Constructing a rhenium complex supported on g-C₃N₄ for efficient visible-light-driven photoreduction of CO₂ to CO via a novel Z-scheme heterojunction

Abstract

Visible-light-driven photocatalytic CO₂ reduction is a promising approach to addressing the problem of global warming and the energy crisis. A Z-scheme photocatalyst comprising a Re(I) complex and a polymeric semiconductor (bulk graphitic carbon nitride (g-C₃N₄)) converted CO₂ to CO even under the irradiation of low-intensity visible light. The electronic interaction between Re(I) and g-C₃N₄ units and the injection of electrons from g-C₃N₄ to the Re(I) complex improved the photocatalytic efficiency. The lower recombination of electron–hole pairs and the prolonged emission decay (the average emission lifetime τ = 5.8 ms) contributed to the increased overall efficiency of the hybrid system. The maximum turnover number (TON) of CO formation reached 28.56 after 240 min with a high CO selectivity (99.8%). Compared to the homogeneous photocatalyst Re(I), the TON of CO formation was increased by 5.6 times and CO selectivity was higher. Density Functional Theory (DFT) was employed to investigate the impact of the substrate-supported Re complex (Re(bpy-COOH)/g-C₃N₄) on the CO₂ Reduction Reaction (CO2RR) activity. The results indicate that Re(bpy-COOH)/g-C₃N₄ presents a lower energy barrier for *CO₂ to *COOH conversion, promoting *COOH formation more effectively than pure g-C₃N₄. Moreover, the energy barrier for *CO desorption is lower in Re(bpy-COOH)/g-C₃N₄ than in Re(bpy-COOH), suggesting a reduction in the strong binding of adsorbed CO intermediates, highlighting the intermediate ensemble effect at the active site. The improved CO2RR activity in Re(bpy-COOH)/g-C₃N₄ can be attributed to the balance between the enhancement of reaction activity and the binding energy of intermediates at the active site. The study of photocatalytic CO₂ reduction (PCO2R) under low-intensity-visible-light irradiation and DFT investigation has reveal an insight into the CO2RR activity for CO₂ reduction under various irradiation conditions.

---

1. Introduction

Carbon dioxide is known as a heat-trapping gas, which causes global heating. Heavy industries and human activities are the main CO₂-emission sources that primarily impact global climate change and relate to severe natural events, including wildfires, floods, and droughts. Concerning climate change, several ways have been attempted to reduce CO₂ concentration, such as using sustainable energy sources instead of burning fossil fuels and utilizing CO₂ as a new chemical feedstock for producing energy-rich carbon compounds. Solar energy conversion of CO₂ into carbon compounds has been a promising research field that can solve global warming, shortage of fossil fuels, energy crisis, etc. To this end, numerous investigations have been carried out to develop many different kinds of visible-light-driven photocatalysts. Transition metal complexes (e.g., Ru, Mn and Re) have been known to be potential homogeneous photocatalysts for CO₂ reduction with high selectivity and activity.^1–3 They possess a long-lived excited state,^4,5 tunable structure via changing the coordination ligand to the metal center,^6,7 and intense emission properties.⁸ However, transition metal complexes, which have low quantum efficiency⁹ and strong absorption in the UV region,⁴ are high-cost and scarce, and could suffer from progressive structural degradation because of photochemical instability that may severely reduce catalytic performances. For those reasons, various hybrid photocatalytic materials including molecular/semiconductor hybrid materials^10–15 and transition metal complex/nanoparticle¹⁶ hybrid materials have been introduced to overcome the drawbacks of homogeneous photocatalytic systems. [Re(2,2′-bipyridine-4,4′-bisphosphonic acid) (CO)₃(L)]ⁿ⁺ (ReP; L = 3-picoline or bromide) immobilized on TiO₂ nanoparticles¹⁷ was reported for a visible-light-driven hybrid displaying improvement in CO₂ reduction with the turnover number (TON) increasing 24 times in comparison with a ReP homogeneous photocatalyst (the TON of the ReP–TiO₂ hybrid was observed to be 48 mol_CO mol_Re⁻¹ in DMF). The metal–organic framework (MOFs supported Ru carbonyl complex¹⁸ for PCO2R under visible light showed twofold enhanced catalytic activity compared to the pure Ru carbonyl complex.

To date, the incorporation of metal complexes with a semiconductor to achieve visible-light-driven photocatalysts¹⁹ has attracted attention, and the metal complexes/semiconductor hybrid system¹⁵ can utilize the advantages of both metal complexes and semiconductors, as well as reducing the usage of noble metals. To design a metal complexes/semiconductor system, the conduction band of semiconductors and the reduction potential of metal complexes need to be compatible to maximize electron transfer between the two components for efficient CO₂ photocatalysis conversion. The rational structure can be achieved by adsorption or by covalent links between metal complexes and semiconductors. Generally, semiconductors play the role of a light absorber and host oxidation reaction sites. Depending on the conduction-band (CB) potential of the semiconductor and the lowest occupied molecular orbital (LUMO) of the metal complexes, the system can be categorized into a hybrid system in which semiconductors function as photosensitizers and metal complexes catalyze CO₂ when the CB potential of semiconductors is more negative than the LUMO of metal complexes, and a Z-scheme system when the CB potential of semiconductors is less negative than the LUMO of metal complexes. In the Z-scheme system, both the semiconductor and metal complex undergo photo-excitation together, resulting in electron transfer and enhancement in the driving force of interfacial electron transfer.

Graphitic carbon nitride has been known as a polymeric semiconductor with high chemical stability and a narrow band gap (∼2.7 eV, corresponding to an absorption edge of ≈460 nm), low cost, and ease of preparation.^20,21 C₃N₄ has been applied for photocatalytic reactions, including water reduction/oxidation²² and environmental treatment.^23–25 However, g-C₃N₄ still has common issues during a photocatalytic process such as insufficient visible-light absorption,²⁶ low quantum yield and fast recombination of photogenerated carriers^27,28 that restrict the photocatalytic performance. Therefore, heterojunctions based on g-C₃N₄ with two or three semiconductor materials have been tailored to integrate their own advantages. The g-C₃N₄ based heterojunctions have been used for PCO2R where the formed products were non-selective and abundant, containing long-chain hydrocarbons (CO, HCOOH, acetone, methanol, and acetaldehyde).^29,30 g-C₃N₄ has a conduction band and valence band at −1.4 eV and 1.34 eV (vs. NHE, pH = 7),^31,32 respectively, which are suitable for architecting a hybrid photocatalyst with metal complexes, and have thus been reported for the high selectivity for formed products. In a hybrid CO₂-reduction system, bulk g-C₃N₄ was explored as a semiconductor in conjunction with Ru complexes^11,33,34 through carboxylic or phosphonic acid anchoring groups for converting CO₂ to formic acid (HCOOH) under visible light (λ > 400 nm). The combination of g-C₃N₄ and Ru complexes maximized the CO₂-reduction efficiency process and improved photocatalytic performance and reached the highest TON (>1000). A manganese (Mn) complex with phosphonic acid anchoring groups in conjunction with bulk g-C₃N₄ forming a Z-scheme system³⁵ was reported for efficient visible-light photoreduction of CO₂ to CO, where the TON of the hybrid system was increased two-fold in comparison with the base Mn complex. Other than Ru and Mn complexes, Ni and Co complexes have been used to combine with g-C₃N₄ to design hybrid materials for PCO2R.^14,36,37 Meanwhile, Re(CO)₃(N–N)Cl complexes have not been reported for catalytic CO₂ reduction in metal complexes/semiconductor hybrid systems. They are appropriate to incorporate with g-C₃N₄ to form an efficient visible-light-driven photocatalyst for CO₂ conversion, because Re(CO)₃(N–N)Cl complexes have efficient CO₂-capturing ability, better than that of Mn complexes in the lower CO₂ concentration in the presence of TEOA solvent.³⁸ Re(CO)₃(bpy-COOH)Cl with a dicarboxylic acid anchor and more negative LUMO potential than the CB of g-C₃N₄ will form a Z-scheme system, which can enhance efficient CO₂ reduction. The adsorption is likely via hydrogen bonding between NH₂ groups on g-C₃N₄ and –COOH anchors. Additionally, the hybrid system can efficiently convert CO₂ under low-intensity irradiation which is important for investigating photocatalytic systems working under various illumination.

In this study, we report the straightforward synthetic process of a hybrid material Re(bpy-COOH)/g-C₃N₄. The characterization of chemical functional groups, chemical states, optical properties, and the morphology was performed by FTIR, XPS, UV-vis DRS, PL, and PL decay spectroscopy, and using SEM and TEM images. PCO2R to CO of this hybrid material was studied in DMF/TEOA mixed solution under low-intensity visible light irradiation (21 mW cm⁻²) for the first time. Utilizing density functional theory (DFT) calculations, we examined the influence of Re(bpy-COOH)/g-C₃N₄ on CO₂ Reduction Reaction (CO2RR) activities, including factors such as activation energies and charge transfer between intermediates and substrates. Our findings indicate that CO2RR in Re(bpy-COOH)/g-C₃N₄ is favorable, facilitated by an increase in reaction activity at the active site and a balanced binding energy of both *COOH and *CO intermediates. These results are critical in understanding the photocatalytic activity of this system under sunlight irradiation under various daytime conditions, providing a pathway for further research in this area.

2. Experimental section

Urea (NH₂CONH₂), pentacarbonylchlororhenium(I) [Re(CO)₅Cl], 2,2′-bipyridine-4,4′ dicarboxylic acid (bpy-COOH), and anhydrous N,N-dimethylformamide (DMF) were purchased from Sigma-Aldrich. Triethanolamine (TEOA – (HOCH₂CH₂)₃N), toluene, methanol, acetone, and hexane solvents were of HPLC grade and were obtained from Merck. All solvents were used without further purification. CO₂ and argon (Ar) gases used in all experiments were 99.999% purified.

Bulk g-C₃N₄ was prepared by calcining urea (20 g) under air at 550 °C for 2 hours at a temperature ramp rate of 2 °C min⁻¹ in a muffle furnace. The urea was placed in an alumina crucible with a cover. Then, the muffle furnace was cooled to room temperature, and the resulting light yellow powder was collected as bulk g-C₃N₄, which was ground finely. The Re(CO)₃(bpy-COOH)Cl complex was synthesized following a previous report.³⁹ The complex was synthesized by refluxing the mixture of [Re(CO)₅Cl] (200 mg, 0.55 mM) and 0.55 mM bpy-COOH ligand in 50 mL toluene for 6 hours under an Ar atmosphere to give a red-orange solution ([Re(CO)₃Cl(2,2′-bipyridine-4,4′ dicarboxylic acid)] Re(bpy-COOH)). Re(bpy-COOH) was recrystallized from acetone-hexane. Yield: 70%. ν_CO/cm⁻¹ (in MeCN): 2041, 1956, 1897, ¹H NMR (500 MHz, d₆ – acetone, ppm) δ = 9.33 (2H, d, bpy H₆, H_6′), 9.18 (2H, s, H₃, H_3’), 8.24–8.26 (2H, m, H₅, H_5′). ESI-MS (in MeCN): m/z: (M⁺ for [C₁₅H₈N₂O₇Re]⁺), 512.9 (M³⁺ for [C₁₅H₆N₂O₇Re]⁺. UV/Vis (in MeCN): 270–350 nm (ligand-to-ligand charge transfer (LLCT)), 350–475 nm (metal-to-ligand charge transfer (MLCT)) (ESI-MS and UV/Vis spectra are shown in Fig. S1b, S2, and S3†).

The hybrid photocatalyst Re(bpy-COOH)/g-C₃N₄ was prepared following a previously reported process.^33,35 The synthesized g-C₃N₄ (50 mg) was added to NaOH (50 mL, 0.1 M), stirring the mixture for 10 minutes. Then, the mixture was ultrasonically cleaned with ionized water (100 mL) and separated by centrifugation. Re(bpy-COOH)/g-C₃N₄ with a mol ratio of 1 : 100 was prepared by dispersion of g-C₃N₄ (46 mg) in methanol (20 mL) and the suspension was ultrasonicated for 1 hour. Then, 2.75 mg of Re(bpy-COOH) was added to g-C₃N₄ suspension and the mixture was stirred for 24 hours in the dark and under room temperature. Re(bpy-COOH)/g-C₃N₄ was collected by filtration and washed with methanol.

After Re(bpy-COOH)/g-C₃N₄ was prepared and collected by filtration. The filtrate was kept in a closed vial and its volume was known. The adsorbed amount (Ads_A) of Re(bpy-COOH) on g-C₃N₄ was calculated based on the UV/Vis absorption spectrum of the filtrate, using the following equation:

where A₁ and A₂ are the absorbance value of the Re(bpy-COOH) solution before and after the adsorption process, respectively, and C is the initial concentration of Re(bpy-COOH). The adsorbed amount of Re(bpy-COOH) on the surface of g-C₃N₄ in the prepared Re(bpy-COOH)/g-C₃N₄ was 31.5 μmol g⁻¹.

The crystalline structure of the prepared g-C₃N₄ was confirmed by using powder X-ray diffraction (XRD) patterns using the D8 Advance XRD with Cu K_α radiation (λ = 0.15418 nm), scanning from 10–70°. The chemical structure of the Re(bpy-COOH) complex was determined by using the proton nuclear magnetic resonance (¹H NMR) spectrum obtained using a Bruker Avance II (500 MHz) in acetone-D6. Electrospray ionization mass spectrometry (EIS-MS) of Re(bpy-COOH) was performed in acetonitrile (CH₃CN) using an X500R QTOF. The chemical functional groups were determined by Fourier transform infrared (FTIR) spectroscopy in KBr pellets using a PerkinElmer MIR/NIR Frontier instrument in the range of 4000–800 cm⁻¹. The chemical states of the materials were tested using a Thermo Scientific X-ray photoelectron spectroscopy (XPS) instrument with Al Kα radiation. The morphology was observed by using a scanning electron microscope (SEM – Hitachi SM- 4800) and transmission electron microscope (TEM- HRTEM, Tecnai G2 F20 S-Twin). The absorption of g-C₃N₄ and Re(bpy-COOH)/g-C₃N₄ was measured by UV-vis diffuse reflectance spectroscopy (UV-vis DRS) performed on a UV-2600 (Shimadzu) in the wavelength range of 300–700 nm. The photoluminescence (PL) spectra were identified on a Cary Eclipse Fluorescence Spectrophotometer (Varian, USA) at 370 nm at room temperature. The emission decay profiles of Re(bpy-COOH)/g-C₃N₄ and g-C₃N₄ at 450 nm were recorded by using a 0.6 m grating monochromator (Jobin-Yvon HRD1) and detected with a fast photomultiplier (Hamamatsu model H733, with a rise time of 700 ps), using a 355 nm laser diode as the excitation source.

Photoelectrochemical experiments: Photocurrent transient response and electrochemical impedance spectroscopy (EIS) measurements were carried out on a potentiostat Biologic VSP-300 using a three-electrode system cell with a working electrode, a Ni mesh counter electrode, and an Ag/AgCl as the reference electrode in a 0.5 M Na₂SO₄ electrolyte. The working electrode was prepared by coating the as-prepared samples (g-C₃N₄ and Re(bpy-COOH)/g-C₃N₄) on fluorine-doped tin oxide (FTO) glass. The FTO glass with a size of 2.5 × 2.5 cm was washed in the following order with acetone, ethanol, and DI water under sonication, and then dried in a N₂ flow. Dispersing the mixture of a certain amount of sample and polyvinyl alcohol (PVA) in distilled water, the obtained slurry was coated on a FTO glass substrate (photoactive area of 1.2 × 1.2 cm) by the spin-coating technique, and dried at 70 °C in 30 min. Photocurrent transient response was recorded under an applied bias of 0.35 V and irradiated with visible light (15 W energy-saving light bulk, center wavelength λ = 408 nm, and light intensity at a distance of 5 cm of 210 W m⁻², see Fig. S4†) with light chopping every 30 s. Electrochemical impedance spectroscopy (EIS) plots were obtained over the frequency range from 0.1 to 10⁵ Hz in the dark.

The photocatalytic CO₂ reaction was carried out in a reaction solution (0.75 mL) containing a photocatalyst (Re(bpy-COOH) (0.6 mM) and Re(bpy-COOH)/g-C₃N₄ (1 mg)) in DMF : TEOA mixed solvent (5 : 1 v/v). The reaction solution was placed in a visible-light-transparent vial (5 mL) and purged with Ar to remove all air, and then CO₂ gas was purged for 10 min. Then, the reaction vial was irradiated with a light-source system (15 W energy-saving light bulk, λ = 408 nm, and light intensity at a distance of 5 cm of 210 W m⁻², see Fig. S4†) in the photoreaction box. During irradiation, the solution was stirred slowly with a magnetic bar and the temperature of the solution was controlled at room temperature by using a constant temperature system. The gaseous reaction products were identified by using a gas chromatograph (GC Clarus 680- PerkinElmer), which was equipped with a TCD detector, molecular sieve column (L × I.D.: 30 2 m × 0.32 mm), and Ar carrier gas. The turnover number TON of CO and H₂, and the selectivity of CO were calculated by using eqn (2)–(4), respectively.

Spin-polarized density functional theory (DFT) calculations were performed using the Vienna ab initio simulation package (VASP).^40,41 The generalized gradient approximation (GGA) with parameterization by Perdew, Becke, and Ernzerhof (PBE)⁴² was employed to describe the exchange–correlation interactions. The interaction between core electrons and valence electrons was treated with the projector-augmented wave (PAW)⁴³ method. The DFT-D3 correction method in Grimme's scheme⁴⁴ was used to describe the long-range van der Waal interactions between atoms. A cutoff energy of 400 eV was chosen for wave function expansion. For Brillouin zone integration, only a Γ k-point with a Gaussian broadening of 0.05 eV was used. Crystal orbital Hamilton population (COHP) and charge transfer analyses were performed using the Lobster program.⁴⁵ To investigate the thermodynamics of CO2RR, the Gibbs free energy (ΔG) was calculated using the equation proposed by Nørskov's group:⁴⁶

ΔG = ΔE + ΔE_ZPE − TΔS + ΔG_pH + ΔG_U (5)

where ΔE is the total energy of reactions obtained from DFT calculation, and ΔE_ZPE and ΔS are the zero-point energy and entropic changes, respectively. T is the temperature, ΔG_pH is the correction on the pH in the electrolyte and ΔG_U is the electrode potential. In this study, ΔG_pH and ΔG_U were set to zero.

In our simulation model, the Re(bpy-COOH) molecule was placed in a cubic unit cell with a = 20 Å, while g-C₃N₄ and Re(bpy-COOH)/g-C₃N₄ were constructed using a 2 × 2 hexagonal unit cell with an optimized lattice constant a = 13.98 Å. A vacuum spacing of 15 Å was selected to minimize the artificial interaction between the neighboring slabs. All atomic coordinates are fully relaxed with a force criterion of 1 meV Å⁻¹.

3. Results and discussion

The 1H NMR and ESI-MS confirmed the high purity and the well-structured molecule of the Re(bpy-COOH) complex, as presented in Fig. S1 and S2.† The XRD pattern showed the characteristic diffraction peaks in the crystalline structure of g-C₃N₄ (Fig. S1(a)†). The two diffraction peaks centered at 13.10 and 27.32°ᵒ correspond to the (100) and (002) crystal planes of graphite phase carbon nitride, respectively, which were in accordance with the planar structure stacking peak of aromatic systems and inter-layer structural packing.⁴⁷

The hybrid material Re(bpy-COOH)/g-C₃N₄ was optically characterized by FTIR, UV-vis DRS, PL spectroscopy, and TRPL measurement, as shown in Fig. 1. As seen in Fig. 1a, the FTIR spectra showed that the characteristic peaks of g-C₃N₄ in the range of 1650–800 cm⁻¹ did not change after combining with Re(bpy-COOH). The appearance of C=O stretching modes in the FTIR spectrum of the hybrid material, ranging from 2100 to 1800 cm⁻¹, confirmed the successful loading of the Re(bpy-COOH) complex on g-C₃N₄.

Fig. 1 Fourier transform infrared spectra (a), UV-vis diffuse reflectance spectra (b), band gap energy of the corresponding UV-vis DRS determined from the Tauc plot (c), photoluminescence spectra under 365 nm of irradiation (d), and emission decay profiles of Re(bpy-COOH)/g-C₃N₄ and g-C₃N₄ (e) at λ_ex = 355 nm and λ_em = 450 nm without the quenching factor.

The UV-vis DRS spectra and the calculated optical band gaps of the hybrid Re(bpy-COOH)/g-C₃N₄ and g-C₃N₄ are presented in Fig. 1(b and c). The g-C₃N₄ showed broad band absorption in the visible region with a bandgap edge of approximately 2.77 eV which was assigned to the π–π* transition of conjugated heptazine units. This result was consistent with previous reports.⁴⁸ The absorption edge of the hybrid material Re(bpy-COOH)/g-C₃N₄ exhibited an insignificant blue-shift from 2.77 to 2.83 eV (∼447–4.38 nm), and the increase of the absorption from 250–315 nm could account for the strongest absorption of Re(bpy-COOH) in the UV region from 250–350 nm (Fig. S3†). Both of the hybrid Re(bpy-COOH)/g-C₃N₄ and g-C₃N₄ had the tailing absorption extended to 550 nm, which was ascribed to n–π* transitions involving lone pairs on the edge nitrogen atoms of the heptazine rings.⁴⁹

Fig. 1c presents the PL of the hybrid Re(bpy-COOH)/g-C₃N₄ and g-C₃N₄, which had strong luminescence emission in the range of 420–550 nm. Anchoring Re(bpy-COOH) on g-C₃N₄, the PL emission peak of the hybrid material was slightly narrow and showed two separated emission peaks at 435 and 455 nm. The PL intensity of Re(bpy-COOH)/g-C₃N₄ was lower than that of g-C₃N₄, which indicated that Re(bpy-COOH)/g-C₃N₄ underwent less recombination of photo-induced electron–hole pairs than g-C₃N₄ did. The lower recombination rate would lead to the higher photocatalytic activity of Re(bpy-COOH)/g-C₃N₄. The emission decay profiles of Re(bpy-COOH)/g-C₃N₄ and g-C₃N₄ under excitation of 355 nm, and emission of 450 nm without an electron donor source (TEOA), and the fitting curve are shown in Fig. 1e. The emission decay of (bpy-COOH)/g-C₃N₄ and g-C₃N₄ had the timescale of nanoseconds, which agreed with a previous report from Zhidong Wei.^49,50 The complex [Re(CO)₃(N–N)Cl] (N–N: polypyridine) displayed its emission lifetime on the nanosecond timescale (as shown in Fig. S7 in the ESI†).⁵¹ The emission lifetime of the hybrid and g-C₃N₄ was fitted to a triple exponential function and the fitting data are summarized in Table S1.† The lifetime of the hybrid was prolonged (average emission lifetime τ = 4.34 ns) compared to g-C₃N₄ (τ = 3.47 ns). Under 355 nm excitation, g-C₃N₄ and Re(bpy-COOH) underwent dual-photoexcitation, resulting in interfacial electron transfer from the conduction band of g-C₃N₄ to the HOMO of Re(bpy-COOH), minimizing recombination with holes in the valence band. It is noted that the COOH anchor was reported as intrinsically superb in terms of electron transfer due to the greater degree of nonadiabatic coupling,¹² which was important for the overall efficiency of the hybrid system. These migrated electrons localized on the HOMO and prevented the combination of photo-induced electrons of Re(bpy-COOH) with holes. That could account for the prolong decay time and the decrease of emission intensity of the PL spectrum. The longer emission lifetime was favorable for promoting PCO2R.

The morphological surface of the materials was observed by using SEM and TEM images, as shown in Fig. 2. The SEM images of g-C₃N₄ showed the planar structure stacking while Re(bpy-COOH) exhibited a crystal-like morphology with sizes varied from a few dozen to a hundred nanometers, as presented in Fig. 2(a) and (b). The TEM image (Fig. 2(c) and (e)) showed that Re(bpy-COOH) had a particle shape with the size of a few hundred nm and g-C₃N₄ possessed a sheet-like morphology. Fig. 2(d) and (f) show the morphological images of the hybrid with Re(bpy-COOH) successfully loaded on the surface of g-C₃N₄.

Fig. 2 SEM images of g-C₃N₄ (a) and Re(bpy-COOH) (b), and TEM images of g-C₃N₄ (c), Re(bpy-COOH)/g-C₃N₄ (d and f), and Re(bpy-COOH) (e).

XPS spectra determined the surface chemical state of g-C₃N₄, Re(bpy-COOH), and Re(bpy-COOH)/g-C₃N₄ and the interaction or bonding between Re(bpy-COOH) and g-C₃N₄ (Fig. 3). Fig. 3a shows the full scan of g-C₃N₄, Re(bpy-COOH), and Re(bpy-COOH)/g-C₃N₄. The XPS survey spectrum of g-C₃N₄ showed the typical structure of graphitic carbon nitride which showed the existence of the elements C, N, and O on the surface. Element O can be the impurity species absorbed on the surface of g-C₃N₄. The full scan of Re(bpy-COOH) consisted of the peaks of Re (Re 4f and Re 4d), C, N, Cl and O, clearly from the structure of the Re complex, and the obtained spectrum reached an agreement with a previous report of this complex.⁵² The surface state of Re(bpy-COOH)/g-C₃N₄ was composed of peaks of Re (Re 4f and Re 4d), C, N, and O. The high-resolution scan of the Re 4f spectrum of the hybrid material Re(bpy-COOH)/g-C₃N₄ included Re 4f_5/2 and 4f_7/2 peaking at 44.78 eV and 47.18 eV, respectively; meanwhile the Re 4f spectrum of Re(bpy-COOH) showed those 4f peaks at 45.23 eV and 48.05 eV (Fig. S5(a)†). The peak displacement of Re 4f_5/2 and 4f_7/2 indicated the interaction between Re with the surrounding elements.

Fig. 3 XPS survey scan of g-C₃N₄, Re(bpy-COOH), and Re(bpy-COOH)/g-C₃N₄ (a), high resolution scan of Re 4f of Re(bpy-COOH)/g-C₃N₄ (b), C 1s (c), and N 1s (d) of Re(bpy-COOH)/g-C₃N₄ (top) and g-C₃N₄ (bottom).

Fig. 3c presents the high-resolution XPS spectra for C 1s of Re(bpy-COOH)/g-C₃N₄ and g-C₃N₄. The C 1s spectrum of Re(bpy-COOH)/g-C₃N₄ showed the peak changes in the position and intensity, and new peaks appeared at binding energies 285.6 and 288.2 eV assigned to C=O and O–C=O, respectively. C=O and O–C=O were the featured groups characterized for Re(bpy-COOH), and hence their appearance on the C 1s spectrum of Re(bpy-COOH)/g-C₃N₄ indicated the successful linking between Re(bpy-COOH) and g-C₃N₄. The blue-shift of O–C=O at 285.6 eV in Re(bpy-COOH)/g-C₃N₄ from 286.0 to 285.6 eV (compared to O–C=O in Re(bpy-COOH), see Fig. S5(b)†) determined the chemical bonding of the anchor group –OOC of Re(bpy-COOH) on g-C₃N₄. The binding energy at 287.6 and 284.3 eV corresponded to typical sp² bonded-carbon N=C–N and C=C/C–C of g-C₃N₄, respectively, which reached agreement with previous reports.⁵³

The N 1s spectra were deconvoluted into three peaks at 398.0, 399.2, and 400.5 eV, which were assigned to sp²-hybridized pyridine N (C–N=C), tertiary pyrrolic N (N-(C)₃), and graphitic N, respectively^54,55 (Fig. 3d). The binding energy of the corresponding peaks obtained in N 1s of Re(bpy-COOH)/g-C₃N₄ (397.8, 398.9, and 400.1 eV) were shifted by 0.2, 0.3, and 0.4 eV, respectively. The peak shifts revealed the interfacial interaction between Re(bpy-COOH) and g-C₃N₄.

Fig. 4 represents the (EIS) results in the form of Nyquist, Bode, and transient photocurrent response of g-C₃N₄ and Re(bpy-COOH)/g-C₃N₄. The reduction of the arc diameter as shown in Fig. 4a suggests a decrease of film and charge transfer resistance, indicating an enhanced charge transfer of g-C₃N₄ by Re(bpy-COOH) addition. Fig. 4b indicates the local surface defects that are detected at the high-frequency data, the pore and substrate/film interface can be detected at the medium and low-frequency data, respectively. It also indicates a significant reduction in the impedance aperture and phase angles when Re(bpy-COOH) was doped in g-C₃N₄. The data indicate two phase constants in an equivalent circuit inserted in Fig. 4a. This equivalent circuit was combined with the ZSimpWin program to determine the optimized charge transfer resistance values of the investigated specimens. The fitting results indicated that the charge transfer resistance values reached 1.645 × 10⁵ and 1.182 × 10⁵ for g-C₃N₄ and Re(bpy-COOH)/g-C₃N₄ specimens, respectively. The significant reduction in charge transfer resistance values suggested that Re(bpy-COOH) doping increases the resistivity of g-C₃N₄. The current response upon light irradiation of g-C₃N₄ and Re(bpy-COOH) was undetermined, whereas that of Re(bpy-COOH)/g-C₃N₄ showed a large (approximately 1 (μA cm⁻²)) and fast rising of the current (Fig. 4c). The slightly attenuation of the current density was observed after the second cycle. As expected, linking Re(bpy-COOH) and g-C₃N₄ enhanced electronic communication and reduced the recombination of photo-induced carriers.

Fig. 4 Electrochemical impedance spectroscopy (EIS) results in the form of Nyquist (insert equivalent circuit) (a) and Bode plots (b), and photocurrent under visible light in Na₂SO₄ of g-C₃N₄ and Re(bpy-COOH)/g-C₃N₄ (c).

PCO2R of g-C₃N₄, Re(bpy-COOH) and Re(bpy-COOH)/g-C₃N₄ was conducted in DMF : TEOA mixed solution under low-intensity visible irradiation. Generally, photocatalytic activity was tested under standard test conditions of AM 1.5G, 100 mW cm⁻².⁵⁶ This investigation studied the photocatalytic CO₂ conversion and the selectivity of the hybrid system under low-light intensity. DMF solvent was the best choice for the highest selectivity of CO formation. TEOA was used as an effective sacrificial electron donor that scavenges holes generated in the valence band of g-C₃N₄. As shown in Fig. 5a and b, for the photocatalytic system with g-C₃N₄ as the photocatalyst, there was no CO and H₂ formation found, which met agreement with a previous report.⁵⁷ The CO generation of the hybrid system reached the highest TON = 28.56 at 240 min and then reached saturation, while the highest TON of Re(bpy-COOH) was 5.6 at 120 min before saturation, as shown in Fig. 5a. Ono et al. reported the total amount of CO production of Re(bpy-COOH) reached TON_CO = 6.59 for 2 h.⁵⁸ In our experiment, the TON_CO of Re(bpy-COOH) obtained was lower than that in the Ono report, which can be explained by the low-intensity irradiation used. The amount of CO formation increased significantly when Re(bpy-COOH) was anchored to g-C₃N₄ (increased by 5.6 times at 240 min of irradiation), indicating that Re(bpy-COOH) and g-C₃N₄ combination enhanced the efficiency of the PCO2R.

Fig. 5 The formation of Co (a) and H₂ (b) during photocatalytic reduction by time of g-C₃N₄, Re(bpy-COOH), and Re(bpy-COOH)/g-C₃N₄, the comparison of Co and H₂ formation of Re(bpy-COOH) (a – Co: turquoise blue and H₂: pink), and Re(bpy-COOH)/g-C₃N₄ (Co: green and H₂: orange) (c), and three consecutive 240 min irradiation cycles using the same hybrid material (d).

Hydrogen formation of the three materials is presented in Fig. 5b. Namely, the TON of H₂ formation reached 0.1 and 0.05 for Re(bpy-COOH)/g-C₃N₄ and Re(bpy-COOH) at 240 min, respectively. The comparison of CO and H₂ formation of Re(bpy-COOH)/g-C₃N₄ and Re(bpy-COOH) is shown in Fig. 5c, in which the amount of CO was dominant. Table 1 shows the TON of CO and H₂ formation, and the selectivity of CO of Re(bpy-COOH) and Re(bpy-COOH)/g-C₃N₄ at 60 min. The selectively of CO formation was determined to be 99.4 and 99.8% for Re(bpy-COOH) and Re(bpy-COOH)/g-C₃N₄, respectively (as shown in Table 1). [Re(CO)₃(N–N)Cl] complexes presented high CO selectivity (greater than 90%), and the CO selectivity of Re(bpy-COOH) under low-intensity irradiation was evaluated to be 99.4% that agreed with previous studies.⁵⁸ The CO selectivity of the hybrid system was greater than that of the homogeneous catalyst, and thus combining Re(bpy-COOH) with g-C₃N₄ improved the photocatalytic performance and the selectivity of CO.

Table 1 Photocatalytic activities of g-C₃N₄, Re(bpy-COOH), and Re(bpy-COOH)/g-C₃N₄ for CO₂ reduction under visible light (λ > 400 nm)

Photocatalyst	TONCO	TONH2	CO selectivity (%)
g-C3N4	—	—	—
Re(bpy-COOH)	5.14	0.03	99.4
Re(bpy-COOH)/g-C3N4	22.44	0.05	99.8

---

To study the stability, the PCO2R of the same hybrid material was re-employed for three successive cycles. After each cycle, the hybrid material was washed with acetonitrile and then dried at 70 °C. After that, the hybrid material was dispersed into DMF : TEOA solution as the preparation process for PCO2R measurement in the experimental section and then started the new cycle. The results (Fig. 5d) showed the stability of CO production in the three cycles. The CO formation rate and generation remained stable in three cycles; however, in the second and third cycles, a slightly insignificant reduction of CO yield was observed. That could be explained by the photo-degradation of Re(bpy-COOH) under irradiation and the loss of the catalyst due to collecting and washing the catalyst after each cycle run.

This hybrid system and a similar hybrid system Mn(I)/g-C₃N₄ (Mn(I) : Mn(bipyridineCOOH)(CO)₃Br) possessed similar photocatalytic properties such as photoreduction CO₂ products (PCO2P) (CO and H₂) and photocatalytic structure (Z-scheme model and g-C₃N₄ surface with the –COOH anchor). The comparison of the two systems showed that the enhancement of PCO2P of the Re(bpy-COOH)/g-C₃N₄ hybrid was lower than that of the Mn(I)/g-C₃N₄ system.⁵⁷ Namely, TON_CO was increased approximately 10 times, whereas TON_CO in this study was increased 5.6 times. It should be noted that Re(bpy-COOH)/g-C₃N₄ worked under the low-intensity visible light, and meanwhile Mn(I)/g-C₃N₄ was studied under a 300 W Xe lamp (100 mW cm⁻²). The different light-irradiation intensities in the two research conditions might affect the CO generation. Referring to the PCO2P, the hybrid was evaluated to be more dominant in the number of products formed. PCO2P of Ru complex/C₃N₄ included HCOOH (80% selectively), CO, and H₂,¹¹ whereas that of Re(bpy-COOH)/g-C₃N₄ generated only two products CO (more than 90% selectively) and H₂.

To clarify the mechanism of PCO2R of the hybrid system, the reduction potential of Re(bpy-COOH) and the energy level of g-C₃N₄ need to be mentioned. Re(bpy-COOH) is a good electrocatalyst for CO₂ reduction, with LUMO potential³⁹ at −1.94 vs. NHE (or one-electron reduction potential, E_red = −1.69 V vs. Ag/AgNO₃) and the HOMO potential for absorbing visible light is approximately 0.75 eV vs. NHE (MLCT band, see the UV/Vis absorbance spectrum of Re(bpy-COOH) in Fig. S3†). Bulk g-C₃N₄ has a conduction band and valence band at −1.4 and 1.34 eV (vs. NHE, pH = 7),^31,32 respectively, which is sufficiently negative for the injection of one electron from the CB to the HOMO of Re(bpy-COOH). The energy structure of Re(bpy-COOH) and g-C₃N₄ forms a Z-scheme hybrid system^59–61 as illustrated in Scheme 1.

Scheme 1 The illustration of the photocatalytic CO₂ reduction process.

Under visible-light irradiation, both g-C₃N₄ and Re(bpy-COOH) undergo photoexcitation that generates electron–hole pairs. Due to the high potential difference between the CB of g-C₃N₄ and the HOMO of Re(bpy-COOH) interfacial photo-induced electron transfer occurred, which reduces electron–hole pair recombination on Re(bpy-COOH). The photo-induced electrons generated on the LUMO on the catalyst side reduce CO₂ molecules while the holes in the VB of g-C₃N₄ are scavenged by TEOA. The efficient interfacial electron transfer and the decrease in electron–hole pair recombination manifest via photoluminescent decay and PL spectra sufficiently boost the photocatalytic performance of the system.

To enhance our understanding of the ensemble effect within the Re(bpy-COOH)/g-C₃N₄ catalyst and its implications for the critical intermediate adsorption in CO2RR, we performed Density Functional Theory (DFT) calculations. Fig. 6 depicts the computed Gibbs free energy (ΔG) progression for CO2RR across g-C₃N₄, Re(bpy-COOH), and Re(bpy-COOH)/g-C₃N₄. This includes a focus on four fundamental steps: (1) CO₂ adsorption onto the N atom of g-C₃N₄ or the Re atom of Re(bpy-COOH); (2) formation of COOH through hydrogenation; (3) creation of CO via disproportionation; (4) CO desorption. Our computations reveal that the ΔG of both Re(bpy-COOH) and Re(bpy-COOH)/g-C₃N₄ is less than that of g-C₃N₄, suggesting a more favorable environment for CO2RR processes in the former materials. Remarkably, in Re(bpy-COOH) and Re(bpy-COOH)/g-C₃N₄, the conversion of *CO₂ to *COOH via hydrogenation overcomes the Gibbs energy barrier with just 0.26 and 0.61 eV, respectively, a significant reduction compared to the 1.53 eV needed in g-C₃N₄. Additionally, *CO formation is energetically exothermic at −0.55 eV for Re(bpy-COOH) and −0.62 eV for Re(bpy-COOH)/g-C₃N₄, compared to −0.42 eV for g-C₃N₄. Interestingly, the Gibbs energy barrier for CO desorption is only 0.64 eV in the hybrid material, lower than the 0.92 eV required in the Re(bpy-COOH) system. Our findings suggest that, despite a slightly higher Gibbs energy barrier for *COOH, the CO2RR processes favoring CO formation in the Re(bpy-COOH)/g-C₃N₄ system are more likely, aligning with our experimental results.

Fig. 6 Calculated Gibbs free energy diagram of CO2RR processes with the corresponding intermediates for g-C₃N₄ (blue), Re(bpy-COOH) (green), and Re(bpy-COOH)/g-C₃N₄ (red). The asterisk (*) denotes the adsorption site.

For a deeper understanding of the CO2RR catalytic activity, we delved into the charge transfer analysis between *COOH, *CO, and g-C₃N₄, Re(bpy-COOH), and Re(bpy-COOH)/g-C₃N₄, respectively. The Mulliken charge and corresponding charge transfer for each atom involved in the *COOH and *CO intermediates are detailed in Table S2 (see the ESI†). Positive and negative charge transfer values denote electron loss and gain, respectively. Remarkably, the charge transfers in both Re(bpy-COOH) and Re(bpy-COOH)/g-C₃N₄ scenarios register negative values (−0.51 and −0.48 for *COOH, and −0.15 and −0.13 for *CO, respectively). This contrasts with the negligible charge transfers observed in the case of g-C₃N₄. These findings suggest that the chemical interactions between CO and COOH with Re(bpy-COOH) and Re(bpy-COOH)/g-C₃N₄ are more potent compared to those with g-C₃N₄. It's noteworthy that the charge transfer of Re(bpy-COOH)/g-C₃N₄ is slightly less than that of Re(bpy-COOH) alone, indicating that g-C₃N₄ reduces the charge transfer between Re(bpy-COOH) and the intermediates, thus moderating their bond strength. This observation provides an explanation for the reduced Gibbs energy barrier for CO desorption observed in Re(bpy-COOH)/g-C₃N₄, as depicted in Fig. 6.

Additionally, we employed COHP analysis to unravel the interactions between the orbital pairs of intermediates and substrates. This strategy facilitates an understanding of the bonding and antibonding states emanating from each atomic orbital. In Fig. S6 (in the ESI†), we show the energy-resolved COHP between the 2p-orbitals of adsorbed C and a surface N 2p-orbital of g-C₃N₄, or a Re d-orbital of Re(bpy-COOH) and Re(bpy-COOH)/g-C₃N₄ for *COOH and *CO intermediates, respectively. Our analysis indicates weak C–N interactions in g-C₃N₄, as nearly zero COHP is observed around the Fermi level (E_F). In contrast, the C–Re interactions display a peak of the bonding state at E_F in the case of Re(bpy-COOH) and Re(bpy-COOH)/g-C₃N₄, signifying substantial interaction. These strong chemical interactions between C–Re atoms contribute to the favorable CO2RR observed in Re(bpy-COOH) and Re(bpy-COOH)/g-C₃N₄. Our DFT calculations affirm that Re(bpy-COOH)/g-C₃N₄ acts as a more efficient catalyst for CO2RR to CO conversion compared to Re(bpy-COOH) and g-C₃N₄. In line with this, our composite catalysts achieve a peak TON (Turnover Number) of CO formation and CO faradaic efficiency of 22.44 and 99.8%, respectively.

4. Conclusions

The hybrid photocatalytic material comprising the molecular catalyst Re(bpy-COOH) and the graphitic carbon nitride g-C₃N₄ is among the first examples of PCO2R under low-intensity visible-light irradiation. Re(bpy-COOH)/g-C₃N₄ is highly selective for CO production and has a high TON of CO formation. The combination of Re(bpy-COOH) and g-C₃N₄ formed a Z-scheme structure that improves the quantum efficiency and interfacial electron transfer, and decreases electron–hole pair recombination, hence boosting the photocatalytic efficiency. Utilizing DFT calculations, it was found that the Re(bpy-COOH)/g-C₃N₄ hybrid system exhibits efficient charge transfer and demonstrates lower energy barriers for *CO₂ to *COOH conversion, as well as an increased propensity for *COOH formation compared to pure g-C₃N₄. Furthermore, the desorption activation energy of the *CO intermediate is also lower in the Re(bpy-COOH)/g-C₃N₄ system than in pure Re(bpy-COOH), indicating that the hybrid system achieves a balance between two rate-limiting steps for CO2RR. The unique combination of the two components contributes to the development and tailoring of highly active photocatalysts for low-intensity visible-light irradiation.

Conflicts of interest

The authors declare no competing financial interests.

Acknowledgements

The authors appreciate the financial support from Vietnam Academy of Science and Technology under Grant Number CSCL19.03/23-24. The authors also gratefully acknowledge the Center for Computational Materials Science, Institute for Materials Research, Tohoku University, Japan, for the use of MASAMUNE-IMR, Cray XC50-LC supercomputer facility.

References

1. H. Takeda and O. Ishitani, Development of efficient photocatalytic systems for CO₂ reduction using mononuclear and multinuclear metal complexes based on mechanistic studies, Coord. Chem. Rev., 2010, 254, 346–354.
2. Y. Tamaki, T. Morimoto, K. Koike and O. Ishitani, Photocatalytic CO₂ reduction with high turnover frequency and selectivity of formic acid formation using Ru(II) multinuclear complexes, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 15673–15678.
3. H. Takeda, H. Koizumi, K. Okamoto and O. Ishitani, Photocatalytic CO₂ reduction using a Mn complex as a catalyst, Chem. Commun., 2014, 50, 1491–1493.
4. A. E. Nahhas, C. Consani, A. M. Blanco-Rodríguez, K. M. Lancaster, O. Braem, A. Cannizzo, M. Towrie, T. P. Clark, S. Záliš, M. Chergui and A. Vlček Jr, Ultrafast excited-state dynamics of rhenium(I) photosensitizers [Re(Cl)(CO)3(N,N)] and [Re(imidazole)(CO)3(N,N)]+: diimine effects, Inorg. Chem., 2011, 50, 2932–2943.
5. S. Sato, Y. Matubara, K. Koike, M. Falkenstrom, T. Katayama, Y. Ishibashi, H. Miyasaka, S. Taniguchi, H. Chosrowjan, N. Mataga, N. Fukazawa and S. Koshihara, Photochemistry of fac-[Re(bpy)(CO)₃Cl], Chem.–Eur. J., 2012, 18, 15722–15734.
6. A. Vlček and M. Busby, Ultrafast ligand-to-ligand electron and energy transfer in the complexes fac-[ReI(L)(CO)₃(bpy)]n+, Coord. Chem. Rev., 2006, 250, 1755–1762.
7. C. Gourlaouen, J. Eng, M. Otsuka, E. Gindensperger and C. Daniel, Quantum chemical interpretation of ultrafast luminescence decay and intersystem crossings in rhenium(I) carbonyl bipyridine complexes, J. Chem. Theory Comput., 2015, 11, 99–110.
8. R. C. Evans, P. Douglas and C. J. Winscom, Coordination complexes exhibiting room-temperature phosphorescence: evaluation of their suitability as triplet emitters in organic light emitting diodes, Coord. Chem. Rev., 2006, 250, 2093–2126.
9. J. N. Demas and G. A. Crosby, Quantum efficiencies of transition-metal complexes. I. d–d luminescence, J. Am. Chem. Soc., 1970, 92, 7262–7270.
10. A. Nakada, T. Nakashima, K. Sekizawa, K. Maeda and O. Ishitani, Visible-light-driven CO₂ reduction on a hybrid photocatalyst consisting of a Ru(II) binuclear complex and a Ag-loaded TaON in aqueous solutions, Chem. Sci., 2016, 7, 4364–4371.
11. R. Kuriki, K. Sekizawa, O. Ishitani and K. Maeda, Visible-light-driven CO₂ reduction with carbon nitride: enhancing the activity of ruthenium catalysts, Angew. Chem., Int. Ed., 2015, 54, 2406–2409.
12. K. Maeda, Metal-complex/semiconductor hybrid photocatalysts and photoelectrodes for CO₂ reduction driven by visible light, Adv. Mater., 2019, 31, 1808205.
13. R. Kuriki and K. Maeda, Development of hybrid photocatalysts constructed with a metal complex and graphitic carbon nitride for visible-light-driven CO₂ reduction, Phys. Chem. Chem. Phys., 2017, 19, 4938–4950.
14. H. Kasap, C. A. Caputo, B. C. M. Martindate, R. Godin, V. W. H. Lau, B. V. Lotsch, J. R. Durrant and E. Reisner, Solar-driven reduction of aqueous protons coupled to selective alcohol oxidation with a carbon nitride–molecular Ni catalyst system, J. Am. Chem. Soc., 2016, 138, 9183–9192.
15. A. Nakada, H. Kumagai, M. Robert, O. Ishitani and K. Maeda, Molecule/semiconductor hybrid materials for visiblelight CO₂ reduction: design principles and interfacial engineering, Acc. Mater. Res., 2021, 2, 458–470.
16. C. Quintana, M. P. Cifuentes and M. G. Humphrey, Transition metal complex/gold nanoparticle hybrid materials, Chem. Soc. Rev., 2020, 49, 2316–2341.
17. C. D. Windle, E. Pastor, A. Reynal, A. C. Whitwood, Y. Vaynzof, J. R. Durrant, R. N. Perutz and E. Reisner, Improving the photocatalytic reduction of CO₂ to CO through immobilisation of a molecular Re catalyst on TiO₂, Chem.–Eur. J., 2015, 21, 3746–3754.
18. D. Sun, Y. Gao, J. Fu, X. Zeng, Z. Chen and Z. Li, Construction of supported Ru complex on bifunctional MOF-253 for photocatalytic CO₂ reduction under visible light, Chem. Commun., 2015, 51, 2645–2648.
19. B. Shan, S. Vanka, T. T. Li, L. Troian-Gautier, M. K. Brennaman, Z. Mi and T. J. Meyer, Binary molecular-semiconductor p–n junctions for photoelectrocatalytic CO₂ reduction, Nat. Energy, 2019, 4, 290–299.
20. F. Paquin, J. Rivnay, A. Salleo, N. Stingelin and C. Silva, Multi-phase semicrystalline microstructures drive exciton dissociation in neat plastic semiconductors, J. Mater. Chem. C, 2015, 3, 10715–10722.
21. Q. Gu, Y. Liao, J. Long, X. Wang and C. Xue, Template-free synthesis of porous graphitic carbon nitride microspheres for enhanced photocatalytic hydrogen generation with high stability, Appl. Catal. B Environ., 2015, 165, 503–510.
22. X. Zhang, Q. Wu, Z. Du, Y. Zheng and Q. Li, Green synthesis of g-C₃N₄ -Pt catalyst and application to photocatalytic hydrogen evolution from water splitting, Fullerenes, Nanotub. Carbon Nanostruct., 2018, 26, 688–695.
23. W. Zhang, L. Zhou and H. Deng, Ag modified g-C₃N₄ composites with enhanced visible-light photocatalytic activity for diclofenac degradation, J. Mol. Catal. A Chem., 2016, 423, 270–276.
24. K. Qi, Y. Li, Y. Xie, S. Y. Lui, K. Zheng, Z. Chen and R. Wang, Ag loading enhanced photocatalytic activity of g-C₃N₄ porous nanosheets for decomposition of organic pollutants, Front. Chem., 2019, 7, 1–9.
25. D. Hao, J. Ren, Y. Wang, H. Arandian, M. Garbrecht, X. Bai, H. K. Shin, W. Wei and B. J. Ni, A green synthesis of Ru modified g-C₃N₄ nanosheets for enhanced photocatalytic ammonia synthesis, Energy Mater. Adv., 2021, 2021, 1–12.
26. G. Mamba and A. K. Mishra, Graphitic carbon nitride (g-C₃N₄) nanocomposites: a new and exciting generation of visible light driven photocatalysts for environmental pollution remediation, Appl. Catal. B Environ., 2016, 198, 347–377.
27. S. Patnaik, S. Martha and K. M. Parida, An overview of the structural, textural and morphological modulations of g-C₃N₄ towards photocatalytic hydrogen production, RSC Adv., 2016, 6, 46929–46951.
28. J. Fu, J. Yu, C. Jiang and B. Cheng, g-C₃N₄-based heterostructured photocatalysts, Adv. Energy Mater., 2018, 8, 1–31.
29. Z. Sun, H. Wang, Z. Wu and L. Wang, g-C₃N₄ based composite photocatalysts for photocatalytic CO₂ reduction, Catal. Today, 2018, 300, 160–172.
30. N. Q. Thang, A. Sabbah, L. C. Chen, K. H. Chen, L. V. Hai, C. M. Thi and P. V. Viet, Localized surface plasmonic resonance role of silver nanoparticles in the enhancement of long-chain hydrocarbons of the CO₂ reduction over Ag-gC₃N₄/ZnO nanorods photocatalysts, Chem. Eng. Sci., 2021, 229, 116049.
31. A. Alaghmandfard and K. Ghandi, A comprehensive review of graphitic carbon nitride (g-C₃N₄)–metal oxide-based nanocomposites: potential for photocatalysis and sensing, Nanomaterials, 2022, 12, 294.
32. X. Ma, Y. Lv, J. Xu, Y. Liu, R. Zhang and Y. Zhu, A strategy of enhancing the photoactivity of g-C₃N₄ via doping of nonmetal elements: a first-principles study, J. Phys. Chem. C, 2012, 116, 23485–23493.
33. K. Maeda, R. Kuriki and O. Ishitani, Photocatalytic activity of carbon nitride modified with a ruthenium(II) complex having carboxylic- or phosphonic acid anchoring groups for visible-light CO₂ reduction, Chem. Lett., 2016, 45, 182–184.
34. K. Maeda, D. An, R. Kuriki, D. Lu and O. Ishitani, Graphitic carbon nitride prepared from urea as a photocatalyst for visible-light carbon dioxide reduction with the aid of a mononuclear ruthenium(II) complex, Beilstein J. Org. Chem., 2018, 14, 1806–1812.
35. X. Ma, C. Hu and Z. Bian, Hybrid photocatalytic systems comprising a manganese complex anchored on g-C₃N₄ for efficient visible-light photoreduction of CO₂, Inorg. Chem. Commun., 2020, 117, 107951.
36. J. J. Walsh, C. Jiang, J. Tang and A. J. Cowan, Photochemical CO₂ reduction using structurally controlled g-C₃N₄, Phys. Chem. Chem. Phys., 2016, 18, 24825–24829.
37. B. Ma, G. Chen, C. Fave, L. Chen, R. Kuriki, K. Maeda, O. Ishitani, T. C. Lau, J. Bonin and M. Robert, Efficient visible-light-driven CO₂ reduction by a cobalt molecular catalyst covalently linked to mesoporous carbon nitride, J. Am. Chem. Soc., 2020, 142, 6188–6195.
38. H. Koizumi, H. Chiba, A. Sugihara, M. Iwamura, K. Nozaki and O. Ishitani, CO₂ capture by Mn(i) and Re(i) complexes with a deprotonated triethanolamine ligand, Chem. Sci., 2019, 10, 3080–3088.
39. P. N. Nguyen, T. B. N. Dao, T. T. Tran, T. A. Nguyen, T. D. L. Phan, L. P. Nguyen, V. Q. Dang, T. M. Nguyen and N. Nguyen Dang, Electrocatalytic CO₂ reduction by [Re(CO)₃Cl(3-(pyridin-2-yl)-5-phenyl-1,2,4-triazole)] and [Re(CO)₃Cl(3-(2-pyridyl)-1,2,4-triazole)], ACS Omega, 2022, 7, 34089–34097.
40. G. Kresse and J. Furthmüller, Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169–11186.
41. G. Kresse and J. Furthmüller, Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set, Comput. Mater. Sci., 1996, 6, 15–50.
42. J. P. Perdew, K. Burke and M. Ernzerhof, Generalized gradient approximation made simple, Phys. Rev. Lett., 1996, 77, 3865–3868.
43. D. Joubert, From ultrasoft pseudopotentials to the projector augmented-wave method, Phys. Rev. B: Condens. Matter Mater. Phys., 1999, 59, 1758–1775.
44. S. Grimme, J. Antony, S. Ehrlich and H. Krieg, A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu, J. Chem. Phys., 2010, 132, 154104.
45. V. L. Deringer, A. L. Tchougréeff and R. Dronskowski, Crystal orbital Hamilton population (COHP) analysis as projected from plane-wave basis sets, J. Phys. Chem. A, 2011, 115, 5461–5466.
46. A. A. Peterson, F. Abild-Pedersen, F. Studt, J. Rossmeisl and J. K. Nørskov, How copper catalyzes the electroreduction of carbon dioxide into hydrocarbon fuels, Energy Environ. Sci., 2010, 3, 1311–1315.
47. X. Wang, K. Maeda, A. Thomas, K. Takanebe, G. Xin, J. M. Carlsson, K. Domen and M. Antonietti, A metal-free polymeric photocatalyst for hydrogen production from water under visible light, Nat. Mater., 2009, 8, 76–80.
48. J. Liu, T. Zhang, Z. Wang, G. Dawson and W. Chen, Simple pyrolysis of urea into graphitic carbon nitride with recyclable adsorption and photocatalytic activity, J. Mater. Chem., 2011, 21, 14398–14401.
49. H. Zhang and A. Yu, Photophysics and photocatalysis of carbon nitride synthesized at different temperatures, J. Phys. Chem. C, 2014, 118, 11628–11635.
50. Z. Wei, J. Liu, W. Fang, M. Xu, Z. Qin, Z. Jiang and W. Shangguan, Photocatalytic hydrogen evolution with simultaneous antibiotic wastewater degradation via the visible-light-responsive bismuth spheres-g-C₃N₄ nanohybrid: waste to energy insight, Chem. Eng. J., 2019, 358, 944–954.
51. A. E. Nahhas, A. Cannizzo, F. V. Mourik, A. M. Blanco-Rodriguez, A. Vlcek Jr. and M. Chergui, Ultrafast excited-state dynamics of [Re(L)(CO)₃(bpy)] n complexes: involvement of the solvent, J. Phys. Chem. A, 2010, 114, 6361–6369.
52. A. Zhanaidarova, S. C. Jones, E. Despagnet-Ayoub, B. R. Pimentel and C. P. Kubiak, Re(tBu-bpy)(CO)₃Cl supported on multi-walled carbon nanotubes selectively reduces CO₂ in water, J. Am. Chem. Soc., 2019, 141, 17270–17277.
53. L. Tan, J. Xu, X. Zhang, Z. Hang, Y. Lia and S. Wang, Synthesis of g-C₃N₄/CeO₂ nanocomposites with improved catalytic activity on the thermal decomposition of ammonium perchlorate, Appl. Surf. Sci., 2015, 356, 447–453.
54. Z. Huang, Q. Sun, K. Lv, Z. Zhang, M. Li and B. Li, Effect of contact interface between TiO₂ and g-C₃N₄ on the photoreactivity of g-C₃N₄/TiO₂ photocatalyst: (001) vs (101) facets of TiO₂, Appl. Catal. B Environ., 2015, 164, 420–427.
55. S. W. Cao, Y. P. Yuan, J. Barber, S. C. J. Loo and C. Xue, Noble-metal-free g-C₃N₄/Ni(dmgH)₂ composite for efficientphotocatalytic hydrogen evolution under visible light irradiation, Appl. Surf. Sci., 2014, 319, 344–349.
56. S. Roy and E. Reisner, Visible-light-driven CO₂ reduction by mesoporous carbon nitride modified with polymeric cobalt phthalocyanine, Angew. Chem., 2019, 131, 12308–12312.
57. X. Ma, L. Zheng and Z. Bian, Visible-light-driven CO₂ reduction with g-C₃N₄-based composite: enhancing the activity of manganese catalysts, Chem. Eng. Sci., 2021, 229, 116042.
58. Y. Ono, J. Nakamura, M. Hayashi and K. I. Takahashi, Effect of substituent groups in rhenium bipyridine complexes on photocatalytic CO₂ reduction, Am. J. Appl. Chem., 2014, 2, 74.
59. R. Guo, Z. Bi, Z. Lin, X. Hua, J. Wang, X. Chen and W. Pan, Carbon quantum dots-modified Z-scheme Bi₁₂O₁₇Cl₂/NiAl-LDH for significantly boosting photocatalytic CO₂ reduction, J. Colloid Interface Sci., 2022, 627, 343–354.
60. X. Ji, R. T. Guo, Y. Miao, Z. Lin, L. Hong, Y. Yuan, Z. Li and W. Pan, Construction of full solar-spectrum-driven Cu_2-xS/Ni-Al-LDH heterostructures for efficient photocatalytic CO₂ reduction, ACS Appl. Energy Mater., 2022, 5, 2862–2872.
61. Y. Miao, R. Guo, J. Gu, Y. Liu, G. Wu, C. Duan and W. Pan, Z-Scheme Bi/Bi₂O₂CO₃/layered double-hydroxide nanosheet heterojunctions for photocatalytic CO₂ reduction under visible light, ACS Appl. Nano Mater., 2021, 4, 4902–4911.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3ta01502e

---