Ligand-Engineered Mn-MOFs Derived Mn₂O₃ for Enhanced Carbon Dioxide Conversion to Ethylene Urea

Abstract

The utilization of carbon dioxide (CO₂) for synthesizing value-added chemicals represents a promising environmentally sustainable strategy. Herein, we synthesized a series of Mn₂O₃ catalysts derived from metal-organic frameworks (MOFs) incorporating three different ligands—homophthalic tricarboxylic acid (H₃BTC), 1,4-benzenedicarboxylic acid (H₂BDC), and a combination of polyvinyl pyrrolidone (PVP) with H₃BTC—via a hydrothermal method for ethylene urea (EU) production from CO₂ and ethylenediamine (EDA). The Mn₂O₃ catalyst derived from the H₃BTC+PVP ligand system (designated MnTP) demonstrated superior catalytic performance, achieving 97% EDA conversion and 97% EU selectivity under mild conditions (100 °C, 1 min), which surpassed all previously reported catalysts under comparable conditions. The enhanced activity originated from structural improvements induced by the H₃BTC+PVP precursors, particularly the promotion of oxygen vacancies and Mn³⁺ species, thereby facilitating efficient CO₂ activation and binding. This work establishes a novel strategy for the sustainable conversion of CO₂ into high-value cyclic ureas through rational catalyst design.

1. Introduction

The buildup of atmospheric carbon dioxide (CO₂), primarily from fossil fuel combustion, has led to irreversible environmental impacts such as global warming and climate change [1,2,3]. In response, conversion and fixation strategies are broadly divided into two types: reductive conversion, which yields products like CO [4], methanol [5], ethanol [6], and hydrocarbons [7], and non-reductive conversion, where CO₂ serves as a carbonyl source in reactions with alcohols or amines to form carbonates [8], carbamates [9], and urea derivatives [10]. Among these, non-reductive conversion stands out for its greater industrial promise, owing to its far lower energy demands and potential as a safe, sustainable alternative to highly toxic phosgene.

Cyclic ureas are recognized for their structural stability, versatile reactivity, environmental compatibility, and high tunability, which make them attractive for applications in green synthesis and molecular engineering [11]. As the most fundamental cyclic urea, ethylene urea (EU) represents a key intermediate in the production of pharmaceuticals, agrochemicals, and polymer additives [12,13]. In contrast to conventional synthesis routes, the direct production of EU from methanol and CO₂ offers a promising alternative that mitigates CO₂ buildup and eliminates the need for toxic raw materials, with water being the sole byproduct [14]. This approach is highly appealing for EU synthesis and CO₂ utilization due to its exceptional atom economy and eco-friendly characteristics. However, the inherent stability of the C–O bond in CO₂ renders it chemically inert and challenging to activate [15]. Therefore, a major obstacle in the direct synthesis of EU lies in achieving efficient activation of the reactants, particularly CO₂ and ethylene diamine (EDA).

Significant efforts have been devoted to designing efficient catalysts that enhance reaction kinetics. Recent research has predominantly centered on CeO₂, ZrO₂, and Mn-based oxide systems, often coupled with surface modification strategies to improve catalytic activity. Previous studies indicated that high catalyst performance often stemmed from a high concentration of surface oxygen vacancies, which promote CO₂ adsorption and activation [16]. Wang et al. [17] prepared a series of Zr-doped CeO₂ catalyst (Ce₁Zr_x, x = 0.01, 0.05 and 0.2) for the catalytic production of ethylene urea (EU) from ethylenediamine (EDA) and CO₂. The results show that the Ce₁Zr_0.01 catalyst revealed excellent catalytic activity with 91.5% EDA conversion and 100% EU selectivity at 160 °C for 3 h in comparison with undoped CeO₂. Recently, our group [18] reported that as-prepared Mn₂O₃ catalysts exhibited the best catalytic activity with a maximal EDA conversion of 82% and EU selectivity of 99% at 160 °C for 2 h compared to MnO₂ and Mn₃O₄ catalysts. Nevertheless, the requirement for harsh reaction conditions (temperatures exceeding 140 °C and time exceeding 2 h) remained a challenge. There was a need to develop catalysts with enhanced capabilities for substrate activation and coupling to improve the efficiency of EU synthesis under milder conditions.

In recent years, metal–organic framework (MOF) derivatives, particularly MOF-derived metal oxides, have garnered considerable interest due to their high specific surface areas (SSAs) and well-developed pore structures inherited from the parent MOFs. As a result, MOFs are increasingly employed as templates for synthesizing metal oxides or composite materials, representing a promising direction for MOF-based applications [19,20]. For example, Jiang et al. [21] reported the synthesis of high-performance MnO_x-CeO₂ catalysts using Ce-MOF precursors and examined the correlation between catalytic activity in ethyl acetate oxidation and surface fluorine-induced oxygen vacancies (F-O_Vs). The unique crossed-channel structure of the Ce-BTC-derived support contributed to its high SSA and ability to regenerate F-O_Vs under high temperatures. The design and synthesis of MOFs typically begin with selecting suitable organic linkers and metal nodes. The use of rigid, geometrically well-defined ligands often enables the construction of frameworks with predetermined topologies [22,23]. Nonetheless, the effect of varying organic ligands on the structure of MOF-derived Mn₂O₃—especially regarding oxygen vacancy formation—remains largely unexplored in the literature.

Based on these findings, we have developed a series of highly active Mn₂O₃ catalysts derived from three distinct MOF ligands—homophthalic acid (H₃BTC), 1,4-benzenedicarboxylic acid (H₂BDC), and a composite ligand system comprising H₃BTC and polyvinyl pyrrolidone (PVP)—via a hydrothermal method. The Mn₂O₃ catalyst obtained from the H₃BTC+PVP ligand system (denoted as MnTP) exhibited a higher concentration of surface oxygen vacancies, which significantly enhanced CO₂ adsorption and activation. The MnTP catalyst demonstrated outstanding performance, achieving 97% conversion of ethylenediamine (EDA) and 97% selectivity toward ethylene urea (EU) under mild conditions (T = 100 °C, t = 1 min). This study offers valuable guidance for the design and preparation of highly efficient MOF-derived metal oxide catalysts for CO₂ conversion applications.

2. Results

2.1. Catalysts Characterization

Three types of Mn-MOF precursors were calcined at 400 °C to obtain the corresponding metal oxides, denoted as MnBTC, MnBDC, and MnTP. The crystal structures of these materials were initially characterized by XRD, as shown in Figure 1. The diffraction patterns of all three samples matched well with the standard phase of cubic bixbyite-type Mn₂O₃ (JCPDS No. 73-1826) [24]. Characteristic peaks corresponding to the (211), (222), (400), (332), (431), (440), and (622) crystal planes were observed at 2θ = 23.5°, 33.2°, 38.5°, 45.1°, 50.2°, 55.4°, and 66.2°, respectively. No residual MOF diffraction peaks were detected [25], confirming complete decomposition of the organic frameworks and formation of pure Mn-based oxides. However, several weak additional peaks appeared at 18.5° and 36.3°, which could be assigned to the (101) and (211) planes of tetragonal Mn₃O₄ (JCPDS No. 24-0734) [26]. The XRD results clearly indicated the coexistence of both Mn₂O₃ and Mn₃O₄ phases in the three samples, with Mn₂O₃ being the dominant phase. Furthermore, the intensity and full width at half maximum (FWHM) of the diffraction peaks differed among MnBTC, MnBDC, and MnTP, suggesting that the choice of organic ligand influenced crystallite size and overall crystallinity. In particular, the MnTP catalyst exhibited the lowest crystallinity among the three, implying smaller crystal size and a higher density of defects, which were considered beneficial for enhancing gas adsorption and facilitating lattice oxygen mobility [27].

Figure 2 displayed the FTIR spectra of the MnBTC, MnBDC, and MnTP catalysts over the range of 400–4000 cm⁻¹. All MOF-derived Mn₂O₃ samples showed similar absorption features between 500 and 650 cm⁻¹, corresponding to Mn–O vibrational modes. The peaks observed at 511 cm⁻¹ and 608 cm⁻¹ were assigned to the stretching vibrations of Mn–O bonds in the tetrahedral and octahedral sites, respectively, within the Mn₂O₃ crystal structure, confirming the formation of the Mn₂O₃ phase [28]. A broad absorption band around 3374 cm⁻¹ was associated with the O–H stretching vibration of surface hydroxyl groups [29]. Additionally, two distinct peaks at 1490 cm⁻¹ and 1363 cm⁻¹ were indicative of chemisorbed carbonates and bicarbonates, which resulted from the interaction of Mn₂O₃ with atmospheric CO₂ and water [30].

The morphology and microstructure of the MnBDC, MnBTC, and MnTP catalysts were examined using scanning electron microscopy (SEM). The SEM results revealed markedly different morphologies dictated by the choice of MOF precursor. As shown in Figure 3a,b, MnBDC, derived from H₂BDC, comprised densely aggregated polyhedral particles approximately 0.5 μm in diameter, with well-defined edges and limited porosity. In contrast, the H₃BTC-derived MnBTC sample (Figure 3c,d) formed a three-dimensional porous framework constructed from irregular, thin nanosheets. Most notably, the MnTP catalyst, synthesized with the H₃BTC+PVP precursor (Figure 3e,f), exhibited a three-dimensional microflower architecture consisting of nanoflakes approximately 1 μm in size. This microflower morphology was constructed from uniformly arranged Mn₂O₃ nanoflakes that cluster into sheet-like assemblies, an outcome attributed to the role of PVP as a chelating agent promoting the formation of a stable and ordered structure. SEM analysis revealed that all three catalysts comprised micron-sized particles, a common trait of MOF-derived metal oxides. While the primary particle size showed no major variation, significant differences were observed in their nanoscale crystallite size, surface chemistry, and porosity, which were identified as the key factors governing catalytic performance.

The N₂ adsorption-desorption isotherms of the MnTP, MnBDC, and MnBTC catalysts displayed type IV characteristics with an H3 hysteresis loop (Figure 4), indicative of mesoporous structures with irregular pore geometries [31]. As summarized in

Table 1. Comparative summary of the physical structure and chemical properties of MnTP, MnBDC, and MnBTC catalysts.

Catalyst	Surface Area (m2·g−1)	Pore Volume (cm3·g−1)	Average Pore Size (nm)	Mn3+ (%)	OV (%)
MnBDC	45.5	0.15	11.2	38.2	26.1
MnBTC	17.8	0.07	16.2	45	27.5
MnTP	19.7	0.08	16.2	49.4	31.2
MnTP-R5	-	-	-	41.5	18.8

, the specific surface area of MnBDC was 45.5 m²/g, significantly greater than those of MnBTC (17.8 m²/g) and MnTP (19.7 m²/g). This suggested that the ligand type in the Mn-MOF precursor played a critical role in determining the surface area of the pyrolyzed product. Furthermore, the pore volumes of MnBTC and MnTP were identical (0.08 cm³/g) and lower than that of MnBDC.

X-ray photoelectron spectroscopy (XPS) was used to analyze the surface elemental valence states of the synthesized catalysts. The Mn 2p and O 1s core-level spectra were presented in Figure 5. As shown in Figure 5a, the Mn 2p spectra were fitted to quantify the oxygen defect content, revealing three characteristic peaks at binding energies of 640.9 eV, 642.2 eV, and 643.8 eV, which were assigned to Mn²⁺, Mn³⁺, and Mn⁴⁺ species, respectively [32]. As summarized in Table 1, the proportion of Mn³⁺ followed the order MnBDC < MnBTC < MnTP, suggesting an increasing concentration of oxygen defects [33]. The O 1s spectra (Figure 5b) displayed three components at 529.1 eV, 530.5 eV, and 531.8 eV, attributed to lattice oxygen (O_l), oxygen vacancies (O_v), and chemisorbed oxygen (O_a), respectively [34]. The abundance of O_v was commonly regarded as an indicator of oxygen defect density [35]. Quantitative fitting results indicated that the MnTP catalyst possessed the highest O_v concentration (31.2%), exceeding those of MnBTC (27.5%) and MnBDC (26.1%) (Table 1). These findings were consistent with the trends observed in the Mn 2p spectral analysis. Therefore, these results from XPS analysis revealed that the MnTP catalyst possesses the highest concentrations of both Mn³⁺ Lewis acid sites and O_v. This unique surface structure was anticipated to be highly conducive to the catalytic reaction, as the abundant O_v sites were expected to efficiently adsorb and activate CO₂ molecules, while the Mn³⁺ sites were likely to stabilize key reaction intermediates and facilitate the cyclization step. This structure-function relationship was further elaborated in the proposed reaction mechanism.

2.2. Catalytic Performance

A comparative evaluation of MnTP, MnBDC, and MnBTC catalysts was conducted to examine the effect of MOF ligand selection on their catalytic performance in the synthesis of ethylene urea (EU) from CO₂ and ethylenediamine (EDA). As illustrated in Figure 6a,b, the MnTP catalyst exhibited the highest activity, achieving 97% EDA conversion and 97% EU selectivity within just 1 min of reaction, surpassing nearly all previously reported catalysts for this transformation (

Table 2. Comparison of catalytic performance of our prepared catalysts with the reported catalysts for the synthesis of EU from CO₂ and EDA.

Entry	Catalyst	T (°C)	t (min)	PCO2 (MPa)	Yield (%)	Ref.
1	Ph3SbO/P4S10	150	720	4.9	85	[36]
2	Cp2Ti(OTf)2	170	900	-	99	[37]
3	CeO2	160	480	0.7	37	[38]
4	CeO2	160	720	0.5	96	[39]
5	2.4ZnO/0.49KF/Al2O3	180	240	1	86	[40]
6	Sn1.1-Ni-O-600	160	240	0.25	84	[41]
7	MOF-derived CeO2	160	720	0.5	94	[42]
8	MnO2	160	120	0.6	22	[18]
9	Mn3O4	160	120	0.6	4.4
10	Mn2O3	160	120	0.6	81
11	Mn2O3-NC	140	20	0.6	70	[43]
12	Mn2O3-NO	140	20	0.6	55
13	Mn2O3-NS	140	20	0.6	90
14	MnBDC	100	1	0.6	48	this work
15	MnBTC	100	1	0.6	71
16	MnTP	100	1	0.6	94

) [18,36,37,38,39,40,41,42,43]. The only detectable byproduct was identified as N-(2-aminoethyl)ethylenediamine, a linear urea derivative, which was formed in trace amounts (<3% selectivity). Its formation was slightly more pronounced at extended reaction times, indicating that it may arise from the reversible ring-opening of EU or further reaction of an intermediate with EDA. In contrast, MnBDC showed the lowest performance, with only 62% EDA conversion and 77% EU selectivity. Notably, both MnTP and MnBTC reached their maximum activity within 1 min, after which their performance gradually decreased over time. Conversely, MnBDC displayed slower reaction kinetics, requiring 30 min to attain 79% EDA conversion and 89% EU selectivity, with both values increasing steadily over the reaction period, reflecting its comparatively lower reaction rate. The overall catalytic activity followed the order MnTP > MnBTC > MnBDC (Figure 6c). These results suggested that the abundant surface O_v on MnTP offered additional active sites that promoted the intramolecular cyclization of isocyanate intermediates to form EU. Furthermore, Arrhenius plots of ln k versus 1/T were used to determine the apparent activation energy (Ea) for each catalyst. As shown in Figure 6d, the Ea values increased in the order MnTP (43.75 kJ·mol⁻¹) < MnBTC (49.51 kJ·mol⁻¹) < MnBDC (53.45 kJ·mol⁻¹). Both the catalytic performance and activation energy results confirmed that MnTP possessed the highest catalytic activity among the three catalysts.

The influence of reaction parameters on the synthesis of EU from CO₂ and EDA over the MnTP catalyst is summarized in Figure 7. First, reaction temperature played a critical role by affecting both substrate activation and thermodynamic constraints. As depicted in Figure 7a, performance was evaluated between 80 °C and 120 °C under a fixed initial CO₂ pressure of 0.6 MPa and catalyst mass of 0.2 g. EDA conversion increased from 58% to 91%, and EU selectivity increased from 73% to 89%, as the temperature rose from 80 °C to 100 °C. This improvement suggested that higher temperatures provide reactant molecules with sufficient kinetic energy to overcome the activation barrier. However, a further increase to 120 °C resulted in a slight decline in performance. The effect of CO₂ pressure was examined at 100 °C using 0.2 g of catalyst (Figure 7b). Raising the initial pressure from 0.2 MPa to 0.6 MPa enhanced EDA conversion from 77% to 91% and EU selectivity from 79% to 89%, which could be attributed to improved CO₂ solubility in the reaction mixture and strengthened interaction with active sites. Further increasing the pressure to 1.5 MPa, however, led to a noticeable decrease in both conversion and selectivity, likely due to excessive CO dissolution reducing effective contact between EDA and the catalyst. As shown in Figure 7c, catalyst dosage also influenced the reaction outcome. Increasing the amount of MnTP from 0.08 g to 0.16 g resulted in a rise in EDA conversion and EU selectivity to 91% and 89%, respectively, reflecting enhanced catalytic efficiency. A further increase to 0.32 g led to a slight reduction, possibly due to promoted side reactions resulting from an overabundance of active sites. Finally, the stability of the MnTP catalyst was assessed under optimal conditions over five consecutive cycles (Figure 7d). The catalyst was recovered by centrifugation, washed with ethanol, and reused directly. Although EDA conversion decreased from 92% to 59% after five runs, EU selectivity remained stable at approximately 80%.

To investigate the origin of catalyst deactivation, fresh and used MnTP catalysts were compared using XRD and XPS (Figure 8). The XRD patterns (Figure 8a) revealed that the crystal structure of MnTP remained unchanged after reaction, demonstrating its high structural stability under catalytic conditions. However, as shown in Figure 8b,c, XPS analysis of Mn 2p and O 1s core levels indicated a decrease in the concentration of Mn³⁺ species from 49.4% to 41.5% and a reduction in O_v from 31.2% to 18.8%. According to previous studies [44], this decline could be attributed to the inherent thermodynamic instability and high reactivity of surface O_v, wherein part of the oxygen atoms involved in CO₂ adsorption occupy and refill the vacancy sites.

Based on the experimental results and previous reports [18,43,45], a reaction mechanism was proposed for the synthesis of EU from CO₂ and EDA over the MnTP catalyst (Figure 9). The process began with the adsorption of EDA onto the catalyst surface through hydrogen bonding between its amino hydrogen atoms and surface oxygen atoms. Meanwhile, the abundant O_v sites facilitated CO₂ adsorption and activation. The activated CO₂ then reacted with adsorbed EDA to form carbamate and carbamic acid intermediates. The carbamic acid further decomposed into adsorbed amine species and carbamate. Owing to the spatial proximity of these intermediates, the -NH₂ group underwent a nucleophilic attack on the carbamate carbon center. This step was promoted by the Mn³⁺ Lewis acid sites, which helped stabilize key intermediates and facilitate cyclization, alongside weak Mn–O interactions and hydroxyl groups near oxygen vacancies. These hydroxyl groups then combined with hydrogen atoms from N–H bond cleavage, releasing water and regenerating the O_v sites. Finally, EU desorbed from the catalyst, closing the catalytic cycle. The continuous regeneration of O_v was essential for maintaining high catalytic activity.

3. Materials and Methods

3.1. Materials

Manganese (II) nitrate tetrahydrate (Mn(NO₃)₂·4H₂O), manganese(II) acetate, 1,3,5-benzenetricarboxylic acid (H₃BTC, 98%), 1,4 benzene dicarboxylic acid (H₂BDC, 98%), polyvinylpyrrolidone (PVP), and dimethylformamide (DMF) were purchased from Aladdin Chemical Technology Co., Ltd., Shanghai, China. Ethanol (C₂H₆O, AR), n-Propanaol (C₃H₈O, AR) were purchased from Sinopharm Chemical Reagent Co., Ltd., Shanghai, China. All the reagents were analytical grade and used without further purification.

3.2. Preparation of MOF-Derived Mn₂O₃ Catalysts

MnBDC: A mixture was prepared by dissolving 0.833 g of H₂BDC in 50 mL of a DMF–ethanol–water mixed solvent (8:1:1, v/v/v). Then, 1.255 g of Mn(NO₃)₂·4H₂O were added, and the mixture was ultrasonicated at 25 °C for 20 min. The resulting solution was transferred to a 100 mL polypropylene bottle and heated at 100 °C for 24 h. After cooling naturally to room temperature, a white precipitate was collected by filtration, washed three times with DMF and ethanol, and dried at 80 °C to obtain MnBDC. The as-synthesized MnBDC was then calcined in air at 400 °C for 2 h with a heating rate of 5 °C/min.

MnBTC: Solution A was prepared by dissolving 4.2 g of H₃BTC in 50 mL of DMF. Solution B consisted of 10 mL deionized water and 3.87 mL of 50% Mn(NO₃)₂·4H₂O aqueous solution. Solution B was added dropwise to Solution A under ultrasonication at 25 °C for 20 min. The mixture was transferred to a 100 mL polypropylene bottle and heated at 120 °C for 8 h. After cooling to room temperature, the white precipitate was filtered, washed three times with DMF and ethanol, and dried at 80 °C for 5 h. The product was calcined in air at 400 °C for 2 h with a ramp rate of 5 °C/min to yield MnBTC.

MnTP: Solution A was formed by mixing 0.049 g of manganese(II) acetate, 0.3 g of PVP, 5 mL of anhydrous ethanol, and 5 mL deionized water under stirring. Solution B was prepared by dissolving 0.09 g of H₃BTC in a blend of 5 mL anhydrous ethanol and 5 mL deionized water. Solution B was added dropwise to Solution A, and the mixture was stirred continuously for 24 h. The white precipitate was collected by centrifugation, filtered, washed three times with DMF and ethanol, and dried at 60 °C to obtain MnTP.

3.3. Catalytic Performance Test

The catalytic performance for the conversion of CO₂ and ethylenediamine (EDA) was assessed in a 50 mL stainless steel batch reactor. A typical reaction mixture comprising 19.6 g of isopropanol, 0.6 g of EDA, and a specified amount of catalyst was charged into the reactor. The system was pressurized to 0.6 MPa, heated to a temperature between 80 and 120 °C under stirring at 500 rpm, and maintained for 30 min. After reaction, the products were analyzed using a gas chromatograph (Agilent Nuoxi GC 8890, Santa Clara, CA, USA) equipped with a KT-TTAP capillary column (50 m × 0.32 mm, 0.40 µm film thickness). The oven temperature was programmed as follows: hold at 80 °C for 1 min, ramp to 180 °C at 20 °C/min, and then maintain at 180 °C for 4 min. High-purity nitrogen (99.99%) was used as the carrier gas at a flow rate of 30 mL/min. Product quantification was performed via the area normalization method. The main components identified included EDA, isopropanol, and ethylene urea (EU). The conversion of EDA (X_EDA) and product selectivity (S) were calculated based on the consumption of EDA.

$${\mathrm{X}}_{\mathrm{E}\mathrm{D}\mathrm{A}}(\%)=(1-{\displaystyle \frac{\mathrm{A}\mathrm{m}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t} \mathrm{o}\mathrm{f} \mathrm{E}\mathrm{D}\mathrm{A} \mathrm{a}\mathrm{f}\mathrm{t}\mathrm{e}\mathrm{r} \mathrm{r}\mathrm{e}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}{\mathrm{A}\mathrm{m}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t} \mathrm{o}\mathrm{f} \mathrm{E}\mathrm{D}\mathrm{A} \mathrm{b}\mathrm{e}\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{e} \mathrm{r}\mathrm{e}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}})\times 100\%$$

$$\mathrm{S}(\%)={\displaystyle \frac{\mathrm{A}\mathrm{m}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t} \mathrm{o}\mathrm{f} \mathrm{e}\mathrm{a}\mathrm{c}\mathrm{h} \mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}}{\mathrm{A}\mathrm{m}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t} \mathrm{o}\mathrm{f} \mathrm{r}\mathrm{e}\mathrm{a}\mathrm{c}\mathrm{h}\mathrm{e}\mathrm{d} \mathrm{E}\mathrm{D}\mathrm{A}}}\times 100\%$$

The solid catalyst was effectively separated from the post-reaction mixture by centrifugation, washed thoroughly with ethanol, and dried under vacuum at 100 °C overnight for recycling tests. Following the initial reaction carried out as described above, a sample of the reaction mixture was withdrawn using a pipette for GC analysis. The remaining solution was centrifuged to separate the solid catalyst from the liquid phase. After filtration to remove the liquid, the recovered solid was washed five times with ethanol and dried overnight under vacuum at 100 °C. The recycled catalyst was then employed directly in a subsequent reaction cycle. The quantities of substrate and MnTP catalyst were adjusted in proportion to the amount of recovered catalyst used in each reuse experiment.

4. Conclusions

In summary, a series of Mn₂O₃ catalysts were synthesized from metal–organic frameworks employing different organic ligands—homophthalic acid (H₃BTC), 1,4-benzenedicarboxylic acid (H₂BDC), and a composite ligand system of H₃BTC with polyvinyl pyrrolidone (PVP) via a hydrothermal route—and evaluated for ethylene urea production from CO₂ and ethylenediamine. The Mn₂O₃ catalyst derived from the H₃BTC+PVP combination (labeled as MnTP) demonstrated exceptional performance, reaching 97% EDA conversion and 97% EU selectivity under mild conditions (100 °C, 1 min), surpassing all previously reported catalysts for this transformation under comparable conditions. This high activity was ascribed to the structural benefits offered by the mixed-ligand system, which facilitated the generation of oxygen vacancies and increased the concentration of surface Mn³⁺ species, thereby improving CO₂ activation and substrate adsorption. This study provides an effective strategy for developing high-performance catalysts toward sustainable CO₂ utilization for synthesizing value-added cyclic ureas through rational design of MOF precursors.