ) [23], and CeO2/CNTs-M (72.97 kJ mol −1 ) [1] reported in the literature. Previous studies revealed that low Ea is beneficial for better catalytic activity [23,46]. Therefore, the low Ea obtained by CeOx is expected to play a key role during the DMF conversion.  Furthermore, the catalytic performance of the as-prepared catalysts was compared based on the specific reaction rate and the apparent activation energy, as illustrated in Figure 9, which were calculated using the Arrhenius equation in the DMF conversion range within 15% [44,45]. The following equations were used to calculate the reaction rate and the apparent activation energy: k = A exp (−Ea/RT) (6) r (mol g −1 s −1 ) = F0 X/Wcat (7) where k is the rate constant, Ea is the apparent activation energy, R is the gas constant, T is the temperature, A the pre-exponential factor, F0 refers to the molar flow rate, X is the conversion achieved at a certain temperature, and Wcat is the weight of the catalyst. DMF's oxidation started as the reaction rate increased with a gradual temperature increase, as shown in Figure 9a. The Ea in this work was found to be 33.30 ± 3 kJ mol -1 , which is smaller than those of CeO2-nanorods (37.48 kJ mol −1 ) [23], CeO2-nanocubes (47.32 kJ mol −1 ) [23], and CeO2/CNTs-M (72.97 kJ mol −1 ) [1] reported in the literature. Previous studies revealed that low Ea is beneficial for better catalytic activity [23,46]. Therefore, the low Ea obtained by CeOx is expected to play a key role during the DMF conversion. Furthermore, the catalytic performance of the as-prepared catalysts was compared based on the specific reaction rate and the apparent activation energy, as illustrated in Figure 9, which were calculated using the Arrhenius equation in the DMF conversion range within 15% [44,45]. The following equations were used to calculate the reaction rate and the apparent activation energy: where k is the rate constant, E a is the apparent activation energy, R is the gas constant, T is the temperature, A the pre-exponential factor, F 0 refers to the molar flow rate, X is the conversion achieved at a certain temperature, and W cat is the weight of the catalyst. DMF's oxidation started as the reaction rate increased with a gradual temperature increase, as shown in Figure 9a. The E a in this work was found to be 33.30 ± 3 kJ mol −1 , which is smaller than those of CeO 2 -nanorods (37.48 kJ mol −1 ) [23], CeO 2 -nanocubes (47.32 kJ mol −1 ) [23], and CeO 2 /CNTs-M (72.97 kJ mol −1 ) [1] reported in the literature. Previous studies revealed that low E a is beneficial for better catalytic activity [23,46]. Therefore, the low E a obtained by CeO x is expected to play a key role during the DMF conversion. Furthermore, it is widely recognized that catalytic stability is an important parameter for industrial applications. The stability test on the DMF conversion was performed over 100 h at a constant temperature setting of 400 °C, as shown in Figure 10. The catalyst showed good durability during a continuous DMF conversion, with no significant deactivation after 100 h time-on-stream at 86%. However, there is a negligible performance decrease around 14% with continuous time-on-stream. The slight deactivation of the CeOx during the stability process might be due to the following reasons: (1) active oxygen vacancy sites decrease due to catalyst sinter and/or (2) disintegration of the active CeO2 crystal phases, which might cause the catalyst deactivation. Alternatively, coke formation over solid catalysts from DMF could lead to continuing oxidation at high temperature.  Furthermore, it is widely recognized that catalytic stability is an important parameter for industrial applications. The stability test on the DMF conversion was performed over 100 h at a constant temperature setting of 400 • C, as shown in Figure 10. The catalyst showed good durability during a continuous DMF conversion, with no significant deactivation after 100 h time-on-stream at 86%. However, there is a negligible performance decrease around 14% with continuous time-on-stream. The slight deactivation of the CeO x during the stability process might be due to the following reasons: (1) active oxygen vacancy sites decrease due to catalyst sinter and/or (2) disintegration of the active CeO 2 crystal phases, which might cause the catalyst deactivation. Alternatively, coke formation over solid catalysts from DMF could lead to continuing oxidation at high temperature. Furthermore, it is widely recognized that catalytic stability is an important parameter for industrial applications. The stability test on the DMF conversion was performed over 100 h at a constant temperature setting of 400 °C, as shown in Figure 10. The catalyst showed good durability during a continuous DMF conversion, with no significant deactivation after 100 h time-on-stream at 86%. However, there is a negligible performance decrease around 14% with continuous time-on-stream. The slight deactivation of the CeOx during the stability process might be due to the following reasons: (1) active oxygen vacancy sites decrease due to catalyst sinter and/or (2) disintegration of the active CeO2 crystal phases, which might cause the catalyst deactivation. Alternatively, coke formation over solid catalysts from DMF could lead to continuing oxidation at high temperature.  The DMF oxidation reaction pathway over CeO x is proposed based on the experimental data, as shown in Figure 11, in accordance with the previous literature [20,22,47]. The first step is the breakage of the C-N bond in DMF via hydrothermal decomposition, which leads to H-attraction, producing HCOOH and HN(CH 3 ) 2 . Then, the HCOOH will be further oxidized to CO 2 and H 2 O, and the HN(CH 3 ) 2 will also gradually be oxidized using the lattice oxygen to produce NO 2 , H 2 O, and CO 2 , which agree with the final products detected (i.e., NO 2 , H 2 O, and CO 2 ). The DMF oxidation reaction pathway over CeOx is proposed based on the experimental data, as shown in Figure 11, in accordance with the previous literature [20,22,47]. The first step is the breakage of the C-N bond in DMF via hydrothermal decomposition, which leads to H-attraction, producing HCOOH and HN(CH3)2. Then, the HCOOH will be further oxidized to CO2 and H2O, and the HN(CH3)2 will also gradually be oxidized using the lattice oxygen to produce NO2, H2O, and CO2, which agree with the final products detected (i.e., NO2, H2O, and CO2).

Conclusions
A cerium-based nanocatalyst for DMF oxidation was synthesized in this work using the sol-gel method. Structure, morphology, elemental distribution and the optical properties of the synthesized CeOx were characterized using XRD, TEM-EDS, XPS, and UVvisible spectroscopy. The crystal structure shows that CeOx has a smaller grain size, and the aggregation of spherical and fine particles is observed via TEM analysis. XPS analysis showed that CeOx samples exhibited a large amount of beneficial adsorbed oxygen. Due to the combined effect of small grain size, high OAds content, and highly dispersed Ce 4+ on the catalyst surface, the DMF conversion reached ⁓82 ± 2%. CeOx possessed a reaction rate of 1.44 mol·g -1 cat·s -1 and apparent activation energy of 33.30 ± 3 kJ mol −1 . CeOx is stable after 100 h time-on-stream of DMF conversion. A mechanism was proposed to understand the decomposition of DMF to form products such as NO2 and CO2. The overall results of this work show that CeOx provides a more suitable active site for DMF oxidation and could be a potential catalyst for further industrial applications.
Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1, Section S1: Synthesis method of as-prepared samples; Section S2: Experimental setup for catalytic tests; Section S3: Catalyst structure; Section S4: Catalytic performance. References [26,48] are cited in the supplementary materials.

Conclusions
A cerium-based nanocatalyst for DMF oxidation was synthesized in this work using the sol-gel method. Structure, morphology, elemental distribution and the optical properties of the synthesized CeO x were characterized using XRD, TEM-EDS, XPS, and UV-visible spectroscopy. The crystal structure shows that CeO x has a smaller grain size, and the aggregation of spherical and fine particles is observed via TEM analysis. XPS analysis showed that CeO x samples exhibited a large amount of beneficial adsorbed oxygen. Due to the combined effect of small grain size, high O Ads content, and highly dispersed Ce 4+ on the catalyst surface, the DMF conversion reached~82 ± 2%. CeO x possessed a reaction rate of 1.44 mol·g −1 cat ·s −1 and apparent activation energy of 33.30 ± 3 kJ mol −1 . CeO x is stable after 100 h time-on-stream of DMF conversion. A mechanism was proposed to understand the decomposition of DMF to form products such as NO 2 and CO 2 . The overall results of this work show that CeO x provides a more suitable active site for DMF oxidation and could be a potential catalyst for further industrial applications.

Conflicts of Interest:
The authors declare that no competing financial or personal interests could have appeared to influence the work.