Highly Efficient Conversion of Methane to Methanol on Fe-Cu/ZSM-5 Under Mild Conditions: Effective Utilization of Free Radicals by Favorable Valence Ratios

Abstract

The selective oxidation of methane to methanol under mild conditions remains a significant challenge due to its stable C-H bond and the propensity for overoxidation of products. Herein, we investigated the Fe- and Cu-modified ZSM-5 catalysts using H₂O₂ as an oxidant for the selective oxidation of methane. It was found that the Fe/Cu ratio had a great impact on methanol yield. The Fe₃Cu₁ displayed the highest methanol yield of 29.7 mmol g_cat⁻¹ h⁻¹ with a selectivity of 80.9% at 70 °C. Further analysis revealed that Fe₃Cu₁ showed the highest Fe³⁺ and Cu⁺ contents. The optimal dual valence cycle not only facilitates the efficient utilization of H₂O₂, promoting the activation of methane to •CH₃ at the Fe site, but also suppresses the deep oxidation caused by the Fenton-like effect of Fe/H₂O₂, thus maintaining the high yield and high selectivity of methanol.

1. Introduction

Methane is the main component of natural gas and combustible ice [1,2]. However, most natural gas reserves are located in remote or offshore areas, making direct utilization challenging [3]. Converting methane into liquid fuels like methanol offers a promising solution for transportation-related issues. Currently, industrial methane utilization relies heavily on indirect conversion processes, involving methane reforming followed by Fischer–Tropsch synthesis, which typically requires high temperatures, high pressures, and significant energy inputs [4,5,6]. As a result, there is a pressing need for direct oxidative conversion of methane under mild conditions [7]. However, the stable C-H bond of methane (438.8 kJ mol⁻¹) makes it difficult to activate under mild conditions. Furthermore, the desired oxidation products (C1 oxygenates) are often more reactive than methane itself, leading to over-oxidation and the formation of CO₂ [8,9]. Thus, achieving efficient methane conversion while maintaining high selectivity for C1 oxygenates remains a major challenge.

Direct catalytic conversion of methane offers an environmentally friendly and energy-efficient method, enabling methane-to-methanol transformation at room temperature with the presence of a catalyst, which plays a critical role in facilitating the process. By simulating the structure of natural methane monooxygenase (MMO), loading noble metals [10,11] and transition metals [12,13,14] on zeolites shows great promise in methane conversion [15]. The binuclear iron center formed on Fe/ZSM-5 resembles the active center of the diiron species in soluble methane monooxygenase (sMMO) [16], and the extra-framework Fe³⁺ species are key species for methane transformation [14,16], which promote the dissociation of C-H bonds and effectively activate methane. Meanwhile, the Fenton-like reaction of Fe with H₂O₂ can generate a significant amount of free radicals, which promote the cleavage of the C-H bond in methane and enhance its conversion efficiency [17]. However, the large number of free radicals also leads to the over-oxidation of C1 oxygenates, resulting in reduced selectivity of methanol. Consequently, a key research focus is to simultaneously improve methane conversion activity and enhance methanol selectivity.

Studies have demonstrated that Cu-based catalysts can enhance methanol yield in methane conversion. The presence of Cu species consumes •OH radicals, thereby preventing the over-oxidation of methanol and suppressing the formation of formic acid and CO₂ [1,18]. CuO-loaded zeolite can significantly improve methanol selectivity by decomposing excess H₂O₂ on CuO species during the reaction, which inhibits further oxidation of methanol [10]. Additionally, research reports that Cu species promote the generation of •OH radicals rather than •OOH radicals from H₂O₂, facilitating methanol production. These •OH radicals further react with •CH₃ intermediates to form methanol, thus improving the selectivity [19].

Given the ability of Cu to effectively suppress the over-oxidation of methane, this study focuses on enhancing methane activation by loading Fe onto ZSM-5 while introducing Cu to mitigate the over-oxidation effects associated with Fenton-like reactions. In this work, a series of high-loading Fe-Cu/ZSM-5 catalysts were prepared using the incipient wetness impregnation method, and the effects of varying Fe/Cu ratios on methane oxidation activity and selectivity were systematically investigated. On one hand, the addition of Cu to Fe-based catalysts alters the structural and surface properties [20]. On the other hand, the synergistic interaction between Fe and Cu bimetallic species on ZSM-5 facilitates redox cycling and promotes the activation of H₂O₂, enabling the continuous generation of active radicals [21,22], thereby achieving efficient methane conversion. The metal loaded on Fe₃Cu₁ showed good dispersion with the highest Fe³⁺ and Cu⁺ content, and its bimetallic valence cycle efficiently utilized H₂O₂, exhibited the best methane conversion activity, achieving a methanol yield of 29.7 mmol g_cat⁻¹ h⁻¹ with a selectivity of 80.9%, and good cyclic stability.

2. Experimental Sections

2.1. Materials and Reagents

Fe(NO₃)₂ 9H₂O, Cu(NO₃)₂ 3H₂O, sulfuric acid (H₂SO₄), acetic acid, ammonium acetate, acetylacetone, isopropanol (IPA), p-benzoquinone (p-BQ), and 5,5-dimethyl-1-pyrroline N-oxide (DMPO) were purchased from Aladdin Chemical Reagent Co., Ltd. (Shanghai, China) and were analytical grade without further purification. HZSM-5 (25) was purchased from the Nankai University Catalyst Plant. Deionized water was used throughout the whole experiment.

2.2. Synthesis of Catalysts

A series of Fe-Cu/ZSM-5 catalysts with a fixed metal content of 10 wt.% and variable Fe and Cu loading amounts were prepared by incipient wetness impregnation. In a typical synthesis, the support of 4 g HZSM-5 was calcined in air at 550 °C for 4 h as pretreatment. Then, 2 mL of aqueous solution with different Fe(NO₃)₃ and Cu(NO₃)₂ concentrations were slowly added to the support with thorough stirring at room temperature. After ultrasonic treatment for 10 min, the support was dried at 30 °C for 48 h and then calcined in air at 550 °C for 4 h. The Fe and Cu loading amounts were determined by ICP-OES (Table S1) and labeled as Fe/ZSM-5, Cu/ZSM-5, Fe₃Cu₁, Fe₁Cu₁, and Fe₁Cu₃.

2.3. Characterizations

X-ray diffraction (XRD) patterns were determined by a Bruker D8 Advance X-ray instrument equipped with Cu-Kα radiation at 2θ of 20–60° with a scanning speed of 5°/min. High-resolution transmission electron microscope (HRTEM) images were captured on the Talos F200X instrument (FEI Co., Ltd. Shanghai, China). The N₂ adsorption–desorption isotherms were obtained at 77 K using a Micromeritics ASAP 2460 instrument (Shanghai, China). X-ray photoelectron spectroscopy (XPS) spectra were collected on the Thermo Scientific EscaLab Xi⁺ (Waltham, MA, USA) equipped with an Mg-Kα radiation source (1253.6 eV). Fourier transform–infrared spectroscopy (FT-IR) was carried out on a Nicolet iS50 spectrometer (Thermo Scientific, Waltham, MA, USA). The absorbance of the liquid mixture was recorded on visible spectroscopy (722N, Shanghai Jinghua Technology Instrument Co., Ltd. Shanghai, China). CH₄ temperature-programmed desorption (CH₄-TPD) experiments were performed on a Micromeritics AutoChem II 2920 chemisorption analyzer (Shanghai, China). UV–vis diffuse reflection spectra (UV–vis DRS) were recorded on a Shimadzu UV-3600 instrument (Shimadzu Instruments, Suzhou, China). Reactive free radicals, including methyl radicals (•CH₃), hydroxyl radicals (•OH), and hydroperoxyl (•OOH), were monitored by electron paramagnetic resonance (EPR, Bruker E500 spectrometer, Billerica, MA, USA) instrument by using DMPO as the radical trap reagent.

2.4. Catalytic Activity Test

The methane conversion and determination process are illustrated in Figure S1. Methane oxidation was conducted in an 80 mL stainless-steel autoclave with a Teflon vessel. Typically, 25 mg of catalyst and 20 mL of 0.5 M H₂O₂ were added to the reactor, which was then purged with pure CH₄ (>99.9%) three times and pressurized to 30 bars. The autoclave was heated to 70 °C and stirred at 800 rpm. After the reaction, the products were rapidly cooled to room temperature using ice water. The gas components (CH₄ and CO₂) were determined by gas chromatograph equipped with a flame ionization detector (FID) and a thermal conductivity detector (TCD) (GC9720 Plus, Zhejiang Fuli Analytical Instrument Co., Ltd. Wenling, China). After centrifugation, the liquid products, including CH₃OOH and CH₃OH, were quantitatively analyzed with ¹H-NMR and gas chromatography with an FID (GC9720 Plus, Zhejiang Fuli Analytical Instrument Co., Ltd.). For ¹H-NMR, typically, 0.5 mL of liquid product was mixed with 0.1 mL of deuterium oxide (D₂O), and 0.05 μL of dimethyl sulfoxide (DMSO, 99.99%, Shanghai Sinopharm, Shanghai, China) was added as an internal standard. Calibration curves were constructed between ¹H-NMR area and concentrations of standard product solutions. The concentration of HCHO was determined using a colorimetric method, which relies on the reaction of acetylacetone with HCHO in the presence of acetic acid and ammonium acetate [23,24]. Meanwhile, the H₂O₂ concentration was quantified through the conventional titanium oxalate potassium method and analyzed using visible spectroscopy [11,25].

The yields of catalysts and selectivity were calculated using the following equation:

Yield of C1 oxygenates (mmol g_cat⁻¹ h⁻¹):

$${\mathrm{Y}}_{products}={\displaystyle \frac{{\mathrm{n}}_{products}}{{\mathrm{m}}_{catalyst}\times \mathrm{t}}}$$

Selectivity of CH₃OH (%):

$${\mathrm{S}}_{C{H}_{3}OH}={\displaystyle \frac{{\mathrm{n}}_{C{H}_{3}OH}}{{\mathrm{n}}_{C{H}_{3}OH}+{\mathrm{n}}_{C{H}_{3}OOH}+{\mathrm{n}}_{HCHO}+{\mathrm{n}}_{C{O}_2}}}\times 100\mathrm{\%}$$

3. Results and Discussion

3.1. Structural and Morphological Characteristics

The states of Fe and Cu on ZSM-5 were analyzed by HRTEM. The hexagonal prism of ZSM-5 remained unchanged, with Cu and Fe particles dispersed on its surface. Lattice fringes with lattice spacings of 0.22 nm and 0.36 nm, corresponding to the (113) and (012) planes of Fe₂O₃, respectively (Figure S2a), were observed on Fe/ZSM-5. Similarly, lattice fringes with lattice spacing of 0.36 nm and 0.25 nm identified on Fe₃Cu₁ and Fe₁Cu₃ were assigned to the (012) and (110) planes of α-Fe₂O₃, respectively (Figure 1a,b). The energy dispersive spectrometer (EDS) shows that Cu was more uniformly distributed on the ZSM-5 surface, as well as better dispersion of Fe on Fe₃Cu₁, compared to Fe₁Cu₃, where Fe was more aggregated on the ZSM-5 exterior.

The structure of the synthesized samples was characterized by XRD (Figure 1c). The XRD patterns after loading Fe were similar to that of pure HZSM-5, implying that both the Fe- and Cu-loaded catalysts maintain the structure of the ZSM-5. When the content of Cu or Fe increased, peaks of Fe₂O₃ and CuO were observed on Fe/ZSM-5 and Cu/ZSM-5, respectively. The peaks located at 2θ of 33.16, 35.63, 49.46, and 54.07 correspond to Fe₂O₃ (JCPDS 99-0060) [26]; 2θ of 35.25, 38.54, and 48.71 are attributed to CuO (JCPDS 89-2530); and 2θ of 30.17 and 35.6 are attributed to Cuprospinel (JCPDS 25-0283 CuFe₂O₄). The bimetallic catalysts showed no obvious oxide peaks, suggesting that Fe and Cu species are finely dispersed on ZSM-5 as amorphous oxides or nanocomposites. It can be inferred that the Fe-Cu interaction favored the dispersion of its oxides.

Additionally, an increase in the Fe/Cu ratio promoted the dispersion of Cu species and reduction of both CuO and Fe₂O₃ [27]. Figure 1d shows the N₂ adsorption–desorption isotherms, where distinct type IV adsorption–desorption isotherms were identified, suggesting that these catalysts are predominantly microporous in structure [28]. The number of micropores decreased significantly after metal loading, indicating that the active component blocks some of the pores of zeolite. At a relatively high pressure (P/P₀) of 0.5–1.0, the N₂ adsorption–desorption isotherm showed an H4 hysteresis loop, reflecting a mesoporous structure [29]. The microporous pore sizes of the catalysts were mainly distributed in 1.5–2 nm, with a minor presence of mesopores between 3.5 and 4.5 nm (Figure S3).

Table 1. Specific surface area and pore volume of catalysts.

Catalyst	Surface Area (m2/g)	t-Plot Micropore Volume (cm3/g)	Pore Diameter (nm)
Fe/ZSM-5	310.20	0.11	1.82
Fe3Cu1	261.64	0.10	1.92
Fe1Cu1	272.62	0.10	1.87
Fe1Cu3	286.02	0.11	1.85
Cu/ZSM-5	298.93	0.12	1.88
HZSM-5	391.17	0.13	1.87

shows that the specific surface area of the bimetallic catalyst increased with the Cu loading ratio, likely due to the better dispersion of Cu on the zeolite [27]. Additionally, the decrease in Fe loading contributes to mitigating the agglomeration phenomenon.

As shown in Figure 2a, the CH₄-TPD tests revealed that the introduction of Cu and Fe had a significant effect on methane adsorption. The desorption temperature decreased after loading Cu alone, while a broad desorption band was observed for Fe₃Cu₁. The higher desorption temperature and significantly larger desorption peak area indicate that Fe₃Cu₁ exhibits a stronger capacity for methane adsorption, which is beneficial for subsequent methane conversion [30]. Subsequently, the valence states of Fe and Cu on the catalyst surface were analyzed by XPS. Figure 2b shows the XPS spectrum of Fe 2p_3/2 and 2p_1/2, signifying the presence of both Fe²⁺ and Fe³⁺ species on the surface of ZSM-5. The peak around 710.7 eV is assigned to Fe²⁺ species, while the peak around 712.8 eV corresponds to Fe³⁺ species [31]. The relative contents and ratios of Fe²⁺ and Fe³⁺ are summarized in Figure 2d. As the Fe proportion increased, the relative content of Fe³⁺ also rose. There are studies [32] using Density Functional Theory (DFT) calculations that found that Fe(IV) clusters generated by the reaction of Fe(III) and H₂O₂ are able to promote homolytic activation of methane, leading to the formation of •CH₃, which means Fe³⁺ plays an important role in methane conversion. Figure 2c shows the XPS spectra of Cu 2p_3/2 and Cu 2p_1/2, with the shakeup satellites centered around 944.5 eV [33]. The absorption peaks of Cu 2p_3/2 and Cu 2p_1/2 can be divided into two peaks with Cu²⁺ located at 934.8 eV and 933.5 eV for Cu⁺ species [34,35]. The relative contents of Cu²⁺ and Cu⁺ are displayed in Figure S4. As the Cu proportion increased, the Cu⁺ content decreased, implying a higher presence of CuO [36], consistent with the CuO peaks observed in the XRD pattern of Cu/ZSM-5. Cu⁺ can effectively promote the selectivity of methanol [37], and Fe₃Cu₁ presented the highest Fe³⁺/Cu⁺ ratio among the bimetallic catalysts.

UV–vis DRS spectroscopy was applied to understand the nature and coordination of copper and iron species in the samples. The absorption bands related to the ligand geometry and environment are related to the ligand-to-metal charge-transfer (LMCT), where the framework Fe adsorption bands are in the range of 200–250 nm, 250–350 nm for isolated and oligomerized extra-framework Fe in zeolite channels, 350–450 nm for larger Fe clusters, and >450 nm for bulk Fe_xO_y particles [12]. The UV–vis DRS spectra were normalized, and it can be seen from the magnified portion of Figure 2e that the isolated Fe³⁺ at the cationic positions on the surface of Cu/ZSM-5 was replaced by Cu compared to Fe/ZSM-5, and therefore the peak of isolated Cu²⁺ appeared (213 nm) [28]. The peaks maintained high intensity over a broad range, suggesting that different Fe species existed [38], which is consistent with the results of the TEM. The absorption bands in the spectral range of 300–450 nm indicate the oligomeric Fe/Cu species. Fe₃Cu₁ displays the highest isolated Fe³⁺ and oligomeric clusters band in the <300 nm range and 300–400 nm range, respectively, and the extra-framework binuclear Fe is usually considered to be the active site for selective oxidation of methane [12,39]. In addition, the band at >450 nm is significantly lower than that of other catalysts, indicating the least amount of bulk Fe₂O₃ particles, which accelerates the catalytic decomposition of H₂O₂ as well as promotes the further oxidation of the C1 oxygenates, unfavorable for methane to methanol [40]. Appropriate proportions of Cu can modulate the ratios of different Fe species and thus affect their catalytic properties.

3.2. Catalytic Performance Evaluation

The direct oxidation reaction of methane was carried out in a stainless-steel autoclave as described in the experimental section. As shown in Figure 3a, pure HZSM-5 performed poorly with a CH₃OH yield of 2.1 mmol g_cat⁻¹ h⁻¹. After introducing Fe, the Fe/ZSM-5 significantly increased the CH₃OH yield to 15.4 mmol g_cat⁻¹ h⁻¹. Although Cu/ZSM-5 showed a similar CH₃OH yield to HZSM-5, more HCHO was produced, demonstrating a stronger methane conversion capability. Fe/ZSM-5 achieved the highest methane conversion with a total C1 oxygenates and CO₂ yield of 38.1 mmol g_cat⁻¹ h⁻¹, but its selectivity of C1 oxygenates was low at 44.4%. The optimal Fe₃Cu₁ catalyst demonstrated a CH₃OH yield of 29.7 mmol g_cat⁻¹ h⁻¹ with a selectivity of 80.9%. In summary, Cu incorporation significantly reduces CO₂ formation, and an optimal Fe/Cu ratio enhances both CH₃OH and selectivity. However, increasing Cu content led to higher HCHO production and lower CH₃OH yield. Fortunately, CO₂ yield can still remain low. Excess Fe tends to cause excessive oxidation of C1 oxygenates, which can be inhibited by the presence of Cu.

The catalytic performance was evaluated by varying the reaction temperature with respect to Fe₃Cu₁. Methane oxidation was observed at room temperature (Figure 3b). Within 30–60 °C, the CH₃OH yield increased with temperature, indicating that the appropriate high temperature is favorable for the activation of methane. At 60 °C, the yield of CH₃OH reached its maximum of 89.1 mmol g_cat⁻¹ h⁻¹, with a selectivity of 91.3%. Further temperature increases led to a decline in CH₃OH yield, likely due to the high temperature that promoted the deep oxidation of CH₃OH to formic acid and CO₂. Additionally, the decomposition of H₂O₂ to O₂ and H₂O at higher temperatures may contribute to the reduced CH₃OH yield. The effect of reaction time on oxidation products was also investigated. As shown in Figure 3c,d, the CH₃OH yield decreased while the HCHO yield increased over time on Fe/ZSM-5, accompanied by a continuous rise in CO₂ selectivity. This suggests that CH₃OH is continuously oxidized over the Fe/ZSM-5 catalyst to HCHO and CO₂. In contrast, CH₃OH yield stayed up on Fe₃Cu₁ and remained high after 5 h reaction (61.0 mmol g_cat⁻¹), with little HCHO production and stable CO₂ selectivity (<25%).

Figure S5 shows the effect of different total pressures on the methane conversion performance of Fe₃Cu₁. Methane conversion exhibited a positive correlation with increasing pressure, with the highest CH₃OH yield observed at 30 bars. Additionally, the effect of H₂O₂ concentration on methane conversion was evaluated (Figure 4a). It was found that CH₃OH yield increased and then decreased with rising H₂O₂ concentration, while CO₂ selectivity continuously increased, suggesting that excessive H₂O₂ tends to lead to deep oxidation of methane. Stability test results (Figure 4b) showed that the Fe₃Cu₁ maintained promising initial stability for oxidation of methane to methanol, with a slight decrease in CH₃OH and a minor increase in CO₂ yield after four cycles. This decline may be due to the leaching of loaded Fe and Cu during the reaction and centrifugation. XPS analysis of Fe₃Cu₁ before and after the reaction (Figure 4c,d) revealed a slight shift of Fe 2p to lower binding energies and Cu 2p to higher binding energies, showing good valence cycle stability.

3.3. Mechanism Investigation

To investigate the roles of different radicals in methane conversion and C1 oxygenate formation, tests were conducted by adding 0.5 mM of p-BQ and IPA, respectively (quenchers for •OH and •OOH, respectively). As shown in Figure 5a, methanol yield decreased significantly, along with a slight increase in CO₂ production. CH₃OOH yield rose after IPA addition, which was thought to be an intermediate in the formation of methanol [39]. These results demonstrate that both •OH and •OOH are crucial radicals in the methane-to-methanol conversion process. The H₂O₂ utilization efficiency is an important indicator of the catalyst on methane partial oxidation performance, which can be assessed by the “gain factor” and defined as the molar ratio of the produced C1 oxygenates to the consumed H₂O₂ [41]. The gain factor for bimetallic catalysts positively correlated with the Fe content, while monometallic Fe/ZSM-5 deviated from this trend (Figure 5b), suggesting synergistic interactions between Fe and Cu that enhance methane conversion. The Fe₃Cu₁ exhibited the highest gain factor (0.065), indicating superior H₂O₂ utilization. In contrast, Cu/ZSM-5 nearly consumed all the H₂O₂ but yielded the lowest C1 oxygenates, which may be due to CuO clusters on Cu/ZSM-5 directly catalyzing the decomposition of the excess H₂O₂ into H₂O and O₂, resulting in a decrease in its gain factor [10].

DMPO was used as a trapping agent to explore the differences and effects of free radicals under different catalyst systems. From Figure 5c, a weak DMPO-OH signal was observed on HZSM-5 but was significantly enhanced after Cu loading. Due to insufficient activation of methane, •CH₃ was not enough to form CH₃OH. At the same time, excess •OH also over-oxidized CH₃OH to produce more HCHO. In contrast to the monometallic Cu/ZSM-5, •CH₃ and •OH were detected on Fe/ZSM-5, and Fe-Cu interactions enhanced the •CH₃ radical formation. The Fe/Cu ratio strongly influenced radical generation: higher Cu content increased •OH levels, while Fe-Cu synergy and dual redox cycling significantly raised •OH production. Therefore, Fe₁Cu₃ in the bimetallic catalyst presented the highest •OH content, and an insignificant •OOH signal was also observed. Its CH₃OH and HCHO yields resembled those of Cu/ZSM-5. The •CH₃ and •OH can quickly combine to form CH₃OH, and since Fe₃Cu₁ had the highest •CH₃ and appropriate •OH, it showed the highest CH₃OH yield.

To investigate the mechanism of direct catalytic oxidation of methane, in situ CH₄-DRIFTS was performed. The catalysts were exposed to H₂O₂ and CH₄ at 70 °C, and the spectra were recorded over 60 min. Figure 6a shows the time-dependent changes in the in situ CH₄-DRIFTS spectrum of Fe₃Cu₁. The peaks at 3614, 3401, and 3214 cm⁻¹ are assigned to *OH₂, *OH, and *OOH [4,42]; the peaks at 3016 cm⁻¹, 1305 cm⁻¹, and 1353 cm⁻¹ are attributed to CH₄, stretching of *CH₃, and *OCH₃ [42,43,44]; the peak at 1643 cm⁻¹ is assigned to adsorbed H₂O* [44,45]; and 1837cm⁻¹ is assigned to Fe(Ⅲ)-(OH) species [46], respectively. After introducing methane, an *OH vibration peak and a gradually increasing *CH₃ peak appeared with reaction time, indicating that *OH formed after the decomposition of H₂O₂ can activate the C-H bond and thus generate the adsorbed *CH₃ species. The source of *OH₂ may be formed by the combination of *OH and H decomposed by CH₄. The appearance of *OCH₃, *OH, and *OOH peaks indicates the formation of CH₃OH and CH₃OOH. The intensities of *OH and *CH₃ peaks increased over time, while the HCOO- peak intensity remained constant, demonstrating its good methane oxidation capability and effective suppression of deep methanol oxidation. Figure 6b shows the time-dependent changes in the in situ CH₄-DRIFTS spectrum of Cu/ZSM-5. The weaker *OH₂, *OH, and *OOH peaks on Cu/ZSM-5 were observed, consistent with their lower methane oxidation performance. Formic acid species are the oxidation products of methanol, but the stretching vibration peak of formic acid species (HCOO-) at 1620 cm⁻¹ was not observed at Cu/ZSM-5 [47], indicating a weak oxidizing capacity.

CO-DRIFTS was employed to analyze the reduction process of Cu²⁺ on the catalyst. CO can be adsorbed on Cu⁺ sites to form stable Cu⁺-CO species without binding to Cu²⁺. The catalyst was treated with Ar at 300 °C, cooled to 50 °C, exposed to CO for 30 min, and then purged with Ar for 30 min. Figure S6 illustrates the spectra of Cu-containing catalysts after Ar purging. The peak at 2157 cm⁻¹ is assigned to the Cu⁺-CO species on the isolated Cu⁺ site, while the weaker absorption peak at 2137 cm⁻¹ is assigned to the Cu⁺-CO-Cu⁺ species on the dimeric Cu⁺ site. Fe₃Cu₁ exhibited the weakest peak at 2137 cm⁻¹ among the catalysts, implying the lowest degree of agglomeration of CuO_x [48]. This is consistent with the XPS results and shows that different Fe/Cu ratios affect the valence and state of Cu species. Figure S7 shows the time-dependent CO-DRIFTS spectrum of Fe₁Cu₃. As the Ar purge time increased, the peak at 2177 cm⁻¹, caused by the polarization of CO by acidic hydroxyls of the zeolites [49], gradually disappeared. In contrast, the peaks at 2157 cm⁻¹ and 2137 cm⁻¹ gradually increased.

Based on the above results and analysis, the reaction path of methane on as-prepared catalysts is shown in Scheme 1. Fe(Ⅲ) species were oxidized by H₂O₂ to [Fe(Ⅳ)-O•] species, and the C-H bond broke by a homolytic mechanism on the Fe(IV) species to form •CH₃, and the [Fe(IV)-O•] species were reduced to [Fe(III)-OH] species (Equation (1)) [32]. Then, •OOH and •OH were generated by the Fenton-like reaction of Fe²⁺/Fe³⁺ cycling and H₂O₂ (Equations (2) and (3)), and similarly on Cu⁺/Cu²⁺ (Equations (4) and (5)). The •CH₃ could also be generated by the •OH and CH₄ reaction (Equation (6)); then, it reacted with •OH and •OOH to produce CH₃OH and CH₃OOH, respectively (Equations (7) and (8)). CH₃OH and CH₃OOH were deeply oxidized by •OH/•OOH/H₂O₂ to form HCHO, CO₂, and H₂O (Equation (9)).

Fe(Ⅲ) + H₂O₂ + CH₄ → Fe(Ⅲ)-OH + •CH₃ + H₂O

(1)

Fe(Ⅱ) + H₂O₂ → Fe(Ⅲ) + •OH + OH⁻

(2)

Fe(Ⅲ) + H₂O₂ → Fe(Ⅱ) + •OOH + H⁺

(3)

Cu(Ⅱ) + OH⁻ + H₂O₂ → •OOH + Cu(Ⅰ) + H₂O

(4)

Cu(Ⅰ) + H⁺ + H₂O₂ → •OH + Cu(Ⅱ) + H₂O

(5)

•OH + CH₄ → •CH₃ + H₂O

(6)

•CH₃ + •OH → CH₃OH

(7)

•CH₃ + •OOH → CH₃OOH

(8)

CH₃OH/CH₃OOH + •OH/H₂O₂ → HCHO → CO₂ + H₂O

(9)

4. Conclusions

In summary, Fe- and Cu-modified ZSM-5 catalysts were synthesized via incipient wetness impregnation and evaluated for the direct oxidation of methane to methanol with H₂O₂ as oxidant. The Fe₃Cu₁ exhibited the highest H₂O₂ utilization efficiency, demonstrating optimal performance on methanol yield of 29.7 mmol g_cat⁻¹ h⁻¹ with a selectivity of 80.9% at 70 °C. In comparison with state-of-the-art catalysts for methane-to-methanol conversion (Table S2), the Fe₃Cu₁ catalyst demonstrates competitive performance, highlighting its exceptional balance of activity and selectivity among non-noble metal catalysts. Characterization results revealed that the superior performance of Fe₃Cu₁ stems from several factors: I. abundant active sites due to high metal loading. II. Fe-Cu interactions preventing oxide cluster aggregation, which is detrimental to methane activation. III. an optimal Fe/Cu ratio maximizing Fe³⁺ and Cu⁺ content, facilitating a robust dual redox cycle. This cycle promotes methane activation to •CH₃ at Fe sites while suppressing deep oxidation induced by the Fenton-like effect of Fe/H₂O₂, thereby maintaining high methanol selectivity. This work provides valuable insights for designing efficient catalysts for methane conversion under mild conditions.