Morphology Dependence of Catalytic Properties of CeO₂ Nanocatalysts for One-Step CO₂ Conversion to Diethyl Carbonate

Abstract

The conversion of CO₂ into value-added chemicals exemplifies an innovative and eco-friendly approach to addressing carbon emissions. In this study, shape-specific CeO₂ nanocrystals (nanorods, nanocubes, and nanoparticles) were successfully synthesized and employed as catalysts to study the structure-dependent behavior and reaction mechanism for one-step CO₂ conversion to diethyl carbonate (DEC). Among the three catalysts, CeO₂ nanorods (Ce-NR) exhibited the best catalytic activity in the synthesis of DEC from CO₂ compared with CeO₂ nanocubes (Ce-NC) and nanoparticles (Ce-NP), which achieved the DEC production of 1.32 mmol_DEC/g_cat at 423 K and 5 MPa for 4 h. Comprehensive characterization further confirmed the enhanced activity of Ce-NR originated from the morphology effect, particularly the promotion of oxygen vacancies and Ce³⁺ species, which promoted reaction activity. Furthermore, the Ce-NR catalyst almost retained 1.32 mmol_DEC/g_cat DEC production of its initial activity after four cycles, underscoring its exceptional stability and promising industrial scalability. These findings provide fundamental insights to guide the rational design of efficient catalysts for CO₂ activation and other critical transformations.

1. Introduction

Escalating levels of atmospheric CO₂, largely a consequence of fossil fuel combustion, pose a significant threat through their contribution to global warming and climate change [1,2,3]. In response, chemical conversion of CO₂ has emerged as a key mitigation strategy, which can be fundamentally classified into reductive and non-reductive pathways. Reductive conversion, often requiring substantial energy input, generates products like CO [4], methanol [5], ethanol [6], and hydrocarbons [7]. Conversely, non-reductive conversion leverages CO₂ as a carbonyl source in reactions with nucleophiles like alcohols and amines to synthesize carbonates [8], carbamates [9], and ureas [10]. Given the thermodynamic stability of CO₂, the non-reductive approach is particularly attractive for its relatively lower energy demands and its potential to replace hazardous reagents such as phosgene in sustainable chemical synthesis, making it a promising direction for green industrial applications.

The direct synthesis of dialkyl carbonates (DAC) from CO₂ and alcohols represents a pivotal green chemistry pathway for carbon valorization, transforming a potent greenhouse gas into valuable compounds like dimethyl carbonate (DMC) and diethyl carbonate (DEC), which are essential for electrolytes, polymers, and green solvents [11,12,13]. However, the inherent stability of the C–O bond in CO₂ renders it chemically inert and challenging to activate [14]. Therefore, the central research focus has shifted to designing efficient catalysts capable of activating CO₂ molecules. Various efficient catalysts have been investigated to modulate the active sites on the surface of the catalyst to address the issue of CO₂ activation, such as ZrO₂ [15], CeO₂ [16], molecular sieve [17], ionic liquids [18], metal–organic framework [19], and so on.

Many metal oxide nanomaterials have been explored as efficient heterogeneous catalysts for the direct synthesis of DAC from CO₂ and alcohol [20]. In particular, CeO₂ stands out as an effective catalyst for this transformation owing to its unique surface properties. The coexistence of acid–base sites and redox properties, linked to the switchable Ce⁴⁺/Ce³⁺ oxidation states, facilitates the adsorption and activation of methanol and CO₂ [21]. Research efforts have focused on finding efficient strategies to modify the strength of the acid–base sites and the redox properties to improve the catalytic activity, specifically targeting higher alcohol conversion and DAC production [22]. For example, Hou et al. [23] introduced abundant electron-enriched lattice oxygen species into CeO₂ catalyst by constructing the point defects and crystal-terminated phases in the crystal reconstruction process. Benefiting from the acid–base properties modulated by the electron-enriched lattice oxygen, the optimized CeO₂ catalyst exhibited a much higher DMC production of 22.2 mmol g⁻¹, surpassing most reported metal-oxide-based catalysts under comparable conditions. Indeed, catalyst morphology and exposed crystal planes are critical determinants of performance in redox reactions, as they dictate surface energy, atomic arrangement, and synergistic interactions [24]. In the case of CeO₂, varied morphologies lead to different densities of acid–base sites and oxygen vacancies, which in turn dictate activity. Wang et al. [25] reported that spindle-like CeO₂ yielded the most DMC, followed by nanorods, nanocubes, and nano-octahedra. Furthermore, Zhang et al. [26] synthesized CeO₂ hollow spheres with controlled sizes and shell numbers, finding a positive correlation between activity and the density of active sites within the cavity volume. Recently, Yang et al. [27] also showed that CeO₂ with high surface area and oxygen vacancy concentration served as efficient sites for CO₂ adsorption/activation, yielding high activity. Thus, ceria morphology is intimately connected to its surface defects and catalytic function, profoundly influencing reactant adsorption and activation [28]. Although prior work has demonstrated the efficacy of morphologically tuned, doped ceria catalysts (e.g., Zr-doped CeO₂) for DEC synthesis [29], a dedicated study on the intrinsic morphology–activity relationship of undoped CeO₂ nanocrystals—specifically, how morphology tailors oxygen vacancy/Ce³⁺ abundance and thereby controls the kinetics of the critical C–O bond formation step—remains an open area of investigation.

Therefore, this study presents the template-free hydrothermal synthesis of ceria catalysts with distinct surface morphologies and evaluates their catalytic performance in the direct synthesis of DEC from CO₂ and ethanol. A combination of characterization techniques, including X-ray diffraction (XRD), Fourier-transform infrared spectra (FTIR), temperature-programmed desorption of ammonia and carbon dioxide (NH₃-TPD and CO₂-TPD), X-ray photoelectron spectroscopy (XPS), and transmission electron microscopy (TEM), was employed to elucidate the morphology and structural properties of the synthesized ceria materials. The findings uncover a fundamental correlation between the morphology of ceria, its acid–base site distribution, surface oxygen species, and the resulting catalytic activity.

2. Results

2.1. Catalysts Characterization

Figure 1 displayed the X-ray diffraction (XRD) patterns of the three CeO₂ morphologies. The diffraction peaks observed at 2 θ = 28.6°, 33.2°, 47.5°, 56.3°, 59.4°, 69.5°, 76.7°, 79.8°, and 88.3° corresponded, respectively, to the (111), (200), (220), (311), (222), (400), (331), and (420) crystal planes of the fluorite cubic structure of CeO₂ (JCPDS No. 34-0394) [30]. No impurity-related diffraction peaks were detected, confirming the high phase purity of the as-synthesized CeO₂ catalysts. The average crystallite sizes of Ce-NR, Ce-NP, and Ce-NC, calculated using the Scherrer–Debye equation, were 8.46 nm, 5.92 nm, and 25.44 nm, respectively (

Table 1. The crystallite size of different CeO₂ morphologies.

Catalyst	Peak Position (2 θ)	FWHM	Crystallite Size D (nm)	Lattice Constant (A)
Ce-NR	28.540	0.659	8.46	5.412
Ce-NP	28.581	1.369	5.92	5.425
Ce-NC	28.556	0.318	25.44	5.411

). Notably, the three CeO₂ morphologies exhibited distinct differences in peak intensity and full width at half maximum (FWHM), primarily reflecting variations in their crystallite size and crystallinity. The broad and weak diffraction peaks of the Ce-NP catalyst were consistent with its smallest crystallite size (5.92 nm), which led to significant peak broadening according to the Scherrer–Debye equation. In contrast, the sharp and intense peaks of the Ce-NC catalyst correspond to its substantially larger crystallite size (25.44 nm), indicating superior crystallinity. These crystallite-size-dependent effects were the main contributors to the observed differences in the XRD line profiles.

Figure 2 shows the Fourier-transform infrared (FTIR) spectra of the three CeO₂ catalysts with distinct morphologies. A broad absorption band observed around 3460 cm⁻¹ corresponded to the O–H stretching and bending vibrations of water molecules adsorbed on the catalyst surface or surface hydroxyl groups [31]. The peak near 1675 cm⁻¹ was attributed to the bending vibration of adsorbed water (H–O–H), while the weak feature at approximately 2882 cm⁻¹ was likely associated with C–H stretching vibrations from residual organic species or environmental adsorption. An additional band at about 1054 cm⁻¹ could be assigned to C–O stretching vibrations. In the region below 1000 cm⁻¹, the characteristic absorption peak located at 550 cm⁻¹ represented the stretching vibration of the Ce–O bond, confirming the formation of the CeO₂ fluorite structure [32]. The formation of CeO₂ was facilitated by the alkaline conditions provided by NaOH, which promoted the generation of OH⁻ groups around Ce³⁺ ions, leading to precipitation and subsequent formation of Ce–O bonds [33]. Notably, the strength of the surface Ce-O bond was influenced by the catalyst’s crystallite size and lattice strain. Enhanced Ce-O bond strength often correlated with a decreased lattice constant and increased crystallite size [34]. Catalysts with larger crystalline sizes and lower surface areas, such as Ce-NC, tended to undergo compressive strain, which strengthened the surface Ce–O bonding. In contrast, smaller crystallites, such as Ce-NP, accumulated point defects that induced tensile strain, thereby weakening the Ce–O bonds on the surface. The absence of additional peaks in the spectra confirmed the high phase purity of the synthesized CeO₂ materials, indicating no significant surface contamination or functionalization.

Figure 3 presents transmission electron microscopy (TEM) images of the as-synthesized CeO₂ nanomaterials, illustrating distinct morphologies including nanorods (Ce-NR), nanoparticles (Ce-NP), and nanocubes (Ce-NC) achieved by precisely controlling key synthesis parameters such as alkali concentration, reaction temperature, and hydrothermal conditions. As shown in Figure 3a,b, Ce-NR samples consisted of uniform nanorods with lengths ranging from 50 to 120 nm and diameters of 8–10 nm. These structures were obtained via hydrothermal processing at 453 K with a high NaOH concentration (6 mol/L), which promoted anisotropic growth along specific crystallographic directions. In contrast, Ce-NP (Figure 3c,d) exhibited irregular polyhedral nanoparticles with an average size of approximately 10 nm. These were synthesized through room-temperature precipitation using a lower NaOH concentration (2 mol/L), which favored rapid nucleation and isotropic growth without pronounced directional crystallization. Meanwhile, Ce-NC (Figure 3e,f) displayed well-defined cubic morphology with edge lengths of 30–50 nm. This morphology was achieved by employing the same high alkalinity as Ce-NR but at a lower hydrothermal temperature of 373 K, thereby altering the relative growth rates of crystal facets. These results demonstrate that precise manipulation of synthetic parameters, particularly alkali strength and thermal conditions, directly governs nucleation kinetics and crystal growth habits, enabling the tailored synthesis of CeO₂ nanostructures with specific morphologies for structure-dependent applications.

The surface chemical composition of the as-synthesized CeO₂ catalysts with different morphologies was examined by X-ray photoelectron spectroscopy (XPS). The survey spectra (Figure 4a) confirmed the presence of Ce and O elements, indicating high purity of the prepared materials. The only other element detected was carbon (C), observed as the C 1s peak. This is a common and expected feature attributed to adventitious carbon from the environment, which is routinely used for binding energy calibration (typically referenced to 284.8 eV). Its presence does not indicate sample impurity. Figure 4b displays the high-resolution Ce 3d spectra, which were deconvoluted into eight characteristic peaks corresponding to two spin–orbit doublets: Ce 3d_5/2 (denoted as u) and Ce 3d_3/2 (denoted as v). This spectral feature confirmed the co-existence of both Ce⁴⁺ and Ce³⁺ oxidation states on the catalyst surfaces. The peaks labeled u‴ (917.5 eV), u″ (907.2 eV), u (901.1 eV), v‴ (897.7 eV), v″ (888.6 eV), and v (882.3 eV) were attributed to Ce⁴⁺, while the peaks u′ (902.5 eV) and v′ (884.9 eV) were assigned to Ce³⁺ [35]. The presence of Ce³⁺ was closely associated with the formation of oxygen vacancies, as the reduction of Ce⁴⁺ to Ce³⁺ is accompanied by the loss of lattice oxygen, and it generates defective sites [36]. The concentration of Ce³⁺, calculated from the peak area ratio Ce³⁺/(Ce⁴⁺ + Ce³⁺), followed the order: Ce-NR (21.1%) > Ce-NC (15.9%) > Ce-NP (14.6%) (

Table 2. XPS results of CeO₂ catalysts with different morphologies (values are the average of three measurements ± standard deviation).

Catalyst	Ce3+/(Ce3+ + Ce4+) (%)	OV/(OL + OC + OV) (%)
Ce-NR	21.1 ± 0.6	24.2 ± 0.2
Ce-NC	15.9 ± 0.2	20.5 ± 0.4
Ce-NP	14.6 ± 0.3	19.8 ± 0.3

). This trend suggested that Ce-NR possesses the highest oxygen vacancy density among the three morphologies. The O 1s spectra (Figure 4c) were fitted into three components: lattice oxygen (O_L) at 528.7 eV, oxygen vacancies (O_V) at 530.9 eV, and chemisorbed oxygen (O_C) at 532.8 eV [37]. The relative concentration of O_V, determined as O_V/(O_L + O_V + O_C), decreased in the sequence Ce-NR (24.2%) > Ce-NC (20.5%) > Ce-NP (19.8%), consistent with the Ce³⁺ content derived from Ce 3d analysis. These results clearly demonstrated that the O_V and Ce³⁺ concentrations were strongly influenced by the morphology of CeO₂, with rod-like CeO₂ exhibiting the highest density of defect sites, which is beneficial for gas-phase CO₂ adsorption and activation in catalytic reactions.

The acidic and basic properties of the three CeO₂ catalysts, which played critical roles in activating ethanol and CO₂, respectively [38], were probed by CO₂- and NH₃-temperature-programmed desorption (TPD). As depicted in Figure 5a,b, the TPD profiles were deconvoluted to identify weak (<473 K), moderate (473–673 K), and strong (>673 K) acid/base sites based on the desorption peak areas [39]. Notably, all samples lacked strong acid–base sites characterized by desorption peaks around 723 K. The quantitative amounts of different acidic and basic sites, summarized in

Table 3. Acidic and basic values of as-prepared CeO₂ catalysts with different morphologies determined from NH₃ and CO₂-TPD measurements.

Catalyst	Acidity (mmol/g)			Basicity (mmol/g)
	Weak	Moderate	Total	Weak	Moderate	Total
Ce-NR	0.32	0.31	0.62	0.29	0.27	0.56
Ce-NP	0.44	0.38	0.82	0.45	0.42	0.87
Ce-NC	0.06	0.05	0.11	0.09	0.08	0.17

, revealed that the total concentration of acid–base sites followed the order Ce-NP > Ce-NR > Ce-NC. Specifically, Ce-NR exhibited the highest concentrations of acid (0.82 mmol/g) and basic sites (0.87 mmol/g). According to previous literature [40,41], strong acid sites promote the formation of diethyl ether or dimethyl ether (DEE/DME), which, along with co-produced H₂O, inhibits the synthesis of diethyl carbonate (DEC). Among the three catalysts, Ce-NC and Ce-NP possessed the lowest and highest concentrations of acid–base sites, respectively. In contrast, Ce-NR exhibited an intermediate acid–base site density between that of Ce-NC and Ce-NP. A few studies had previously mentioned that too high a concentration of acidic or basic sites was unfavorable for the DAC synthesis, possibly due to the difficult desorption of the product or the occurrence of side reactions [42]. Therefore, the reaction involved acid–base synergistic catalysis, and a proper balance of acid–base sites was crucial for the synthesis of DAC from CO₂ and alcohol, as it affects the generation of the key intermediates [43].

2.2. Catalytic Performance

The catalytic performance of CeO₂ with different morphologies (Ce-NR, Ce-NP, and Ce-NC) for the synthesis of diethyl carbonate (DEC) from CO₂ and ethanol was evaluated under fixed conditions: temperature of 423 K, CO₂ pressure 5 MPa, reaction time 4 h, and catalyst mass 0.17 g. As summarized in Figure 6, the Ce-NR catalyst yielded the highest DEC production (1.32 mmol_DEC/gcat), outperforming both Ce-NP (0.69 mmol_DEC/gcat) and Ce-NC (0.72 mmol_DEC/gcat). This superior activity could be attributed to the higher concentration of surface oxygen vacancies in Ce-NR, which promoted CO₂ adsorption and facilitated the formation of a bidentate carbonate intermediate—a key step in the DEC formation pathway. In addition, Ce-NR possessed an optimal acidity/basicity ratio, contributing to a balanced distribution of acid–base sites that further enhanced catalytic efficiency. Crucially, control experiments conducted without a catalyst, with inert SiO₂, and with the uncalcined CeO₂ precursor resulted in no detectable DEC production, confirming that the reaction is genuinely catalytic and hinges on the specific surface structure of the shaped CeO₂ nanocrystals.

The influence of reaction parameters on the synthesis of DEC from CO₂ and ethanol over the Ce-NR catalyst is summarized in Figure 7. First, to determine the optimal reaction time for DEC production from CO₂ with ethanol in the presence of a Ce-NR catalyst, the reaction time varied between 2 and 10 h, keeping other parameters constant: catalyst amount (0.17 g), temperature (423 K), and CO₂ pressure (5 MPa) (Figure 7a). The experimental findings indicated that the production of DEC increased from 1.05 to 1.49 mmol_DEC/g_cat as the reaction duration increased from 2 to 8 h. However, there was no significant alteration in DEC production when the reaction time was extended beyond 10 h. This was most likely because the catalyst was deactivated by adsorbed water at this point. Meanwhile, reaction temperature also played a critical role by affecting both substrate activation and thermodynamic constraints. As depicted in Figure 7b, the temperature varied between 383 K and 443 K in the presence of the Ce-NR catalyst at a constant catalyst amount (0.17 g) and constant CO₂ pressure (5 MPa). It was found that the DEC production increased from 0.97 to 1.35 mmol_DEC/g_cat when the temperature was increased from 383 K to 403 K. However, a further increase in temperature to 443 K led to a slight decrease in production to 1.31 mmol_DEC/g_cat. This decline could be attributed to the fundamental thermodynamics of physical absorption. The solubility of CO₂ in ethanol, as in many physical solvents, exhibited an inverse relationship with temperature due to the exothermic nature of dissolution. At elevated temperatures (>403 K), the equilibrium concentration of dissolved CO₂ in the ethanol solvent decreased significantly. Since the dissolved CO₂ was the primary reactant, its reduced availability at the catalyst surface likely became the rate-limiting step, overshadowing the positive effect of temperature on reaction kinetics and leading to the observed decrease in DEC production. Generally, for reactions involving gaseous feedstocks, especially for those in which the total number of molecules is reduced, the increase in reaction pressure can effectively promote the reaction. Therefore, as shown in Figure 7c, increasing the initial CO₂ pressure from 1 g to 5 MPa resulted in a rise in DEC production from 0.71 to 1.32 mmol_DEC/g_cat. Increasing the initial pressure of CO₂ in the reaction system can improve the DEC production from CO₂ and ethanol by strengthening the interaction between CO₂ and acid–base pairs of Ce-NR, which can effectively activate more CO₂. Finally, the effect of catalyst amount on the formation of DEC was shown in Figure 7d. The increase in catalyst dosage, on one hand, provided more catalytically active sites, while limiting the mixing and mass transfer of systems on the other hand. With the increase in catalyst dosage from 0.1 to 0.4 g, the DEC production decreased gradually from 1.56 to 0.43 mmol_DEC/g_cat. This may be attributed to the excessive catalytic active sites that enhanced the side reactions, resulting in a remarkable decrease in the selectivity of the target product DEC.

The reusability of the Ce-NR catalyst, a critical parameter for practical application, was evaluated under the optimized reaction conditions (423 K, 5 MPa CO₂, 4 h, 0.17 g catalyst). After each run, the catalyst was recovered by centrifugation, washed thoroughly with absolute ethanol, and directly reused in the next cycle. As illustrated in Figure 8, the DEC production remained almost unchanged over four consecutive reaction runs, demonstrating the outstanding catalytic stability of the Ce-NR catalyst. To examine the stability of the catalyst after recycling, the used catalyst was characterized by XRD and XPS techniques. As shown in Figure 9a, XRD results confirmed that the fluorite structure of CeO₂ was intact and no new crystal phase had appeared, suggesting that the structure of the catalyst did not change. XPS analysis (Figure 9b,c) also showed that the relative concentrations of Ce³⁺ species and Ov remained largely unchanged compared to the fresh catalyst. This indicated the preservation of active surface chemistry, corroborating the consistent catalytic performance. This high stability supports its potential for future industrial implementation by maintaining consistent performance across multiple uses without significant deactivation.

The reaction pathway detailed below was proposed as a plausible mechanism based on the above experimental observations (e.g., the critical role of oxygen vacancies and Ce³⁺ sites) and consistent with established literature on CeO₂-catalyzed reactions [44,45,46]. A schematic diagram illustrating this proposed pathway is included in Scheme 1. The process was hypothesized to commence with the adsorption of CO₂ molecules onto the Ce-NR catalyst surface. This step was likely facilitated by Lewis acid–base interactions, wherein the oxygen vacancies (acting as Lewis acid sites) interacted with the nonbonding electrons of the oxygen atoms in the linear CO₂ molecule. This interaction potentially led to the activation of CO₂ and the formation of carbonate-like adsorbed intermediates, such as bidentate carbonate. Concurrently, ethanol molecules were probably adsorbed on adjacent Ce³⁺ Lewis acid sites. At these sites, the hydroxyl group of ethanol might interact with neighboring basic low-coordinated surface oxygen anions (O²⁻), resulting in the deprotonation of ethanol and the formation of ethoxyl ions (CH₃CH₂O⁻) and protons (H⁺). Subsequently, the nucleophilic oxygen atom of the adsorbed ethoxyl species attacked the electrophilic carbon atom of the activated CO₂ intermediate, leading to the formation of a monodentate ethyl carbonate (MEC) species (CH₃CH₂OCOO⁻). Almost simultaneously, another ethanol molecule might be activated on a proximal Lewis acid site, facilitating the generation of an ethyl cation (CH₃CH₂⁺) and a hydroxyl anion (OH⁻). The critical C–O bond formation step culminating in DEC production was then proposed to proceed via the reaction between the nucleophilic MEC intermediate (CH₃CH₂OCOO⁻) and the electrophilic ethyl cation (CH₃CH₂⁺), yielding DEC (CO(OCH₂CH₃)₂). The proton (H⁺) generated from the first ethanol molecule then combined with the OH- anion to produce water as a by-product. Finally, the active sites, including the oxygen vacancies, were regenerated on the catalyst surface, completing the proposed catalytic cycle. This proposed mechanism highlighted the vital synergy between the acidic sites (Ce³⁺, oxygen vacancies) and basic sites (O²⁻) on the CeO₂ surface for the simultaneous adsorption and activation of both CO₂ and ethanol. A crucial aspect of this synergy was the postulated role of Ce³⁺ sites adjacent to oxygen vacancies in creating moderate-strength basic sites. These sites were inferred to be optimal for stabilizing key reaction intermediates like monodentate ethyl carbonate, thereby avoiding the pitfalls of either ineffective weak adsorption or catalyst-poisoning strong adsorption. The rod-like morphology (Ce-NR) was concluded to be crucial as it uniquely optimized the surface crystal structure and acid–base properties, providing the highest concentration of oxygen vacancies and a balanced distribution of acid–base pairs. As electron-rich centers, these oxygen vacancies were reasoned to significantly enhance CO₂ adsorption and activation. Thus, the enhancement of surface functionality induced by morphology control, including the increased density of active sites and optimized acid–base properties, was proposed as the fundamental reason for the promoted DEC formation.

3. Materials and Methods

3.1. Materials

Cerium(III) nitrate hexahydrate (Ce(NO₃)₃·6H₂O) and sodium hydroxide (NaOH) were purchased from Aladdin Chemical Technology Co., Ltd. (Shanghai, China). Ethanol (C₂H₆O, AR) was purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All the reagents were analytical grade and used without further purification.

3.2. Preparation of CeO₂ Catalysts

CeO₂ nanorods: 5.2 g of Ce(NO₃)₃·6H₂O was dissolved in 30 mL of deionized water to form a cerium nitrate solution. The 210 mL of 6 mol/L NaOH solution was slowly added dropwise to the cerium nitrate solution under stirring. After the addition was complete, the mixture was stirred for an additional 15 min. The resulting mixture was transferred into a hydrothermal autoclave and subjected to a hydrothermal reaction at 453 K for 24 h. After the reaction, the product was collected, washed with deionized water until the pH of the filtrate reached 7, and then washed 2–3 times with ethanol. The product was dried at 353 K for 8–10 h, ground, and finally calcined in a muffle furnace. The furnace temperature was raised from 293 K to 773 K at a heating rate of 2 K/min and held at 773 K for 4 h to obtain CeO₂ nanorods, denoted as Ce-NR.

CeO₂ nanoparticles: 5.2 g of Ce(NO₃)₃·6H₂O was dissolved in 30 mL of deionized water to form a solution. The 30 mL of 2 mol/L NaOH solution was slowly added dropwise to the cerium nitrate solution under stirring. After complete addition, the mixture was stirred for another 15 min. The precipitate was collected, washed with deionized water until neutral (pH = 7), and then washed 2–3 times with ethanol. The product was dried at 353 K for 8–10 h. The dried material was calcined in a muffle furnace, with the temperature programmed to increase from 293 K to 773 K at a rate of 2 K/min and maintained at 773 K for 4 h, yielding CeO₂ nanoparticles, denoted as Ce-NP.

CeO₂ nanocubes: CeO₂ nanocubes were synthesized using a similar initial procedure to the nanorods: 5.2 g of Ce(NO₃)₃·6H₂O was dissolved in 30 mL of deionized water. Then, 210 mL of 6 mol/L NaOH solution was added dropwise to the cerium nitrate solution with stirring, followed by 15 min of additional stirring after the addition. The mixture was then transferred to a hydrothermal autoclave for a hydrothermal reaction at 373 K for 24 h. After the reaction, the product was collected, washed with deionized water to pH = 7, followed by 2–3 washes with ethanol, and dried at 353 K for 8–10 h. The dried product was ground and calcined in a muffle furnace using the same temperature program as for the nanorods (heating from 293 K to 773 K at 2 K/min and holding for 4 h) to obtain CeO₂ nanocubes, denoted as Ce-NC.

3.3. Characterization

A D/MAX2500/PC powder diffractometer (Rigaku) with a Cu Kα radiation source was operated at 40 kV and 200 mA to record the X-ray Diffraction (XRD) patterns in reflection geometry to identify the crystalline structure of samples (Rigaku Corporation, Tokyo, Japan). Fourier transform infrared (FT-IR) spectra were tested by a Tensor 27 (Bruker, Billerica, MA, USA) spectrometer in the transmission mode with a resolution of 4 cm⁻¹. Each spectrum was based on 32 scans (4000–400 cm⁻¹). The transmission electron microscopy (TEM) images of samples were recorded on a JEOL JEM-2100 (JEOL, Tokyo, Japan) with a 200 kV field emission gun. Samples for TEM analyses were prepared by dispersing the powder products as a slurry in ethanol and then depositing them on a carbon film coated on a copper grid. X-ray photoelectron spectroscopy (XPS) was conducted on a Thermo ESCALAB 250 system (Thermo Fisher Scientific, Waltham, MA, USA), employing AlKα radiation, operating at 150 W with an energy pass of 20 eV. The binding energies of various surface elements were calibrated using the C 1s peak at 284.8 eV. Surface acidity and basicity of the obtained samples were obtained by temperature-programmed desorption of CO₂ and NH₃ experiments (CO₂-TPD and NH₃-TPD) (Autochem 2920, Micromeritics, Norcross, GA, USA) with a TCD detector. For the CO₂-TPD or NH₃-TPD, a 50 mg sample was pretreated under Ar at 473 K for 30 min to remove surface physically adsorbed H₂O and CO₂. CO₂ or NH₃ adsorption was carried out at room temperature by switching Ar flow to a stream of 10 vol% CO₂/Ar or 10% NH₃/Ar (30 mL·min⁻¹) for 1 h. Then the sample was purged with He (50 mL·min⁻¹) for 1 h in order to remove the physically adsorbed CO₂ or NH₃ species. The final TPD profile was recorded under He (30 mL·min⁻¹) in temperatures ranging from room temperature to 1073 K at a heating rate of 10 K·min⁻¹. The concentration of CO₂ or NH₃ was measured by a TCD detector.

3.4. Catalytic Performance Test

The catalytic reaction was carried out in a 25 mL stainless-steel autoclave equipped with a magnetic stirrer. In a typical procedure, 0.17 g of catalyst was added to 0.1 mol of ethanol in the reactor. The air in the reactor was then purged by pressurizing and venting with CO₂ several times, after which the reactor was charged with CO₂ to the desired initial pressure (5 MPa). The reaction mixture was heated to the target temperature (423 K) and maintained for 4 h under constant stirring. After the reaction, the reactor was cooled to room temperature. The liquid products were qualitatively analyzed using a gas chromatograph (Haixin GC-950) equipped with a flame ionization detector (FID) and an HP-5 capillary column. The quantification of DEC and ethanol was performed by gas chromatography (GC) using an internal standard method (n-propanol). A calibration curve was established with known concentrations of authentic DEC and ethanol standard prior to analyzing the reaction mixtures to ensure accurate quantification of the DEC and ethanol amount. Ethanol conversion and diethyl carbonate (DEC) production were calculated based on the chromatographic data.

$$\mathrm{DEC}\mathrm{production}={\displaystyle \frac{{\mathrm{n}}_{\mathrm{DEC}}(\mathrm{mmol})}{{\mathrm{m}}_{\mathrm{catalyst}}(\mathrm{g})}}$$

$$\mathrm{Ethanol}\mathrm{conversion}={\displaystyle \frac{{\mathrm{n}}_{\mathrm{in}}-{\mathrm{n}}_{\mathrm{out}}}{{\mathrm{n}}_{\mathrm{in}}}}\times 100\%$$

n_DEC represents the molar amount of DEC produced; m_catalyst represents catalyst dosage; n_in represents the molar amount of ethanol fed into the reactor; and n_out represents the molar amount of unreacted ethanol remaining after the reaction.

The solid catalyst was separated from the post-reaction mixture by centrifugation, thoroughly washed with ethanol, and dried under vacuum at 373 K overnight for recycling tests. Following the initial reaction, a sample of the reaction mixture was extracted with a pipette for GC analysis. The remaining mixture was centrifuged to separate the catalyst from the liquid phase. The recovered catalyst was washed five times with ethanol and then dried overnight under vacuum at 373 K. The recycled catalyst was directly used in the next reaction cycle. The amounts of the substrate and the Ce-NR catalyst were adjusted proportionally to the amount of recovered catalyst in each reuse experiment.

The catalytic reactions were performed in triplicate, and the values reported represent the mean. The error bars in the figures and the uncertainties in the table correspond to the standard deviation from these independent measurements.

4. Conclusions

In summary, three distinct CeO₂ morphologies, including nanorods (Ce-NR), nanocubes (Ce-NC), and nanoparticles (Ce-NP), were successfully prepared via hydrothermal methods. Crucially, it was found that the catalytic performance for the direct synthesis of diethyl carbonate (DEC) from CO₂ and ethanol was decisively influenced by the surface structure, which was intrinsically linked to the catalyst morphology. Thereby, a definitive correlation was established connecting the morphology with the abundance of oxygen species, the acid–base properties, and ultimately the catalytic activity. Among the three synthesized catalysts, CeO₂ nanorods (Ce-NR) exhibited the superior catalytic performance, achieving a DEC production of 1.32 mmol_DEC/gcat, leading to the performance order: Ce-NR > Ce-NC > Ce-NP. This superior activity was attributed to the synergistic effect of a higher concentration of surface Ce³⁺ species, a greater density of oxygen vacancies, and an optimal balance of acid–base sites on the nanorod morphology. These features collectively promoted the adsorption and activation of CO₂ and ethanol, facilitating the formation of DEC. In addition, the Ce-NR catalyst also demonstrated outstanding catalytic stability. Consequently, this study highlights morphology engineering as a critical tool for tailoring the surface properties of CeO₂ catalysts. The strategy of designing catalysts with specific morphologies to expose defect-rich surfaces possessing suitable acid–base characteristics proves effective for advancing CO₂ utilization. The insights from this work provide a valuable framework for the rational design of next-generation catalytic systems aimed at sustainable synthesis.