ZIF-8/g-C₃N₄ photocatalysts: enhancing CO₂ reduction through improved adsorption and photocatalytic performance

Abstract

Nowadays, the widespread concern over controlling CO₂ emissions and mitigating the adverse effects of greenhouse gases on global climate has attracted significant attention. In this study, g-C₃N₄ was synthesized by thermopolymerizing urea. Subsequently, ZIF-8 was combined with g-C₃N₄ using an in situ deposition method, resulting in the fabrication of ZIF-8/g-C₃N₄ composite photocatalysts at various molar ratios. Effective incorporation of ZIF-8 into g-C₃N₄ suppressed the recombination of photogenerated electrons and holes, thereby enhancing CO₂ capture capacity and preserving light absorption capabilities. The ZIF-8/g-C₃N₄ composite demonstrates excellent photocatalytic performance for CO₂ reduction, where the optimized material exhibited a CO₂ adsorption capacity 1.52 times that of pure g-C₃N₄ and increased the conversion of CO₂ to CH₄ by more than sevenfold. This study harnesses the superior CO₂ adsorption properties of metal–organic frameworks to develop more efficient photocatalysts, enhancing CO₂ conversion efficacy and offering insights for developing efficient photocatalysts that utilize CO₂.

---

1 Introduction

Graphitic carbon nitride (g-C₃N₄) has emerged as a semiconductor material with significant potential in the realm of CO₂ photocatalysis, attributed to its distinctive advantages. This non-metallic semiconductor is composed of carbon and nitrogen, elements abundantly available on Earth. It features a relatively small bandgap of approximately 2.7 eV, enabling a high responsiveness to visible light. Additionally, g-C₃N₄ exhibits superior resistance to acids and bases and maintains its integrity under illumination. Its structure and properties can be readily tailored, offering commendable photocatalytic activity.¹ However, the prevalent issues currently include low photogenerated electron–hole separation efficiency and low CO₂ absorption capacity.² Therefore, the development of new materials or the enhancement of CO₂ photocatalytic reduction through doping, compositing to form heterojunctions, and other methods, represents a current research focus. In fact, several adsorbents, such as graphene, activated carbon, zeolites, and certain mesoporous materials, have been employed in conjunction with specific photocatalysts to augment the reduction of CO₂. However, the improvement achieved through these combinations is often quite limited. The catalysts currently in use commonly suffer from issues such as light shielding, photocapture of charges, occupation of redox active sites, and non-selective adsorption of reactants and products.³

Metal–organic frameworks (MOFs) are a new class of crystalline porous materials with periodic network structures, self-assembled from metal clusters or metal ions and organic ligands through coordination bonds. They hold significant promise in applications such as adsorption and catalysis. MOF materials have demonstrated strong capabilities in CO₂ adsorption. Based on this, our experiment proposes that combining MOFs with semiconductor photocatalysts can significantly enhance CO₂ capture performance, thereby increasing the efficiency of photocatalytic CO₂ reduction.⁴ ZIF-8 (zeolitic imidazolate framework-8) is a type of MOF material with a zeolite topology structure formed by the coordination of Zn²⁺ ions with 2-methylimidazole. ZIF-8 possesses a large specific surface area, high porosity, regular pore channels, good thermal stability, water resistance, CO₂ affinity, and the combined advantages of traditional zeolite molecular sieves and MOFs in terms of chemical stability. Furthermore, it has two other features: one aspect is associated with the inherent transparency of ZIF-8.⁵ The integration of ZIF-8 into semiconductor matrices scarcely detracts from their photonic capture capabilities. Another aspect is associated the selective adsorptive properties of ZIF-8. ZIF-8 exhibits a pronounced preference for adsorbing CO₂, the primary reactant, while maintaining a low affinity towards the resultant reduction products, such as CH₄ and CH₃OH.⁶ Moreover, ZIF-8 can typically be synthesized through three common methods: synthesis in methanol solution;⁷ synthesis in a mixed solution of methanol and aqueous ammonia;⁸ and solvothermal synthesis in DMF.⁹ Research has found that ZIF-8(MeOH) has a larger external surface area and more acid–base sites, thus exhibiting significantly higher catalytic activity than ZIF-8(NH₄OH) and ZIF-8 (DMF).¹⁰ Therefore, this study chose to synthesize ZIF-8 by the stirring method and introduced g-C₃N₄ on this basis, modifying ZIF-8 onto the surface of g-C₃N₄ and investigating its enhanced effect on CO₂ reduction performance. The in situ prepared ZIF-8/g-C₃N₄ composite material has stronger light capture ability, higher charge separation efficiency, and greater CO₂ absorption capacity. As a result, compared to traditional urea-thermal polymerized g-C₃N₄, the yields of CO and CH₄ are significantly improved, with the yield of CH₄ increasing more than sevenfold.

2 Experimental

2.1 Chemical

All reagents with analytical reagent grade were used as received without further purification. Urea, zinc nitrate hexahydrate, 2-methylimidazole, methanol were purchased from Sinopharm Chemical Reagent Co., Ltd. Deionized water was used throughout the experiment.

2.1.1. The synthesis of g-C₃N₄. g-C₃N₄ was prepared using a high-temperature condensation method, with urea as the precursor. The process involved placing urea in a covered Al₂O₃ crucible, heating from room temperature to 550 °C at a rate of 8 °C min⁻¹, maintaining this temperature for 3 hours, and allowing the material to naturally cool, resulting in the formation of g-C₃N₄.¹¹

2.1.2. The synthesis of ZIF-8/g-C₃N₄. Employing an in situ method to fabricate ZIF-8/g-C₃N₄ composites. To enhance the integration of ZIF-8 with g-C₃N₄, the experiment first involves the combination of Zn(NO₃)₂·6H₂O with g-C₃N₄. The specific steps are as follows: 0.63 g of zinc nitrate and 1.26 g of g-C₃N₄ are weighed and added to 60 mL of methanol, followed by ultrasonication for 1 hour, during which the zinc nitrate dissolves. Then, 60 mL of a methanol solution containing 1.6 g of 2-methylimidazole is quickly poured into the above solution, maintained under ultrasonication for 30 minutes, and then mechanically stirred for 8 hours. The mixed solution is centrifuged, and the resultant mixture is dried at 60 °C for 6 hours to obtain the ZIF-8/g-C₃N₄ composite photocatalytic material. To investigate the effects of different ZIF-8 to g-C₃N₄ ratios on the photocatalytic CO₂ reduction performance, experiments were designed with composite photocatalytic materials at different molar percentages of 20%, 40%, 60%, and 80%, denoted as Z20CN, Z40CN, Z60CN, and Z80CN, respectively Pure ZIF-8 was also prepared utilizing the procedure described previously, with the exclusion of g-C₃N₄ addition.

2.2 Characterization of photocatalyst

The phase structure of the catalysts obtained from the experiment was qualitatively analyzed using X-ray diffraction (XRD), with the measurements carried out using a D8-ADVANCE X-ray diffractometer from Bruker, Germany. The micro-morphology of the catalysts was observed using a dual-beam field emission scanning electron microscope (SEM), specifically a GAIA 3 XMN model from TESCAN, Czech Republic, equipped with an energy dispersive X-ray (EDX) scanner for quantitative and qualitative analysis of the elemental distribution within the catalysts. Prior to testing, gold sputtering was performed using a JFC-1600 ion sputter from JEOL Ltd, Japan. High-performance imaging X-ray photoelectron spectroscopy (XPS) was used for qualitative analysis of the composition and chemical states of the catalysts, with measurements taken using an ESCALAB 250Xi from Thermo Fisher Scientific, USA. Binding energies of all samples were calibrated to the C 1s peak at 284.8 eV. FT-IR was employed to analyze the structure and functional groups of the catalysts, using an FT-IR 1200 spectrometer from ST Japan, with pure KBr as the background and a scanning range of 4000 to 500 cm⁻¹. Ultraviolet-visible diffuse reflectance spectroscopy (UV-vis) was used to analyze the light absorption properties of semiconductor materials, their electronic structures, and to calculate the band gap width of semiconductor materials. The measurements were carried out using a U-3900 spectrophotometer from Hitachi, Japan, with a test wavelength range of 200–800 nm. Photoluminescence spectroscopy (PL) was utilized to analyze the migration, transfer, and separation efficiency of photo-generated electron–hole pairs in semiconductors, with the characterization performed using an F-4600 high-performance imaging X-ray photoelectron spectrometer from Hitachi, Japan, obtaining photoluminescence spectra of the samples under an excitation wavelength of 370 nm. Thermogravimetric analysis was conducted in an oxidative atmosphere utilizing a DTG-60H analyzer from Shimadzu, Japan. Nitrogen adsorption–desorption isotherms were quantified using an ASAP2460 from Micromeritics, USA. CO₂ adsorption measurements were performed at room temperature using Micromeritics ASAP 2460 from Micromeritics, USA. Electrochemical testing of the photoelectrochemical properties was performed using a 1260 A + 1287 A electrochemical workstation from Solartron Analytical, UK, with the excitation light source provided by a PLS-SXE300 xenon lamp from Beijing Perfectlight Technology Co., Ltd, and transient photocurrent tests were conducted at 0.1 V for a duration of 540 s.

2.3 Examination of photocatalytic CO₂ reduction performance

In this experiment, a typical gas–solid CO₂ reduction reaction was selected.¹² It is to prevent the impact of water photolysis on the experiment. If a hydrogen source is introduced, the photogenerated charges and energy produced by the catalyst would suffice to meet the demands for both photocatalytic water splitting and CO₂ reduction.¹³ The experimental procedure is as follows: the photocatalytic CO₂ reduction performance was evaluated in a 250 mL photoreactor equipped with an optical quartz window. The opening of the reactor was sealed with a silicone stopper, and a 300 W xenon lamp simulating sunlight served as the light source, positioned approximately 15 cm above the reactor. 0.5 mL of H₂O and 50 mg of catalyst powder were added to the sealed glass reactor, uniformly dispersed on a glass plate. Before illumination, the reactor was evacuated and then filled with 20 mL of high-purity CO₂ (99.999%). After 3 hours of light irradiation, the products were analyzed using a gas chromatograph (GC-2014C, Shimadzu, Japan) equipped with a flame ionization detector (FID) and an automatic sampler. The reported gas volumes represent the average of three experimental trials.

3 Results and discussion

3.1 Characterization of the photocatalysts

X-ray diffraction analysis was conducted to study the crystal structure of each sample, with the results shown in (Fig. 1). g-C₃N₄, Z20CN, and Z40CN all exhibited characteristic diffraction peaks of g-C₃N₄ at 12.8° and 27.6° (JCPDS Card No. 87-1526),¹⁴ corresponding to the (100) and (002) planes of g-C₃N₄, respectively.⁴⁰ The (100) plane reflects the in-plane periodic arrangement of the tri-s-triazine ring structure of the g-C₃N₄ molecular skeleton, while the (002) plane indicates the interlayer stacking of conjugated aromatic ring structures.¹⁵ The strong and pure diffraction peaks of ZIF-8 indicate its high crystallinity. Multiple characteristic crystal planes of ZIF-8, including the (010), (002), (112), and (022) planes, can be observed in the XRD pattern, consistent with previous reports.¹⁶ The standard peaks of the synthesized ZIF-8 were also compared with the simulated ZIF-8 (JCPDS 00-062-1030) and are depicted in Fig. S1.† The diffraction peaks of composites doped with different ratios of ZIF-8 show slight variations. In Z20CN and Z40CN, the (002) plane of g-C₃N₄ is clearly visible, due to the lower doping amount of ZIF-8; however, in Z60CN and Z80CN, the intensity of the (002) peak gradually decreases. Notably, the position of the (002) peak shifts towards a lower angle, indicating an increase in the lattice spacing due to the addition of ZIF-8.¹⁷ As the amount of ZIF-8 increases, the diffraction peaks of ZIF-8 become more prominent while the characteristic peaks of g-C₃N₄ gradually weaken, suggesting an effective composite formation between ZIF-8 and g-C₃N₄.

Fig. 1 XRD patterns of diverse photocatalysts.

Field emission scanning electron microscopy (FESEM) and transmission electron microscopy (TEM) were employed to investigate the microstructure of g-C₃N₄, ZIF-8, and Z60CN. As depicted in (Fig. 2a), g-C₃N₄ exhibits a layered stacking of agglomerated clusters,¹⁸ characterized by a relatively loose structure with irregular morphology and size. (Fig. 2b) shows the rhombic dodecahedral morphology of ZIF-8, with microcrystals measuring 80–100 nm in diameter, demonstrating well-crystallized particles, regular shapes, and uniform dispersion.¹⁹ (Fig. 2c) reveals the uniform distribution of ZIF-8 octahedral cubic particles on the surface of g-C₃N₄, with the agglomerated clusters of g-C₃N₄ in the composite product undergoing fragmentation, thereby enhancing the overall particle dispersion and uniformity. The TEM image of Z60CN (Fig. 2d) further confirms the uniform deposition of ZIF-8 on g-C₃N₄ sheets, indicating a tight integration between the two. The energy-dispersive X-ray spectroscopy (EDS) elemental maps of Z60CN (Fig. 2e–h) clearly demonstrate that, beyond the presence of carbon (C) and nitrogen (N), zinc (Zn) is also uniformly distributed within the composite material.²⁰ These outcomes decisively validate the formation of an effective composite between g-C₃N₄ and ZIF-8, aligning with the X-ray diffraction (XRD) observations detailed earlier.

Fig. 2 SEM images of (a) g-C₃N₄ (b) ZIF-8 (c) Z60CN; (d) TEM images of Z60CN; (e–h) the EDX elemental maps of Z60CN.

Thermogravimetric analysis (TGA) under air was utilized to ascertain the actual ZIF-8 loading in the samples. As Fig. 3a illustrates, the three samples demonstrated excellent thermal stability with less than 10% mass loss between 0–350 °C, likely due to the loss of impurities adhered to the surface or within pores, and water adsorbed on the surface. Decomposition of g-C₃N₄ commenced at approximately 550 °C, culminating in complete sublimation and decomposition. ZIF-8 exhibited a weight loss of around 70% from 350 °C to 600 °C, indicative of the metal–organic framework's collapse, leaving ZnO as the residue. Within the same temperature range, Z60CN showed an 85% weight loss, with ZnO being the remaining product, which correlates with ZIF-8 deposited on the Z60CN surface. With ZIF-8 molecular weight at 227.6 g mol⁻¹, including one Zn atom (65.4 g mol⁻¹), the additional ∼8% mass loss of Z60CN compared to g-C₃N₄ allowed for the calculation of the actual ZIF-8 loading on the composite Z60CN to be 63.3%. This confirms the actual presence of ZIF-8 in the prepared samples aligns with the designed values.

Fig. 3 (a) TG analyses of sample ZIF-8, g-C₃N₄ and Z60CN; (b) the FT-IR spectra of different photocatalysts.

Fig. 3b delineates the Fourier-transform infrared (FT-IR) spectral profiles for a suite of photocatalysts. The spectral signatures of g-C₃N₄ and ZIF-8/g-C₃N₄ composites exhibit congruent stretching vibrational modes, evidencing that the quintessential lattice structure of g-C₃N₄ is retained after surface modification. The absorption peaks observed in the range of 3000–3500 cm⁻¹ for g-C₃N₄ and the ZIF-8/g-C₃N₄ composite samples are attributed to the stretching vibrations of the NH and NH₂ functional groups,^3,21 while the peaks from 1200 to 1630 cm⁻¹ are due to the typical stretching vibrations of the g-C₃N₄ heterocycles. The absorption peak at 815 cm⁻¹ for both g-C₃N₄ and the composite samples corresponds to the bending vibrations of the triazine structure in g-C₃N₄.⁴² The absorption peaks in the range of 1350–900 cm⁻¹ and at 800 cm⁻¹ are assigned to the in-plane and out-of-plane bending vibrations of the imidazole rings.²² Notably, the absorption peak at 815 cm⁻¹ for the ZIF-8/g-C₃N₄ composite samples shifts to a higher wavenumber, which may originate from structural distortions of the triazine ring units in ZIF-8/g-C₃N₄. Additionally, a new characteristic peak at 422 cm⁻¹ appears for ZIF-8 and its composites, corresponding to the stretching vibration of the Zn–N bond,²³ further corroborating the effective composite formation between the two materials by the in situ method.

Fig. 4a–d display the XPS spectra of g-C₃N₄ and Z60CN. As shown in the full scan spectrum in Fig. 4a. C, N, and O elements were detected in both g-C₃N₄ and Z60CN, with the O 1s peak plausibly attributable to the adsorption of CO₂ and H₂O molecules on the sample's surface. The characteristic peak of C 1s at 284.8 eV serves as a charge correction reference for the analysis of the elements. Fig. 4b shows the high-resolution C 1s spectrum, where the characteristic peak in g-C₃N₄ at a binding energy of 287.80 eV corresponds to the sp² hybridized orbitals of aromatic N=C–N.²⁴ In Z60CN, this peak shifts by 0.05 eV to a lower binding energy at 287.75 eV, suggesting effective composite formation between ZIF-8 and g-C₃N₄ and indicating charge transfer or changes in the chemical environment on the surface of the triazine rings. Fig. 4c presents the high-resolution N 1s spectrum, with g-C₃N₄ fitted with two characteristic peaks at 398.45 eV and 400.00 eV, corresponding to the C=N–C within the triazine rings and the bridging nitrogen atoms N–(C)₃ hybridized orbitals,^25,26 respectively. The peaks in Z60CN shift to 398.6 eV and 400.2 eV, indicating that the chemical state of N is influenced by the composite of g-C₃N₄ and ZIF-8. Fig. 4d reveals two distinct peaks at 1022.1 eV and 1045.2 eV, corresponding to Zn 2p_3/2 and Zn 2p_1/2, respectively. The observed chemical shifts in the carbon and nitrogen elements in Z60CN relative to g-C₃N₄ further indicate the effective integration of ZIF-8 into the g-C₃N₄ lattice, in alignment with the findings from X-ray diffraction analysis.

Fig. 4 (a) XPS full scan spectrum of g-C₃N₄ and ZIF-8/g-C₃N₄, (b) C ls, (c) N ls, (d) Zn 2p.

The UV-vis absorption spectra of the samples (Fig. 5a) reveal that the absorption capacity of ZIF-8/g-C₃N₄ in the ultraviolet and visible light regions is enhanced compared to g-C₃N₄, with the absorption edge of each experiencing a red shift. Notably, the absorption edge of Z60CN shifts to 456 nm, indicating a significant improvement in visible light absorption. The bandgap energy (E_g) can be calculated using:

αhv = A(hv − E_g)^(n/2)

Fig. 5 (a) UV-vis and corresponding relationship between optical energy changes of different samples, (b) PL.

The calculated bandgap energies for g-C₃N₄ and Z60CN are 2.76 eV and 2.71 eV, respectively, indicating that the integration of ZIF-8 reduces the bandgap of g-C₃N₄ by 0.05 eV. Thus, the red shift and the narrowing of the bandgap in Z60CN suggest that it can excite electrons from the valence band to the conduction band at lower energies than g-C₃N₄, potentially enhancing the efficiency of CO₂ reduction during photocatalytic processes.

Fig. 5b presents the photoluminescence (PL) spectra of different photocatalysts. It is observed that the fluorescence intensity of ZIF-8/g-C₃N₄ composites is significantly lower than that of g-C₃N₄, indicating a much slower recombination rate of electron–hole pairs in the composites. This slow recombination facilitates more effective separation and migration of electron–hole pairs.²⁷ Therefore, the introduction of ZIF-8 markedly suppresses the recombination of charge carriers within g-C₃N₄. This suppression contributes to a higher utilization rate of electron–hole pairs,⁴¹ thereby enhancing the activity of the photocatalyst and leading to a more efficient photocatalytic reaction.²⁸ ZIF-8 not only improves the light absorption properties of g-C₃N₄ but also boosts the photocatalytic efficiency through the diminution of charge carrier recombination rates.

Fig. 6 displays the transient photocurrent response of g-C₃N₄ and Z60CN. The transient photocurrent intensity of Z60CN is notably higher than that of g-C₃N₄, indicating that a greater number of electrons per unit time pass through the Z60CN electrode, which suggests a higher efficiency of photogenerated carrier separation. This is in agreement with the fluorescence test results discussed earlier.

Fig. 6 Transient photocurrent response graphs of g-C₃N₄ and Z60CN.

3.2 Adsorption performance

Nitrogen adsorption measurements were employed to characterize the pore structure and specific surface area of the synthesized samples. A larger specific surface area in catalysts typically indicates more adsorption and active sites, which may translate to higher catalytic performance. Fig. 7a presents the adsorption–desorption isotherms of g-C₃N₄ and Z60CN, both of which exhibit type IV classification.^29,30 The presence of H₃ hysteresis loop in the relative pressure (P/P₀) range of 0.60–0.95 suggests the presence of mesopores.³¹ The hysteresis loop observed in the isotherms extends across a substantial range of relative pressures, signifying a heterogeneous mesopore size distribution. This is substantiated by the commensurately broad pore-size distributions delineated by the Barrett–Joyner–Halenda (BJH) analysis. Table 1 enumerates the specific surface areas, pore volumes, and average pore diameters for each of the samples. Overall, in this study, the construction of a composite photocatalyst was demonstrated to increase the specific surface area of g-C₃N₄, resulting in more small-sized mesopores.³²

Fig. 7 (a) N₂ adsorption–desorption isotherms and pore size distribution curves of sample CN and Z60CN (b) CO₂ adsorption curves of the sample CN, ZIF-8 and Z60CN.

Table 1 Comparison of the SBET, pore volume, average pore size, CO₂ adsorption

Sample	SBET (m^2 g^−1)	Pore volume (cm^3 g^−1)	Average pore size (nm)	CO2 adsorption (mmol g^−1)
g-C3N4	73.87	0.41	22.40	0.21
Z60CN	363.41	0.43	4.80	0.32
ZIF-8	1182.69	0.78	2.64	0.91

---

There is ample literature documenting that the adsorption of CO₂ on the surface of photocatalysts is a primary key step affecting the pathways and kinetics of photocatalytic CO₂ reduction reactions.³³ An in-depth study of the CO₂ adsorption performance of g-C₃N₄, Z60CN, and ZIF-8 samples was conducted via CO₂ adsorption isotherms shown in Fig. 7b. Notably, all samples exhibit a linear relationship between CO₂ adsorption capacity and relative pressure (P/P₀) in the range of 0.1 to 1.0.³⁴ This relationship suggests that the interactions between CO₂ and the samples are predominantly physical.³⁵ In the relative pressure (P/P₀) range of 0.2–1.0, the CO₂ adsorption capacity of Z60CN is significantly higher than that of pure g-C₃N₄, which is attributable to Z60CN higher specific surface area, thereby exhibiting more pronounced CO₂ adsorption. In summary, the introduction of ZIF-8 to modify the g-C₃N₄ photocatalyst effectively tunes its adsorption properties.³⁶

3.3 Photocatalytic CO₂ reduction performance

In a typical gas–solid system, the photocatalytic CO₂ reduction capabilities of all fabricated samples were assessed under irradiation from a 300 watt xenon lamp without an ultraviolet filter. Control experiments confirmed that in the absence of both photocatalyst and light exposure, byproducts such as CO and CH₄ remained undetected. Although it has been demonstrated in recent years that some MOFs exhibit photocatalytic activity for CO₂ reduction under light irradiation,³⁷ no organic products were detected during prolonged CO₂ photoreduction using pure ZIF-8 as the photocatalyst. This observation that, within the context of this investigation, the presence of both light irradiation and the photocatalyst are essential prerequisites for the effective of CO₂ reduction.

As shown in Fig. 8(a) and (b), as marked increase in the photocatalytic activity for generating CO and CH₄ was achieved with g-C₃N₄ photocatalysts composited with ZIF-8. The total yield of CO and CH₄ increased with the extension of light irradiation time. For Z20CN composited with ZIF-8, the yield of CO production increased by about 70% after 3 hours (Fig. 8a). Meanwhile, for Z60CN composited with ZIF-8, the yield of CH₄ production could be increased by more than 7 times after 3 hours (Fig. 8b). It is important to note that the amount of ZIF-8 composite is crucial for photocatalytic CO₂ reduction. As delineated in earlier discussions, the addition of ZIF-8 substantially amplifies CO₂ capture, serving as an essential preliminary phase to the ensuing photocatalytic conversion of CO₂ into CO and CH₄. It is generally observed that an increase in the CO₂ adsorption capacity of a semiconductor photocatalyst is directly correlated with an improvement in its photocatalytic CO₂ reduction efficacy. Therefore, initially, as the amount of ZIF-8 composite increases, the yield of photocatalytic CH₄ continuously increases. However, the amount of ZIF-8 composite seems to have a dual effect on the performance of photocatalytic CH₄ production. When the molar ratio of ZIF-8 to g-C₃N₄ is 6 : 4 (sample Z60CN), the highest yield of photocatalytic CH₄ production after 3 hours is achieved, reaching 2.56 μmol g⁻¹. Although the introduction of ZIF-8 can enhance CO₂ adsorption, the poor conductivity of ZIF-8 may inhibit surface charge transfer, as also confirmed by the analysis of fluorescence spectra and transient photocurrent responses.⁴³ To assess the durability of the ZIF-8/g-C₃N₄ composite photocatalyst, we recovered the catalyst after the reactions and conducted XRD comparisons, as shown in Fig. 8(c). The results indicate that the ZIF-8/g-C₃N₄ composite maintains excellent stability for the photocatalytic reduction of CO₂, showing no significant changes in the crystal structure, which suggests its high durability in photocatalytic applications.

Fig. 8 (a) Comparison of CO production rates for the prepared samples. (b) Comparison of CH₄ production rates for the prepared samples. (c) XRD patterns of Z60CN before and after the catalytic reactions.

In this investigation, across the spectrum of synthesized photocatalysts, CO₂ is predominantly converted into CO and CH₄ as the primary outputs in the photocatalytic reduction process. The possible mechanism for the formation of CH₄ can be described by the following reaction formula,³⁸ based on the presence of different intermediate products:

H₂O + H⁺ → OH⁻ + H⁺

2CO² + 4e⁻ + 4h⁺ → 2CO² + O₂

CO +6e⁻ + 6H⁺ → CH₄ + H₂O

Based on the results above, a possible degradation mechanism was proposed in Fig. 9. Although the ZIF-8 in the composite cannot be excited by the visible light, the g-C₃N₄ component can be excited. Under illumination, g-C₃N₄ can be excited to generate electron–hole pairs, with the excited electrons transitioning from the valence band (VB) to the conduction band (CB). The conduction band of ZIF-8 is more negative compared to that of g-C₃N₄, prompting the electrons to migrate towards the energetically favorable conduction band of ZIF-8. Given ZIF-8's large surface area and high porosity, it exhibits strong CO₂ adsorption properties. Thus, the electrons flowing to ZIF-8 interact with the adsorbed CO₂. Since the CB potential of ZIF-8 (−2.22 eV) is more negative than the potential of the CO₂/CH₄ redox couple,³⁹ the accumulated electrons in the CB of ZIF-8 can effectively reduce CO₂ to produce CH₄ and CO.

Fig. 9 Mechanism diagram of ZIF-8/g-C₃N₄ photocatalytic CO₂ reduction.

4 Conclusions

The ZIF-8/g-C₃N₄ series of composite photocatalytic materials was synthesized using an in situ deposition method. The designed composite photocatalytic materials largely inherit the optical absorption characteristics of g-C₃N₄, enhance photocatalytic redox capability, improve efficiency of charge separation, and possess excellent CO₂ adsorption capacity. The findings demonstrate that the Z60CN composite photocatalyst displays enhanced photocatalytic performance, reaching a methane production of 2.56 μmol g⁻¹ in 3 hours, which is 7 times that of g-C₃N₄, without the addition of a sacrificial agent. Z60CN shows lower photoluminescence (PL) intensity and higher photocurrent density than g-C₃N₄, suggesting that this structure is conducive to interfacial charge transfer. Moreover, ZIF-8's limited inherent conductivity impedes charge transfer within the system. Notably, as the proportion of ZIF-8 in the composite increases, the photocatalytic efficiency of CO₂ reduction for Z80CN declines compared to Z60CN. This illustrates a trade-off between charge transfer efficiency and CO₂ capture capabilities in the composites. Therefore, the precise quantification of ZIF-8 is pivotal in enhancing photocatalytic efficacy. This research offers fresh perspectives on the development of effective CO₂ photocatalysts, leveraging the synergistic interactions between semiconductor nanostructures and metal–organic frameworks.

Author contributions

Y. H. S. and L. Y. synthesized the samples and performed the performance tests. R. F. W and S. L. A. conceived the idea, coordinated all stages of this research, and supervised the work. Y. H. S. and Z. H. wrote the manuscript. All authors contributed to the manuscript.

Conflicts of interest

The authors declare no competing interests.

Acknowledgements

This work was supported by the Inner Mongolia Autonomous Region Science and Technology Program (2021GG0042); Inner Mongolia Autonomous Region Youth Science and Technology Excellence in Higher Education (NJYT22064); Inner Mongolia Autonomous Region Natural Science Foundation Program (2022MS05018) Basic research business cost project in universities (0406082217).

References

1. Y. Zhang, J. Zhou and Q. Feng, et al., Visible light photocatalytic degradation of MB using UiO-66/g-C₃N₄ heterojunction nanocatalyst, Chemosphere, 2018, 212, 523–532.
2. S. Liu, F. Chen and S. Li, et al., Enhanced photocatalytic conversion of greenhouse gas CO₂ into solar fuels over g-C3N4 nanotubes with decorated transparent ZIF-8 nanoclusters, Appl. Catal., B, 2017, 211, 1–10.
3. J. Wen, J. Xie and X. Chen, et al., A review on g-C₃N₄-based photocatalysts, Appl. Surf. Sci., 2017, 391, 72–123.
4. A. Dhakshinamoorthy, A. M. Asiri and H. Garcia, Metal–organic framework (MOF) compounds: photocatalysts for redox reactions and solar fuel production, Angew. Chem., Int. Ed., 2016, 55, 5414–5445.
5. S. Liu, J. Wang and J. Yu, ZIF-8 derived bimodal carbon modified ZnO photocatalysts with enhanced photocatalytic CO₂ reduction performance, RSC Adv., 2016, 6, 59998–60006.
6. Z. Y. Yeo, P. Y. Tan and S. Chai, et al., Continuous polycrystalline ZIF-8 membrane supported on CO₂-selective mixed matrix supports for CO₂/CH₄ separation, RSC Adv., 2014, 4, 52461–52466.
7. S. R. Venna, J. B. Jasinski and M. A. Carreon, Structural evolution of zeolitic imidazolate framework-8, J. Am. Chem. Soc., 2010, 132, 18030–18033.
8. C. Chizallet, S. Lazare and D. Bazer-Bachi, et al., Catalysis of transesterification by a nonfunctionalized metal−organic framework: acido–basicity at the external surface of ZIF-8 probed by FTIR and ab initio calculations, J. Am. Chem. Soc., 2010, 132, 12365–12377.
9. U. P. Tran, K. K. Le and N. T. Phan, Expanding applications of metal−organic frameworks: zeolite imidazolate framework ZIF-8 as an efficient heterogeneous catalyst for the knoevenagel reaction, ACS Catal., 2011, 1, 120–127.
10. Y. Lee, M. Jang and H. Cho, et al., ZIF-8: A comparison of synthesis methods, Chem. Eng. J., 2015, 271, 276–280.
11. F. Saman, H. Bahruji and A. H. Mahadi, et al., Pd/g-C₃N₄ photocatalyst for hydrogen production: Role of experimental condition for Schottky barrier, Fuel, 2023, 349, 128725.
12. K. M. Kamal, R. Narayan and N. Chandran, et al., Synergistic enhancement of photocatalytic CO₂ reduction by plasmonic Au nanoparticles on TiO₂ decorated N-graphene heterostructure catalyst for high selectivity methane production, Appl. Catal., B, 2022, 307, 121181.
13. W. Dong, J. Jia and Y. Wang, et al., Visible-light-driven solvent-free photocatalytic CO₂ reduction to CO by Co-MOF/Cu₂O heterojunction with superior selectivity, Chem. Eng. J., 2022, 438, 135622.
14. J. Wang, J. Huang and H. Xie, et al., Synthesis of g-C₃N₄/TiO₂ with enhanced photocatalytic activity for H₂ evolution by a simple method, Int. J. Hydrogen Energy, 2014, 39, 6354–6363.
15. Y. Li, S. Wu and L. Huang, et al., Synthesis of carbon-doped g-C₃N₄ composites with enhanced visible-light photocatalytic activity, Mater. Lett., 2014, 137, 281–284.
16. N. Hassan, A. Shahat and A. El-Didamony, et al., Equilibrium, Kinetic and Thermodynamic studies of adsorption of cationic dyes from aqueous solution using ZIF-8, Moroccan J. Chem., 2020, 8, 3–8.
17. X. Guo, S. He and Z. Meng, et al., Ag@ZIF-8/g-C₃N₄ Z-scheme photocatalyst for the enhanced removal of multiple classes of antibiotics by integrated adsorption and photocatalytic degradation under visible light irradiation, RSC Adv., 2022, 12, 17919–17931.
18. X. Li, B. Wang and W. Yin, et al., Cu²⁺ modified g-C3N4 photocatalysts for visible light photocatalytic properties, Acta Phys. Chim. Sin., 2020, 36, 1902001.
19. B. Liu, C. Bie and Y. Zhang, et al., Hierarchically porous ZnO/g-C₃N₄ S-scheme heterojunction photocatalyst for efficient H₂O₂ production, Langmuir, 2021, 37, 14114–14124.
20. Y. Zhao, Y. Li and L. Chang, et al., Bimetal doped Cu–Fe-ZIF-8/g-C₃N₄ nanocomposites for the adsorption of tetracycline hydrochloride from water, RSC Adv., 2024, 14, 4861–4870.
21. M. Tahir, C. Cao and F. K. Butt, et al., Large scale production of novel g-C₃N₄ micro strings with high surface area and versatile photodegradation ability, CrystEngComm, 2014, 16, 1825–1830.
22. Y. Hu, H. Kazemian and S. Rohani, et al., In situ high pressure study of ZIF-8 by FTIR spectroscopy, Chem. Commun., 2011, 47, 12694–12696.
23. L. Lin, T. Zhang and H. Liu, et al., In situ fabrication of a perfect Pd/ZnO@ZIF-8 core–shell microsphere as an efficient catalyst by a ZnO support-induced ZIF-8 growth strategy, Nanoscale, 2015, 7, 7615–7623.
24. C. Chen, Y. He and G. Xiao, et al., Graphic C₃N₄-assisted dispersion of graphene to improve the corrosion resistance of waterborne epoxy coating, Prog. Org. Coat., 2020, 139, 105448.
25. H. Cai, D. Han and X. Wang, et al., High specific surface area defective g-C₃N₄ nanosheets with enhanced photocatalytic activity prepared by using glyoxylic acid mediated melamine, Mater. Chem. Phys., 2020, 256, 123755.
26. C. Saka, Surface modification with oxygen doping of g-C₃N₄ nanoparticles by carbon vacancy for efficient dehydrogenation of sodium borohydride in methanol, Fuel, 2022, 310, 122444.
27. L. Jiang, X. Yuan and Y. Pan, et al., Doping of graphitic carbon nitride for photocatalysis: a review, Appl. Catal., B, 2017, 217, 388–406.
28. X. Liu, J. Zhang and Y. Dong, et al., A facile approach for the synthesis of Z-scheme photocatalyst ZIF-8/g-C₃N₄ with highly enhanced photocatalytic activity under simulated sunlight, New J. Chem., 2018, 42, 12180–12187.
29. K. S. Sing, Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity (Recommendations 1984), Pure Appl. Chem., 1985, 57, 603–619.
30. L. Wang, P. Jin and S. Duan, et al., In-situ incorporation of Copper (II) porphyrin functionalized zirconium MOF and TiO₂ for efficient photocatalytic CO₂ reduction, Sci. Bull., 2019, 64, 926–933.
31. G. Tzvetkov, M. Tsvetkov and T. Spassov, Ammonia-evaporation-induced construction of three-dimensional NiO/g-C₃N₄ composite with enhanced adsorption and visible light-driven photocatalytic performance, Superlattices Microstruct., 2018, 119, 122–133.
32. Y. Zhou, L. Zhang and W. Huang, et al., N-doped graphitic carbon-incorporated g-C₃N₄ for remarkably enhanced photocatalytic H2 evolution under visible light, Carbon, 2016, 99, 111–117.
33. Q. Huang, J. Yu and S. Cao, et al., Efficient photocatalytic reduction of CO₂ by amine-functionalized g-C₃N₄, Appl. Surf. Sci., 2015, 358, 350–355.
34. J. Yu, S. Wang and B. Cheng, et al., Noble metal-free Ni (OH)₂-g-C₃N₄ composite photocatalyst with enhanced visible-light photocatalytic H₂-production activity, Catal. Sci. Technol., 2013, 3, 1782–1789.
35. Z. Sun, Z. Yang and H. Liu, et al., Visible-light CO₂ photocatalytic reduction performance of ball-flower-like Bi₂WO₆ synthesized without organic precursor: effect of post-calcination and water vapor, Appl. Surf. Sci., 2014, 315, 360–367.
36. K. K. Han, C. C. Wang and Y. Y. Li, et al., Facile template-free synthesis of porous g-C₃N₄ with high photocatalytic performance under visible light, RSC Adv., 2013, 3, 9465–9469.
37. D. Li, M. Kassymova and X. Cai, et al., Photocatalytic CO₂ reduction over metal-organic framework-based materials, Coord. Chem. Rev., 2020, 412, 213262.
38. O. K. Varghese, M. Paulose and T. J. LaTempa, et al., High-rate solar photocatalytic conversion of CO₂ and water vapor to hydrocarbon fuels, Nano Lett., 2009, 9, 731–737.
39. C. Chen, J. Jin and S. Chen, et al., In situ growth of ultrafine ZnO on g-C₃N₄ layer for highly active and selective CO₂ photoreduction to CH4 under visible light, Mater. Res. Bull., 2021, 137, 111177.
40. H. He, Z. Luo and C. Yu, Diatomite-anchored g-C₃N₄ nanosheets for selective removal of organic dyes, J. Alloys Compd., 2020, 816, 152652.
41. Y. Yang, J. Wu and B. Cheng, et al., Enhanced photocatalytic H₂-production activity of CdS nanoflower using single atom Pt and graphene quantum dot as dual cocatalysts, Chin. J. Struct. Chem., 2022, 41, 2206006–2206014.
42. C. Yu, H. He and W. Zhou, et al., Novel rugby-ball-like Zn₃ (PO₄)₂@C₃N₄ photocatalyst with highly enhanced visible-light photocatalytic performance, Sep. Purif. Technol., 2019, 217, 137–146.
43. F. Hong-Ru, Y. Jiang and L. Jia-Hua, et al., A robust heterometallic Cd(II)/Ba(II)-organic framework with exposed amino group and active sites exhibiting excellent CO₂/CH₄ and C₂H₂/CH₄ separation, Chin. J. Struct. Chem., 2022, 41, 2203287–2203292.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ra02548b

---