Size Effect of Cu Particles on Interface Formation in Cu/ZnO Catalysts for Methanol Synthesis

Abstract

Cu/ZnO/Al₂O₃ catalysts are extensively utilized in methanol synthesis from CO and CO₂, which is a vital industrial process and a promising strategy for mitigating CO₂ emissions when renewable green hydrogen is employed. Despite the considerable efforts to study CO₂ hydrogenation over Cu/ZnO, understanding the structure of active sites on Cu/ZnO has remained a major challenge. We studied a series of Cu/ZnO catalysts with various Cu particle sizes and found a volcano-like pattern in methanol selectivity with respect to the Cu particle size. TEM, XPS, and TPD measurements demonstrated the migration of ZnO_x species onto the Cu particle surface and showed a correlation between the ZnO_x-Cu interface and methanol yield. The size of supported Cu particles affects the migration of Zn species onto Cu particle surfaces. Our study has thus explicated the role of the ZnO_x-Cu interface in catalyzing CO₂ hydrogenation to methanol.

1. Introduction

The transformation of CO₂ into value-added chemicals presents a promising approach to achieving a carbon-neutral process. Methanol (CH₃OH) serves as an energy storage agent and is frequently employed as a chemical feedstock to produce other chemicals, solvents, and gasoline additives [1,2]. The hydrogenation of CO₂ for CH₃OH production is considered an effective strategy for CO₂ utilization [3], and Cu/ZnO/Al₂O₃ (CZA) catalysts are widely used in industrial methanol production [4], with Cu/ZnO proposed as catalytically active components. So far, the design of efficient catalysts for CO₂ hydrogenation has aimed to find appropriate configurations to improve the dissociative adsorption of CO₂ and the selectivity towards methanol. Optimizing the interfacial sites between Cu and ZnO has shown promise in enhancing methanol synthesis [5,6,7].

The structure of active sites in CZA catalysts varies significantly based on the size of the supported metal particles, likely due to interfacial evolution and migration phenomena [8,9]. Regarding the synergistic effects between Cu and ZnO, previous studies on the Cu/ZnO/Al₂O₃ catalysts have shown clear evidence in HRTEM images for the formation of disordered ZnO_x overlayer [10] or metastable “graphite-like” ZnO layers [11] during the reductive activation of catalysts. In addition, Kattel et al. and Chen et al. reported that the hydrogenation of CO₂ to methanol on Cu/ZnO catalysts might mainly take place at the interface between ZnO and Cu [5,12]. Meanwhile, CO₂ hydrogenation is a reaction sensitive to catalyst structure [13], which is critically influenced by the size of Cu particles and the migration ability of zinc oxide, leading to the formation of active interfacial sites.

Regarding the size effect of Cu particles in methanol synthesis, recent studies have presented controversial results. For instance, Arena et al. reported a volcano-like trend in the metal surface area of Cu-ZnO/ZrO₂ catalysts with respect to the Zn/Cu ratio, favoring methanol production with poorly dispersed Cu particles [14]. Conversely, Natesakhawat et al. found that incorporating Ga₂O₃ and Y₂O₃ into CuZnZr catalysts increased the specific surface area of Cu, resulting in higher methanol synthesis rates due to the presence of more coordinatively unsaturated atoms in smaller Cu particles [15]. The controversy could be complicated by the influence of supports [4] and additives [16] in catalyzing the methanol synthesis process. Thus, it is necessary to examine the role of differently sized Cu nanoparticles (NPs) in CO₂ hydrogenation with simpler Cu/ZnO model catalysts. While the influence of ZnO migration onto Cu particles has been considered, investigating the relationship between ZnO coating capacity and Cu particle size is still warranted.

In this study, we synthesized Cu NPs with varying sizes on ZnO by adjusting the Cu loading using the co-precipitation method. Through a combination of characterization techniques and reaction measurements, we established a correlation between the catalyst structure and its catalytic performance in methanol synthesis. Our results suggested a strong relationship between the production rate of methanol and the interfacial area between Cu and ZnO. This interfacial area was attributed to the migration of zinc species, forming a coating layer on the Cu particle surface during the reduction process. The migration phenomenon was significantly influenced by metal–support interactions (MSIs) and particle size. Our study provides valuable insights into the formation of active sites in methanol synthesis and offers strategies for enhancing Cu/ZnO-based catalysts for methanol production.

2. Results

2.1. Structural Characterization of As-Prepared Cu/ZnO Catalyst

Cu/ZnO catalysts (CZ catalysts) with different Cu mass loadings of 0.5, 2, and 5 wt% were synthesized and named 0.5-CZ, 2-CZ, and 5-CZ, respectively. XRD analysis on the as-prepared 0.5-CZ, 2-CZ, and 5-CZ catalysts (Figure 1a) showed peaks corresponding to the (111) facet of CuO (JCPDS card no. 45-0937) [17], and the (100), (002), and (101) facets of ZnO (JCPDS card no. 36-1451) [18]. The surface chemical state of 0.5-CZ, 2-CZ, and 5-CZ catalysts was also measured using XPS. The Zn 2p_3/2 peak at 1021.6 eV suggested the chemical state of Zn species was mainly ZnO [19] (Figure S1a). The O 1s spectra showed the presence of carbonate or hydroxylate species on the catalyst surface after calcination [20] (Figure S1b). As the Cu loading increased, the diffraction peaks of ZnO exhibited a shift towards higher angles, indicating lattice contraction, which could be attributed to the incorporation of Cu into the ZnO lattice. Additionally, the peak width of CuO(111) and ZnO(110) decreased with increasing Cu loading, suggesting the growth of Cu and ZnO crystals and implying that the Cu loading influenced the stability of ZnO.

Table 1. Structural properties of fresh catalysts.

Sample	Pore Volume a/(cm3·g−1)	SBET b/(m2·g−1)	Pore Size/(nm)	dCuO c/(nm)	dZnO d/(nm)
0.5-CZ	0.282	42.6	25.8	8.4	11.3
2-CZ	0.226	33.6	17.3	15.0	17.7
5-CZ	0.127	25.4	8.7	20.1	20.8

^a Total pore volume and pore size calculated using Barrett–Joyner–Halenda (BJH) model; ^b Surface area using Brunauer–Emmett–Teller (BET) method; ^c,d CuO and ZnO particle size using the Scherrer equation calculated from the X-ray diffraction of CuO (111) (2θ = 38.7°) and ZnO (110) (2θ = 56.6°).

presented the average particle size of CuO, calculated from the diffraction signal of CuO(111) using the Scherrer equation; the BET surface area and total pore volume of the calcined samples are also summarized. In agreement with XRD results, the increased Cu loading is accompanied by a decrease in the specific surface area of CZ catalysts. In addition, larger Cu metal particles formed on the surface blocked the ZnO pores according to the pore size data [19].

The metal–support interaction and reducibility of Cu species in the catalysts were investigated using TPR analysis (Figure 1b). Since ZnO is difficult to reduce at below 700 °C [21,22], the observed peaks in the graph are attributed to the reduction of copper species. Peaks in the low-temperature region correspond to the reduction in highly dispersed CuO, and the peaks in the high-temperature region correspond to the reduction in the bulk phase of CuO [23]. The reduction peak temperature decreases as the Cu loading increase. In H₂-TPR, the 0.5-CZ catalyst exhibited reduction peaks with the highest peak temperatures, which could be attributed to a stronger interaction between Cu and ZnO at lower Cu loading [24]. Since the size of Cu particles increased with Cu loadings, the larger Cu nanoparticles would show weaker interactions with ZnO, which facilitated the reduction process [25]. Consistent with XRD measurements, Figure 1c shows XPS of Cu 2p, from which the peaks at 962.4 and 933.6 eV are found to correspond to the Cu 2p_1/2 and Cu 2p_3/2 together with intense satellite peaks at 942.4 and 944.6 eV, these two peaks overlapping, and XPS Cu 2p spectra also showed that Cu species on as-prepared CZ catalysts are mainly in the state of Cu²⁺, as evidenced by the presence of shake-up satellite peaks [26,27]. Accordingly, SEM images of as-prepared CZ catalysts (Figure S2) also showed particle aggregation, with larger particles formed as the Cu loading increased.

2.2. Structural Characterization of Cu/ZnO Catalyst after H₂ Reduction

The absorption edge observed in the Cu K-edge XANES corresponds to the primary 1s → 4p transition. Cu⁰ and Cu⁺, both having a d⁰ configuration, lack a hole in the 3d orbital, while Cu²⁺ adopts a d⁹ configuration. Consequently, Cu²⁺ exhibits a weak pre-edge peak, indicating the allowed 1s → 3d transition and serving as a significant characteristic for divalent copper. Figure 2a showed that the Cu K-edge XANES spectrum of the as-prepared sample exhibited a weak pre-edge peak around 8977.6 eV, similar to that of CuO reference [28] and indicating a divalent oxidation state for the as-prepared catalyst [29,30]. The XANES spectrum of Cu foil exhibits the edge absorption (the first inflexion point) at 8979 eV together with a well-resolved doublet in the post-edge region [31]. In the Cu₂O reference compound, a distinct peak at approximately 8981.4 eV is assigned to the 1s → 4p_xy transition, followed by another peak at around 8995.7 eV, which arises from the 1s → 4p_z transition due to the influence of ligand field effect [32].

In the H₂ gas, XANES spectra of the 2-CZ catalyst gave a pre-edge feature at 8977.6 eV [33], suggesting the presence of ionic Cu at below 150 °C (Figure 2a). After H₂ reduction at 250 °C, XANES features of the 2-CZ catalyst resemble that of the Cu foil, implying the reduction of Cu²⁺ into metallic Cu⁰. LCF analysis in Figure S7 plotted the changes in the relative concentration of different Cu species, illustrating the reduction of CZ catalysts in H₂. Note that, the small fraction of ionic Cu species might be due to the limited reduction time in XAS measurements. As shown in Figure 2b, patterns at 2θ of 43.3° and 50.4°, assigned to Cu(111) and Cu(200) phases (JCPDS card no. 04-0836), XRD patterns of CZ catalysts after reduction showed also the emergence of Cu(111) and Cu(200) phases, indicating the reduction of CuO to Cu following H₂ treatment.

Since XRD peak intensity is influenced by a number of factors, such as reduced crystallinity and crystalline size effects, we have normalized ZnO(002) peak intensity relative to the strongest peak intensity according to previous reports [34,35]. A change in the peak intensity of ZnO(002) indicated that a strong interaction between the polar ZnO(002) facet and Cu induced significant morphological changes in ZnO during CO₂ hydrogenation [36,37,38]. HRTEM images on the Cu/ZnO/Al₂O₃ catalysts have demonstrated the presence of disordered ZnO_x overlayer on Cu nanoparticles after reductive activation [10], which could reduce the relative intensity of ZnO(002) diffraction peaks. Consistently, our HRTEM studies on the reduced Cu/ZnO catalysts showed also the covering of Cu nanoparticles by ZnO_x overlayers. We have thus attributed the decrease in the relative intensity of ZnO(002) to the reduction of ZnO(002) and the migration of ZnO_x species onto Cu nanoparticles forming overlayers. Among the three catalysts, Figure 2c suggested a most significant reduction in ZnO(002) peak intensity of 2-CZ due to the substantial formation and migration of ZnO_x species onto Cu particles after H₂ reduction, in agreement with our HRTEM measurements.

HR-TEM was further employed to study the CZ catalysts (Figure 3), which showed the presence of ZnO(002) and Cu(111) phases, with lattice spacings of 0.26 nm and 0.21 nm [39], respectively. Fast Fourier transform (FFT) of the bright-field HR-TEM images showed diffraction spots linked to various crystal planes of ZnO and Cu.

With increased Cu loading, the brightness of diffraction spots corresponding to the Cu(111) and ZnO(002) planes markedly enhanced, indicating the growth of Cu(111) and ZnO(002) accompanying the growth of Cu and ZnO particles. By selecting the specific region of crystal diffraction spots in the FFT images, spatial chemical distributions (in dark-field TEM images) of Cu and ZnO species for CZ catalysts were obtained (see details in Supplementary Materials, Figures S3–S6). Figure 4 showed that reduced 0.5-CZ and 5-CZ catalysts display smaller contact regions between Cu and ZnO than those on the 2-CZ catalyst, suggesting the migration behavior of ZnO layers is dependent on the size of Cu particles during the reduction process (Figure S6).

Quasi-in situ XPS experiments on the CZ catalysts after H₂ reduction showed the binding energy (BE) of Cu 2p_3/2 peak at 932.5 eV (Figure 5a), indicating the presence of Cu⁺ or Cu⁰ species [22,40]. The Zn 2p_3/2 peak (Figure S1c) at around 1021.6 eV corresponds to Zn²⁺ species [19]. The Zn/Cu atomic ratios from XPS measurements increased significantly after H₂ reduction compared to as-prepared CZ catalysts. This implies that the coating of Cu particles by ZnO layers increased the surface density of ZnO-Cu interfacial sites. We also estimated the surface density of oxygen vacancies (O_V) by fitting the O 1s peak with three components (O_α, O_β, and O_γ) and calculating the peak intensity ratio of I_Oβ/(I_Oα + I_Oβ + I_Oγ) [41]. The three peaks of O_α, O_β, and O_γ are located at 530.3 eV, 531.6 eV, and 532.3 eV, which could be attributed to lattice oxygen (O_L), oxygen defect (O_V), and weakly bound species (O_C, such as hydroxyl groups or carbonates), respectively (Figure 5b) [20,42,43]. O_V was found to increase significantly for all catalysts after H₂ reduction, with the 2-CZ catalyst exhibiting a higher concentration (~20.4%) than the 0.5-CZ (~16.5%) and 5-CZ (~17.7%) catalysts.

Furthermore, the atomic ratios of Zn/Cu obtained via XPS measurements increased significantly after H₂ reduction compared to as-prepared CZ catalysts (Figure 5c), and the increase in Zn/Cu ratio was generally observed for all three catalysts after reductive activation. Although a number of factors, such as the sintering of Cu nanoparticles, the penetration of Cu particles into the ZnO bulk, and the encapsulation of Cu particles by ZnO_x layers could contribute to the changes in the Zn/Cu ratio, we found the encapsulation of Cu particles by ZnO_x overlayers most probable, combining XPS, XRD and HRTEM measurements. XRD patterns of the three catalysts after reduction exhibited no significant change in peak intensities of Cu(111) compared to as-prepared catalysts (Figure S8), suggesting that the agglomeration of Cu particles was not significant. The penetration of Cu particles into bulk ZnO was also not obvious since the process would lead to the decomposition of Cu particles and homogeneous distribution of Cu ions within the ZnO bulk, which was not evident from XRD and HRTEM measurements. Rather, the encapsulation of Cu nanoparticles by ZnO_x overlayers, as evidenced from HRTEM images, could prevent the agglomeration of Cu particles, consistent with XRD results. Figure 5c showed the evolution of the surface atomic ratio of Zn/Cu, as calculated from the XPS Zn 2p and Cu 2p spectra on the CZ catalysts after H₂ reduction. A significant increase was observed in the 2-CZ catalyst, indicating a distinct migration of ZnO_x species onto Cu particles and a simultaneous increase in O_V. In contrast, the 5-CZ catalyst exhibited a limited increase in the Zn/Cu atomic ratio, implying that ZnO_x migration onto Cu particles was hindered by the larger particle size of Cu. The increase in the Zn/Cu atomic ratio in the 0.5-CZ catalyst was also less than that in the 2-CZ catalyst, indicating the migration of ZnO_x species is strongly dependent on the Cu particle size.

2.3. Catalytic Activity and CO₂ Adsorption Ability of Cu-ZnO Catalysts

The steady-state catalytic activities of the CZ catalysts were measured in a fixed-bed reactor, and the 2-CZ catalyst displayed the highest reaction activity across the entire temperature range (Figure 6a and Figure S9). As the temperature increased, the conversion of CO₂ gradually rose while the selectivity towards methanol significantly decreased. The conversion of CO₂ increased with increasing Cu loading, while the selectivity towards methanol behaved differently and showed a volcano shape with regard to Cu loading. CO production increased with the rise of reaction temperature due to the competitive reaction of the reverse water–gas shift reaction. The differences between the three catalysts used in this study in reaction selectivity could originate from the various distribution of Cu sites and Cu-ZnO interfacial sites, in addition to their differences in structure and geometry. From a mass-specific reaction rate perspective (Figure 6b), the 2-CZ catalyst exhibited the highest specific production rate of CH₃OH compared to the other catalysts within the temperature range of 200 °C to 260 °C. Considering the loading of Cu, the 2-CZ catalyst was found more catalytically efficient than other CZ catalysts, which could be attributed to the more significant migration of ZnO_x species onto the surface of Cu particles in the 2-CZ catalyst. The formation of ZnO_x-Cu interfacial sites greatly enhanced the selectivity towards methanol.

In order to gain further insights into the superior performance of the 2-CZ catalyst, CO₂-TPD was employed to investigate the CO₂ adsorption properties of the CZ catalysts, as depicted in Figure 6c. The TPD profiles can be deconvoluted into several peaks, indicating the desorption from different types of basic sites. Peaks α, β, and γ correspond to CO₂ desorption from weakly basic sites associated with the hydroxyl group, moderately basic sites attributed to Zn–O pairs, and strongly basic sites linked to low-coordinated oxygen atoms [44], respectively. Among these, moderately basic sites have been suggested as active sites for CO₂ hydrogenation [45]. Based on the TPD profiles,

Table 2. The amount and distribution of basic sites of Cu/ZnO catalysts.

Sample	Total Amount of Basic Sites/(μmol∙g–1)		Peak Position/(°C)
	β	γ	β1	β2	β3
0.5-CZ	4.7	6.4	289	420	516
2-CZ	8.7	3.8	279	413	515
5-CZ	8.1	4.9	286	376	498

presents the quantitative measurements of basic sites on the CZ catalysts. The CO₂ desorption peaks in the range of 350 °C~600 °C shift to lower temperature with increasing Cu loading, probably due to the interface structural changes of the catalysts. The β desorption peak was found to be the largest for the 2-CZ catalyst, and the β₂ and β₃ peaks could be attributed to CO₂ desorption at the interfacial sites and ZnO overlayers. The peak area of β₂ and β₃ can be, therefore, regarded to quantify the enhancement of CO₂ adsorption. From above, a correlation between the migration of ZnO_x species onto Cu particles, the adsorption property of CO₂, and methanol selectivity could be built.

From above, it is evident that ZnO_x species migrated onto the surface of Cu nanoparticles, forming overlayers partially covering the Cu surface during the reduction process. This migration led to the formation of ZnO_x-Cu interfaces, which played a crucial role in significantly enhancing the selectivity for methanol production. Notably, the size of supported Cu particles exerted an influence on the migration of ZnO_x species onto the surfaces of Cu particles. Moreover, a positive correlation was observed between the selectivity of CH₃OH and the sum of β₂ and β₃ peaks. This correlation implies that the ZnO_x-Cu interfaces could serve as active sites in the CO₂ hydrogenation to methanol reaction, thereby enhancing the adsorption and hydrogenation of CO₂, as depicted in Figure 7.

3. Materials and Methods

3.1. Chemicals and Starting Materials

Zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O, AR, CAS 10196-18-6), copper(II) nitrate trihydrate (Cu(NO₃)₂·3H₂O, AR, CAS 10031-43-3), deionized water (DI water) and sodium carbonate (Na₂CO₃, AR, CAS 497-19-8). All the starting materials and chemicals were used without further purification.

3.2. Preparation of Catalysts

Cu/ZnO powder catalysts with Cu loading of 0.5, 2, and 5 wt% (feed ratio) were prepared using a co-precipitation method. A total of 1.0 g of Zn(NO₃)₂·6H₂O (AR) and corresponding Cu(NO₃)₂·3H₂O (AR) were dissolved in 150 mL of deionized water to obtain the precursor solution. At room temperature, saturated Na₂CO₃ (AR) solution was added to the solution at a rate of 3 mL/min with vigorous stirring to adjust the pH to 9.5. The solution was stirred for another 3 h. The products were collected with vacuum filtering with DI water and washed 3 times. After washing and drying, the catalysts were calcined in air atmosphere at 300 °C for 2 h.

Cu/ZnO catalysts with different Cu loadings of 0.5, 2, and 5 wt% were synthesized according to the method prepared above, named 0.5-CZ, 2-CZ, and 5-CZ, respectively.

3.3. Catalyst Characterization

The N₂ adsorption measurements were carried out at 77 K using an Autosorb-iQ-MP-AG instrument (Autosorb-iQ, Quantachrome, Boynton Beach, FL, USA). Then, 100 mg sample was degassed in vacuum (<0.1 Torr) for 8 h at 200 °C prior to analysis. The surface area, average pore size, and total pore volume were determined with the Brunauer–Emmett–Teller (BET) method and Barrett–Joyner–Halenda (BJH) model, respectively.

The X-ray diffraction (XRD, D8 ADVANCE, Bruker, Mannheim, Germany) pattern was operated at 40 kV and 40 mA with a Cu Kα radiation source (λ = 0.15418 nm). XRD patterns were collected with 2θ value from 10° to 90° at a scanning rate of 0.02°/s. Identification of the phases was carried out using the ICDD-JCPDS database. The Scherrer equation was applied to calculate the grain size.

Inductively coupled plasma optical emission spectrometry (ICP-AES, Optima 2000 DV, Perkin Elmer, Waltham, MA, USA) was employed to measure the total metal elements composition of the catalysts. The samples were dissolved using concentrated nitric acid prior to ICP-AES analysis. Each sample was measured at least three times to acquire the average value.

X-ray photoelectron spectroscopy (XPS, ESCALAB 250Xi, Thermo Fisher Scientific, Waltham, MA, USA) analysis was performed with an Al Kα Anode (1486.6 eV) and a spot size of 500 μm. All the spectra were fitted via Casa XPS Software using a Gaussian–Lorentzian spectral line shape and Shirley-type background. The surface molar ratios of catalysts were quantified by O 1s, C 1s, Cu 2p_3/2, and Zn 2p_3/2 peak area with inelastic mean free path (IMFP). For XPS measurements, the samples were sealed and transferred with a pure N₂-protected chamber to avoid surface chemical state changes in the air after the reaction. Raw XPS data were used with no preliminary smoothing. Gaussian–Lorentzian product functions were used to simulate the line shapes of the fitting components. Atomic ratios were computed from peak intensity ratios normalized with atomic sensitivity factors [46].

High-resolution TEM (HRTEM, JEOL-2100Plus, JEOL, Tokyo, Japan) images were recorded on a JEOL-2100Plus operated at an accelerating voltage of 200 kV. Cu and ZnO species in HRTEM images were distinguished using lattice spacing analysis. By performing FFT transformation on HRTEM images and applying Fourier filters on the lattice, we could locate the filtered lattice fringe with their corresponding locations in HRTEM images after inversion. As such, the distribution of Cu and ZnO species could be identified.

SEM images were recorded using a JSM-IT500 HR/LA (JEOL, Tokyo, Japan) operated at an accelerating voltage of 10 kV.

The reduction evolution profiles of catalysts were conducted on a Quantachrome Chemstar TPx Instrument (Quantachrome, Boynton Beach, FL, USA) equipped with a thermal conductivity detector (TCD) to detect H₂ consumption. Then, 100 mg of catalysts were placed in a U-type quartz tube and were heated (5 °C/min) from room temperature to 800 °C in 10% H₂/Ar (30 mL/min). Fresh samples were pretreated in flowing Ar (30 mL/min) at 300 °C for 1 h to remove absorbed impurities prior to tests.

The temperature-programmed CO₂ desorption (CO₂-TPD) experiment was also conducted on a Quanta Chrome Chemstar TPx Instrument. Then, 150 mg of sample was pretreated at 300 °C for 2 h in a flow of 10% H₂/Ar (20 mL/min). After cooling down to room temperature and saturating CO₂ adsorption in flowing 5% CO₂/He (20 mL/min) for 1 h, the sample was purged in pure He for 30 min and raised to 800 °C. CO₂ desorption was monitored using a TCD detector.

The in situ energy dispersive X-ray absorption (In situ XAS) spectroscopy measurements on catalysts were performed on the BL-05U Beamline of Shanghai Synchrotron Radiation Facility (SSRF), operated at 3.5 GeV with a constant current of 220 mA. The spectra were recorded in a transmission mode with a curved crystal-based Si (111) Bragg polychromator with a XIMEA camera. The energy resolution is 0.14 eV. Cu K-edge (8979 eV) X-ray absorption near-edge structure (XANES) spectra for samples 2-CZ and 5-CZ were collected by heating the samples from room temperature to 250 °C in pure H₂. The energy was calibrated using measuring X-ray absorption spectrum of Cu metal foil and by assigning the first inflection point in the rising portion of the absorption spectra as 8979 eV. The obtained data were analyzed using IFEFFIT suite of software programs [47].

3.4. Catalyst Characterization

Catalytic activity tests were conducted using a fixed-bed reactor with a quartz tube (inner diameter: 8 mm). Then, 500 mg sample was pretreated in a H₂ flow (1 bar, 30 mL/min) at 250 °C for 2 h and subsequently cooled to ambient temperature under N₂ flow (30 mL/min). The reaction gas was a mixture of 24% CO₂, 72% H₂, and 4% N₂ as an internal standard. The reaction was carried out at 200/220/240/260 °C, keeping thermal stability for 3 h in each temperature phase to achieve steady-state reactivity. During the catalytic evaluation process, when reaching the desired temperatures, data were collected at least four times, and the average data were calculated. Product analysis was performed using an online gas chromatograph (Agilent GC-8890) outfitted with a thermal conductivity detector (TCD) and a flame ionization detector (FID). The CO₂ conversion (denoted as X_CO2) was calculated according to an internal standard method (4% N₂ as an internal standard), assuming that the amount of N₂ remained constant in the reaction process. The equation is as follows:

$$X_{{\mathrm{C}\mathrm{O}}_2}\left(\mathrm{\%}\right)=\left(\frac{\frac{{{A^{in}}_{\mathrm{C}\mathrm{O}}}_2 }{{{A^{in}}_{\mathrm{N}}}_2 }-\frac{{{A^{out}}_{\mathrm{C}\mathrm{O}}}_2 }{{{A^{out}}_{\mathrm{N}}}_2 }}{\frac{{{A^{in}}_{\mathrm{C}\mathrm{O}}}_2 }{{{A^{in}}_{\mathrm{N}}}_2 }}\right)\times 100=\left(1-\frac{{{A^{in}}_{\mathrm{N}}}_2{{A^{out}}_{\mathrm{C}\mathrm{O}}}_2}{{{A^{in}}_{\mathrm{C}\mathrm{O}}}_2{{A^{out}}_{\mathrm{N}}}_2}\right)\times 100$$

(1)

$$S_i\left(\mathrm{\%}\right)=\frac{f_i{A^{out}}_i}{\sum f_i{A^{out}}_i}\times 100$$

(2)

where A represents the chromatographic peak area of the corresponding component, f is the corresponding molar correction factor of the component, and i represents products such as CH₃OH, CO, and CH₄.

The unit mass formation rate of CH₃OH and CO R_w (μmol·g_cat⁻¹·s⁻¹) is calculated as follows:

$$R_{w \left(\mathrm{C}\mathrm{O} or \mathrm{C}\mathrm{H}3\mathrm{O}\mathrm{H}\right)}=\frac{X_{{\mathrm{C}\mathrm{O}}_2}\times F_{{\mathrm{C}\mathrm{O}}_2}\times S_{\mathrm{C}\mathrm{O} or \mathrm{C}\mathrm{H}3\mathrm{O}\mathrm{H}}}{m}$$

(3)

$X_{{\mathrm{C}\mathrm{O}}_2}$ is the conversion rate of CO₂ (%), $F_{{\mathrm{C}\mathrm{O}}_2}$ is the molar flow rate of CO₂ (mol·s⁻¹), $S_{\mathrm{C}\mathrm{O} or \mathrm{C}\mathrm{H}3\mathrm{O}\mathrm{H}}$ is the selectivity of CO or CH₃OH (%), and m is the catalyst weight (g).

4. Conclusions

In summary, we have utilized the coprecipitation method to prepare CZ catalysts with varying amounts of Cu. By inducing ZnO_x migration onto Cu particles via H₂ reduction treatment, the CZ catalysts showed a significant increase in ZnO_x-Cu interfacial sites. XRD, TEM, and TPD measurements confirmed the dynamic evolution of ZnO_x migration onto Cu particles, which were found dependent on Cu particle sizes. Our results showed a volcano-like relationship between Cu particle size and methanol selectivity, highlighting the significant role played by the Cu/ZnO interface in the CO₂ hydrogenation-to-methanol reaction. Our study has thus offered valuable insights into understanding the fundamental size effect in catalysts.