Generating Strongly Basic Sites on C/Fe₃O₄ Core–Shell Structure: Preparation of Magnetically Responsive Mesoporous Solid Strong Bases Catalysts

Abstract

Novel solid strong base catalysts have attracted considerable attention in fine chemical synthesis owing to their unique advantages. In this work, a magnetic solid strong base catalyst with controlled morphology and porous carbon shell structure was successfully fabricated using low-cost carbon sources combined with Fe₃O₄ nanoparticles. KOH was used to introduce strong basic sites through ultrasonic-assisted impregnation. The carbon shell acted as a protective barrier to suppress detrimental interactions between basic species and the support while maintaining structural integrity after high-temperature activation without morphology degradation. The obtained K/C/Fe₃O₄ catalyst exhibits excellent catalytic performance and near-ideal superparamagnetic behavior. In the transesterification reaction for dimethyl carbonate (DMC) synthesis, the K/C/Fe₃O₄ catalyst provides superior performance than conventional solid base catalysts and maintains stable activity over six consecutive cycles. Notably, efficient solid–liquid separation was achieved successfully via magnetic separation, demonstrating practical applicability for the K/C/Fe₃O₄ catalyst.

1. Introduction

With the growing emphasis on green chemistry, the development of environmentally friendly catalysts has become a focal point of the research [1,2,3,4,5,6,7]. Among diverse heterogeneous catalytic systems, solid base catalysts have obtained widespread attention in fine chemical synthesis owing to their distinct advantages, including mild reaction conditions, facile product separation, excellent recyclability, minimized equipment corrosion, and superior environmental compatibility [8,9,10,11,12,13,14,15,16]. Mesoporous silica is considered as an ideal support for solid base catalysts. However, strong basic species tend to react with siliceous support, leading to structure collapse [17,18,19]. Furthermore, the separation and recovery of solid base catalysts remain challenging, primarily relying on filtration or centrifugation, which are not only costly but also time-consuming, limiting the practical application of solid base catalysts [20,21,22,23,24,25].

Non-silicon-based carbon materials have a wide range of applications and are indispensable in fields such as adsorption, composite materials, fuel cells, and new energy sources, due to unique surface chemical properties and physical structures [17,25]. It has been confirmed that the sol–gel process is able to carbonize resorcinol and formaldehyde polymers (RF) into porous carbon. Subsequently, various methods have been explored to control the structure and morphology of the products for different applications, such as battery anodes, ion exchange resins, gas diffusion electrodes in fuel cells, and electrodes for capacitive deionization in aqueous solutions.

Magnetic carbon materials have consistently been a prominent research focus. Zhang et al. successfully synthesized Fe₃O₄@C nano-microspheres through hydrothermal/solvothermal carbonization of glucose in the presence of Fe₃O₄ nanoparticles, improving the electrochemical performance of nanostructured transition metal oxides as superior anode materials for lithium-ion batteries [26]. Lei et al. employed dopamine as both carbon and nitrogen source to fabricate nitrogen-doped core–shell structured magnetic particles [27]. Furthermore, Li et al. developed an innovative approach involving a polystyrene (PS) coating on Fe₃O₄ particles followed by high-temperature carbonization, resulting in the formation of carbon-coated nano-microspheres [28]. Li et al. reported a new and robust magnetically recoverable catalyst specifically engineered for the green synthesis of benzothiazoles and benzoxazoles [29]. If carbon-based solid base catalysts are endowed with magnetism, separation and regeneration would become highly convenient, significantly reducing operational costs [30,31]. Nevertheless, research on magnetic carbon-based mesoporous solid base materials remains relatively scarce in the literature to date.

This study presents the synthesis of magnetic solid base catalyst with well-defined morphology and porous carbon shell structure, achieved by integrating RF-derived porous carbon with Fe₃O₄ nanoparticles. The KOH, as a strong alkaline precursor, significantly enhances the basicity of the catalyst, while the incorporation of the carbon shell effectively prevents reactions between the highly basic species and the support. Figure 1 illustrates the preparation process of the K/C/Fe₃O₄ catalyst, highlighting that the porous structure of the support is initially well maintained. The results further indicate that the developed magnetic solid base catalysts outperform typical solid base catalysts in terms of catalytic activity. Notably, the catalysts exhibit superparamagnetic behavior, facilitating rapid separation via an external magnetic field after the reaction.

Dimethyl carbonate (DMC) represents a pivotal green chemical building block in modern organic synthesis, acclaimed for its versatile reactivity, favorable physicochemical properties, and low environmental impact. Its exceptional utility stems from its capacity to replace highly toxic reagents (e.g., phosgene, methyl chloroformate) in methylation/carbonylation reactions while also serving as an eco-friendly substitute for halogenated solvents (e.g., trichloroethane) and aromatic hydrocarbons (e.g., benzene, xylene) in coatings applications. Owing to these multifaceted applications, DMC is recognized as a new cornerstone of sustainable chemistry [12]. Although sodium methoxide exhibits high catalytic activity in the transesterification synthesis of DMC from ethylene carbonate and methanol, its homogeneous nature imposes severe limitations: challenging product-catalyst separation, stringent equipment requirements, high energy consumption, and significant environmental contamination. These drawbacks have spurred intensive research into heterogeneous alternatives, including anion-exchange resins, organic/inorganic solid bases, and hybrid catalysts. Against this backdrop, we selected the transesterification reaction as a model system to validate the performance of our designed K/C/Fe₃O₄ catalyst. This magnetic solid base uniquely integrates a protective carbon shell that stabilizes active sites while enabling rapid magnetic separation—addressing critical gaps in existing catalytic technologies.

2. Results and Discussion

2.1. Structural Characteristics

Figure 2 displays the TG profile of C/Fe₃O₄ magnetic spheres under an air atmosphere. Data correspond to the as-synthesized C/Fe₃O₄ sample before pyrolysis at 600 °C for 3 h under N₂ atmosphere. The TG curve exhibits two distinct mass loss stages. The initial mass loss occurring at approximately 100 °C corresponds to the evaporation of physiosorbed water molecules. The subsequent mass reduction, commencing around 320 °C, results from the oxidative decomposition of carbonaceous components into CO₂. Visual inspection confirmed the complete transformation of the final product into a reddish hue, demonstrating the exhaustive combustion of carbon constituents and the exclusive formation of Fe₂O₃ as the product. Quantitative analysis determined the carbon content in the C/Fe₃O₄ composite to be 47.8 wt%.

Table 1. Physicochemical properties of different samples.

Sample	Composition (wt%) a		SBETb (m2·g−1)	Vp b (cm3·g−1)	Micropore Size (nm)	Mesopore Size (nm)	Amount of Basic Sites (mmol∙g−1)
	Fe3O4	K2O
Fe3O4	99.6	0	59	0.07	-	-	0.03
C/Fe3O4	51.4	0	264	0.17	1.2	2.0/3.5	0.07
K/C/Fe3O4	48.6	5.5	258	0.17	1.1	4.8/7.1	1.03
							1.17 c
The leachate							0.024 d

^a The composition was determined by XRF. ^b S_BET: BET surface area, Vp: total pore volume. ^c theoretical value was calculated based on the data from XRF. ^d theoretical value was calculated based on the data from ICP.

provides the composition of the magnetic mesoporous strong base materials. The content of Fe₃O₄ in the C/Fe₃O₄ sample is 51.4 wt%. Further analysis reveals that this result is consistent with the carbon mass fraction of 47.8 wt% calculated from the TG data mentioned earlier, indicating that the porous carbon has been successfully coated onto the Fe₃O₄ particles. For the K/C/Fe₃O₄ sample, the content of K₂O is 5.5 wt%, which suggests that the basic species have been successfully introduced onto the C/Fe₃O₄ carrier.

Figure 3 presents the SEM and TEM images of Fe₃O₄ particles, C/Fe₃O₄, and K/C/Fe₃O₄ microspheres. The SEM (Figure 3a) and TEM (Figure 3b) images of pristine Fe₃O₄ particles demonstrate well-defined spherical morphology with a textured surface topography. These particles exhibit monodisperse size distribution with an average diameter of approximately 300 nm. Figure 3c further illustrates the irregular surface features of the Fe₃O₄ particles. The selected area electron diffraction (SAED) pattern of the Fe₃O₄ particles, depicted in Figure 3d, exhibits multiple concentric diffraction rings, which confirm the polycrystalline nature of the Fe₃O₄ particles. These diffraction rings correspond to the (220), (311), (400), (422), (511), and (440) crystal planes, which are characteristic of the Fe₃O₄ structure. Figure 3e,f present the SEM and TEM images of C/Fe₃O₄ composite particles, revealing a well-developed core–shell architecture. The micrographs confirm the successful formation of a continuous porous carbon shell through the polymerization and subsequent carbonization of resorcinol-formaldehyde precursors on the Fe₃O₄ core. In contrast to the bare Fe₃O₄ particles, the carbon-coated spheres display a smooth surface morphology with significantly increased particle dimensions. Quantitative analysis indicates a uniform carbon shell thickness of about 90 nm. The SEM and TEM images of K/C/Fe₃O₄ microspheres (Figure 3g,h) demonstrate that the introduction of basic species preserves the core–shell configuration and spherical morphology. After the introduction of the basic species, the sample did not aggregate significantly, only showing slight adhesion, thus indicating that the basic species are uniformly dispersed within the carbon shell layer. We have further examined the recovered catalyst by powder SEM. The micrographs (Figure S1) reveal that the micro spherical morphology remains intact after five catalytic cycles; no discernible aggregation, fracture, or surface roughening is observed.

EDX-mapping analysis was conducted to further investigate the surface elemental composition of the K/C/Fe₃O₄ composite. As illustrated in Figure 4, the K/C/Fe₃O₄ microspheres are predominantly composed of four distinct elements: Fe, O, C, and K. The elemental distribution of Fe matches the shape of the magnetic nanoparticles, corresponding to the spherical morphology of Fe₃O₄. The C element demonstrates a characteristic shell-like distribution pattern, with higher density observed at the periphery and lower density in the central region. The K element is uniformly dispersed throughout the entire sample, confirming the homogeneous incorporation of basic species within the carbon shell without any noticeable aggregation. Notably, the O element displays a concentration gradient, with higher intensity in the core region corresponding to the Fe₃O₄ phase and lower intensity at the edges. The residual O signal detected at the periphery can be attributed to the presence of K₂O, providing further evidence for the successful integration of basic oxide species into the carbon shell structure. This comprehensive elemental mapping not only validates the core–shell architecture of the composite but also confirms the effective functionalization of the carbon shell with potassium-based basic sites.

Figure 5 presents the XRD patterns of Fe₃O₄, C/Fe₃O₄, and K/C/Fe₃O₄ microsphere samples, providing detailed insights into their crystalline structures. The Fe₃O₄ magnetic spheres (JCPDS No. 65-3107) display distinct and well-defined diffraction peaks at 2θ values of 30.2°, 35.7°, 43.3°, 53.5°, 57.2°, and 62.8° [17,23,25]. These peaks are indexed to the (220), (311), (400), (422), (511), and (440) crystal planes of the face-centered cubic (Fd3m space group) structure of magnetite, respectively. The observed diffraction features are in excellent agreement with the selected area electron diffraction (SAED) pattern analysis of Fe₃O₄, as previously shown in Figure 3d, thereby confirming the accuracy and consistency of the structural characterization. Upon the deposition of a carbon layer and the incorporation of K, the characteristic diffraction peaks of Fe₃O₄ remain clearly observable in the XRD patterns of both C/Fe₃O₄ and K/C/Fe₃O₄ microspheres. This observation confirms the preservation of the cubic Fe₃O₄ crystal structure throughout the synthesis process, even with the introduction of additional components. Notably, the diffraction peaks of the C/Fe₃O₄ and K/C/Fe₃O₄ microspheres exhibit enhanced intensities and reduced peak widths compared to those of the pristine Fe₃O₄ particles. This enhancement in peak intensity and narrowing of peak widths can be attributed to the increased particle size of C/Fe₃O₄ and the carbonization process, which collectively contributes to an improved degree of crystallization of Fe₃O₄. The carbon layer not only enhances the structural stability of the Fe₃O₄ particles but also promotes a more ordered arrangement of the Fe₃O₄ crystallites, thereby resulting in the observed improvements in the XRD patterns.

The Scherrer analysis of the Fe₃O₄ (311) reflection yields a coherent-scattering (crystallite) dimension of about 28 nm. Transmission electron microscopy, on the other hand, reveals spherical entities with an overall diameter of 300 nm. This apparent difference arises because the two techniques probe distinct structural levels. X-ray diffraction is sensitive only to the size of single-crystal domains; the value of 28 nm, therefore, reflects the dimension of the individual magnetite crystallites. Electron microscopy images the secondary particles, which are dense agglomerates composed of many such crystallites. These agglomerates form spontaneously during the hydrothermal synthesis because of oriented attachment and van der Waals interactions, leading to the 300 nm spheres observed in TEM.

The N₂ adsorption–desorption isotherms and corresponding pore size distributions of Fe₃O₄, C/Fe₃O₄, and K/C/Fe₃O₄ samples are presented in Figure 6, with their structural parameters summarized in Table 1. As depicted in Figure 6a, the Fe₃O₄ particles exhibit negligible hysteresis and demonstrate a relatively low specific surface area of 59 m²·g⁻¹ along with a pore volume of 0.07 cm³·g⁻¹, which can be attributed to interparticle voids formed during nanoparticle packing. Upon carbon coating, the C/Fe₃O₄ sample displays a substantial increase in specific surface area to 263 m²·g⁻¹, accompanied by a distinct hysteresis loop, characteristic of a Type I isotherm. The observed hysteresis in the intermediate pressure region confirms the presence of mesoporous structures. Following K functionalization, the K/C/Fe₃O₄ sample exhibits a slight reduction in both specific surface area and total pore volume, while retaining a diminished yet discernible hysteresis loop, indicating the preservation of mesoporous features. Additionally, the notable adsorption in the low-pressure region for both carbon-containing samples suggests the formation of a well-developed microporous structure after porous carbon coating. From the pore size distribution analysis (Figure 6b, Table 1), it is evident that the C/Fe₃O₄ sample exhibits predominant pore sizes at 1.2 nm, 2.0 nm, and 3.5 nm. In addition, after K incorporation, an enlargement in pore size is observed (4.8 nm, 7.1 nm), likely due to the high-temperature reaction between K and carbon, which partially etches the carbon framework, thereby modifying the pore structure. While micropores govern surface area, the persistent hysteresis and 2.0–8.0 nm pore modes confirm coexisting mesoporosity critical for mass transport.

Figure 7 presents the infrared spectra of Fe₃O₄, RF/Fe₃O₄, C/Fe₃O₄, and K/C/Fe₃O₄ microspheres. For Fe₃O₄ particles, the peaks at 1652 and 1396 cm⁻¹ are attributed to the carboxyl groups, which result from the addition of sodium citrate during synthesis. The broad band spanning 3300–3800 cm⁻¹ corresponds to surface-adsorbed water molecules. For carbonized samples (C/Fe₃O₄ and K/C/Fe₃O₄), the attenuation of this broad band suggests the hydrophobic nature of the porous carbon surface. In contrast, the RF/Fe₃O₄ spectrum exhibits distinct vibrational features: the absorption bands in the 2800–3000 cm⁻¹ arise from −CH₂ stretching vibrations of polymeric species, while the broad peak centered at 3380 cm⁻¹ comprises contributions from aromatic −OH groups of resorcinol. The absorption band at 1614 cm⁻¹ is assigned to aromatic ring stretching vibrations, and those at 1222 and 1092 cm⁻¹ correspond to methylene ether linkages in the resorcinol-formaldehyde polymer [32]. The relatively weak intensity of these bands implies that the polymerization form is not predominant. Notably, the carbonization process effectively eliminates all organic functional groups, confirming complete thermal decomposition of the polymer matrix. Furthermore, the introduction of K species does not generate any new absorption features in the FTIR spectrum, indicating that K is not chemically bonded to the carbon framework in any detectable form.

2.2. Magnetic Properties

The magnetic properties of the synthesized samples were thoroughly characterized using VSM to evaluate their performance in magnetic separation and recovery processes. As shown in Figure 8a, the saturation magnetization (Ms) values of the Fe₃O₄, C/Fe₃O₄, and K/C/Fe₃O₄ samples were determined to be 44.0, 28.6, and 26.7 emu g⁻¹, respectively. These results confirm the excellent superparamagnetic behavior of the synthesized materials, which is crucial for their efficient separation and recovery using an external magnetic field. The introduction of basic species in the K/C/Fe₃O₄ sample leads to a slight decrease in saturation magnetization compared to the pristine Fe₃O₄ and C/Fe₃O₄ materials. However, the K/C/Fe₃O₄ sample still maintains a sufficiently high level of magnetization (26.7 emu g⁻¹) to ensure its magnetic responsiveness to an external magnetic field. The value remains comfortably above the 5–10 emu g⁻¹ range typically cited as sufficient for conventional magnetic separation devices. Nevertheless, the retained magnetization suggests that magnetic recovery should not become limiting when eventual scale-up is pursued. This balance between basicity and magnetism is essential for the practical application of the catalyst in various chemical processes, where both catalytic activity and ease of separation are required.

As depicted in Figure 8c, the K/C/Fe₃O₄ sample is uniformly dispersed in the reactant solution (Figure 8b). When an external magnetic field is applied, as shown in Figure 8d, the well-dispersed K/C/Fe₃O₄ particles rapidly migrate and adhere to the vial wall. This migration results in a clear and transparent supernatant solution, which closely resembles the initial reactant solution depicted in Figure 8b. In addition, the K/C/Fe₃O₄ catalyst exhibits remarkable redispersibility upon the removal of the magnetic field, with gentle agitation facilitating its rapid rehomogenization in the reactant solution. The combination of strong superparamagnetic and excellent redispersibility makes these magnetic solid base catalysts highly suitable for practical applications, particularly in catalyst recovery and reuse, which are essential for economic and environmental sustainability.

2.3. Basicity and Catalytic Performances

The alkaline properties of the prepared magnetic solid-base catalysts were assessed through quantitative analysis of surface basic site, and the results were compiled in Table 1. Titration analysis reveals that the Fe₃O₄ and C/Fe₃O₄ samples exhibit minimal basic site contents of 0.03 and 0.07 mmol·g⁻¹, respectively, demonstrating the absence of significant basic species on their surfaces. Following the incorporation of KOH as a basic modifier, the K/C/Fe₃O₄ sample achieves a substantially enhanced basicity of 1.03 mmol·g⁻¹. This value shows good agreement with the theoretical basicity of 1.17 mmol·g⁻¹ calculated from XRF data, thereby verifying the successful loading of basic species onto C/Fe₃O₄.

Figure 9 displays the CO₂-TPD profiles of C/Fe₃O₄ and K/C/Fe₃O₄ samples. The CO₂-TPD deconvolution further quantifies basic strength distribution (Table S1). In the case of the C/Fe₃O₄, no CO₂ desorption peak is detected even at elevated temperatures up to 500 °C, which can be attributed to the lack of basic species. In contrast, the K/C/Fe₃O₄ sample exhibits a prominent desorption peak centered at approximately 300 °C, characteristic of medium-strength basic sites. Furthermore, an additional CO₂ desorption peak emerges at 460 °C, demonstrating the formation of strong basic sites in the K/C/Fe₃O₄ sample compared to the C/Fe₃O₄ [33]. These findings are in excellent agreement with the previously obtained basicity measurements, thereby providing robust evidence for the successful incorporation of strong basic sites in K/C/Fe₃O₄ catalyst.

The catalytic performance of the magnetic solid base catalysts was thoroughly evaluated through the transesterification of ethylene carbonate for the synthesis of DMC, following the established protocols reported in the literature [34,35]. Ethylene carbonate undergoes transesterification with methanol via a nucleophilic substitution pathway, as illustrated in Figure 10, the catalyst’s basic sites first deprotonate CH₃OH to generate CH₃O⁻, which is then attacked by the electrophilic carbonyl carbon of vinyl carbonate. This nucleophile then attacks the electrophilic carbonyl carbon of ethylene carbonate, cleaving the C–O bond of the cyclic carbonate and yielding the transient alkoxide intermediate CH₃OCOOCH₂CH₂O⁻ (step I). Rapid proton exchange with a second methanol molecule converts this intermediate into its neutral form CH₃OCOOCH₂CH₂OH while regenerating the active CH₃O⁻ nucleophile (step II). A subsequent intramolecular nucleophilic substitution releases dimethyl carbonate (DMC) and deposits ethylene glycolate (CH₂OHCH₂O⁻) on the surface (step III). Final proton transfer with methanol restores the catalytic base and liberates ethylene glycol (CH₂OHCH₂OH), completing the cycle (step IV).

We have performed the transesterification in the complete absence of any catalyst at 65 °C for 6 h. Under identical conditions, the DMC yield was only 1.0 mol% (Figure S2a), confirming that the reaction is truly catalytic and that no appreciable background activity exists. When Fe₃O₄ and C/Fe₃O₄ are employed as catalysts for 6 h, the DMC yields are relatively low of 3.7 mol% and 4.5 mol%, respectively. The results are presented in Figure 11a. These results indicate the limited presence of active basic sites on the surfaces of these catalysts, which is consistent with the basicity data summarized in Table 1. Notably, the catalytic performance of the K/C/Fe₃O₄ sample was significantly superior to that of the other catalysts. When K/C/Fe₃O₄ is utilized as the catalyst, the DMC yield reached a maximum of 30.7 mol% after 6 h reaction. This substantial increased yield highlights the effectiveness of the K/C/Fe₃O₄ catalyst in promoting the transesterification reaction.

To further contextualize the performance of the K/C/Fe₃O₄ catalyst, a comparative analysis was conducted with several conventional solid base catalysts, in the synthesis of DMC, as detailed in

Table 2. The yield of DMC under the catalysis of different samples.

Catalyst	Reaction Conditions	DMC Yield (%)	TOF (h−1)	Ref.
K/C/Fe3O4	0.50 mol methanol, 0.10 mol EC, catalyst (0.5 wt% of methanol), 65 °C, 6 h	30.7	54.6	This work
MgO	0.50 mol methanol, 0.10 mol EC, catalyst (0.5 wt% of methanol), 65 °C, 6 h	7.6	0.95	[36]
MgO/NPC	0.50 mol methanol, 0.10 mol EC, catalyst (0.5 wt% of methanol), 65 °C, 6 h	11.4	/	[12]
Na2O/Al2O3	0.50 mol methanol, 0.10 mol EC, catalyst (0.5 wt% of methanol), 65 °C, 6 h	28.2	36.2	[24]
CaO/G	0.50 mol methanol, 0.10 mol EC, catalyst (0.35 wt% of methanol), 65 °C, 6h	13.5	46.2	[37]
CaO/NPC	0.50 mol methanol, 0.10 mol EC, catalyst (0.5 wt% of methanol), 65 °C, 6 h	13.5	36.2	[23]
CsX	0.50 mol methanol, 0.10 mol EC, catalyst (0.5 wt% of methanol), 65 °C, 6 h	6.1	2.9	[38]
KNO3/C	0.50 mol methanol, 0.10 mol EC, catalyst (0.1 g), 65 °C, 6 h	32.5	35.2	[25]
CeO-meso-400	0.50 mol methanol, 0.10 mol EC, catalyst (0.5 wt% of methanol), 65 °C, 6 h	33.6	16.5	[39]
MgO/G	0.50 mol methanol, 0.10 mol EC, catalyst (0.5 wt% of methanol), 65 °C, 4 h	16.1	12.7	[40]
Na2O/G	0.50 mol methanol, 0.10 mol EC, catalyst (0.5 wt% of methanol), 65 °C, 4 h	25.8	38.1	[41]
K2O/C/SBA-15	0.50 mol methanol, 0.10 mol EC, catalyst (0.5 wt% of methanol), 65 °C, 4 h	33.1	34.5	[24]

. The catalytic activities of metal oxides such as MgO (7.6%), MgO/NPC (11.4%), CaO/G (13.5%) and Na₂O/Al₂O₃ (28.2%) were found to be significantly lower than that of K/C/Fe₃O₄. Notably, under identical reaction conditions, our K/C/Fe₃O₄ catalyst demonstrates markedly superior activity relative to functionalized carbon-based base catalysts (MgO/G, Na₂O/G and K₂O/C/SBA-15). Even CsX zeolite, which is known for its strong basicity, exhibited a catalytic activity of only 6.0%. To further assess the activity based on active sites, turnover frequency (TOF) values for different catalysts were calculated. The formula and assumptions used for TOF are provided in Supporting Information Note S1. Although KNO₃/C and CeO-meso-400 achieved slightly higher DMC yields than K/C/Fe₃O₄, the latter exhibited a significantly higher TOF value of 57.6 h⁻¹. In contrast, the TOF values for KNO₃/C and CeO-meso-400 were only 35.2 h⁻¹ and 16.5 h⁻¹, respectively. This indicates that K/C/Fe₃O₄ demonstrated superior catalytic activity in terms of reaction rate. These comparative results clearly demonstrate that the magnetic solid base catalysts K/C/Fe₃O₄ samples developed in this study exhibit superior catalytic performance compared to typical solid base catalysts.

To investigate the stability of the basic sites within the catalyst, recycling experiments were conducted. The K/C/Fe₃O₄ catalyst could be rapidly and conveniently recovered after reaction via the application of an external magnetic field. No significant catalyst loss was observed during the recovery process. As shown in Figure 12, the regenerated catalyst largely retained its initial catalytic activity even after six consecutive reaction cycles, exhibiting only a marginal decline in performance. It is noteworthy that powder XRD of the catalyst recovered after six consecutive cycles (Figure S3) exhibits identical reflections to those of the fresh sample, confirming that the spinel Fe₃O₄ lattice and the carbon scaffold remain structurally intact. Furthermore, leaching experiments were performed. An equivalent mass of the K/C/Fe₃O₄ catalyst was stirred in methanol for 24 h, and the potassium ion concentration in the liquid phase was subsequently analyzed by ICP-OES. Results (Table 1) indicated that the leached potassium accounted for merely 2.1% of the total potassium loading on the catalyst, demonstrating the effective confinement of basic species by the carbon shell layer. To elucidate the catalytic mechanism, the leachate was employed as a catalyst for the transesterification reaction. The resulting DMC yield was only 1.8% (Figure 11b). In conjunction with the significantly higher yield (30.7%) achieved using the original catalyst system, this finding strongly indicates that the reaction proceeds predominantly via heterogeneous catalysis. Consequently, the K/C/Fe₃O₄ catalyst synthesized in this study demonstrates exceptional stability for the synthesis of DMC via transesterification.

To isolate the role of the carbon shell, we prepared K/Fe₃O₄ by simple incipient-wetness impregnation of KOH onto bare Fe₃O₄. In the first run this material delivered 28.3 mol% DMC yield (Figure S2a), which is close to that of K/C/Fe₃O₄ (30.7 mol%). However, upon a second catalytic cycle the activity collapsed to 3.5 mol% (Figure S2b), whereas the carbon-shielded K/C/Fe₃O₄ retained 30.2% after six cycles. The sharp deactivation observed for K/Fe₃O₄ is attributed to leaching of soluble potassium species. These findings underscore the indispensable function of the porous carbon layer: it stabilizes the alkaline phase and enables robust recycling.

2.4. Mechanism of Stable Basic Site Formation

The formation of catalytically active basic sites was predominantly governed by high-temperature activation. The K/C/Fe₃O₄ sample was activated by KOH at 700 °C for 2 h. During this process, the carbon shell reacted with KOH, forming highly stable potassium species embedded within the carbon matrix.

Existing studies corroborate that thermal activation under alkaline conditions facilitates the development of stable basic functionalities [42,43,44]. Fundamentally, this activation process primarily involves the reduction in hydroxides and the oxidation of carbon, concurrently accompanied by the development of porosity within the carbon support [44,45]. The selection of the 700 °C activation temperature was informed by systematic optimization within this work and validated against established protocols [44,45,46]. This specific thermal regime enables a pronounced redox interaction between KOH and the carbon support, culminating in the creation of active basic sites. The absence of distinct crystalline potassium phases (e.g., K₂O, KOH, or K₂CO₃) in the XRD patterns (Figure 5), combined with the exceptional catalytic stability (Figure 12), strongly suggests that potassium species exist as isolated atoms anchored via K-C coordination bonds within the carbon matrix. This coordination mechanism aligns with recent reports on single-atom solid superbases, where K–C configurations ensure both atomic dispersion and robust stability [24]. The interaction between potassium species and carbon is exceptionally strong. This process yields thermally resilient potassium centers that serve as highly dispersed basic sites. The stabilization mechanism likely involves electron donation from potassium to the unoccupied orbitals of carbon, forming an electron-rich interfacial complex [47]. Moreover, the engineered C/Fe₃O₄ support exhibited a substantial specific surface area (264 m² g⁻¹, Table 1), providing an optimal substrate for the uniform dispersion of potassium species. Collectively, these synergistic effects endow the synthesized K/C/Fe₃O₄ catalyst with sustained catalytic activity and stable basic site performance.

3. Experimental Section

3.1. Synthesis of Fe₃O₄ Nanoparticles

The synthesis of Fe₃O₄ nanoparticles was conducted as follows. It is essential to maintain the reaction system under moisture-free conditions from precursor preparation to final product isolation. Anhydrous FeCl₃ (1.30 g) was promptly combined with 40.0 mL of ethylene glycol and dissolved completely. To ensure homogeneity, sodium citrate (0.50 g) and sodium acetate (2.00 g) were successively added to the solution under constant stirring. The mixture was then subjected to mechanical stirring at a rate of 500 rpm for 0.5 h, yielding a uniform dark-yellow colloidal suspension. The precursor colloidal suspension was subsequently subjected to hydrothermal treatment in a Teflon-lined stainless-steel autoclave maintained at 200 °C for 10 h. Following natural cooling to 25 °C, the resultant magnetite nanoparticles were collected through successive washing with ethanol and deionized water.

3.2. Synthesis of C/Fe₃O₄ Nanoparticles

The fabrication process of C/Fe₃O₄ magnetic particles is conducted as described below. A homogeneous mixture consisting of 0.50 g Fe₃O₄, 280 mL deionized water, 0.4 mL concentrated aqueous ammonia, and 0.8 g resorcinol were introduced into a three-necked flask and subjected to mechanical stirring at 30 °C for 1 h. Subsequently, 1.2 mL formaldehyde was added dropwise to the reaction mixture, followed by continuous stirring for 6 h. Then, the reaction temperature was elevated to 80 °C, and the mixture was maintained under stirring for 5 h. The obtained product, after thorough washing and drying, is denoted as RF/Fe₃O₄. The RF/Fe₃O₄ composite was carbonized in a tubular furnace under nitrogen atmosphere at 600 °C for 3 h, yielding the final product designated as C/Fe₃O₄.

3.3. Synthesis of K/C/Fe₃O₄ Nanoparticles

The alkaline precursor is introduced via the wet impregnation method. Specifically, KOH (0.01 g) was dissolved in deionized water (10 mL), followed by the addition of C/Fe₃O₄ (0.1 g). The mixture is then subjected to ultrasonication for 10 min. Subsequently, the mixture is evaporated under continuous mechanical stirring at 80 °C until complete dryness is achieved. The obtained powder is subsequently calcined in a nitrogen atmosphere at 600 °C for 3 h. The final product is designated as K/C/Fe₃O₄.

3.4. Characterization

3.4.1. Structural and Morphological Analysis

The X-ray diffraction (XRD) analysis of the samples was performed utilizing a Bruker D8 Advance diffractometer (Bruker, Billerica, MA, USA). The instrument was configured with monochromatic Cu Kα radiation, operating at 40 kV and 30 mA. The 2θ scan was conducted over a range of 5° to 80°.

High-resolution transmission electron microscopy (HRTEM) imaging was performed using an FEI Talos 200X instrument (Thermo Scientific, Waltham, MA, USA). Meanwhile, scanning electron microscopy (SEM) imaging was conducted with a Zeiss Gemini SEM 300 (Zeiss, Oberkochen, Germany).

The N₂ adsorption–desorption isotherms were recorded with a BSD-660 (BSD, Beijing, China). Surface area quantification employed the Brunauer–Emmett–Teller (BET) model, utilizing multipoint adsorption data from the 0.04–0.20 P/P0 pressure region. Total pore volume calculations utilized the maximum adsorbed nitrogen quantity at P/P0 = 0.99. Mesoporous size distribution analysis applied non-local density functional theory (NLDFT) to the adsorption isotherm branch data.

Fourier transform infrared (FTIR) spectroscopy was performed using a Nicolet iS10 spectrometer (Thermofisher Scientific, Waltham, MA, USA) to obtain spectral data. Thermogravimetric (TG) analysis was conducted using a TGA209F1 instrument (Netzsch, Waldkraiburg, Germany) to assess the thermal stability of the samples. TG analysis utilized air to enable complete carbon combustion for quantitative analysis, contrasting with synthesis under N₂ where inert conditions, and the samples were heated from ambient temperature to 1073 K at a constant rate of 10 K·min⁻¹.

3.4.2. Magnetic and Basicity Characterization

The UV-visible (UV-vis) absorption spectra of the samples were recorded using a Shimadzu UV-2401PC spectrophotometer (Shimadzu Corporation, Kyoto, Japan), with barium sulfate (BaSO₄) utilized as the reflectance reference standard. Magnetic properties were characterized using a Lakeshore 7407 vibrating sample magnetometer (VSM) at 300 K. Elemental composition analysis was conducted by X-ray fluorescence (XRF) spectroscopy utilizing an ARL ADVANT’XP wavelength-dispersive spectrometer (Thermo Scientific, Waltham, MA, USA).

CO₂ temperature-programmed desorption (CO₂-TPD) was performed using a BELSORP BEL-CAT-A analyzer (Microtrac BEL Corp., Osaka, Japan). The sample was pretreated under helium atmosphere at 500 °C for 2 h. After cooling to room temperature, CO₂ adsorption proceeded for 1 h, followed by helium purging, with subsequent temperature ramping to 500 °C at 8 °C min⁻¹. The CO₂ liberation was detected by an OmniStar mass spectrometer.

The total basic site density was determined through acid-base back-titration methodology. Precisely 50 mg of catalyst was dispersed in 10.0 mL of 0.05 M hydrochloric acid solution. Following 24 h of shaking agitation, magnetic phase separation was achieved through application of an external magnetic field to the reaction system, enabling precise isolation of the clarified supernatant. Quantitative analysis of unreacted acid in the supernatant was performed via potentiometric titration using 0.01 M sodium hydroxide solution, with endpoint determination conducted through pH metric verification [45,48].

3.5. Catalytic Performances

The production of DMC was achieved through a catalytic ester group exchange process between methanol and ethylene carbonate (EC). In a representative synthesis, a stoichiometric excess of methanol (0.50 mol, 4:1 molar ratio relative to EC) was combined with EC (0.10 mol) and a magnetically recoverable heterogeneous catalyst (K/C/Fe₃O₄, 160 mg) in a three-necked round-bottom flask (50 mL capacity). In a standard procedure, methanol (0.50 mol), ethylene carbonate (0.10 mol), and K/C/Fe₃O₄ catalyst (80 mg) were charged into a 50 mL three-necked round-bottom flask equipped with a water-cooled condenser. The reaction system, equipped with a reflux condenser, was maintained at 65 °C under continuous agitation for a predetermined duration [49]. The K/C/Fe₃O₄ catalyst was subsequently separated from the reaction mixture using magnetic separation. Quantitative compositional analysis of the reaction products was performed using an Agilent 7890A gas chromatograph equipped with a flame ionization detector.

Gas chromatography analysis: high-purity helium (99.999%) served as the carrier gas at a constant linear velocity of 42.3 cm·s⁻¹. The injector and detector temperatures were maintained at 250 °C and 300 °C, respectively. Samples (0.5 μL) were introduced via a 10:1 split ratio. The temperature program initiated at 40 °C (hold 3 min), ramped to 120 °C at 15 °C·min⁻¹, then to 240 °C at 30 °C·min⁻¹ (final hold 2 min). Compounds were quantified using internal standard calibration with decane, achieving a detection limit of 0.01 ppm. Retention times were calibrated against authentic standards: methanol (1.8 min), DMC (4.2 min), ethylene carbonate (7.5 min), ethylene glycol (5.1 min).

4. Conclusions

This study successfully synthesize a magnetically responsive solid strong base catalyst, K/C/Fe₃O₄, utilizing Fe₃O₄ particles as the magnetic core and incorporating strongly basic K₂O species. The K/C/Fe₃O₄ catalyst exhibits a well-defined spherical morphology with a structurally intact carbon shell (90 nm thickness) even after high-temperature activation, providing an optimal platform for the incorporation of basic species. The resulting catalyst demonstrates both strong basic sites and excellent catalytic performance of 30.7 mol% in DMC synthesis, maintaining its catalytic efficacy without significant degradation over six consecutive reaction cycles, surpassing conventional solid strong base materials. Furthermore, the superparamagnetic characteristics enable efficient magnetic separation post-reaction under an external magnetic field.