Facilitation of CO₂ Hydrogenation to Methanol by Spinel ZnGa₂O₄ in Cu-ZnO Catalysts

Abstract

The hydrogenation of CO₂ to methanol is an effective approach for utilizing carbon resources. Cu-ZnO-based catalysts have attracted significant attention due to their ability to activate CO₂; however, improving methanol selectivity remains a challenge. In this study, the incorporation of an appropriate amount of Ga into Cu-ZnO catalysts, resulting in the formation of spinel ZnGa₂O₄ crystals, significantly enhances the conversion of CO₂ to methanol. Ternary CuZnGa composite oxides with varying Ga contents were synthesized, and their effects on CO₂ hydrogenation were investigated. The optimal Cu₆Zn₃Ga₁ catalyst achieved a CO₂ conversion rate of 13% and a methanol selectivity of 59% under reaction conditions of 240 °C, 4 MPa, and a GHSV of 7500 mL⋅g_cat⁻¹⋅h⁻¹. In contrast, the undoped Cu₆Zn₄ catalyst exhibited a lower CO₂ conversion of 9.8% and a methanol selectivity of 38%. Characterization results indicate that the introduction of Ga promotes the formation of oxygen vacancies, enhances CO₂ activation, and facilitates electronic interactions between spinel ZnGa₂O₄ and Cu sites, thereby improving methanol production. Furthermore, the spinel ZnGa₂O₄-modified Cu catalyst demonstrated excellent stability over 90 h of continuous operation. This study presents a novel approach to designing spinel ZnGa₂O₄-modified Cu-ZnO-based catalysts and offers a new strategy for enhancing CO₂ hydrogenation to methanol.

1. Introduction

The burning of fossil fuels releases large amounts of CO₂ into the atmosphere, and the rapid increase in CO₂ has caused many environmental problems, such as ocean acidification, climate warming, and more frequent volcanic eruptions [1,2,3,4,5]. Many researchers are exploring ways to store and change CO₂ into valuable chemicals [6,7]. CO₂ is a safe and non-toxic C1 molecule that can be readily transformed into high-value chemicals, like CH₄ [8], CH₃OH [9,10,11], C₂H₅OH [12], olefins [13], aromatics [14], and others. One promising process is the catalytic hydrogenation of CO₂ to methanol. Methanol serves as a crucial C1 platform molecular, facilitating the production of fuels and other valuable chemicals [15,16]. As an important bridge between CO₂ and high-value-added chemicals, methanol is considered a key component in achieving “carbon neutrality” [17]. The hydrogenation of CO₂ to methanol can help mitigate the conflict between economic development and sustainability, offering significant economic and energy value.

Since methanol synthesis and reverse water–gas shift (RWGS) reactions are competing processes, achieving a balance between high CO₂ conversion and high methanol selectivity is particularly important [18]. CO₂ hydrogenation to methanol is an exothermic reaction, while the competitive RWGS reaction is endothermic. Therefore, from a thermodynamic perspective, low temperature and high pressure favor methanol formation during CO₂ hydrogenation [19]. Cu-ZnO-based catalysts are commonly used for methanol synthesis from CO₂/H₂ and have been extensively studied in the literature [20,21,22,23,24]. However, a significant challenge with these catalysts is that water produced during the reaction accelerates the sintering of Cu, the active component in the CO₂ hydrogenation process, leading to poor stability [25]. Developing rational strategies for catalyst optimization remains a critical challenge.

The spinel structure has shown great potential in CO₂ hydrogenation to methanol due to its unique properties [26]. It has been reported that surface-dissociative ZnO and ZnCr₂O₄ spinel structures promote methanol synthesis rates and inhibit CO formation from the RWGS reaction [19]. Furthermore, spinel CuAl₂O₄ and MgAl₂O₄ have also been found to increase the number of active basic sites for CO₂ adsorption and activation, thereby enhancing methanol selectivity [27]. In addition, spinel ZnFe₂O₄-supported Cu catalysts with tunable Cu nanoparticle sizes have shown increased methanol selectivity due to the generation of additional Cu⁺ sites at Cu–spinel interfaces, which markedly inhibit the reverse water–gas shift reaction during CO₂ hydrogenation [28]. More recently, Cu/ZnAl₂O₄ prepared by the ammonia evaporation method was also reported to improve methanol synthesis in CO₂ hydrogenation by regulating the interaction between spinel ZnAl₂O₄ and Cu sites [29].

Encouraged by these findings, a series of Cu₆Zn_xGa_y (Cu: Zn: Ga = 6: x: y, x + y = 4) catalysts with varying Ga contents were synthesized using the co-precipitation method based on unmodified Cu-ZnO catalysts. All prepared catalysts were tested for their activity in CO₂ hydrogenation to methanol and compared with Cu-ZnO catalysts without Ga. It was discovered that the addition of Ga generated spinel ZnGa₂O₄ crystals, creating oxygen vacancies on the catalyst surfaces. The spinel ZnGa₂O₄ crystals also altered the electronic interactions with Cu sites to some extent. By correlating the physicochemical properties of the catalysts with their catalytic activities, it was determined that the formation of spinel ZnGa₂O₄ in CuZnGa ternary composite oxide catalysts effectively promotes CO₂ hydrogenation to methanol. This work provides valuable insights for understanding and designing highly active catalysts for CO₂ hydrogenation to methanol.

2. Materials and Methods

2.1. Material Preparation

All chemicals that were used in this study were used as received without further purification. Copper nitrate trihydrate (Cu(NO₃)₂⋅3H₂O, 99.9 wt.%) was purchased from Thermo Fisher Scientific. Zinc nitrate hexahydrate (Zn(NO₃)₂⋅6H₂O, 99.9 wt.%) was purchased from Alfa Aesar (Shanghai, China) Chemical Science Co. Gallium nitrate hydrate (Ga(NO₃)₃⋅xH₂O, 99.9 wt.%) was purchased from Beijing Innochem Technology Co., Beijing China. Sodium carbonate (Na₂CO₃) was acquired from Sinopharm Chemical Reagent Co. Ltd., Shanghai, China. We made our own deionized water for the experiments.

2.2. Catalyst Preparation

All catalysts in this experiment were prepared by the co-precipitation method. The metal precursors were the corresponding metal nitrates: Cu(NO₃)₂⋅3H₂O, Zn(NO₃)₂⋅6H₂O, and Ga(NO₃)₃⋅xH₂O. Sodium carbonate solution was used as a precipitant, and the concentration of the solution was controlled to be 0.1 mol/L. In the case of Cu₆Zn₃Ga₁, 7.248 g of Cu(NO₃)₂⋅3H₂O, 4.4621 g of Zn(NO₃)₂⋅6H₂O, and 1.2787 g of Ga(NO₃)₃⋅xH₂O were weighed and mixed with 200 mL of deionized water. A small amount of Na₂CO₃ was also weighed and mixed with 150 mL of deionized water to keep the solution’s concentration at 0.1 mol/L. The catalysts were prepared in accordance with the following conditions, which were controlled to be 0.1 mol/L. Using a dispensing funnel, the base and salt solutions were added at the same time to a beaker with 200 mL of deionized water. The pH was kept at 7 during the titration, and the mixture was stirred while it was being added. The entire co-precipitation process was carried out in a thermostatic water bath at 70 °C, and after the precipitation was completed, the resulting suspension continued to be aged with stirring at this temperature for 2 h. After aging, it was pump-filtered and washed with large amounts of deionized water to remove the impurity Na⁺ and finally dried at 110 °C overnight. The obtained precursor was calcined in a muffle furnace at a ramp rate of 2 °C/min for 3 h at 500 °C to obtain the sample, noted as Cu₆Zn₃Ga₁. Cu₆Zn₄, Cu₆Zn_3.5Ga_0.5, and Cu₆Zn_2.5Ga_1.5 composite oxides were prepared by the same method.

2.3. Characterization

The crystal structure of various catalysts was tested by an X-ray powder diffractometer (XRD, Bruker D8 Advance). The scanning range was 5°–80°, and the scanning rate was 5 °/min. Scanning transmission electron microscopy (STEM) and energy-dispersive X-ray spectroscopy (EDS) analyses were carried out on a JEOL JEM-F200 filed-emission electron microscope with an acceleration voltage of 200 kV. X-ray photoelectron spectroscopy (XPS) was carried out on a Thermo Fisher Scientific (Waltham, MA, USA) K-Alpha+ apparatus equipped with a mono Al Kα radiation. The binding energy shift due to surface charging was adjusted to the standard of the C 1s line at 284.8 eV. Temperature-programmed reduction of H₂ (H₂-TPR) and temperature-programmed desorption of CO₂ (CO₂-TPD) were conducted on a Micromeritics Autochem II chemisorption analyzer. During TPR, the reduction profile of the catalyst was recorded from 50 to 800 °C (heating rate: 10 °C/min) in a 5% H₂/Ar (in volume, 30 mL/min) mixed gas. For TPD, the catalyst was pretreated at 400 °C with 5% H₂/Ar flow for 60 min and then cooled to 50 °C. Subsequently, it was switched to argon purge for 30 min, followed by the introduction of a 10% CO₂/Ar gas mixture adsorption for 60 min. Then, argon was switched, and the gas-phase H₂ was purged from the catalyst for 60 min. Finally, the desorption profile from 50 to 500 °C was recorded in the argon stream after waiting for the baseline to stabilize.

2.4. Catalytic Activity Test

All prepared catalysts were tested and evaluated for CO₂ hydrogenation catalytic activity in a high-pressure fixed-bed reactor. Typically, 0.24 g of catalyst was loaded into a quartz tube prior to reaction testing, and pure H₂ was introduced at a rate of 30 mL/min for 3 h at 400 °C for reduction. After the end of reduction and waiting for the reactor to cool down to room temperature, a reaction gas mixture with an H₂/CO₂ ratio of 3/1 was introduced into the reactor to start the experiment. The flow rate of all gases was controlled by a mass flow controller, and the line temperature was maintained at 150 °C to avoid condensation of the gas products. The products were analyzed using online gas chromatography (GC2060) equipped with a flame ionization detector (FID) and a thermal conductivity detector (TCD) for online detection. TCD used He as a carrier gas to analyze Ar, CO, and CO₂, and FID used N₂ as a carrier gas to analyze alcohols and other carbon-containing compounds, with 8% Ar as an internal standard. All test data were collected 2 h after the start of the reaction, and each catalyst sample was tested three times to ensure the reproducibility of the experiments.

3. Results and Discussion

3.1. Phases of Catalyst Samples

The XRD patterns of the catalyst phase structures prepared by the co-precipitation method are shown in Figure 1. The Cu₆Zn₄ catalyst exhibits characteristic peaks corresponding to ZnO (JCPDS 79-2205) and CuO (JCPDS 80-1916), with no additional crystalline phases detected. Due to the higher Cu content compared to Zn in this catalyst, the XRD pattern is dominated by the characteristic peaks of CuO. Upon the introduction of Ga, a ZnGa₂O₄ (JCPDS 38-1240) spinel structure appears [30], while some ZnO diffraction peaks disappear. The main diffraction peak of ZnGa₂O₄ is observed at 2θ = 35.7°, which partially overlaps with the primary diffraction peak of CuO. This may indicate the formation of a ZnGa₂O₄ spinel structure with a small amount of Ga and suggests that the ZnGa₂O₄ crystallite size is below the detection limit. The absence of ZnGa₂O₄ as a dominant diffraction peak also implies that the spinel phase is well-dispersed within the catalyst. Further increasing the Ga content results in no significant changes in the XRD patterns of Cu₆Zn₃Ga₁ and Cu₆Zn_2.5Ga_1.5, with CuO and ZnO remaining as the predominant phases. The simultaneous presence of CuO, ZnO, and ZnGa₂O₄ diffraction lines indicates the formation of a mixed crystalline phase, which is further confirmed by HRTEM analysis. However, regardless of the Ga content, no apparent structural changes to the Cu₆Zn₄ catalyst are observed in the XRD patterns.

To investigate the morphology and crystalline phase transition of the catalysts, TEM and HRTEM scanning were performed for Cu₆Zn₄ and Cu₆Zn₃Ga₁ catalysts. As shown in Figure 2a,b,d,e, both Cu₆Zn₄ and Cu₆Zn₃Ga₁ exhibit well-defined particle structures. In Cu₆Zn₄, the lattice spacing of 0.232 nm and 0.261 nm corresponds to the CuO (111) and ZnO (002) facets, respectively. The CuO (111) facet is also observed on Cu₆Zn₃Ga₁, while ZnO is present in the form of the (111) facet with a lattice spacing of 0.246 nm. Additionally, the ZnGa₂O₄ (222) facet with a lattice spacing of 0.241 nm is identified. However, the (222) facet is not the strongest characteristic diffraction peak, which is typically observed at the (311) facet. This could be due to the small amount of Ga added, which results in the formation of a partial ZnGa₂O₄ spinel phase that deviates from the standard stoichiometric ratio. Therefore, the observed (222) peak is not the dominant diffraction peak. EDX mapping analysis of both catalysts, shown in Figure 2c,f, reveals that the elements are uniformly distributed, indicating the high stability of the catalysts and the effectiveness of the sample preparation method.

3.2. Electronic Structure of Catalysts

The surface chemical environments of four elements (O, Cu, Zn, and Ga) were investigated using XPS. As shown in Figure 3a, the surface oxygen vacancies of the catalysts were analyzed using O 1s XPS spectra. Oxygen vacancies were quantified by deconvoluting the O 1s XPS spectra into three components: lattice oxygen (~530.0 eV), oxygen near surface vacancies (~531.0 eV), and hydroxyl groups (~532.0 eV) [31,32]. The introduction of Ga significantly increased the oxygen vacancies, with the Cu₆Zn₃Ga₁ catalyst showing the highest vacancy concentration (36.1%) at an optimal Ga content. However, the oxygen vacancy concentration decreased when excess Ga was added, suggesting that oxygen vacancies play a critical role in CO₂ adsorption and activation, thereby promoting CO₂ conversion [33]. Subsequently, XPS analysis of Cu, Zn, and Ga was performed. As shown in Figure 3b, the binding energies of Cu 2p_1/2 and Cu 2p_3/2 were approximately 953.22 eV and 933.3 eV, respectively. With increasing Ga content, the binding energies of Cu shifted first to higher and then to lower binding energies, indicating that Ga introduction alters the electronic state of Cu in the catalyst. Initially, Cu becomes electron-deficient, and later, it becomes electron-rich. In contrast, as shown in Figure 3c, Ga addition had minimal impact on the electronic structure of Zn, which remained in an electron-rich state. Figure 3d shows that with increasing Ga content, the electronic binding energy of Ga shifted to lower values, indicating that the surface became more electron-rich. These electronic interactions between Cu, Zn, and Ga significantly affect the adsorption and activation of CO₂ and H₂, thereby influencing CO₂ hydrogenation to methanol [34]. Notably, rather than a single electronic interaction between two elements, the combined electronic interactions of all three elements (Cu, Zn, and Ga) govern the CO₂ hydrogenation process. The changes in the electronic binding energies of these elements suggest that appropriate electronic interactions favor methanol production, whereas too-strong or too-weak interactions hinder the reaction.

3.3. Adsorption and Reduction Behavior on Catalyst Surfaces

Surface adsorption and desorption of CO₂ play a crucial role in CO₂ hydrogenation to methanol [35]. As shown in Figure 4a, the Cu₆Zn_xGa_y catalysts, after the addition of Ga, all exhibit a low-temperature desorption peak around 105 °C, corresponding to physical desorption. There is no significant difference in the peak position across the catalysts, but the intensity and area of the peaks increase with the further addition of Ga. The Cu₆Zn₄ catalyst without the addition of Ga shows no low-temperature desorption peaks and only exhibits a large desorption peak near 300 °C, corresponding to chemical desorption. This suggests that modifying the Cu₆Zn₄ catalyst by adding Ga enhances the adsorption of CO₂.

As shown in Figure 4b, H₂-TPR was performed to study the reduction behavior of the catalysts. All the catalysts show a large reduction peak centered around 300 °C, with the largest peak area observed for Cu₆Zn₄. Upon the addition of Ga, the reduction peak temperature shifted to lower values, indicating that surface Ga reduces the reduction temperature and effectively promotes catalyst reduction.

3.4. Catalytic Performance

The methanol synthesis reaction (CO₂ + 3H₂ → CH₃OH + H₂O) and the RWGS reaction (CO₂ + H₂ → CO + H₂O) are competing reactions that usually occur simultaneously [36]. Therefore, CO is considered a byproduct in activity evaluation. We conducted a detailed study on the Ga ratio, as shown in Figure 5a–d. The Cu₆Zn₄ catalyst without Ga catalyzed the generation of CO, which is undesired. However, upon adding Ga, the overall catalytic activity and selectivity followed a “volcano-type” trend with increasing Ga content, showing improved CO₂ conversion and CH₃OH selectivity compared to Cu₆Zn₄. The highest activity and selectivity were achieved when the Ga-to-Cu ratio formed Cu₆Zn₃Ga₁, as shown in Tables S1–S4.

The activity of four catalysts with different Ga ratios at varying temperatures was tested. As expected, CO₂ conversion increased with temperature, while CH₃OH selectivity decreased to some extent. This behavior aligns with thermodynamic principles, where increased temperature promotes the forward reaction of CO₂ hydrogenation to methanol. At 240 °C, we compared the catalytic activity of four catalysts. The Cu₆Zn₄ catalyst showed a CO₂ conversion of 9.83%, with CO selectivity reaching 62.2%. When a small amount of Ga was added, Cu₆Zn_3.5Ga_0.5 exhibited a CO₂ conversion of 11.1% and a CH₃OH selectivity of 58.8%. Increasing the Ga content further to form Cu₆Zn₃Ga₁ led to a CO₂ conversion of 12.5% and the highest CH₃OH selectivity of 59.2%. However, when the Ga content exceeded this amount, as in Cu₆Zn_2.5Ga_1.5, the catalytic performance decreased. Although CH₃OH selectivity reached 61.3%, the CO₂ conversion dropped significantly to 8.9%, far below that of Cu₆Zn₃Ga₁. This indicates that an optimal Ga ratio is essential for maximizing CO₂ conversion and CH₃OH selectivity. Therefore, Cu₆Zn₃Ga₁ demonstrates the best performance in terms of both CO₂ conversion and CH₃OH selectivity with optimal Ga incorporation.

The reaction process conditions were explored, as shown in Figure 6a,b. First, we investigated the effect of gas hourly space velocity (GHSV) on catalytic performance. GHSV is a crucial factor influencing the residence time of reactant gas molecules on the catalyst surface and the extent of the catalytic reaction. The catalytic activity of Cu₆Zn₃Ga₁ catalysts varied under different GHSVs. As the GHSV increased from 4500 mL∙g_cat⁻¹∙h⁻¹ to 9000 mL∙g_cat⁻¹∙h⁻¹, CO₂ conversion decreased. Similarly, CO selectivity declined with increasing GHSV, while CH₃OH selectivity exhibited the opposite trend. Considering both CO₂ conversion and CH₃OH selectivity, the highest methanol yield was achieved at a GHSV of 7500 mL∙g_cat⁻¹∙h⁻¹. Both higher and lower gas velocities resulted in reduced CH₃OH yield. Therefore, the optimal GHSV for this experiment was determined to be 7500 mL∙g_cat⁻¹∙h⁻¹.

Pressure, a key factor influencing reaction kinetics, also plays a significant role in catalytic performance [36]. In this study, the optimal reaction pressure was 4 MPa. When the pressure was lower than 4 MPa, both CO₂ conversion and CH₃OH selectivity were less favorable compared to the results at 4 MPa. Increasing the pressure beyond 4 MPa led to a slight decrease in CH₃OH selectivity, even though CO₂ conversion reached 13.7% at 5 MPa, where the CH₃OH selectivity dropped to 57.6%. Given the potential impact of high pressure on reaction equipment and the limited improvement in catalytic performance with higher pressure, we selected 4 MPa as the optimal reaction pressure. At the same time, we summarized the performance of the four catalysts under the best test conditions and found that Cu₆Zn₃Ga₁, the catalyst with the highest TOF, demonstrated the best performance (

Table 1. Catalytic properties of the Cu₆Zn_xGa_y catalyst.

Catalysts	CO2 Conversion (%)	CH3OH Selectivity (%)	TOFCu (10−3s−1)
Cu6Zn4	12.0	32.5	6.4
Cu6Zn3.5Ga0.5	14.5	56.4	9.7
Cu6Zn3Ga1	15.8	56.5	10.2
Cu6Zn2.5Ga1.5	11.2	58.2	8.1

).

After optimizing these conditions, a stability test under optimal conditions—240 °C, reaction pressure of 4 MPa, and air velocity of 7500 mL∙g_cat⁻¹∙h⁻¹—was conducted, as shown in Figure 7. The catalytic performance showed only a slight decrease over the 90 h test, with the best performance occurring approximately 5 h after the reaction began. Overall, the Cu₆Zn₃Ga₁ composite oxide catalyst demonstrated high stability.

3.5. Discussion

The enhancement in catalytic activity observed upon Ga incorporation can be attributed to both structural and electronic modifications introduced by the spinel ZnGa₂O₄ phase. O 1s XPS analysis confirms an increase in surface oxygen vacancies, which facilitates CO₂ adsorption and activation. At the same time, changes in the Cu 2p binding energies and the downward shift in H₂-TPR reduction temperatures indicate electronic interactions between Ga species and Cu, promoting hydrogen activation and spillover. These dual effects—oxygen vacancy generation and electronic modulation—are likely to operate synergistically, although separating their individual contributions experimentally remains a challenge. Furthermore, the volcano-shaped trend in catalytic performance with increasing Ga content can be explained by an optimal balance between these factors. Excess Ga may lead to phase segregation or structural changes that disrupt the active Cu–Zn–Ga interface, as suggested by XRD and catalytic data, thereby lowering activity. Thus, integrating physical structure, electronic environment, and performance results supports the conclusion that Ga incorporation enhances CO₂ hydrogenation to methanol through a combined mechanism.

4. Conclusions

In this work, we prepared spinel ZnGa₂O₄-modified Cu-ZnO catalysts using the co-precipitation method and evaluated their performance and stability for CO₂ hydrogenation to methanol. The results demonstrated that the introduction of Ga facilitated the formation of the spinel crystal structure, enhancing CH₃OH production and increasing CO₂ conversion compared to the undoped Cu-ZnO catalyst. Characterization studies revealed that the incorporation of Ga promoted the formation of oxygen vacancies on the catalyst surface, which enhanced CO₂ adsorption and the activity of carbon dioxide and increased the conversion of CO₂. Additionally, spinel ZnGa₂O₄ altered the electron interactions, effectively driving the CO₂ hydrogenation process to methanol, thereby further improving the selectivity of CH₃OH. The catalysts prepared using this method exhibited excellent performance and stability during CO₂ hydrogenation to methanol.