MOF-808 as Effective Support for Cu-Based Catalyst for CO₂ Hydrogenation to Methanol

Abstract

The thermocatalytic hydrogenation of CO₂ to methanol offers a promising route for reducing greenhouse gas emissions (GHG) and producing valuable chemicals and fuels. In this study, copper–zinc bimetallic catalysts supported on a zirconium-based MOF-808 framework were synthesized via a facile deposition–precipitation method and compared to a conventional Cu/ZnO/Al₂O₃ (CZA) catalyst. MOF-808 was selected due to its high surface area and porous structure, which enhance metal dispersion. Characterization through X-ray diffraction (XRD) and N₂ physisorption showed significant changes in surface area and pore structure after Cu-Zn incorporation and calcination. The 50-CuZn MOF-808 catalyst achieved the best catalytic performance at 260 °C and 40 bar, demonstrating a high STY of 193.32 gMeOH·Kgcat⁻¹ h⁻¹ and a turnover frequency (TOF) of 47.44 h⁻¹, surpassing traditional CZA catalysts. The strong Cu-Zn-Zr interactions within the MOF-808 framework played a crucial role in promoting CO₂ activation and methanol formation. This study underscores the potential of MOF-808-supported Cu-Zn catalysts as viable alternatives to traditional systems for CO₂ hydrogenation to methanol.

1. Introduction

The rapid increase in atmospheric CO₂ levels, primarily driven by the extensive combustion of fossil fuels, has led to significant environmental challenges, including global warming, ocean acidification, and polar ice melting. The concentration of CO₂ in the atmosphere has risen from pre-industrial levels of around 270 ppm to over 414 ppm today, highlighting the urgent need for solutions that mitigate these emissions while supporting sustainable energy practices [1,2,3]. As efforts to address climate change intensify, the transformation of CO₂ into valuable products through catalytic processes has emerged as one of the promising strategies. Among the various products that can be derived from CO₂, methanol stands out due to its dual role as a clean fuel and as a versatile chemical feedstock. It plays a critical role in the synthesis of light olefins, aromatics, and gasoline, and can serve as a hydrogen carrier, offering a practical solution for hydrogen storage and transportation [4,5,6,7,8]. Consequently, the hydrogenation of CO₂ to methanol presents an attractive platform for addressing both environmental and energy challenges simultaneously.

The catalytic conversion of CO₂ to methanol has also long been recognized as a viable approach to carbon capture and utilization (CCU). However, the process is highly dependent on the development of effective catalysts. Copper-based catalysts, particularly the Cu/ZnO/Al₂O₃ (CZA) system, have been the industrial standard for methanol synthesis from syngas for several decades [9,10,11]. The catalytic activity of CZA is largely attributed to the synergistic interaction between copper and zinc oxide, which creates active sites that promote the adsorption and subsequent hydrogenation of CO₂ [12]. However, the stability of this interface as well as its effectiveness for CO₂ conversion remains a significant challenge.

One of the primary issues with traditional Cu-based catalysts for CO₂ conversion is the gradual agglomeration of copper nanoparticles (NPs) during the reaction, which leads to a decrease in the number of active Cu/ZnO interfaces [13,14,15,16]. As the copper particles separate from the zinc oxide support, they catalyze the reverse water–gas shift (RWGS) reaction, which produces CO rather than methanol, thereby reducing the selectivity of the catalysts for methanol [17]. This phase separation, exacerbated by high pressures and reaction conditions that typically occur at 250–300 °C, results in the deactivation of the catalysts over time, limiting its practical application in industrial settings [18]. As a result, it is important to focus on developing strategies to stabilize these active interfaces and improve the long-term performance of Cu-based catalysts.

Apart from traditional support such as such as ZnO [19,20], ZrO₂ [21,22,23], CeO₂ [24,25,26], and Al₂O₃ [17,27,28,29], recent developments in catalyst design have focused on utilizing metal–organic frameworks (MOFs) as supports for Cu-based catalytic systems. MOFs are a class of porous materials characterized by their high surface areas, tunable pore structures, and the ability to incorporate various metal nodes and organic linkers [30]. These properties make MOFs ideal candidates for stabilizing metal NPs in catalytic reactions due to their ability to provide strong metal–support interactions (SMSIs), which can prevent the agglomeration of copper NPs and maintain the integrity of the Cu/ZnO interface.

Various studies have illustrated the effectiveness of using Cu-based catalysts within MOFs for CO₂ hydrogenation, particularly by optimizing metal–support interactions. An et al. encapsulated Cu/ZnO nanoparticles within UiO-66-bpy MOFs, preventing NP agglomeration and achieving 100% methanol selectivity with a space time yield (STY) of 2.59 gMeOH·kgCu⁻¹·h⁻¹ [31]. Liu et al. developed a Cu@ZrO₂ catalyst via the in situ reconstruction of Cu@UiO-66, showcasing excellent activity and stability due to the robust three-dimensional ZrOx framework and the presence of abundant Cu⁺-ZrO₂ interfaces, which are crucial for its high performance [32]. Mitsuka et al. demonstrated the potential of amorphous UiO-66 frameworks, with the Cu/amUiO-66 catalyst outperforming its crystalline counterpart by threefold in methanol production. They suggested that the amorphous nature of UiO-66 facilitates a more homogeneous dispersion of active sites [33]. Zhu et al. demonstrated the importance of copper–zirconia (Cu-ZrO₂) interfaces within UiO-66, where Cu NPs were incorporated to enhance CO₂ adsorption and methanol synthesis. Two methods, ion exchange and impregnation, were explored, with ion exchange preserving Cu-O-Zr bonds for better catalytic performance [34].

Stawowy et al. explored a different route by partially substituting Zr⁴⁺ with Ce⁴⁺ in UiO-66. This modification increased methanol selectivity, revealing the potential of Ce incorporation in enhancing catalytic performance [35]. Marcos et al. investigated the physiochemical characteristics of Cu NPs supported on UiO-66 and noted the significant influence of copper loading on catalytic performance. The 20Cu/MOF catalyst, with the highest copper loading, demonstrated a turnover frequency (TOF) of 7.4 mol/molCu∙h⁻¹ at 250 °C and 32 bar, attributed to the increased metallic copper surface area and the presence of basic sites related to Zr clusters [36]. Song et al. explored the impact of the MOF coordination environment on the catalytic behavior of Cu@UiO-66 catalysts, showing superior methanol selectivity compared to traditional Cu/ZnO/Al₂O₃ systems, achieving a space time yield (STY) of 2.86 gMeOH·kgCu⁻¹·h⁻¹ at 260 °C. The superior performance was explained by its structural integrity and that the MOF structure undergoes an in situ transformation into a mixed phase of amorphous and tetragonal ZrO₂ as well as strong Cu-ZrO₂ interactions facilitating abundant Cu⁺ sites crucial for methanol synthesis [37].

Similarly, Wang et al. reported an STY of 2.649 gMeOH·gCu⁻¹·h⁻¹ using CuO/s-UiO-66 with strong Cu-ZrO₂ interactions. This superior performance is attributed to the unique properties of the Si-infused ZrO₂ framework, such as high surface area and enlarged pore size, which ensure optimal Cu dispersion and accessibility [38]. Yu et al. optimized Cu/Zn mole ratios in Cu-Zn@UiO-66, demonstrating an optimal ratio of 2.5 for maximum CO₂ conversion and stable methanol selectivity [39]. Yaghi et al. highlighted the formation of effective Cu-Zr interface sites through the interaction between Cu NPs and Zr metal nodes, which played a critical role in improving catalyst performance. The Cu⊂UiO-66 catalyst exhibited remarkable 100% methanol selectivity and an eightfold increase in methanol yield when compared to conventional Cu/ZnO/Al₂O₃ catalysts at 175 °C. Furthermore, the study compared two methods for producing Cu/ZnO nanoparticles: In situ generation within the MOF resulted in a finer Cu-Zn mix and the formation of ultrasmall NPs, boosting catalytic performance significantly. In contrast, the ex situ approach, where NPs were synthesized before being introduced into the MOF, led to larger particles with less efficient distribution, which ultimately diminished the catalyst’s activity [40]. However, the intricate and time-consuming nature of the preparation methods for these catalysts significantly limits their potential for large-scale industrial applications.

In this study, we systematically investigate the effect of varying copper loadings on the catalytic performance of Cu-loaded MOF-808 catalysts by a facile deposition–precipitation method for CO₂ hydrogenation to methanol. MOF-808, a Zr-based MOF with large, open pores, offers a highly tunable platform for the incorporation of Cu NPs and has not been investigated as support for Cu-based methanol synthesis catalysts. By optimizing the copper content, we aim to maximize the formation of stable Cu/ZnO interfaces, thereby enhancing the catalysts’ activity and selectivity for methanol production. Additionally, we examine the structural properties of the Cu-loaded MOF-808 catalysts to gain insights into the mechanism underlying the observed catalytic performance.

2. Results and Discussion

2.1. Catalyst Characterization

2.1.1. Inductively Coupled Plasma (ICP) Analysis and N₂ Physisorption

Table 1. Copper and Zinc loading, specific surface area, pore volume, and mean pore size of different catalysts.

Catalysts	Theoretical Loading (wt.%)			Actual Loading a (wt.%)			Surface Area (m2/g)	Pore Volume (cm3/g)	Pore Size (nm)
	Cu	Zn	Al	Cu a	Zn a	Al a
CZA	50	30	10	46	21	7	84.52	0.177	8.4
20-CZ MOF	20	10		22	11	-	31.15	0.092	7.0
30-CZ MOF	30	15		29	14	-	24.23	0.063	6.2
50-CZ MOF	50	25		43	20	-	15.64	0.026	6.0
50-C-CZ MOF *	50	25		49	24	-	15.18	0.033	8.4
MOF-808	-	-	-	-	-	-	2143.20	0.911	1.5

*—“C” in 50-C-CZ MOF denotes the MOF has been calcined at 250 °C before loading of Cu and Zn; ^a—determined from ICP-MS.

summarizes the theoretical and actual Cu loadings measured by ICP-MS for various catalysts, including a 50% Cu-loaded CZA catalyst synthesized for comparison. It can be observed that the actual Cu and Zn loadings for all catalysts are close to their theoretical values. This indicates quantitative incorporation of metals during synthesis.

The N₂ physisorption of pristine MOF-808 shown in Figure S1 exhibits type 1 Langmuir behavior with a high surface area of 2143.20 m²/g and a pore volume of 0.911 cm³/g, and a microporous structure with an average pore size of 1.5 nm. These characteristics are ideal for catalytic applications, offering ample adsorption sites and facilitating the diffusion of small molecules [41]. Figure 1a,b show the N₂ physisorption isotherms and pore size distribution of the Cu-Zn-loaded catalysts, respectively, highlighting significant structural changes upon metal incorporation. After the introduction of Cu and Zn, there is a significant reduction in both surface area and pore volume, as shown in Table 1. The catalysts exhibit a type IV isotherm with hysteresis loops, which suggests that these catalysts possess a mesoporous structure [39]. Despite the decrease in surface area, the Cu-Zn-loaded MOF samples exhibit a transition from microporosity to mesoporosity, as indicated by the increase in average pore size as well as the adsorption isotherm, seen in Figure 1.

For the 20-CZ MOF, the surface area decreases to 31.15 m²/g, with a corresponding reduction in pore volume to 0.092 cm³/g. This trend continues with increasing metal loadings, as seen in the 50-CZ MOF, where the surface area drops to 15.64 m²/g, and the pore volume decreases to 0.026 cm³/g. This is due to the structural decomposition of MOF-808, leading to mesoporosity formation after calcination. Specifically, the organic linkers decompose, causing the collapse of micropores and the formation of larger mesopores. The further reduction can also be partly attributed to the partial blockage of the internal pore structure by the deposited Cu and Zn species, which fill the micropores and reduce the available space. The 20-CZ MOF shows an average pore size of 7.0 nm, a significant increase compared to the pristine MOF-808. This shift toward mesoporosity is also apparent at higher loadings, as demonstrated by the 50-CZ MOF, with an average pore size of 6.0 nm. The larger pores in the Cu-Zn-modified MOF-808 catalysts allow for easier diffusion of reactants and products, reducing the diffusion limitations often encountered with microporous materials [42].

2.1.2. XRD Characterization

Figure 2a,b presents the XRD patterns for CZA, corresponding Cu-Zn-loaded MOF catalysts and the pristine MOF-808, respectively. The MOF-808 sample exhibited its characteristic diffraction peaks of the structure [41]. After the loading of Cu and Zn onto MOF-808 and calcination, the characteristic diffraction peaks of the pristine MOF-808 structure were no longer observed. This suggests that the introduction of metal species and the subsequent calcination caused significant disruption to the MOF framework, due to the distortion, i.e., loss of linked ligands, which leads to the loss of long-range order in the MOF structure. The XRD patterns reveal the appearance of distinct diffraction peaks corresponding to the CuO phase, with characteristic peaks at 2θ of 35.45°, 35.55°, and 38.73° (PDF#00-005-0661). Diffraction peaks at 31.75°, 34.4°, and 36.25° align with the characteristic reflections of ZnO (JCPDS#00-005-0664). As we increased the Cu loading, the peak intensity of CuO and ZnO gradually increased, indicating the growth of CuO and ZnO crystallites. Furthermore, slight shifts in both CuO and ZnO diffraction peaks were also observed, signifying a different degree of interaction of the oxides with Al₂O₃ or ZrO₂ support. Similar trends of peak shifts and peak intensity change have been reported by Zhao et al. [43]. The crystal sizes of the catalyst were calculated using the Scherrer equation (

Table 2. Copper surface area, TOF, dispersion, and CuO size of different catalysts.

Catalysts	Scu (m2/g) a	TOF (h−1)	DCu (%) a	D (nm) b
CZA	26.07	4.61	8.42	9.39
20-CZ MOF	6.14	24.43	4.14	7.71
30-CZ MOF	9.40	26.22	4.70	8.14
50-CZ MOF	15.10	47.44	5.21	9.91
50-C-CZ MOF	10.37	38.64	3.14	9.96

^a. Determined from N₂O chemisorption. ^b. Calculated by Scherrer equation.

). CZA showed an average size of 9.39 nm, while the MOF-derived catalysts had sizes ranging from 7.71 to 9.96 nm. Among these, 20-CZ MOF showed the smallest size, while the 50-CZ MOF and 50-C-CZ MOF had the largest size of 9.96 nm.

2.1.3. SEM and TEM Characterization

Figure 3a–c show SEM images of MOF-808, 50-CZ MOF before calcination, and 50-CZ MOF after calcination, respectively. The MOF-808 sample displays a characteristic octahedral morphology, attributed to the intrinsic crystalline structure and symmetry of the zirconium-based framework. In the uncalcined 50-CZ MOF, the MOF particles are visible, with Cu-Zn species appearing to be well dispersed. However, upon calcination, the MOF structure undergoes significant change, leading to the loss of its distinctive octahedral shape due to ligand decomposition and framework collapse under thermal treatment. Figure 3d–f shows the representative TEM images for the 50-CZ MOF catalyst at different magnifications. Characteristic lattice fringes corresponding to various metal oxides can be identified. The fringes at 2.60 and 2.41 Å are characteristic of the CuO (002) and CuO (101) planes, respectively. This indicates the presence of copper oxide as a major phase within the catalyst. The lattice fringes at 2.80 and 1.47 Å are attributed to the ZnO (100) and ZnO (103) planes, respectively, confirming the formation of zinc oxide within the catalyst. ZnO not only acts as a structural stabilizer but also enhances the interaction between Cu and Zn, promoting the formation of active sites at the Cu-Zn interface. The appearance of a 2.72 Å fringe, corresponding to the ZrO₂ (200) plane, is particularly noteworthy, and originates from the Zr node in the MOF structure.

2.1.4. Temperature-Programmed Reduction (TPR) and N₂O Chemisorption

The H₂-TPR profiles of various catalysts are shown in Figure 4. For the CZA catalyst, the reduction behavior is characterized by two distinct peaks: an α peak at 185 °C and a β peak at 216 °C. The α peak at 185 °C represents the reduction of surface CuO species, slightly elevated compared to other CuZn systems due to the stabilizing interaction between CuO and Al₂O₃. The β peak at 216 °C indicates the reduction of bulk CuO. The reduction temperatures for the catalysts are as follows: 20-CZ MOF > 30-CZ MOF > 50-C-CZ-MOF > 50-CZ-MOF. With increasing Cu loading, the reduction peak temperature tends to decrease, implying that larger Cu particles exhibit weaker interactions with ZnO, which facilitates their reduction. This behavior aligns with findings from XRD characterization that CuO particle size increases with Cu loading.

N₂O chemisorption was conducted following H₂ TPR to determine the metallic copper surface area, as summarized in Table 2. Among the catalysts, the CZA sample exhibits the highest copper surface area (26.07 m²/g). As the Cu loading increases, there is a corresponding increase in copper surface area (20-CZ MOF < 30-CZ MOF < 50-C-CZ MOF < 50-CZ MOF). This trend suggests that the MOF based catalysts effectively accommodates increasing amounts of Cu species while maintaining a relatively high dispersion. The 50-C-CZ MOF sample exhibits a reduction in surface area compared to its non-calcined counterpart.

2.1.5. XPS Analysis

The XPS analysis of the CZ-MOF catalysts provides a detailed understanding of the surface composition and chemical states of the elements present (Figure 5). The Cu 2p spectra exhibit strong peaks at binding energies of 932.30 eV for Cu 2p₁/₂ and 952.15 eV for Cu 2p₃/₂, which are characteristic of Cu²⁺. Figure 5a shows the presence of shake-up satellite peaks between 938 and 945 eV, which further confirms that copper remains in the Cu²⁺ oxidation state [44]. As the Cu content increases from 20 to 50 wt.%, the intensity of the Cu 2p peaks increases, reflecting the higher surface concentration of copper. Compared with the CZA catalyst, the peaks of MOF-supported catalysts shifted to lower binding energies, indicating stronger Cu-Zr interactions compared to Cu-Al and electron density accumulation on Cu species. This shift suggests charge transfer from ZrO₂ to Cu, stabilizing Cu species, which are known to enhance methanol selectivity [22]. This is consistent with Moafor et al., where Co–zeolite interactions induced Co binding energy shifts due to electronic effect [45]. Additionally, a shift in the binding energy of the Cu 2p₃/₂ and Cu 2p₁/₂ peaks is observed for MOF-loaded catalysts compared to the CZA catalyst. This shift suggests a different electronic environment for Cu species, likely due to stronger Cu–support interactions in the MOF framework.

Figure 5b shows Zn 2p spectra peaks at 1020.74 eV for Zn 2p₃/₂ and 1043.69 eV for Zn 2p₁/₂. The consistency in these peaks across different CZ MOF catalysts indicates that zinc remains in its Zn²⁺ state after calcination, while the peak intensity increases with increasing loading. Figure 5c shows Zr 3d spectra reveal binding energies at 181.36 eV for Zr 3d₅/₂ and 183.86 eV for Zr 3d₃/₂, indicating Zr⁴⁺ in the form of ZrO₂. Notably, the intensity of these peaks decreases with increasing CuZn loading, suggesting increasing surface coverage of the zirconium by the deposited metal species. Figure 5d represents the survey for the 50-CZ MOF, where Cu 2p, Zn 2p, and Zr 3d are present, although the Zr 3d peak is minuscule because of the low amount of MOF-808. In the XPS spectra of the CZA catalyst (Figure S3), the Cu 2p peaks confirm the presence of Cu²⁺, while the Zn 2p peaks indicate Zn²⁺. Additionally, Al 2p peaks are observed, though their low intensity suggests a small presence of Al₂O₃.

2.2. Activity Tests

Table 3. CO₂ hydrogenation performance on the CZA and CZ MOF catalysts with different copper loadings.

Catalyst	STY (gMeOH·Kgcat−1 h−1)	CO2 Conversion (%)	CH3OH Selectivity (%)	CO Selectivity (%)	CH3OH Yield (%)
CZA	94.11	19.57	11.17	88.83	2.23
20-CZ MOF	117.32	5.78	48.90	51.10	2.83
30-CZ MOF	192.78	6.29	59.76	40.24	3.75
50-CZ MOF	193.32	8.60	43.93	56.07	3.78
50-C-CZ MOF	94.91	4.54	48.49	51.51	2.20

Reaction conditions: T = 260 °C, P = 40 bar, H₂/CO₂ = 3, GHSV = 18,000 h⁻¹; data taken after 4 h when steady state reached.

summarizes the catalytic performance of the CZA catalyst and the CZ MOF series for CO₂ hydrogenation at 260 °C and 40 bar. Figure 6a presents CO₂ conversion, CH₃OH selectivity, and yield of different catalysts. The CZA catalyst exhibits the highest CO₂ conversion of 19.57% due to its high Cu surface area, though it demonstrates the lowest CH₃OH selectivity of 11.17%. In contrast, the CZ MOF catalysts show an increase in CO₂ conversion with higher Cu loading, reaching a maximum of 8.60% for the 50-CZ MOF. A comparison of activity data with the literature is listed in Supplementary Information (SI) Table S1, and it can be concluded that our catalysts performed better than many of the previous reported catalysts.

The Cu surface area (Table 2) significantly influences catalytic performance as observed. The CZA catalyst achieves the highest Cu surface area (26.07 m²/g), which contributes to its superior CO₂ conversion but lower methanol selectivity. Among the CZ MOF catalysts, 50-CZ MOF displays a higher Cu surface area (15.10 m²/g). This correlates with its higher CO₂ conversion and methanol yield. However, the 30-CZ MOF catalyst already achieved a good methanol yield due to its high methanol selectivity. Therefore, the Cu loading could be optimized for the catalytic performance. The calcined 50-C-CZ MOF with a moderate Cu surface area exhibits relatively low CO₂ conversion (4.54%) and methanol yield (2.20%). This can be attributed to the weakened Cu-Zr interface, as calcination of the MOF framework prior to metal deposition likely disrupts the interaction between copper species and the Zr nodes. The reduced interaction diminishes the synergistic effect required for efficient CO₂ activation and methanol synthesis, underscoring the critical role of the Cu-Zr interface in achieving high catalytic performance.

The space time yield (STY) data of the CZA and CZ MOF catalysts at 260 °C and 40 bar are presented in Figure 6b. The 50-CZ MOF exhibited the best performance, showing a STY of 193.32 gMeOH·Kgcat⁻¹ h⁻¹. The results underscore the crucial role of the MOF-808 system as a viable alternative support to conventional CZA catalyst. The CZA catalyst is well regarded for its excellent methanol production performance due to the optimal dispersion of Cu active sites and the strong synergetic interaction between Cu and ZnO [46,47]. When compared to the CZA catalyst, the MOF-based catalyst exhibits competitive performance under the same conditions. The superior performance of the 50-CZ MOF catalyst indicates that the MOF structure enhances the Cu-Zr synergy, enabling more efficient CO₂ activation due to the presence of ZrO₂, even though the Cu surface area is lower than that of the CZA.

The turnover frequence (TOF) of the CZA and CZ MOF catalysts at 260 °C and 40 bar are also calculated and summarized in Table 2. The 50-CZ MOF exhibited the highest TOF of 47.44 h⁻¹, possibly due to its optimized Cu/Zn-Zr interfaces [48]. It is also obvious that all the MOF-based catalysts have higher TOFs than the CZA system, again manifesting the potential of MOF-808 as catalyst support for CuZn in CO₂ hydrogenation to methanol.

The effect of temperature on catalytic performance of the best performing 50-CZ MOF catalyst is shown in Figure 6c, where the STY of methanol increases with temperature, peaking at 260 °C, followed by a slight decline at 280 °C, indicating temperature-dependent reaction behavior. Kinetically, the CO₂ conversion increases with temperature as expected. However, the thermodynamic nature of the reactions involved dictates that at low temperatures, the methanol synthesis reaction is favored, while the RWGS reaction is favored at higher temperatures, which limits the extent of methanol formation [49]. The CO selectivity of 56.07% for 50-CZ MOF at 260 °C indicates that the RWGS reaction is pronounced under the reaction conditions, as Cu is a well-known WGS reaction catalyst. Although the methanol selectivity is only moderate at 43.93%, CO₂ conversion at this temperature contributes to a relatively high methanol yield (3.78%). This dual-pathway behavior is not unusual for Cu-based systems, particularly when ZnO is present as a promoter, which tends to stabilize intermediate species conducive to both reactions [49].

The effects of pressure for the 50-CZ MOF are shown in Figure 6d. As pressure increases from 10 to 40 bar, both CO₂ conversion and CH₃OH selectivity show an upward trend, consistent with thermodynamic expectations [50]. The GHSV from 7500 to 18,000 h⁻¹ led to a decline in CO₂ conversion, as expected, while simultaneously enhancing methanol selectivity (Figure 6e). Additionally, the 50-CZ MOF catalyst demonstrated excellent stability during long-term activity test, maintaining stable performance for over 200 h of time-on-stream (Figure 6f). This stability is particularly important for industrial applications, where catalysts are prone to deactivation due to sintering or water poisoning. The MOF framework contributes to this stability by providing a well-defined porous structure that prevents agglomeration of active Cu particles while facilitating the diffusion of reactants and products, including water. The combination of enhanced catalytic activity and long-term stability makes the Cu-Zn MOF-808 system a promising candidate for CO₂ hydrogenation to methanol.

3. Materials and Methods

3.1. Catalyst Preparation

The materials used in the experiments were of high purity and sourced from various suppliers. Copper (II) nitrate trihydrate (Cu(NO₃)₂∙3H₂O, >99.5%) was provided by Acros Organics (Geel, Belgium). Zinc (II) nitrate hexahydrate (Zn(NO₃)₂∙6H₂O, >99.9%) and silicon carbide (SiC, 40 mesh, powder) were sourced from Alfa Aesar (Ward Hill, MA, USA). Aluminum (III) nitrate nonahydrate (Al(NO₃)₃∙9H₂O, >98.5%) and sodium carbonate (Na₂CO₃, >99.9%) were bought fromMerck Millipore(Burlington, MA, USA). Zirconium dichloride oxide octahydrate (ZrOCl₂·8H₂O) 98% and 1,3,5-Benzenetricarboxylic acid 98% were purchased from Thermo Scientific (Waltham, MA, USA).

a.

Synthesis of Cu/ZnO/Al₂O₃ (CZA) catalyst

The Cu/ZnO/Al₂O₃ (CZA) catalyst was synthesized through a modified co-precipitation method adapted from Behrens et al. [51]. Specifically, 10.57 g of copper (II) nitrate trihydrate (Cu(NO₃)₂·3H₂O), 6.50 g of zinc nitrate hexahydrate (Zn(NO₃)₂∙6H₂O), and 7.35 g of aluminum nitrate nonahydrate (Al(NO₃)₃∙9H₂O) were dissolved in distilled water, along with 5 mL of concentrated nitric acid to ensure complete dissolution of the metal salts. A 1.5 M solution of sodium carbonate (Na₂CO₃) was prepared as the precipitating agent. Both the metal precursor and sodium carbonate solutions were injected dropwise into a reaction flask containing 100 mL of distilled water maintained at 60 °C. During the co-precipitation process, the pH of the solution was carefully controlled within the range of 6 to 7. The resulting precipitate was aged under continuous stirring for 3 h. The precipitate was then filtered and washed thoroughly with distilled water to remove any residual ions. The filtered solid was dried overnight at 70 °C, followed by calcination in atmosphere at 350 °C for 3 h.

b.

MOF-808 and CZ MOF catalyst synthesis

The MOF-808 catalyst was synthesized using zirconium (IV) oxychloride octahydrate as the metal precursor, trimesic acid as the organic linker, and acetic acid as the modulator to enhance crystallinity and stability. The precursors were dissolved in water, and the solution was refluxed at 95 °C for synthesis. The resulting product was separated, thoroughly washed with water and ethanol, and dried to obtain a fine powder. CZ MOF catalysts with varying copper loadings of 10 wt.%, 20 wt.%, and 50 wt.% were synthesized using the deposition–precipitation method adapted from Yu et al. [39].Firstly, 0.5 M Cu(NO₃)₂∙3H₂O, 0.5 M Zn(NO₃)₂∙6H₂O, and 0.5 M Na₂CO₃ aqueous solutions were prepared, respectively. The metal nitrate solution was added and stirred until complete dissolution was achieved. Subsequently, 1.0 g of activated MOF-808 powder was dispersed into the metal nitrate solution, and the mixture was stirred at 60 °C for 1 h. A 0.5 M aqueous solution of Na₂CO₃ was added dropwise to the suspension until the pH of the solution reached 7.5. The resulting suspension was stirred at 60 °C for an additional 5 h to ensure adequate deposition of the metal species. After aging, the mixture was allowed to cool for 1 h at room temperature. The solid was collected by filtration and washed thoroughly with deionized water. The washed catalyst was dried at 110 °C overnight in an oven. Finally, the dried product was calcined in air at 250 °C for 4 h. The calcined material was then ground and sieved to a particle size of 40–60 mesh. Additionally, to investigate the effects of calcining MOF-808 (in air) and subsequently loading of Cu-Zn, the MOF was first calcined to produce ZrO₂ (linkers turned to carbon). Following this, Cu-Zn was deposited using the same procedure. This approach allowed for us to study how the transformation from MOF to ZrO₂ impacts the performance of the resulting Cu-Zn catalysts in CO₂ hydrogenation reactions.

3.2. Catalyst Characterization

Nitrogen adsorption–desorption measurements were carried out at 77 K using a Micromeritics (Norcross, GA, USA), TriStar II instrument to evaluate the textural properties of the catalysts. Prior to analysis, the samples underwent vacuum degassing using a Micromeritics (Norcross, GA, USA), VacPrep 061 system. The specific surface area (SSA) was determined following the Brunauer–Emmett–Teller (BET) method, while the pore size distribution was analyzed using the Barrett–Joyner–Halenda (BJH) model. The total pore volume (PV) was derived from the nitrogen adsorption capacity at a relative pressure (P/P₀) of 0.99.

The crystalline phases of the catalysts were examined using powder X-ray diffraction (XRD) on a Bruker-AXS D8 ADVANCE diffractometer (Billerica, MA, USA), employing Cu Kα radiation (λ = 1.5406 Å) at an operating voltage of 40 kV and a current of 25 mA. Data were collected over a 2θ range of 3° to 90° at a scan rate of 5°/min. Phase identification was performed using the Joint Committee on Powder Diffraction Standards (JCPDS) database, and the average crystallite size was estimated via the Scherrer equation.

The morphological and structural characteristics of the catalysts were investigated through transmission electron microscopy (TEM) using a JOEL (Peabody, MA, USA) JEM-2100F instrument operated at 200 kV. TEM samples were prepared by dispersing the catalyst powders in ethanol via ultrasonication, followed by depositing a droplet of the suspension onto a holey carbon-coated copper grid. Additionally, scanning electron microscopy (SEM) was performed using a ZEISS (Oberkochen, Germany) Gemini Supra 35VP instrument to assess the morphology of the precursor powders, with samples being fixed onto carbon tape before imaging.

The elemental composition of the catalysts was quantified by inductively coupled plasma mass spectrometry (ICP-MS) using an Agilent 7850 instrument (Santa Clara, CA, USA). Prior to measurement, the samples were digested in a heated mixture of nitric acid and hydrochloric acid (1:3), followed by dilution and filtration.

Temperature-programmed reduction (H₂-TPR) was performed using a Micromeritics (Norcross, GA, USA), Autochem II ASAP 2920 system to evaluate the reducibility of the catalysts. The samples underwent pretreatment under helium flow at 200 °C for 30 min before the experiment. During the TPR analysis, a 7 vol% H₂/Ar gas mixture was passed over the sample at a flow rate of 50 mL/min, with the temperature ramped from ambient to 300 °C at a heating rate of 10 °C/min.

The copper surface area was determined by nitrous oxide (N₂O) chemisorption using the Micromeritics Autochem II ASAP 2920 instrument. The catalyst was first pretreated by heating to 200 °C under helium flow for 30 min, followed by reduction in 7 vol% H₂/Ar at 300 °C for 2 h. After purging with helium until the sample cooled to 50 °C, N₂O adsorption was carried out in a 1% N₂O/He gas mixture at 50 °C for 1 h. The system was then flushed with helium for another hour to remove any physisorbed N₂O. Quantification of N₂O consumption was conducted through a subsequent H₂-TPR analysis from 50 to 300 °C at a heating rate of 10 °C/min in a 7 vol% H₂/Ar atmosphere. The exposed copper surface area (SA_Cu) was estimated from the hydrogen consumption using Equation (1).

$$\mathrm{S}{\mathrm{A}}_{\mathrm{Cu}}\left({\displaystyle \frac{{\mathrm{m}}^2}{{\mathrm{g}}_{\mathrm{cat}}}}\right)={\displaystyle \frac{\mathrm{Y}\times \mathrm{SF}\times {\mathrm{N}}_{\mathrm{A}}}{{\mathrm{C}}_{\mathrm{M}}\times {\mathrm{W}}_{\mathrm{cat}}}}\times 100$$

(1)

where Y is the moles of H₂ consumed in the TPR following N₂O chemisorption, SF is the stoichiometric factor (2), N_A is Avogadro’s number (6.022 × 10²³ mol⁻¹), C_M is the number of surface Cu atoms per unit surface area (1.47 × 10¹⁹ atoms∙m⁻²), and Wcat is the amount of catalyst (g).

The dispersion (D_Cu) is calculated from Equation (2), where n_surfaceCu is number of Cu atoms on the surface and n_totalCu is total number of Cu atoms in the catalyst.

$${\mathrm{D}}_{\mathrm{Cu}}={\displaystyle \frac{{\mathrm{n}}_{\mathrm{surfaceCu}}}{{\mathrm{n}}_{\mathrm{totalCu}}}}\times 100\%$$

(2)

The turnover frequency of methanol (TOF_Methanol) is calculated according to Equation (3).

$$\mathrm{T}\mathrm{O}{\mathrm{F}}_{\mathrm{Methanol}}({\mathrm{h}}^{-1})={\displaystyle \frac{{\mathrm{R}}_{\mathrm{methanol}}\times {\mathrm{N}}_{\mathrm{A}}}{\mathrm{S}{\mathrm{A}}_{\mathrm{Cu}}\times {\mathrm{C}}_{\mathrm{M}}}}$$

(3)

where SA_Cu is obtained from the N₂O chemisorption and R_methanol is the rate of methanol production in (mol∙g⁻¹∙s⁻¹).

The elemental composition and oxidation states of surface species were determined by X-ray photoelectron spectroscopy (XPS) using an Thermo Fisher Scientific (Waltham, MA, USA), ESCALAB Xi+ system equipped with Al Kα radiation 1486.6 eV, and a scan range from 1 to 1300 eV.

3.3. Catalyst Evaluation

The CO₂ hydrogenation experiments were conducted using a high-pressure continuous flow fixed-bed reactor. For each test, 0.2 g of the catalyst was thoroughly mixed with 0.6 g of quartz sand (both sieved to 40–60 mesh) and then packed between layers of quartz wool in the reactor tube. The catalyst bed was positioned centrally along the reactor axis, and a thermocouple placed near the bed was used to monitor the reaction temperature. Prior to initiating the reaction, the catalyst underwent a prereduction process at 250 °C for 2 h under a flow of 10 vol% H₂/N₂. Following this reduction step, the reactor was cooled to 60 °C, and the gas feed was switched to a mixture containing 60% H₂, 20% CO₂, and 20% N₂. The hydrogenation reaction was carried out at a pressure of 40 bar with the reactor temperature controlled between 200 and 280 °C and a gas hourly space velocity (GHSV) ranging from 7500 to 18,000 h⁻¹. To prevent condensation of the gaseous products, all lines and valves were maintained at 150 °C throughout the experiment. The products were analyzed online using an Agilent 8860 gas chromatograph (GC) using HP-5 column with flame ionization detector (FID), along with a Molecular Sieve 5A column to thermal conductivity detector (TCD). The CO₂ conversion XCO₂, selectivity of product (Si), and space time yield (STY) of methanol were calculated using N₂ as internal standard according to Equations (4)–(6).

$${{\mathrm{X}}_{\mathrm{CO}}}_2(\%)=\left[1-{\displaystyle \frac{\mathrm{moles}\mathrm{C}{{\mathrm{O}}_2}_{\mathrm{out}}}{\mathrm{moles}\mathrm{C}{{\mathrm{O}}_2}_{\mathrm{in}}}}\times {\displaystyle \frac{\mathrm{moles}{{\mathrm{N}}_2}_{\mathrm{in}}}{\mathrm{moles}{{\mathrm{N}}_2}_{\mathrm{out}}}}\right]\times 100$$

(4)

$${\mathrm{S}}_{\mathrm{i}}\left(\%\right)={\displaystyle \frac{\mathrm{moles}\mathrm{produc}{{\mathrm{t}}_{\mathrm{i}}}_{\mathrm{out}}}{\mathsf{\Sigma }\mathrm{moles}\mathrm{produ}{{\mathrm{ct}}_{\mathrm{i}}}_{\mathrm{out}}}}\times 100$$

(5)

$$\mathrm{ST}{\mathrm{Y}}_{\mathrm{C}{\mathrm{H}}_3\mathrm{OH}}\left({\displaystyle \frac{\mathrm{g}}{\mathrm{K}{\mathrm{g}}_{\mathrm{cat}}.\mathrm{h}}}\right)={\displaystyle \frac{\frac{{{\mathrm{X}}_{\mathrm{CO}}}_2\times {\mathrm{S}}_{\mathrm{i}}}{10000}\times \frac{{{\mathrm{F}}_{\mathrm{CO}}}_2\times 60}{22400}\times {\mathrm{M}}_{\mathrm{CH}3\mathrm{OH}}\times 1000}{{\mathrm{W}}_{\mathrm{cat}}}}$$

(6)

where FCO₂ in is the molar flowrate of CO₂ at the inlet of the reactor, M_CH3OH is molar mass of CH₃OH and W_cat (g) is the amount of catalyst.

4. Conclusions

In this study, Cu-Zn catalysts derived from a Zr-based MOF-808 framework were successfully synthesized and systematically characterized to evaluate their performance in CO₂ hydrogenation to methanol. XRD and TEM confirmed the structural transformation of MOF-808 into a composite of CuO, ZnO, and ZrO₂. N₂O chemisorption and XPS analyses highlighted the increase in Cu surface area with increasing Cu loading and stronger Cu-Zn-Zr interaction, with higher TOFs for the MOF-supported catalysts. Among the catalysts investigated, the 50-CZ MOF demonstrated outstanding catalytic activity, achieving high TOF, methanol yield, and selectivity. This superior performance can be attributed to the well-dispersed copper species and strong Cu-Zn-Zr interactions, which collectively enable efficient CO₂ activation and hydrogenation.

The unique structural properties of the MOF-808 framework, including its high surface area, tunable porosity, and the existence of Zr node with ability to stabilize active metal sites, played a crucial role in facilitating optimal Cu-Zn-Zr synergy. These findings highlight the potential of MOF-based catalysts as a robust and efficient platform for methanol synthesis, offering a promising alternative to traditional CZA systems.