Surface Reduced CeO₂ Nanowires for Direct Conversion of CO₂ and Methanol to Dimethyl Carbonate: Catalytic Performance and Role of Oxygen Vacancy

Abstract

Ultralong 1D CeO₂ nanowires were synthesized via an advanced solvothermal method, surface reduced under H₂ atmosphere, and first applied in direct synthesis of dimethyl carbonate (DMC) from CO₂ and CH₃OH. The micro morphologies, physical parameters of nanowires were fully investigated by transmission electron microscopy (TEM), X-ray diffraction (XRD), N₂ adsorption, X-ray photoelectron spectrum (XPS), and temperature-programmed desorption of ammonia/carbon dioxide (NH₃-TPD/CO₂-TPD). The effects of surface oxygen vacancy and acidic/alkaline sites on the catalytic activity was explored. After reduction, the acidic/alkaline sites of CeO₂ nanowires can be dramatically improved and evidently raised the catalytic performance. CeO₂ nanowires reduced at 500 °C (CeO₂_NW_500) exhibited notably superior activity with DMC yield of 16.85 mmol gcat⁻¹. Furthermore, kinetic insights of initial rate were carried out and the apparent activation energy barrier of CeO₂_NW_500 catalyst was found to be 41.9 kJ/mol, much tiny than that of CeO₂_NW catalyst (74.7 KJ/mol).

1. Introduction

As an environmentally benign compound and unique intermediate of versatile chemical products, dimethyl carbonate (DMC) is widely applied in polymer industry and pharmaceutical as well as detergent, surfactant, and softener additives [1,2]. In addition, DMC is important raw material when serving as a non-toxic substitute for poisonous phosgene and dimethyl sulfate in sustainable chemistry of carbonylation, methylation, and polymer synthesis [3,4]. As an additive, DMC can improve the octane number and oxygen content of fuels, thereby enhancing its antiknock [1]. Furthermore, DMC can be used as a cleaning solvent in coating paints and the important composition of electrolyte [5]. Considering the wide applications, DMC is known as the “new cornerstone” for synthesis chemistry nowadays and lots of efforts have been made in finding appropriate routes to meet the demand of DMC industrial production since it is far from satisfaction until now. Several approaches including the methanolysis of phosgene [6], the oxidative carbonylation of methanol [7], the transesterification of alkene carbonates [8], and the alcoholysis of urea [3], have been developed, but it is still limited with strict operation conditions, highly toxicity, and corrosivity up to now.

Using carbon dioxide (CO₂) in DMC synthesis is particularly attractive since CO₂ is known as a recyclable and naturally abundant raw materials for the production of plentiful chemical reagents. Meanwhile, the emissions of CO₂ have significantly increased and contributed to global warming, thus the utilization of CO₂ has attracted more and more attention in the last decades [9,10]. In this regard, the direct synthesis of DMC from CO₂, and methanol (Scheme 1) is considered as one of the most attractive and effective methods since such an approach is environmentally benign not only for reduction of greenhouse gas emissions but also for development of a new carbon resource [11,12]. However, such a sustainable route also exists significant challenges due to facts including the highly thermodynamically stability of CO₂, as well as the kinetically inert and deactivation of catalysts induced by water formation in the reaction process [13,14,15].

Several methods, such as adding co-reagents and dehydrants in the reaction systems, have been developed [16,17]. Furthermore, some new technologies, such as photo-assistant [14], electro-assistant [18], membrane separation [19], and supercritical CO₂ technology [20,21] have been introduced to boost the production of DMC in former reports. Even then, the reactions are preferred at strict conditions and the yield of DMC is relatively low. Though the efforts to these approaches are devoted today, the explorations of advanced heterogeneous catalysts are still regarded as the most effective route [22,23,24]. In particular, CeO₂ based catalysts have been transplanted in the direct synthesis of DMC and show much better catalytic activity as an excellent heterogeneous catalyst [25,26]. Plenty of references has employed CeO₂ as competent catalysts in DMC formation involving dehydration [27]. Furthermore, previously studies have revealed that the different crystal facets exposed on the surface of CeO₂ nanostructures were strongly controlled by its morphology, leading to differential physicochemical properties and further effecting the catalytic performance [28]. In this context, 1D structured CeO₂ nanorods catalyst demonstrated superior DMC yield (0.906 mmol DMC/mmol cat) from CO₂ and methanol when compared to CeO₂ nanocubes (0.582 mmol DMC/mmol cat) and CeO₂ nano-octahedrons (0.120 mmol DMC/mmol cat) [25,29]. However, major drawbacks of CeO₂ nanorods are the extremely low yield and high cost of hydrothermal method, preventing it from being used in practical applications [30]. In the meantime, the low aspect ratio of nanorods limits the specific surface area of catalysts, which would affect the catalytic performance further [25,26].

In this respect, we were especially interested in new trials for ultralong 1D CeO₂ nanostructure. Furthermore, oxygen deficiency of the CeO₂ based catalyst has been proved playing important roles in CO₂ and methanol activation in former research [26,31]. Thus, we conducted further research on surface reduced CeO₂ nanowires catalyst. Herein, CeO₂ nanowires with a diameter of 10 nm and an aspect ratio of more than 50 was successfully prepared by the refluxing approach established by Yu et al. [32] and then simply surface reduced under hydrogen atmospheres, followed by their application in the direct synthesis of DMC from CO₂ and methanol. Moreover, the influence of surface oxygen-deficiency and acid-basic sites were fully investigated. The catalytic recyclability was also detected. Finally, we conducted a detailed kinetic investigation for the direct formation of DMC in an autoclave reactor over catalysts.

2. Results and Discussion

2.1. Morphology and Microstructure of the Prepared Catalysts

CeO₂ nanowires catalyst was prepared using a solvothermal method in a mixed water/ethanol solvents (v/v = 1:1). Figure 1a,b show the morphology of the unreduced CeO₂ nanowires catalyst, exhibiting an intact nanowire structure with an average length of around 500 nm and a uniform diameter of less than 10 nm. After reduced with H₂, the nanowire structure was kept undestroyed, and the size of nanowires has made almost no change (Figure 1c). The crystal structures of CeO₂ nanowires catalysts were investigated by XRD, and the spectra are shown in Figure 2. For unreduced CeO₂ nanowires, the diffraction peaks of 2θ can be ascribed to the fluorite-structured CeO₂ (JCPDS 34-0394, 28.6° (111), 33.1° (200), 47.6° (220), and 56.4° (311)). After reduction under H₂ atmosphere as a function of temperature (450–700 °C), the spectra of nanowires remained almost unchanged, indicating that the crystalline stucture of the nanowires was not destroyed.

Physical and chemical parameters of the as-prepared catalysts are summarized in

Table 1. Textural data of as-prepared catalysts basing on BET, XRD, and XPS investigation.

Catalysts	BET Surface Area (m2g−1)	Sites Amount (μmol g−1)		Oxygen Vacancy (%)
		Moderate Acidity	Moderate Alkalify
CeO2_NW	116.3	82	25	5.1
CeO2_NW_450	104.9	190	52	14.2
CeO2_NW_500	98.1	282	82	20.5
CeO2_NW_550	70.4	235	65	22.7
CeO2_NW_600	55.2	192	48	25.1
CeO2_NW_650	42.5	171	47	27.3
CeO2_NW_700	28.7	115	39	31.2

. The specific surface area of the nanowires was acquired from BET method and seemed to decrease slightly from 116.33 m²g⁻¹ to 98.1 m²g⁻¹ upon increasing the reduction temperature up to 500 °C, indicating that the low-temperature reduction only can influence the specific surface area within tolerable extent. When elevating the reduction temperature, specific surface area of reduced CeO₂ nanowires drops abruptly, which inevitably lead to the covered up of efficient active sites and eventually cause the worse catalytic performance.

Further investigation on the surface acidic/alkaline properties of as-prepared catalysts was acquired by NH₃/CO₂-TPD. The amount of moderate acidic and alkaline sites is also summarized in Table 1. For CeO₂_NW_450 and CeO₂_NW_500, more plentiful moderately acidic and alkaline sites are generated with the elevating of reduction temperature, which is bond to benefit the DMC formation according to former research. [33] While for other nanowires reduced under higher temperature, the amount of moderate acidity and alkalinity lessens. It can be ascribe to the fast-declining specific surface area upon elevating the reduction temperature, and then result in the covered up of efficient active sites. CeO₂_NW_500 was determined to possess both the richest acidity and alkalify, which is mainly because of the enriching of oxygen vacancy on catalysts surface, further providing much richer active sites when compared to unreduced nanowires.

Surface chemical state of prepared catalysts was investigated by XPS. Based on the calculated result, the XPS result forecasts that the oxygen vacancy on the surface of as-prepared nanowires varies along the reduction temperature from 5.1% in CeO₂_NW to 31.2% in CeO₂_NW_700. Due to the Ce valent state partly shift from +4 to +3, reduction of CeO₂ nanowires leads to the formation of oxygen vacancy on the catalyst surface.

Based on the aforesaid result, a certain relationship between the surface active sites (mainly the moderate acidic and alkaline sites), specific surface area and surface oxygen vacancy was established. Both the moderate acidic and alkaline sites showed a linear relationship contrast specific surface area multiply surface oxygen vacancy (Figure 3). The combination of TPD, XPS, and BET reveals that the oxygen vacant structure of nanowires contributes to the formation of moderately acidic and basic sites, which is also influenced by the specific surface area.

2.2. Catalytic Performance

The effects of reduction temperature for nanowires on the catalytic activity were probed and the catalytic reaction was conducted in a stainless autoclave micro-reactor with high-speed stirring. The DMC yield of as-prepared catalysts with different catalysts are demonstrated in Figure 4 and serves as the basis for original selection of reduction temperatures.

Unreduced CeO₂ nanowires (CeO₂_NW) catalyst obtained much inferior DMC yield when compared with the surface reduced CeO₂ nanowires (CeO₂_NW_x). DMC yield enhanced with elevating the reduction temperature of CeO₂ nanowires, reached a maximum at 500 °C and then declined with further temperature rise. We observed the catalytic performance of all nanowires catalysts with loading amount of 0.2 g and 0.5 g respectively. The reaction found to reach saturated and catalytic performance was influenced by leveling effect when loading 0.5 g catalyst, while the catalyst was efficiently utilized at 0.2 g. Among the catalysts examined, CeO₂_NW_500 catalyst achieves excellent DMC yield of 16.85 mmol gcat⁻¹, superior than the catalytic activity of CeO₂_NW catalyst (5.45 mmol gcat⁻¹) under the same condition. Associating with the specific surface area of as-prepared catalysts, CeO₂_NW_x (x > 500) catalysts with tinier surface area proof inferior catalytic activity, illustrating that the catalytic performance is directly related to the specific surface area of as-prepared catalysts. Smaller specific surface area necessarily leads to the covering of efficient active sites and causes lower catalytic activity [34,35].

Further research on direct synthesis of DMC from CO₂and methanol over CeO₂_NW_500 was conducted. The effects of different catalytic conditions was fully investigated. Figure 5 shows the DMC amount with different reaction time and reaction temperatures over CeO₂_NW_500 catalysts catalyst. The generation rate of the destination product DMC enhanced when elevated the catalytic temperature, while the final yield of DMC constantly decreased due to the limitations of thermodynamic and generation of side product. Under 140 °C, the yield of DMC reached the maximum value at 75 min, and then seemed to be almost unchanged. However, the formation amount of DMC even trends increasing after 5 h at 120 °C.

Further investigation for the recyclability of CeO₂_NW_500 was carried out and the used nanowires catalyst was thermal reduced under H₂ atmosphere before re-catalyze the direct synthesis of DMC under the same reaction conditions. BET specific surface area and catalytic performance of the recovered CeO₂_NW_500 catalysts are demonstrated in Figure 6. Both specific surface area and the catalytic performance were found mildly falling as the number of reuses accumulates, which is on account of the slight surface collapse during the retreatment of the catalysts. Anyway, CeO₂_NW_500 shows favorable stability for the direct formation of DMC from CO₂ and methanol.

2.3. Kinetic Analysis

Initial rate kinetic insights in the direct synthesis of DMC over CeO₂_NW_500 catalyst were conducted. Based on the result of Figure 5, the yield data within 60 min was selected as the initial rate region. In addition, similar initial reaction was carried out on CeO₂_NW and compared with that of CeO₂_NW_500. The linear fitting of Arrhenius plot in Figure 7 gives a slope at −5.04 (±0.41), indicating the apparent activation energy at 41.9 ± 3.4 kJ/mol for CeO₂_NW_500 catalyst, which is lower than CeO₂_NW catalyst (74.7 kJ/mol). It suggests that the surface reduction of CeO₂ nanowires has reduced the activation energy barriers and improved the catalytic performance by enriching the surface active sites.

Furthermore, based on the former proposed mechanism for the direct synthesis of DMC, surface adsorption and activation of CO₂ and methanol occurs on the alkaline sites and acidic sites, respectively [36]. As a consequence, the inferior catalytic performance of unreduced CeO₂_NW catalyst in this study should mainly ascribe to much poorer surface acidic and alkaline sites. While for CeO₂_NW_x catalysts, more abundant moderately acidic and alkaline sites generate along with the surface reduction process, thus resulting in more favorable catalytic performance than CeO₂_NW. Figure 8 shows the Ln–Ln curves of initial rate for each catalysts contrast the concentration of moderately acidic/alkaline sites. There is a positive liner relationship of these parameters, suggesting the initial rates of this reaction influenced by the activation of both CO₂ and methanol. This result corresponds to the deduction of Langmuir–Hinshelwood mechanism [37,38].

3. Materials and Methods

3.1. Materials

Cerium (III) nitrate hexahydrate Ce(NO₃)₃·6H₂O, methanol, and dimethyl carbonate (DMC) were purchased from Aladdin Co., Ltd. (Shanghai, China). Ammonium hydroxide NH₃·H₂O (25 wt %) and ethanol was purchased from Guangdong Chemical Reagent Factory (Guangzhou, China). All the reactants were of analytical purity and used without any further treatment.

High purity CO₂ (>99.9999%) and H₂ (>99.999%) were obtained from Guangqi Gas Co., Ltd. (Guangzhou, China).

3.2. Catalysts Preparation

Ceria nanowires catalyst was prepared using a solvothermal method in a mixed water/ethanol solvents [32]. Briefly, stoichiometric Ce(NO₃)₃·6H₂O was dissolved in a flash with water/ethanol mixed solution (v/v = 1:1), following by oil bath heating up to 140 °C. Then the NH₃·H₂O was added into the flash and the reaction mixture was refluxed for 12 h under stirring. After cooling to room temperature, the resulting mixture was separated by centrifugation. Afterward, the solid product was bathed with a mixture of ethanol and water (v/v = 1:1) for several times. After that, the pre-synthesized nanowires were freeze-dried in a lyophilizer (Four-Ring Science Instrument Plant Beijing Corporation, Beijing, China) at a vacuity of 3 mbar and a frigorific temperature of −40 °C. Finally, as-prepared nanowires catalyst were thermal reduced under H₂ atmosphere for 4 h. The samples were named after CeO₂_NW and CeO₂_NW_x, in which x represented the reduced temperature.

3.3. Catalyst Characterization

Micromorphology measurement was carried out on a transmission electron microscope (TEM, JSM-2010HR, JEOL Ltd., Tokyo, Japan) at a high voltage of 200 kV. Samples were ultrasonic dispersed into ethanol absolute, and then dropwise loaded onto the micro copper grid, followed by drying under air condition at room temperature.

Powder X-ray diffraction (XRD) was measured on a XRD diffractometer (Dmax 2200, Rigaku Ltd., Tokyo, Japan) at a scan rate of 5°/min. High voltage of 40 kV and current of 30 mA were employed in this measurement and Cu Kα radiation target (λ = 0.154178 nm) was used.

N₂ adsorption characterization was acquired on a nitrogen adsorption apparatus (ASAP-2020, Micrometrics Ltd., Cumming, GA, USA) and the specific surface area was calculated through the Brunauer–Emmett–Teller method from the adsorption results. Samples were pre-treated under nitrogen atmosphere at 200 °C for 2 h. After cooling, N₂ at a flow rate of 110 mL/min was adsorbed on the samples surface in a U tube surrounded by liquid nitrogen.

Temperature programmed desorption (TPD) was conducted on a chemical adsorption apparatus (Chem-BET 3000, Quantachrome Ltd., Boynton Beach, FL, USA). Firstly, samples were pre-treated under nitrogen atmosphere at 200 °C for 1 h. Then, a mixture standard gas of 10%CO₂/90%N₂ or 10%NH₃/90%N₂ saturated with the samples for 30 min at a flow rate of 60 mL/min in a U tube. After that, surface physical adsorption of CO₂ or NH₃ was dislodged by bathing with 30%N₂/70% He standard gas for 2 h at a flow rate of 50 mL/min. Then, the samples were thermal treated under N₂/He 30%N₂/70% He from room temperature up to 600 °C with 8 °C/min heating rate. Finally, the total desorption of NH₃/CO₂ was determined through back-titration method. HCl/NaOH (0.01 mol/L) was employed as an adsorbent for NH₃/CO₂, NaOH/HCl (0.01 mol/L) was served as titrant together with a mixed indicator reagent, which consisted of bromocresol green ethanol solution (1%, 3 equivalent volumes) and methyl red ethanol solution (2%, 1 equivalent volume) [39].

X-ray photoelectron spectrum (XPS) was acquired on an X-ray photoelectron spectrometer (ESCALAB250, Thermo Fisher Scientific Ltd., Waltham, MA, USA) with a scan survey of 1100-0 eV binding energy range. Monochromatized Al-Ka source at 1486.6 eV and 150 w was applied in the characterization with a voltage of 15 kV. Surface oxygen vacancy can be roughly calculated according to the equation

$$O_{Vac} (\%) =\frac{C\left(Ce\right) -0.5 C\left(O\right)}{C\left(Ce\right)}$$

(1)

3.4. Catalytic Performance Measurement

Direct synthesis of DMC from CO₂ and methanol was carried out in a stainless steel autoclave with a volume of 50 mL and high-velocity stirring. As-prepared catalyst and a certain amount of absolute methanol were added into the reactor, following be purging CO₂ for several times to evacuate the air inside and obtained the strict oxygen-free and water-free condition. Reaction pressure of CO₂ was set at 5 MPa and the reaction was conducted at 120 °C for 5 h if no otherwise specified. The final products were measured and quantified by a gas chromatograph (GC-7900II, Techcomp Ltd., Beijing, China) equipped with a flame ionization detector (FID) after filtrating with PES membrane with a pore size of 0.45 um.

4. Conclusions

Ultralong 1D CeO₂ nanowires were synthesized via an advanced solvothermal method, surface reduced under H₂ atmosphere, and firstly applied in direct synthesis of dimethyl carbonate (DMC) from CO₂ and CH₃OH. The influences of reduction temperatures for the nanowires and different operating conditions for the catalysis reactivity were fully explored. The catalysis reactivity of ceria nanowires was founded to be greatly improved after surface reduction by generating more surface acidic-alkaline sites. Among the catalysts investigated, CeO₂_NW_500 obtains the most favorable catalytic activity for DMC formation than CeO₂_NW and all of the other CeO₂_NW_x catalysts. Under optimal reaction conditions, CeO₂_NW_500 catalyst achieves the best catalysis reactivity with DMC yield of 16.85 mmol gcat⁻¹ in an autoclave reactor. Based on the approach of initial rates method, the kinetic insight were conducted for the direct synthesis of DMC over CeO₂_NW_500 catalyst and the activation energy barrier is determined to be 41.9 kJ/mol, tinier than 74.7 kJ/mol for unreduced CeO₂ nanowires. Moreover, a certain relationship between the initial rate and the surface acidity/alkalify was found, which is identical to the deduction of former proposed Langmuir–Hinshelwood mechanism where the initial rates of this reaction are influenced by the activation of both CO₂ and methanol.