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

Abstract

Cerium oxide nanopowder (CeO_x) 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 CeO_x, playing a key role during DMF oxidation. The catalyst had a reaction rate of 1.44 mol g⁻¹_cat 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.

1. 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 high conversion efficiency. It involves reaction at lower temperatures and thus could be highly effective in oxidizing NVOC pollutants into harmless products, such as CO₂, H₂O, and N₂ [11,19]. Therefore, this method is very suitable for the conversion of DMF. In recent years, some researchers have studied the catalytic degradation of NVOCs. Noble metals, such as Rh-, Ru-, Pd-, and Pt-based catalysts, have often been utilized for DMF elimination due to their superior low-temperature activity [10]. Among them, Ru has been reported to show the best performance [20,21]. However, because of the significant leaching of the active material induced by the coordination with the lone-pair electrons in the intermediate products, such as amine, that occurs during the oxidation reaction, these catalysts are not suggested for the combustion of DMF [1,22]. Nevertheless, noble metals are limited resources, and high costs have severely restricted their industrial applications. As transition metal oxides, cerium oxides appear to be good potential metal oxides because of their high oxygen storage capacity and rate of oxygen transfer, which are advantageous for their shape, giving CeO_x a positive effect during the oxidation reaction [23,24,25].

The main purpose of this study is to synthesize CeO_x nanocatalyst for DMF oxidation using the sol–gel technique. Particular attention has been paid to the structure, redox properties, and catalytic activity of the CeO_x nanopowder. XRD, BET, TEM, XPS, and UV–Vis were used to characterize the as-synthesized catalyst. Temperature-programmed and FTIR methods were utilized to investigate the thermal stability and redox properties. The catalytic performance of the powder for the DMF oxidation was evaluated in a fixed bed quartz reactor at room temperature. This aimed to correlate the activity, tailored structure, and physicochemical characteristics of CeO_x.

2. Experimental

2.1. 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(NO₃)₃.6H₂O, 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.

2.2. 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 (H₂-TPR) and temperature-programmed oxidation ((O₂-TPO) AutoChem Ⅱ 2920, Micromeritics). The optical properties were evaluated with a SHIMADZU UV−2600i visible–near-infrared spectrophotometer.

2.3. Catalytic Test

The catalytic performance of prepared Ce₂O₃ 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% O₂ 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:

DMF conversion (at temperature T) = [(X_in − X_out)/X_in] × 100

(1)

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.

$${\mathrm{CO}}_2/\mathrm{NO}\mathrm{selectivity}(\%)=\frac{{\mathrm{CO}}_2/\mathrm{NO}\left(\mathrm{vol}.\%\right)}{\mathrm{Total}\mathrm{products}\mathrm{C}\left(\mathrm{vol}.\%\right)}\times 100$$

(2)

3. 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 (CeO₂) (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).

$$D=\frac{0.9\lambda }{\beta cos\theta }$$

(3)

$$\epsilon =\frac{\beta }{2cos\theta }$$

(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. Grain size and microstrain of the catalyst.

Catalyst	hkl	FWHM (β)	2θ (º)	Crystallite Size D (nm)	D Average (nm)	Microstrain $\mathit{\epsilon }$ (%)	$\mathit{\epsilon }$ Average (%)
CeOx	111	0.42	28.55	19.52	20.07	0.14	0.12
	220	0.43	47.55	20.19		0.13
	311	0.44	56.38	20.49		0.11

Note: hkl refers to Miller indicies, θ refers to Bragg’s angle, and FWHM refers to the full width at half maximum of the peak.

. 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 m² g⁻¹, with a pore volume of 0.01 ± 0.001 cm³ g⁻¹ and a pore size of 25.7 ± 0.2 nm at the relative pressure of P/P₀ = 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 CeO₂ (53.48 nm) [1] and CeO₂-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.

3.2. 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-CeO₂ [11,23] and the large particles of CeO₂ 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.

3.3. 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. Deconvoluted XPS spectrum of Ce 3d distribution in the catalyst.

Catalyst	Parameters			Ce 3d5/2				Ce 3d3/2
CeOx	Species	Ce4+	Ce3+	Ce4+	Ce3+/Ce4+	Ce3+	Ce4+	Ce3+	Ce4+	Ce3+	Ce4+
	BE (eV)	882.2	883.9	888.6	0.42	896.4	898.1	900.6	900.8	904.0	907.2
	RA (%)	28.97	29.58	41.45		3.58	27.84	5.51	36.78	6.04	20.25

Note: BE refers to binding energy, and RA refers to the relative area of the peak.

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 1.3.1 software (Copyright 2001-2015, Marcin Wojdyr). Ce 3d_5/2 was deconvoluted into three peaks at 882.2, 883.9, and 888.6 eV, while Ce 3d_3/2 was fitted into six peaks at 896.4, 898.1, 900.6, 900.8, 904.0, and 907.2 eV. The peaks at 883.9, 896.4, 900.6, and 904.0 eV are assigned to Ce³⁺, and the peaks at 882.2, 888.6, 898.1, 900.8, and 907.2 eV are assigned to Ce⁴⁺ [29,30]. Ce 3d mainly exists in the form of Ce⁴⁺ 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 Ce³⁺/Ce⁴⁺ to decrease, which could be due to Ce³⁺ being converted to Ce⁴⁺.

Figure 4 is an XPS spectrum of the valence state and distribution of elements before the catalytic test, and

Table 3. Deconvoluted XPS spectrum of O 1s distribution in the catalyst.

Catalysts	Parameter	O 1s
CeOx	Species	O2−	OH−	H2O	OLat/OAds
	BE (eV)	529.1	532.5	534.7	0.21
	RA (%)	17.31	73.38	9.31

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 (O²⁻) at 529.1 eV. The higher binding energy peak corresponds to the adsorbed oxygen (O_Ads), expressed as OH^-, and H₂O 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 H₂O, NO₂, and CO₂, which improves the mass transfer driving force of product diffusion, thus improving the reaction rate.

3.4. Redox Properties

H₂-TPR studies the reducibility of the prepared catalysts, and the results of the H₂ consumption by CeO_x as a function of temperature are shown in Figure 5a and

Table 4. H₂-TPR profiles and O₂-TPO peaks data.

Cat.	H2 Consumptions (mmol/g)				O2 Consumptions (mmol/g)
	Peak 1	Peak 2	Peak 3	Total	Peak 1	Total
CeOx	4.03	1.31	6.52	11.86	7.08	7.08
	575.6 °C	589.9 °C	641.8 °C		693.4 °C

. 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 Ce⁴⁺ to Ce³⁺ on the cerium oxide surface with oxygen vacancy, while the one > 600 °C represents a reduction of Ce⁴⁺ to Ce³⁺ by H₂ inside the bulk oxygen from cerium oxides species [1,31]. Therefore, the H₂-TPR profiles show that the CeO_x could result from the removal of bulk oxygen of cerium oxide.

The oxygen mobility in the CeO_x 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 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.

3.5. 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:

αhν = A (hν − E_g)ⁿ

(5)

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 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.

3.6. 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—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, 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 Ce⁴⁺ 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 H₂-TPR and O₂-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 (−E_a/RT)

(6)

r (mol g⁻¹ s⁻¹) = F₀ X/W_cat

(7)

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, F₀ 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 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 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 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 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(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).

4. 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 Ce⁴⁺ 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 NO₂ and CO₂. 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.