Cu₉S₅/Gel-Derived TiO₂ Composites for Efficient CO₂ Adsorption and Conversion

Abstract

Engineering phase-selective gel composites presents a promising route to enhance both CO₂ adsorption and conversion efficiency in photocatalytic systems. In this work, Cu₉S₅/TiO₂ gel composites were synthesized via a hydrazine-hydrate-assisted hydrothermal method, using TiO₂ derived from a microwave-assisted sol–gel process. The resulting materials exhibit a porous gel-derived morphology with highly dispersed Cu₉S₅ nanocrystals, as confirmed by XRD, TEM, and XPS analyses. These structural features promote abundant surface-active sites and interfacial contact, enabling efficient CO₂ adsorption. Among all samples, the optimized 0.36Cu₉S₅/TiO₂ composite achieved a methane production rate of 34 μmol·g⁻¹·h⁻¹, with 64.76% CH₄ selectivity and 88.02% electron-based selectivity, significantly outperforming Cu₉S₈/TiO₂ synthesized without hydrazine hydrate. This enhancement is attributed to the dual role of hydrazine: facilitating phase transformation from Cu₉S₈ to Cu₉S₅ and modulating the interfacial electronic environment to favor CO₂ capture and activation. DFT calculations reveal that Cu₉S₅/TiO₂ effectively lowers the energy barriers of critical intermediates (*COOH, *CO, and *CHO), enhancing both CO₂ adsorption strength and subsequent conversion to methane. This work demonstrates a gel-derived composite strategy that couples efficient CO₂ adsorption with selective photocatalytic reduction, offering new design principles for adsorption–conversion hybrid materials.

1. Introduction

The rising concentration of atmospheric carbon dioxide (CO₂), driven by intensified fossil fuel consumption and anthropogenic emissions, poses a critical threat to global climate stability. As the primary greenhouse gas, CO₂ contributes significantly to global warming, environmental degradation, and energy imbalance [1,2,3]. In light of these challenges, developing sustainable and efficient CO₂ conversion strategies has become an urgent imperative for achieving carbon neutrality and promoting renewable energy technologies [4,5].

Among various emerging technologies, photocatalytic CO₂ reduction (CO₂RR) offers a green, solar-driven route to convert CO₂ into value-added fuels such as methane (CH₄), ethylene (C₂H₄), methanol (CH₃OH), and carbon monoxide (CO) [5,6,7,8,9,10] This artificial photosynthetic process utilizes semiconductor photocatalysts to activate and reduce CO₂ under light irradiation, thereby enabling both carbon recycling and solar-to-chemical energy conversion. However, the practical application of photocatalytic CO₂RR remains hindered by several intrinsic limitations, particularly those related to low quantum efficiency, poor CO₂ adsorption, and product selectivity [11,12].

Titanium dioxide (TiO₂), a widely studied photocatalyst, suffers from a wide bandgap (~3.3 eV) that restricts its optical response to the ultraviolet spectrum, as well as the rapid recombination of photogenerated charge carriers [13,14]. To overcome these drawbacks, composite photocatalysts have been extensively explored. By coupling TiO₂ with visible-light-active materials such as metal oxides, sulfides, or nitrides, the resulting heterostructures exhibit improved light harvesting, interfacial charge transfer, and redox kinetics [15,16].

Copper-based sulfides (e.g., CuS, Cu₉S₈, Cu₉S₅, Cu₂S) have attracted particular attention due to their narrow bandgaps, strong CO₂ adsorption affinity, and suitable conduction band positions for CO₂RR [17,18,19,20]. Among them, Cu₉S₅ exhibits enhanced visible light absorption and possesses favorable catalytic sites for deep reduction pathways toward methane [21,22,23]. Controlling the phase, dispersion, and interfacial interaction between Cu₉S₅ and the support remains essential for optimizing CO₂ adsorption and selective CH₄ formation [24,25,26]. However, despite extensive studies on Cu-based sulfides, no prior report has systematically combined phase-selective modulation (Cu₉S₈ → Cu₉S₅) with a gel-derived TiO₂ matrix to simultaneously enhance CO₂ adsorption and CH₄ selectivity.

Hydrazine hydrate (N₂H₄·H₂O), a strong reducing agent, offers unique advantages in tailoring the phase composition and surface electronic states of metal sulfide catalysts. When used during hydrothermal synthesis, the accurate addition of hydrazine can promote a desired phase transition and modulate the catalyst–support interface [27,28,29,30]. For example, Yang et al. demonstrated that hydrazine can effectively regulate defect states and promote charge separation in composite photocatalysts for hydrogen evolution, highlighting its potential in photocatalytic systems. Similarly, Sattigeri et al. reported that a hydrazine–sulfur complex can stabilize ZnS quantum dots and enhance their photocatalytic activity, underlining the role of hydrazine in controlling phase purity and nanostructure. In addition, Li et al. showed that hydrazine hydrate reduction could induce oxygen vacancies in Co₃O₄ nanosheets, thereby tuning the electronic structure and improving catalytic properties. Zhang et al. further confirmed that hydrazine-induced oxygen vacancies in MoO₃ nanobelts significantly accelerated charge transfer and boosted electrochemical kinetics, offering a broader insight into the versatility of hydrazine in phase and defect engineering. Inspired by these findings, hydrazine treatment is expected to have great potential in facilitating the phase transition from Cu₉S₈ to Cu₉S₅, thereby enabling precise phase control and interfacial optimization in Cu-based sulfide photocatalysts. Furthermore, using gel-derived TiO₂ prepared via a microwave-assisted sol–gel route provides a porous and high-surface-area matrix, ideal for CO₂ adsorption and intimate interfacial contact with Cu₉S₅ domains [31].

Subsequently, in this study, we present a novel strategy for synthesizing Cu₉S₅/TiO₂ composites by employing a hydrazine-hydrate-assisted hydrothermal method. This approach not only induces the phase transition from Cu₉S₈ to Cu₉S₅ but also enhances the dispersion and crystallinity of Cu₉S₅ on TiO₂. Additionally, we introduce a gel-derived composite framework for TiO₂, prepared via a microwave-assisted sol-gel method, which improves both CO₂ adsorption and the interfacial contact between Cu₉S₅ and TiO₂. This strategy integrates the advantages of both CO₂ adsorption and selective conversion functionalities into a single composite material, addressing two critical challenges in photocatalytic CO₂ reduction. Furthermore, density functional theory (DFT) calculations are employed to investigate the electronic structure and reaction pathways of Cu₉S₅/TiO₂ composites, offering insights into the enhanced methane selectivity observed in these materials. The combination of hydrazine-induced phase transformation and gel-based composite design provides a highly efficient, selective, and stable photocatalyst for CO₂ reduction.

2. Results and Discussion

2.1. Structural Features of Gel-Derived Cu₉S₅/TiO₂ Composites

The phase composition and crystallinity of the gel-derived CuS_x/TiO₂ composites, both with and without the inclusion of hydrazine hydrate (N₂H₄), were thoroughly examined using X-ray diffraction (XRD), as illustrated in Figure 1. The XRD pattern of pure TiO₂ exhibits characteristic peaks at 25.3°, 37.8°, 48.1°, and 55.1°, corresponding to the anatase phase (PDF#78-2486). The XRD peaks of TiO₂ in the 0.36CuS_x/TiO₂ sample synthesized with hydrazine hydrate are sharper and more defined, indicating an enhancement in the crystallinity of TiO₂. This improvement in TiO₂ crystallinity is likely due to the better phase formation and reduced defects induced by the presence of hydrazine hydrate, which helps to stabilize the TiO₂ structure during the hydrothermal synthesis.

For the 0.36CuS_x/TiO₂ sample synthesized without hydrazine hydrate, additional diffraction peaks appear at 29.15°, which correspond to the Cu₉S₈ phase (PDF#36-0379). The main peak at 47.9° overlaps with those of TiO₂, suggesting that the TiO₂ structure remains largely intact while Cu₉S₈ is formed. This confirms that the Cu₉S₈ phase is the dominant copper sulfide species when hydrazine hydrate is not used in the synthesis. In contrast, the 0.36CuS_x/TiO₂ sample synthesized with hydrazine hydrate shows distinct peaks at 27.9°, 32.4°, and 46.4°, which can be indexed to Cu₉S₅ (PDF#26-0476). These peaks do not overlap with TiO₂, indicating the formation of Cu₉S₅ and confirming that hydrazine hydrate influences the phase transition from Cu₉S₈ to Cu₉S₅. Importantly, no CuS species were observed in any of the samples, highlighting that the formation of Cu₉S₈ or Cu₉S₅ is strictly controlled by the presence of hydrazine hydrate under the hydrothermal conditions. These results emphasize the critical role of hydrazine hydrate in modulating the copper sulfide phase in CuS_x/TiO₂ composites. In the next sections, the samples synthesized with hydrazine hydrate are referred to as Cu₉S₅/TiO₂, while those synthesized without hydrazine hydrate are referred to as Cu₉S₈/TiO₂.

The morphology and microstructure of the gel-derived Cu₉S₅/TiO₂ and Cu₉S₈/TiO₂ composites were investigated using a combination of SEM, TEM, high-resolution TEM (HRTEM), and energy-dispersive X-ray spectroscopy (EDS) mapping, as illustrated in Figure 2. For the Cu₉S₅/TiO₂ composite synthesized with hydrazine hydrate, the SEM image (Figure 2a) reveals a relatively uniform distribution of Cu₉S₅ particles across the TiO₂ surface. The TEM image (Figure 2b) further confirms that the Cu₉S₅ domains are evenly dispersed without significant agglomeration. High-resolution TEM analysis (Figure 2c) displays distinct lattice fringes with an interplanar spacing of approximately 0.240 nm, which can be indexed to the (101) plane of Cu₉S₅, indicating the formation of a well-crystallized phase. The corresponding FFT pattern (Figure 2d) shows sharp diffraction spots, consistent with a highly ordered crystalline structure. A low-magnification TEM overview (Figure 2i) supports the homogeneous distribution of Cu₉S₅ on the TiO₂ support. In contrast, the Cu₉S₈/TiO₂ composite prepared without hydrazine hydrate exhibits less favorable morphology. As shown in the SEM image (Figure 2e), Cu₉S₈ particles are more aggregated and less uniformly spread across the TiO₂ surface. TEM observation (Figure 2f) also reveals uneven distribution and particle clustering. The HRTEM image (Figure 2g) shows lattice fringes with a spacing of 0.190 nm, corresponding to the (111) plane of Cu₉S₈, confirming the formation of the Cu₉S₈ phase. However, the associated FFT pattern (Figure 2h) is less distinct, suggesting relatively lower crystallinity compared to the Cu₉S₅ counterpart. The overall particle arrangement in the overview TEM image (Figure 2n) further confirms the irregular distribution of Cu₉S₈ domains.

Elemental mapping via EDS was employed to examine the spatial distribution of key elements in both composites. For Cu₉S₅/TiO₂, the EDS maps (Figure 2j–m) clearly show the co-localization of Ti, Cu, and S, with Cu and S signals uniformly distributed across the TiO₂ substrate, consistent with the homogeneous dispersion of Cu₉S₅. The oxygen map further supports the presence of TiO₂ throughout the sample. In comparison, the EDS maps of Cu₉S₈/TiO₂ (Figure 2o–r) show more localized and concentrated Cu and S signals, corresponding to the Cu₉S₈ aggregates, while Ti and O remain uniformly distributed.

Collectively, the X-ray diffraction and transmission electron microscopy analyses confirm the formation of Cu₉S₅ in the composites. The role of hydrazine hydrate in promoting the phase transition from Cu₉S₈ to Cu₉S₅ can be summarized in the following reaction scheme. In the first stage, Cu₉S₈ is initially formed during the synthesis process in the absence of hydrazine hydrate. In the second stage, hydrazine hydrate acts as a strong reducing agent, donating electrons to Cu₉S₈ and reducing Cu²⁺ to Cu⁺, which then combine with sulfur to form Cu₉S₅. The Cu₉S₅/TiO₂ composite exhibits superior microstructural uniformity and phase clarity, which are expected to enhance interfacial contact and charge separation during photocatalysis, aligning with the enhanced performance discussed in later sections.

To gain deeper insight into the surface composition and electronic states of the gel-derived Cu₉S₅/TiO₂ and Cu₉S₈/TiO₂ composites, X-ray photoelectron spectroscopy (XPS) was performed on both materials as shown in Figure 3. The XPS spectra of Ti 2p, O 1s, Cu 2p, and S 2p regions were analyzed to investigate the chemical environment of titanium, oxygen, copper, and sulfur species, which are crucial for understanding the photocatalytic behavior of these composites. The Ti 2p spectra of both composites exhibited a characteristic binding energy of 458.6 eV, which is typical for TiO₂. This confirms that the TiO₂ phase is preserved in both the Cu₉S₅/TiO₂ and Cu₉S₈/TiO₂ composites. The spectra also show a satellite peak at lower binding energies, which is a signature feature of TiO₂, indicating that the introduction of copper sulfides did not significantly alter the TiO₂ structure. There were no significant shifts in the Ti 2p binding energy, further suggesting the stability of the TiO₂ phase. In the O 1s region, two distinct peaks were observed corresponding to lattice oxygen (Lattice O) and vacancy oxygen (Vacancy O), with binding energies around 529 eV and 532 eV, respectively. These peaks are indicative of the oxygen species present in the TiO₂ lattice and the oxygen vacancies that are often present in metal oxides. Both Cu₉S₅/TiO₂ and Cu₉S₈/TiO₂ show similar peak positions and relative intensities, suggesting that the incorporation of Cu₉S₅ or Cu₉S₈ did not cause significant changes in the oxygen species or the oxygen vacancies within the TiO₂ lattice.

The Cu 2p spectra showed notable differences between the two composites. For Cu₉S₅/TiO₂, the Cu 2p spectrum exhibited peaks corresponding to Cu⁰ at 932.4 eV and Cu⁺ at 932.7 eV, indicating the presence of both Cu(I) and Cu(0) species. This suggests that hydrazine hydrate, used during synthesis, played a key role in reducing Cu²⁺ to Cu⁰ and Cu⁺, both of which are important for enhancing photocatalytic activity. In contrast, the Cu₉S₈/TiO₂ composite exhibited peaks corresponding to Cu⁺ at 932.73 eV and Cu²⁺ at 933.6 eV, indicating a higher proportion of Cu²⁺ species. This suggests that in the absence of hydrazine hydrate, Cu²⁺ remains more prevalent, which likely limits the photocatalytic activity compared to Cu₉S₅/TiO₂. This further emphasizes the importance of hydrazine hydrate in reducing copper species and enhancing catalytic efficiency.

In the S 2p region, both Cu₉S₅/TiO₂ and Cu₉S₈/TiO₂ composites exhibited peaks corresponding to sulfate and sulfide species. The presence of sulfate at 168.6 eV suggests that sulfur is bonded to oxygen, possibly forming sulfate species due to surface oxidation or interaction with surface oxygen atoms from TiO₂. The more intense sulfide peak at 161.4 eV indicates that sulfur primarily exists in its reduced state as sulfide. The higher intensity of the sulfide peak in both composites suggests that the sulfide species are more abundant and better dispersed on the surface. This finding is consistent with the energy dispersive spectroscopy (EDS) results, which showed a more homogeneous distribution of Cu₉S₅.

2.2. Optical and Photoelectrochemical Properties of Gel-Derived Cu₉S₅/TiO₂ and Cu₉S₈/TiO₂

Figure 4 presents the optical properties of the gel-derived Cu₉S₅/TiO₂ and Cu₉S₈/TiO₂ composites, shedding light on their potential for photocatalytic applications. Figure 4a,b display the Mott–Schottky plots for Cu₉S₅/TiO₂ and Cu₉S₈/TiO₂ composites, respectively, used to determine the flat-band potential (E_fb) and semiconductor type. The plot for Cu₉S₅/TiO₂ (Figure 4a) reveals a flat-band potential around −0.43 eV, indicative of n-type semiconductor characteristics, which is important for efficient electron transfer during photocatalytic reactions. Similarly, the Mott–Schottky plot for Cu₉S₈/TiO₂ (Figure 4b) shows a flat-band potential of approximately −0.48 eV, confirming the n-type nature of this composite as well.

In Figure 4c,d, the optical properties of the composites are examined through UV–vis absorption spectra and Tauc plots. Figure 4c shows that TiO₂ has a sharp absorption edge around 380 nm, typical of its wide band gap, which limits its absorption to the UV region. Both Cu₉S₅/TiO₂ and Cu₉S₈/TiO₂ composites exhibit enhanced absorption in the visible light range compared to pure TiO₂. Cu₉S₅/TiO₂ demonstrates a more prominent absorption in the visible region, extending well beyond 600 nm and up to 800 nm, indicating significantly improved visible light absorption. On the other hand, Cu₉S₈/TiO₂ also absorbs in the visible region, but the absorption intensity is slightly lower than that of Cu₉S₅/TiO₂, which suggests that Cu₉S₅/TiO₂ has a better capacity for visible light utilization.

Figure 4d presents the Tauc plots, where the optical band gaps (Eg) of the composites are determined by plotting (αhν)² versus hν. The optical band gap of Cu₉S₈/TiO₂ is measured to be 3.28 eV, which is slightly smaller than that of TiO₂ (3.30 eV), indicating enhanced visible light absorption due to the Cu₉S₈ phase. In contrast, the optical band gap of Cu₉S₅/TiO₂ is even smaller, measured at 3.26 eV, confirming that Cu₉S₅ further reduces the band gap and enhances the ability of the composite to absorb visible light more efficiently.

Therefore, the results from the Mott–Schottky analysis and UV–vis absorption spectra show that the incorporation of Cu₉S₅ and Cu₉S₈ into TiO₂ enhances their photocatalytic performance by improving visible light absorption. Cu₉S₅/TiO₂ (3.26 eV) exhibits a smaller band gap than Cu₉S₈/TiO₂ (3.28 eV), which enables better visible light absorption and higher photocatalytic efficiency, making both composites promising candidates for visible-light-driven photocatalysis.

Figure 5 demonstrates the structural and electrochemical properties of TiO₂ and Cu₉S₅/TiO₂ composites, providing insights into their photocatalytic performance. As shown in Figure 5a, the Raman spectrum of TiO₂ shows the characteristic peaks around 143 cm⁻¹ and 445 cm⁻¹, which correspond to the Eg and A1g vibrational modes of the TiO₂ anatase phase. In contrast, the Cu₉S₅ spectrum exhibits an additional peak at approximately 200 cm⁻¹, corresponding to the Cu-S vibrational mode, which confirms the successful incorporation of Cu₉S₅ into the composite.

In Figure 5b, the electrochemical impedance spectroscopy (EIS) results are shown for Cu₉S₅/TiO₂ and Cu₉S₈/TiO₂ composites. The Nyquist plots display the charge transfer resistance (Rct) of both composites. The Cu₉S₅/TiO₂ composite shows a significantly lower Rct compared to Cu₉S₈/TiO₂, suggesting enhanced charge carrier mobility and more efficient charge transfer in Cu₉S₅/TiO₂. This lower impedance indicates better photocatalytic performance, as reduced resistance facilitates more effective electron transport during the photocatalytic process.

Figure 5c presents the photocurrent responses of Cu₉S₅/TiO₂ and Cu₉S₈/TiO₂ under periodic light illumination. The Cu₉S₅/TiO₂ composite shows a stronger and more stable photocurrent response compared to Cu₉S₈/TiO₂, which demonstrates the superior photoelectrochemical performance of Cu₉S₅/TiO₂. This is due to the enhanced light absorption and efficient charge separation promoted by Cu₉S₅, which contributes to higher photocurrent generation under visible light irradiation.

Figure 5d shows the photoluminescence (PL) spectra of TiO₂, Cu₉S₅/TiO₂, and Cu₉S₈/TiO₂ composites. The PL intensity of Cu₉S₅/TiO₂ is significantly lower than that of TiO₂ and Cu₉S₈/TiO₂, indicating a reduced recombination of photogenerated electron–hole pairs. This reduced recombination enhances the photocatalytic efficiency of Cu₉S₅/TiO₂, as more charge carriers are available for photocatalytic reactions.

Obviously, the incorporation of Cu₉S₅ improves charge transfer, reduces charge recombination, and enhances light absorption, making Cu₉S₅/TiO₂ a promising composite for efficient photocatalytic applications.

2.3. Photocatalytic Performance and Selectivity Toward Methane

Figure 6 presents a comprehensive analysis of the photocatalytic CO₂ reduction performance of various Cu₉S₅/TiO₂ composites under different experimental conditions, highlighting their efficiency and stability. Figure 6a shows the production rates of CH₄ and CO from CO₂ reduction by Cu₉S₅/TiO₂ composites with varying Cu₉S₅ content (0.12Cu₉S₅/TiO₂, 0.24Cu₉S₅/TiO₂, 0.36Cu₉S₅/TiO₂, and 0.48Cu₉S₅/TiO₂). The results indicate that the 0.36Cu₉S₅/TiO₂ composite exhibits the highest CH₄ production rate, reaching approximately 34 μmol/g·h, followed by the 0.48Cu₉S₅/TiO₂ composite, which produces 28 μmol/g·h of CH₄. The 0.12Cu₉S₅/TiO₂ and 0.24Cu₉S₅/TiO₂ composites show lower CH₄ production rates, around 20 μmol/g·h and 22 μmol/g·h, respectively. On the other hand, the CO production rates remain relatively consistent across all composites, with 0.36Cu₉S₅/TiO₂ showing 15 μmol/g·h, the highest CO production rate among the composites. This suggests that a moderate amount of Cu₉S₅ (0.36 mmol) enhances the photocatalytic activity for CH₄ production, while maintaining a relatively balanced CO generation.

Figure 6b compares the CO₂ reduction performance of the 0.36Cu₉S₅/TiO₂ composite under various experimental conditions. In Condition 1, under standard experimental conditions with light irradiation, the 0.36Cu₉S₅/TiO₂ composite shows significant production of both CH₄ (around 34 μmol·g⁻¹·h⁻¹) and CO (approximately 15 μmol·g⁻¹·h⁻¹). In Condition 2, when the reaction is conducted in the dark, the production of both CH₄ and CO is almost negligible, demonstrating the essential role of light in driving the photocatalytic process. Condition 3, where no catalyst is used, results in zero CH₄ and CO production, highlighting the need for an efficient catalyst to facilitate the CO₂ reduction. Finally, in Condition 4, where CO₂ is replaced by Ar, the reaction effectively ceases, further confirming that CO₂ is crucial for the reaction to occur.

In Figure 6c, the photocatalytic CO₂ reduction performance of various composites is shown: TiO₂, 0.36Cu₉S₅/TiO₂ with hydrazine hydrate (N₂H₄), 0.36Cu₉S₈/TiO₂ without hydrazine hydrate, and 0.36Cu₉S₅/TiO₂ without TiO₂. The 0.36Cu₉S₅/TiO₂ composite with hydrazine hydrate achieves the highest CH₄ production rate of 34 μmol·g⁻¹·h⁻¹ and CO production rate of 15 μmol/g·h, significantly outperforming TiO₂ (5 μmol·g⁻¹·h⁻¹ CO) and the 0.36Cu₉S₈/TiO₂ without hydrazine hydrate (46 μmol·g⁻¹·h⁻¹ CO). The addition of hydrazine hydrate improves the separation of photogenerated charge carriers and increases the photocatalytic efficiency, leading to a higher CH₄ production rate.

In addition, in order to evaluate the activity of our Cu₉S₅/TiO₂ composites, we compared the results with representative copper sulfide–TiO₂ systems reported in the literature. For example, a 0D/1D Cu_2-xS/TiO₂ S-scheme heterojunction exhibited a CH₄ formation rate of 14.1 μmol·h⁻¹, which was nearly 3.9 times higher than that of pristine TiO₂ under similar conditions [32]. Another study on an in situ constructed Cu₂S/TiO₂ Schottky junction achieved CO and CH₄ production rates of 6.71 μmol·g⁻¹·h⁻¹ and 1.20 μmol·g⁻¹·h⁻¹, corresponding to 1.41- and 5.71-fold enhancements relative to bare TiO₂ [33]. For CuS/TiO₂ composites, a photothermal–photocatalytic system utilizing 2 wt% CuS/TiO₂ under full-spectrum irradiation demonstrated enhanced CO₂ conversion efficiency due to the synergistic effect of CuS-induced photothermal heating [34]. To the best of our knowledge, no prior studies have directly investigated Cu₉S₈- or Cu₉S₅-based composites for CO₂ adsorption or reduction. The only related work involving Cu₉S₅ is its combination with S,C,N-doped TiO₂ for photocatalytic N₂ fixation rather than CO₂ conversion. Therefore, the present study provides a novel strategy by employing hydrazine-assisted phase transformation (Cu₉S₈ → Cu₉S₅) within a gel-derived TiO₂ framework, enabling both enhanced CO₂ adsorption and selective CH₄ photoreduction.

Figure 6d demonstrates the stability of the 0.36Cu₉S₅/TiO₂ composite in four consecutive cycles of CO₂ reduction. The production of CH₄ and CO remains relatively consistent across the cycles, with CH₄ production at 34 μmol·g⁻¹·h⁻¹ and CO production at 15 μmol·g⁻¹·h⁻¹ in the first cycle, and only a slight decrease observed in subsequent cycles. This shows that the 0.36Cu₉S₅/TiO₂ composite exhibits excellent stability and can be used for multiple cycles without significant degradation in photocatalytic performance. The minimal decrease in CH₄ and CO production over the cycles suggests that the composite is highly durable, making it suitable for real-world applications in photocatalytic CO₂ reduction.

In a word, Figure 6 demonstrates that the 0.36Cu₉S₅/TiO₂ composite, especially with hydrazine hydrate treatment, is an efficient and stable photocatalyst for CO₂ reduction, producing substantial amounts of CH₄ and CO. The results also underscore the critical role of light, CO₂, and catalyst in achieving efficient photocatalytic CO₂ reduction. The excellent stability over four cycles further highlights the composite’s potential for practical applications in renewable energy and environmental sustainability.

Figure 7a illustrates the product selectivity for CH₄ and CO over the TiO₂ and Cu₉S₅/TiO₂ composites. For TiO₂, no CH₄ is produced (0%), and the product selectivity is entirely CO (100%). As the Cu₉S₅ content in the composites increases, the selectivity for CH₄ significantly improves. Specifically, the 0.48Cu₉S₅/TiO₂ composite shows the highest CH₄ selectivity at 66.51%, accompanied by a CO selectivity of 33.49%. The 0.36Cu₉S₅/TiO₂ composite also shows a notable CH₄ selectivity of 64.76%, with CO selectivity at 35.24%, indicating that the addition of Cu₉S₅ enhances CH₄ formation. The 0.24Cu₉S₅/TiO₂ composite displays a balanced selectivity of 60.12% for CH₄ and 39.88% for CO, while the 0.12Cu₉S₅/TiO₂ composite exhibits 43.3% CH₄ selectivity and 56.7% CO selectivity. These results demonstrate the progressive role of Cu₉S₅ in shifting the product distribution towards CH₄.

Figure 7b presents the electronic selectivity based on electron consumption for CH₄ and CO. The 0.36Cu₉S₅/TiO₂ composite exhibits the highest electronic selectivity for CH₄ at 88.02%, confirming its role in promoting efficient electron utilization for CH₄ formation. The 0.48Cu₉S₅/TiO₂ composite follows with an electronic selectivity for CH₄ of 88.82% and 11.18% for CO. In contrast, TiO₂ and the other Cu₉S₅/TiO₂ composites demonstrate significantly higher electronic selectivity for CO, reflecting the greater electron demand for CO formation. Specifically, the 0.12Cu₉S₅/TiO₂ composite shows an electronic selectivity for CH₄ of 75.34%, and 0.24Cu₉S₅/TiO₂ shows 85.77% for CH₄, indicating an improvement in electron efficiency with increasing Cu₉S₅ content.

These findings highlight the significant effect of Cu₉S₅ content on both product selectivity and electronic efficiency in CO₂ photoreduction. The 0.36Cu₉S₅/TiO₂ composite achieves the best balance of CH₄ and CO selectivity with 64.76% CH₄ selectivity and 35.24% CO selectivity, while demonstrating the highest electronic selectivity for CH₄ at 88.02%. This makes the 0.36Cu₉S₅/TiO₂ composite a promising photocatalyst for selective CO₂ reduction towards methane, offering an efficient electron utilization and favorable product distribution.

2.4. DFT Calculations and Heterojunction Analysis

To clarify the origin of the enhanced CH₄ selectivity in Cu₉S₅/TiO₂ composites, we combined DFT calculations and band structure characterization to reveal how interfacial energetics and charge dynamics modulate CO₂ reduction pathways. Figure 8 presents the DFT-calculated reaction free energy profiles and band structure diagrams for TiO₂ and Cu₉S₅/TiO₂ composites, offering a detailed insight into the photocatalytic CO₂ reduction mechanism, particularly focusing on the intermediates COOH, CO, and CHO, which are critical to methane (CH₄) formation.

Figure 8a illustrates the reaction free energy profiles for CO₂ photoreduction over TiO₂ and Cu₉S₅/TiO₂ composites, highlighting the key intermediates that drive the CO₂ reduction process. The energy profiles for COOH, CO, and CHO intermediates play a significant role in determining the overall efficiency of the reaction, particularly in methane generation. For TiO₂ (red line), the energy barrier for the COOH intermediate is relatively high at 0.8 eV, indicating a less favorable formation of CO and CHO, which are essential intermediates for methane production. The CO intermediate has a high energy value of 0.78 eV, while the CHO intermediate is similarly high, further limiting the pathway toward efficient methane formation.

In contrast, the Cu₉S₅/TiO₂ composite (blue line) significantly reduces the energy barriers for the critical intermediates. The COOH intermediate’s energy is lowered to 0.3 eV, facilitating easier conversion to CO and CHO. This reduced barrier encourages the formation of CO (with an energy value of 0.32 eV) and CHO (with an energy value of 0.12 eV), both of which are key precursors to methane formation. The CHO intermediate is particularly important, as its strong adsorption is a key step in methane production. Additionally, the strong adsorption of CO and CHO intermediates at the active sites on the Cu₉S₅/TiO₂ composite provides an ideal pathway for efficient methane production. This is evident in the lower energy profile for CH₄ formation on Cu₉S₅/TiO₂ compared to TiO₂. The lower energy for CH₄ formation on Cu₉S₅/TiO₂ directly supports its higher photocatalytic activity for CO₂ reduction to methane.

Figure 8b shows the band structure of TiO₂ and Cu₉S₅/TiO₂ composites, further supporting the enhanced photocatalytic performance. TiO₂ exhibits a conduction band edge at 0.54 eV, while the Cu₉S₅/TiO₂ composite has a lower conduction band edge at 0.43 eV, which facilitates more efficient electron transfer. This lowering of the conduction band edge enhances electron mobility, promoting better separation of charge carriers and leading to more efficient reduction of CO₂ to CH₄.

Together, these results demonstrate that the Cu₉S₅/TiO₂ composite not only facilitates the efficient formation of key intermediates such as CO and CHO but also significantly lowers the energy barriers for CH₄ production. This makes Cu₉S₅/TiO₂ a superior photocatalyst for CO₂ reduction, especially for selective methane production. The strong adsorption of CO and CHO, coupled with the improved electronic structure, is crucial for its enhanced photocatalytic activity.

3. Conclusions

In summary, hydrazine-engineered Cu₉S₅/TiO₂ gel composites were successfully fabricated via a hydrothermal method using gel-derived TiO₂ as a porous framework. The resulting materials exhibited well-dispersed Cu₉S₅ nanocrystals and enhanced crystallinity, which significantly promoted CO₂ adsorption and photocatalytic activity. The optimized 0.36Cu₉S₅/TiO₂ composite achieved an outstanding methane production rate of 34 μmol·g⁻¹·h⁻¹ and CO production of 15 μmol·g⁻¹·h⁻¹, with a methane selectivity of 64.76% and an electron-based CH₄ selectivity of 88.02%. In contrast, the Cu₉S₈/TiO₂ control sample prepared without hydrazine hydrate generated only CO, indicating the critical role of hydrazine in phase regulation and activity enhancement. The improved performance of Cu₉S₅/TiO₂ is attributed to synergistic effects between the gel-derived TiO₂ matrix and hydrazine-modulated Cu₉S₅ domains. The hydrazine-assisted synthesis not only facilitated Cu₉S₅ phase formation and dispersion, but also optimized interfacial electronic structure, thereby enhancing charge carrier separation and intermediate adsorption. DFT calculations confirmed that Cu₉S₅/TiO₂ lowered the energy barriers for key intermediates (*COOH, *CO, and *CHO), enabling efficient methane formation via a deep reduction pathway. This study highlights the potential of gel-based composite photocatalysts for coupling CO₂ adsorption and selective conversion. The design strategy demonstrated here provides valuable insights into phase and interface engineering for efficient CO₂-to-CH₄ photoreduction, contributing to the development of sustainable energy and environmental technologies.

4. Materials and Methods

4.1. Materials

All reagents used in this study were of analytical grade and used without further purification. Titanium isopropoxide (TTIP), glacial acetic acid, nitric acid (HNO₃), and isopropyl alcohol (IPA) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Sodium hydroxide (NaOH, 0.1022 mol/L) was self-prepared in the laboratory. Thiourea (C₂H₅NS, AR, 99%) was purchased from Shanghai Maike Chemical Technology Co., Ltd. (Shanghai, China). Hydrazine hydrate (N₂H₄·H₂O, AR) was obtained from the Pharmaceutical Group Chemistry Reagent Co., Ltd. (Beijing, China). Deionized water (H₂O) was used for all synthesis, dispersion, and washing procedures. Ethanol (C₂H₅OH, AR) was purchased from Guangdong Linmin Chemical Reagent Co., Ltd. (Guangzhou, China). Copper Citrate (C₆H₅Cu₂O₇, AR) was obtained from Shanghai Maike Chemical Technology Co., Ltd. (Shanghai, China).

4.2. Microwave-Assisted Sol-Gel TiO₂ and CuSx/TiO₂

4.2.1. Synthesis of Gel-Derived TiO₂

The synthesis of TiO₂: The TiO₂ was prepared using a microwave-assisted sol–gel method to enhance uniformity [31]. Firstly, a homogeneous solution was prepared by mixing 4.8 mL of TTIP, 0.8 mL of glacial acetic acid, and 4 mL of IPA. Then, an acidic solution was prepared by heating 11 mL of Milli-Q water and 0.20 mL of concentrated nitric acid to 90 °C. The TTIP-containing solution was quickly added to the acidified water under vigorous stirring, resulting in a homogeneous milky white reaction mixture. This reaction mixture was transferred to a 60 mL microwave reaction vessel, and water was added to dilute the solution to a final volume of 44 mL. The mixture was irradiated in a microwave reactor (Sineo Microwave Chemistry Technology (ShangHai) Co., ltd. (Shanghai, China), UWave-2000) at 90 °C with a stirring rate of 1200 rpm for 10 min, resulting in a transparent TiO₂ hydrosol. Finally, the synthesized TiO₂ hydrosol was freeze dried to obtain loosely packed TiO₂ aggregates as the final product. The synthesis of TiO₂ involves the hydrolysis and condensation of titanium alkoxide (such as titanium isopropoxide) in the presence of an acid or base. It may follow the chemical reaction below:

TTIP + H₂O → TiO₂ + IPA

In this reaction, TTIP undergoes hydrolysis in the presence of water, forming TiO₂ and IPA as a byproduct.

4.2.2. Synthesis of Cu₉S₅/TiO₂ Composites

A series of Cu₉S₅/TiO₂ composites with varying Cu loadings (0.12, 0.24, 0.36, and 0.48 mmol) were prepared. For comparison, a control sample without hydrazine hydrate was also synthesized (yielding Cu₉S₈/TiO₂).

Typically, 12 mmol of the synthesized TiO₂ was dispersed in 50 mL of deionized water and ultrasonicated for 30 min to form a uniform suspension (designated as Suspension A). Meanwhile, aqueous copper citrate solutions of different concentrations were prepared (Solution B) and slowly added dropwise into Suspension A under continuous stirring. The mixture was stirred magnetically for 20 h to ensure sufficient interaction between TiO₂ and Cu species. Subsequently, 40 mL of 16.8 mM NaOH solution was added dropwise to adjust the pH and facilitate precursor precipitation, followed by 20 min of stirring. Afterward, the precipitates were separated by centrifugation, washed several times with deionized water, and redispersed in 60 mL of 8 mM thiourea solution. For Cu₉S₅ formation, 1 mL of 85% hydrazine hydrate was introduced into the dispersion. The resulting mixture was sealed in a 100 mL Teflon-lined reactor and heated at 160 °C for 7 h. Upon cooling, the final solid products were collected by centrifugation, thoroughly washed with anhydrous ethanol and deionized water, and dried in a vacuum oven at 70 °C. The as-obtained products were denoted as 0.12Cu₉S₅/TiO₂, 0.24Cu₉S₅/TiO₂, 0.36Cu₉S₅/TiO₂, and 0.48Cu₉S₅/TiO₂. The control sample synthesized under identical conditions but without hydrazine hydrate was labeled 0.36Cu₉S₈/TiO₂.

In the procedure above, the copper sulfide (Cu₉S₈) was formed through the following reactions:

9Cu²⁺ + 8S²⁻ → Cu₉S₈

Subsequently, hydrazine hydrate was added as a reducing agent, contributing to a small amount of reduced Cu⁺, and then converting Cu₉S₈ into Cu₉S₅. The hydrazine reduction process can be represented by the following reactions:

9Cu²⁺ + N₂H₄ + 5S²⁻ → Cu₉S₅ + N₂ + 4H⁺

4.3. Sample Characterization

The morphology and internal nanostructure of the prepared catalysts were investigated by scanning electron microscopy (SEM; Regulus 8100, Hitachi, (Tokyo, Japan) and high-resolution transmission electron microscopy (HRTEM; Talos F200X, Thermo Fisher Scientific, Waltham, MA, USA). Prior to HRTEM analysis, the samples were dispersed in anhydrous ethanol through ultrasonication, drop-cast onto carbon-supported copper grids, and naturally dried in air. Imaging was conducted at an accelerating voltage of 200 kV. Phase composition and crystallinity were determined via X-ray diffraction (XRD) using a SmartLab diffractometer (Rigaku, Tokyo, Japan) equipped with a Cu Kα radiation source (λ = 1.5406 Å). The measurements were conducted under conditions of 40 kV tube voltage and 30 mA current. Diffraction patterns were collected over a 2θ range of 5–90°, with a step increment of 0.02° and a scan speed of 10° per minute. Photoluminescence (PL) spectroscopy was used to evaluate the recombination characteristics of photogenerated carriers. The spectra were obtained on an FLS980 spectrometer (Edinburgh Instruments, Livingston, UK) under excitation at 325 nm.

Photoelectrochemical performance was assessed using a CHI660D electrochemical workstation (CH Instruments, Shanghai, China) in a conventional three-electrode configuration. For electrode preparation, 5 mg of catalyst was ultrasonically dispersed in a mixture of 2 mL ethanol and 10 μL of 5 wt% Nafion solution. A 100 μL aliquot of the resulting dispersion was deposited onto a 1 cm² indium tin oxide (ITO) glass slide and dried under ambient conditions. The working electrode was coupled with a platinum foil counter electrode and an Ag/AgCl reference electrode. All measurements were carried out in 0.5 M Na₂SO₄ electrolyte under light-off conditions unless otherwise specified.

4.4. Evaluation of CO₂ Photocatalytic Reduction

The photocatalytic CO₂ reduction performance was investigated using a CEL-PAEM-D88P photocatalysis analysis system (Au-light, Beijing China Education Au-light Technology Co., Ltd., (Beijing, China). For each test, 7 mg of photocatalyst was ultrasonically suspended in 25 mL of deionized water to ensure uniform dispersion. The suspension was then vacuum filtered through a Teflon membrane (pore size < 100 μm), and the resulting thin film was dried under vacuum at 60 °C for 30 min.

The prepared catalyst membrane was fixed onto a stainless-steel holder and placed at the center of a gas-tight quartz reaction cell. To provide a humid environment, 5 mL of deionized water was added to the base of the reactor. The chamber was then sealed and purged three times with high-purity CO₂ (99.999%) to eliminate residual air. After degassing, CO₂ was introduced to adjust the internal pressure to ~40 kPa. The reaction system was allowed to stabilize for 30 min using a built-in plunger pump to ensure homogeneous mixing of CO₂ and water vapor. A 300 W xenon arc lamp (PLS-SXE300+, Au-light, China) was employed as the irradiation source to simulate sunlight. The reactor temperature was kept constant at 5 °C throughout the reaction using a circulating water-cooling system to suppress thermal side effects. Evolved gas-phase products were continuously monitored using an integrated online gas chromatograph equipped with one thermal conductivity detector (TCD) and two flame ionization detectors (FIDs). This configuration enabled real-time, simultaneous detection of H₂, CO, CH₄, C₂H₄, and C₂H₆ under ambient pressure.

4.5. Calculation of Product Yield and Selectivity

The generation rates of gaseous products (H₂, CO, CH₄, C₂H₄, and C₂H₆) were quantified based on their concentrations measured by gas chromatography and normalized to catalyst mass and irradiation time. The formation rate ($R$) of a given product was calculated using the following equation:

$$R={\displaystyle \frac{n}{m\times t}}$$

(1)

where $R$ is the product yield (μmol g⁻¹ h⁻¹), $n$ is the amount of product formed (μmol), $m$ is the mass of photocatalyst used (g), and $t$ is the reaction time (h).

To evaluate the distribution of reduction products, the product selectivity (S) toward species was determined by

$$S_x={\displaystyle \frac{n_x}{\sum n_i}}\times 100\%$$

(2)

where $n_x$ is the amount of product, and $n_i$ is the total amount of all carbon-containing reduction products (e.g., CO, CH₄, C₂H₄, C₂H₆).

For further insight into the reduction pathway, the electron selectivity was calculated based on the number of electrons transferred per molecule of each product. The electron-based selectivity toward CH₄, for instance, is expressed as

$$e^{-}\text{-}{Selectivity}_{{CH}_4}={\displaystyle \frac{8 \times n_{{CH}_4}}{\sum \left(e_i \times n_i\right)}} \times 100\%$$

(3)

where $e_i$ is the number of electrons required for the formation of product $i$ (e.g., 2 for CO, 8 for CH₄, 12 for C₂H₄, etc.), and $n_i$ is the molar quantity of each product.

All values were derived from the integrated GC peak areas using calibrated response factors.

4.6. DFT Calculation Method

First-principles calculations based on density functional theory (DFT) were performed using the Vienna Ab initio Simulation Package (VASP, version 5.4.4). The interactions between core and valence electrons were described using the projector-augmented wave (PAW) method [35,36,37]. A plane-wave cutoff energy of 400 eV was applied for all calculations. The electronic self-consistent field (SCF) iterations were considered converged when the total energy change was below 1 × 10⁻⁵ eV. Geometry optimizations were carried out until the residual atomic forces were reduced below 0.03 eV·Å⁻¹. To investigate the interfacial behavior of adsorbed intermediates involved in CO₂ photoreduction, surface models were constructed for the TiO₂ (101) and CuSx/TiO₂ heterojunction interfaces. Each model was built using a four-layer slab structure, where the two bottom layers were fixed during both structural optimization and transition state calculations. A vacuum space of 15 Å was introduced perpendicular to the slab to eliminate interlayer interactions between periodic images. Thermodynamic properties and reaction energetics were obtained by computing the Gibbs free energy (G), which includes electronic, vibrational, enthalpic, and entropic contributions. The Gibbs free energy of a given species A was estimated using the following expression:

$$G_{\mathrm{A}(T, p)} = E_{\mathrm{total}, \mathrm{A}} + E_{\mathrm{ZPE}} + ∆H(0 \to T)-TS(T, P)$$

(4)

in which E_{total, A}, E_ZPE, $∆$H(0 → T), and S(T, P) are the total energy obtained by DFT calculation, the zero-point energy, the enthalpy change from 0 K to temperature T, and the entropy at temperature T and pressure P, respectively. All thermodynamic parameters were evaluated using the VASPKIT 1.3.6 post-processing toolkit [38].