Electron-withdrawing pyridine-functionalized g-C₃N₄-coordinated cobalt phthalocyanine for enhanced photocatalytic CO₂ reduction

Abstract

To enhance the charge separation efficiency of g-C₃N₄ and facilitate electron transfer between photosensitizer and molecular catalyst, polarization and coordination strategies are used by grafting pyridine rings onto the edge of the g-C₃N₄ framework. Herein, electron-withdrawing pyridine edge-functionalized g-C₃N₄ (g-C₃N₄-P_x) was synthesized via facile one-step thermal polymerization of urea and 4-aminopyridine, and employed as visible-light photosensitizer hybridized with cobalt phthalocyanine (CoPc) for efficient photoreduction of CO₂. Both experimental and theoretical results confirm that electron-withdrawing pyridine grafting facilitates in-plane charge separation and directs electron migration toward the edge of g-C₃N₄, narrows its band gap for enhanced visible-light absorption, and provides dynamic coordination sites that significantly boost interfacial electron transfer from g-C₃N₄-P_x to CoPc. A significant increase in the CO yield was achieved with the optimized CoPc/g-C₃N₄-P_1.5 hybrid, reaching 14.95 mmol g⁻¹ after 6 hours of visible-light irradiation—a 6.1-fold improvement over the unmodified CoPc/g-C₃N₄ (2.47 mmol g⁻¹). This work provides a facile approach for developing highly efficient hybrid photocatalysts for CO₂ reduction and improving the charge separation and visible-light absorption in organic semiconductors.

Keywords: CO₂ photoreduction; Cobalt phthalocyanine/g-C₃N₄ hybrid photocatalysts; Polarization engineering; Directional electron transfer; Coordination interaction; Electron-withdrawing effect.

---

1 Introduction

Photocatalytic carbon dioxide (CO₂) reduction using solar energy represents an ideal approach to address energy and environmental challenges, as it operates under ambient conditions and produces value-added chemicals.^1–3 Despite widespread interest, achieving high selectivity and efficiency in CO₂ photoreduction remains challenging due to thermodynamic and kinetic limitations.^4–6 The inherent stability of the CO₂ molecule, competitive hydrogen evolution reactions in aqueous environments, limited solar spectrum absorption, and severe charge carrier recombination—especially in single-component systems—further hinder the progress. Thus, developing highly active and selective hybrid photocatalysts is essential.

A promising strategy involves hybrid systems comprising a visible-light-absorbing semiconductor sensitizer and a metal molecular catalyst with high CO₂ reduction activity.^7–23 Recent studies have explored various hybrids, such as ruthenium(II)-complex-modified graphitic carbon nitride (g-C₃N₄) for formate production, and iron (Fe) or cobalt (Co) molecular catalysts coupled with g-C₃N₄, cadmium sulphide (CdS), or related semiconductors for the photoreduction of CO₂ to carbon monoxide (CO).^9–17 In these systems, spatial separation between the light absorber and catalytic centre promotes charge separation and enhances CO₂ reduction performance. Consequently, enhancing both the semiconductor's light-harvesting capacity and its interaction with the molecular catalyst is critical for increasing photon utilization and facilitating electron transfer.^24,25

Efforts to accelerate electron transfer from photosensitizers to catalysts have led to covalent linker strategies, like direct connection, ethylene or amide bonds;^13,14,26,27 however, these often involve complex synthesis and may suffer from back electron transfer, compromising activity and stability. As a covalent bond, a coordination bond can be an ideal intermolecular force to realize the connection between a photosensitizer and a catalyst. The labile nature of coordination bonds confers dynamic stability. At the same time, the coordination interaction is expected to realize inner shell electron transfer, thus overcoming the limitation of diffusion rate and promoting the electron transfer between molecules. Moreover, dynamic coordination interactions offer reversible and self-correcting binding, which can improve the durability of the hybrids.²⁰

As an organic semiconductor, g-C₃N₄ has attracted significant attention for various applications due to its facile synthesis, suitable band alignment, good stability, visible-light response and layered structure.^28–32 Nonetheless, pristine g-C₃N₄ suffers from rapid charge recombination and restricted visible-light absorption, limiting its photocatalytic activity. Strategies such as defect engineering, elemental doping, heterojunction construction, and metal deposition have been employed to address these issues.^33–35 Regulating molecular polarity has also emerged as an effective route to promote intramolecular charge transfer and narrow the bandgap of g-C₃N₄.^36–38 Additionally, coupling g-C₃N₄ with molecular catalysts like cobalt phthalocyanine (CoPc)—known for its high selectivity in CO₂-to-CO conversion^39,40—offers a pathway toward efficient CO₂ reduction, with CO serving as a key industrial feedstock for synthetic fuels and chemicals.

In this work, we introduce polarization engineering and dynamic coordination strategies to enhance the in-plane charge separation in g-C₃N₄ and provide reversible coordination binding sites for CoPc molecular catalyst. Pyridine edge-functionalized g-C₃N₄ (g-C₃N₄-P_x) was designed and synthesized as a light absorber (Scheme S1). The electron-withdrawing pyridine groups can promote the formation of a donor–acceptor structure, narrow the bandgap and improve the visible-light absorption and charge separation. More importantly, the pyridine edges can offer dynamic axial coordination sites for CoPc,^41–43 facilitating interfacial electron transfer from the light-harvesting part, g-C₃N₄, to the catalytic site, CoPc. Through the synergistic effect of polarization engineering and dynamic coordination, efficient charge separation and transfer are realized, and ultimately, the photocatalytic performance of the CoPc/g-C₃N₄-P_x hybrid catalyst for CO₂ reduction is enhanced.

2 Results and discussion

2.1 Structure and morphology

The pyridine edge-functionalized g-C₃N₄-P_x samples were prepared through a facile one-step homogeneous thermal polymerization of urea with different amounts of 4-aminopyridine.³⁶ A possible polymerization reaction is shown in Scheme S2. Fig. 1a presents the X-ray diffraction (XRD) patterns of pristine g-C₃N₄ and g-C₃N₄-P_x samples, where x is the weight percentage of 4-aminopyridine relative to urea used during synthesis. All XRD patterns exhibit the characteristic diffraction peaks of g-C₃N₄ at 2θ = 12.8° and 27.6°, corresponding to the (100) in-plane structure and the (002) interlayer stacking, respectively. Notably, no significant shift is observed in the positions of these diffraction peaks in the g-C₃N₄-P_x samples compared to pristine g-C₃N₄, indicating that pyridine functionalization at the edges does not alter the crystalline and layered structure of g-C₃N₄. Fig. 1b shows the N₂ adsorption–desorption isotherms of the g-C₃N₄ and g-C₃N₄-P_x samples. All the materials show type IV isotherms with H3 hysteresis, indicative of mesoporous structures. The Brunauer–Emmett–Teller (BET) specific surface areas are measured to be 80.2, 78.0, 75.4, 75.6, 72.7 and 57.6 m² g⁻¹ for g-C₃N₄, g-C₃N₄-P_0.5, g-C₃N₄-P_1.0, g-C₃N₄-P_1.5, g-C₃N₄-P_2.0 and g-C₃N₄-P_3.0, respectively.

Fig. 1 (a) XRD patterns and (b) N₂ adsorption–desorption isotherms of the pristine g-C₃N₄ and g-C₃N₄-P_x (x = 0.5–3.0) samples.

The morphology of pristine g-C₃N₄ and the as-synthesized g-C₃N₄-P_x samples were characterized using scanning electron microscopy (SEM) (Fig. 2a–f) and transmission electron microscopy (TEM) (Fig. 2g–j). All the samples exhibit mesoporous sheet morphology. Specifically, g-C₃N₄, g-C₃N₄-P_0.5, g-C₃N₄-P_1.0, and g-C₃N₄-P_1.5 display predominantly relatively ultrathin, flaky, and porous structures. In contrast, g-C₃N₄-P_2.5 and g-C₃N₄-P_3.0—which feature a higher content of edge-grafted pyridine—show increasingly aggregated and thicker sheet-like structures that are less prone to exfoliation. Representative TEM images of g-C₃N₄, g-C₃N₄-P_0.5, g-C₃N₄-P_1.5 and g-C₃N₄-P_3.0 are presented in Fig. 2g–j. These images reveal that g-C₃N₄, g-C₃N₄-P_0.5 and g-C₃N₄-P_1.5 maintain a flaky and ultrathin nanosheet structure, whereas g-C₃N₄-P_3.0 exhibits slightly thicker sheets. Both SEM and TEM results confirm that the edge functionalization with pyridine does not disrupt the fundamental lamellar porous structure of g-C₃N₄, consistent with the XRD patterns shown in Fig. 1a. Samples with low levels of edge-pyridine functionalization (g-C₃N₄-P_x, x = 0–2.0) exhibit relatively high surface areas, attributed to their flaky and porous morphology, which slightly decreases as the degree of pyridine grafting increases. In contrast, g-C₃N₄-P_3.0, with a high degree of functionalization, shows a marked decrease in surface area (57.6 m² g⁻¹), likely due to the formation of thick aggregates, as observed in SEM and TEM images.

Fig. 2 (a–f) SEM and (g–j) TEM images of the pristine g-C₃N₄ and g-C₃N₄-P_x (x = 0.5–3.0) samples.

The chemical composition and electronic states of the samples were further analysed by X-ray photoelectron spectroscopy (XPS). The XPS survey spectra in Fig. S1a confirm the presence of C and N in all g-C₃N₄-P_x materials, while XPS survey spectra in Fig. S1b confirm the successful hybridisation of CoPc and g-C₃N₄-P_1.5. The small O peak appearing at 532.1 eV is attributed to the adsorbed H₂O. High-resolution C 1s, N 1s and Co 2p spectra are presented in Fig. S2 and 3a and b, respectively. The C 1s spectrum displays two peaks at 284.8 eV (graphitic carbon, C–C/C=C) and 288.1 eV (sp²-hybridized carbon in N–C=N/C–N=C groups).⁴⁴ The N 1s spectrum can be deconvoluted into three components: pyridinic N (C–N=C) at 398.5–399.6 eV, pyrrolic N (tertiary N–C₃) at 399.8–399.9 eV, and terminal amino groups (–NH₂/NH) at 401.2 eV.⁴⁵ While the C 1s spectrum remains unchanged after pyridine modification (Fig. S2), the N 1s peaks corresponding to pyridinic and pyrrolic nitrogen shift slightly to higher binding energy (from 398.5 to 398.7 eV and from 399.8 to 400.1 eV, respectively) after grafting pyridine, suggesting reduced electron density around nitrogen atoms due to the electron-withdrawing effect of grafted pyridine. The Co 2p spectrum displays two peaks at 793.2 eV (Co 2p_1/2) and 777.6 eV (Co 2p_3/2) in CoPc. However, the formation of the CoPc/g-C₃N₄-P_1.5 hybrid causes the Co 2p_1/2 and Co 2p_3/2 peaks to shift to lower binding energies of 792.4 eV and 776.8 eV, respectively. This indicates an increase in electron density at the Co site because of the electrons flowing from g-C₃N₄-P_1.5 to CoPc.

Fig. 3 High-resolution XPS spectra of (a) N 1s of the g-C₃N₄, g-C₃N₄-P_1.5, and g-C₃N₄-P_3.0 samples, and (b) Co 2p of CoPc and the CoPc/g-C₃N₄-P_1.5 hybrid catalyst.

2.2 Optical properties and band structures

The optical properties of pristine g-C₃N₄ and the pyridine-modified g-C₃N₄-P_x samples were investigated using UV-vis diffuse reflectance spectroscopy (UV-vis DRS) measurements. As shown in Fig. 4a, the absorption intensity in the visible region initially increases and then decreases, with g-C₃N₄-P_1.5 exhibiting the strongest absorption between 380 and 450 nm. Concurrently, the absorption edge progressively red-shifts from g-C₃N₄ to g-C₃N₄-P_3.0 as the pyridine content increases. Samples g-C₃N₄-P_1.5, g-C₃N₄-P_2.0 and g-C₃N₄-P_3.0 demonstrate significantly enhanced visible-light absorption compared to unmodified g-C₃N₄.

Fig. 4 (a) UV-vis DRS spectra and Tauc plots (inset), (b) Mott–Schottky plots, and (c) band structure alignments for g-C₃N₄ and g-C₃N₄-P_x (x = 0.5–3.0) samples; (d) CO yield from CO₂ photoreduction over g-C₃N₄ and g-C₃N₄-P_x (x = 0.5–3.0) coupled with CoPc in a CH₃CN : H₂O : TEOA (3 : 1 : 1, 100 mL) solution after 6 h of illumination, 420 nm < λ < 780 nm), 280 mW cm⁻².

The band gaps (E_g) of the samples, determined via the Kubelka–Munk method (Fig. S3), are 2.80 eV (g-C₃N₄), 2.77 eV (g-C₃N₄-P_0.5), 2.74 eV (g-C₃N₄-P_1.0), 2.68 eV (g-C₃N₄-P_1.5), 2.50 eV (g-C₃N₄-P_2.0), and 2.07 eV (g-C₃N₄-P_3.0), indicating a gradual narrowing of E_g with increasing pyridine modification. This narrowing band gap is attributed to the formation of an intramolecular donor–acceptor structure within g-C₃N₄-P_x, resulting from the grafting of electron-withdrawing pyridine onto the g-C₃N₄ framework. This structure promotes intramolecular charge separation, thereby causing a red shift in the absorption edge and enhancing visible-light absorption. Consistent with these changes, the powder colour deepens from light yellow to dark yellow with increasing pyridine content, as shown in Fig. S4.

Mott–Schottky measurements were conducted to determine the flat-band potentials (E_FB) of the synthesized g-C₃N₄ and g-C₃N₄-P_x samples. As shown in Fig. 4b, the positive slopes of the Mott–Schottky plots confirm that all samples exhibit n-type semiconductor behaviour. The E_FB values were obtained by extrapolating the linear region of the plots to the intercept on the x-axis (where 1/C² = 0).^46,47 The calculated E_FB values are −0.99, −1.03, −1.06, −1.06, −1.03 and −1.01 V (vs. the normal hydrogen electrode, NHE) for g-C₃N₄, g-C₃N₄-P_0.5, g-C₃N₄-P_1.0, g-C₃N₄-P_1.5, g-C₃N₄-P_2.0, and g-C₃N₄-P_3.0, respectively. For n-type semiconductors, the conduction band minimum (E_CB) is typically 0.1–0.3 eV higher than E_FB, depending on the electron effective mass and carrier concentration.^48,49 Assuming a difference of 0.2 eV, the estimated E_CB values are −1.19 V, −1.23 V, −1.26 V, −1.26 V, −1.23 V, and −1.21 V (vs. NHE) for g-C₃N₄, g-C₃N₄-P_0.5, g-C₃N₄-P_1.0, g-C₃N₄-P_1.5, g-C₃N₄-P_2.0, and g-C₃N₄-P_3.0, respectively. The Mott–Schottky results indicate that pyridine modification at the edges has a minimal effect on both E_FB and E_CB of the g-C₃N₄-P_x materials.

Based on the E_g values derived from the Tauc plots in Fig. 4a, the valence band energy (E_VB) values of g-C₃N₄, g-C₃N₄-P_0.5, g-C₃N₄-P_1.0, g-C₃N₄-P_1.5, g-C₃N₄-P_2.0 and g-C₃N₄-P_3.0 were calculated to be 1.61, 1.54, 1.48, 1.42, 1.27, and 0.86 V (vs. NHE), respectively. This progressive increase in E_VB indicates that pyridine incorporation at the edges of the g-C₃N₄ framework systematically increases the valence band level. As depicted in the band structure diagram in Fig. 4c, pyridine modification leads to a continuous reduction of the band gap from 2.80 eV to 2.07 eV, accompanied by an upward shift of the E_VB from 1.61 eV to 0.86 eV. In contrast, the conduction band level remains essentially unchanged, with E_CB values confined within a narrow range of −1.19 V to −1.26 V—a variation of only 0.02–0.07 V across all samples. This stability in the conduction band preserves the original charge transfer direction while concurrently suppressing carrier recombination. Although narrow-bandgap semiconductors typically exhibit high charge recombination rates, this work demonstrates an effective strategy for significantly narrowing the band gap without compromising charge separation efficiency.

2.3 Photoreduction of CO₂ over CoPc/g-C₃N₄-P_x hybrid catalysts

First, a series of control experiments was conducted to confirm that the observed products originate exclusively from the photoreduction of CO₂. These experiments were conducted under an argon (Ar) atmosphere rather than CO₂, in the dark without illumination, using only CoPc (without a photosensitizer), using only g-C₃N₄-P_x (without CoPc), and without any hybrid photocatalyst. Table S1 lists the results of the corresponding control experiments for CO₂ photoreduction based on the CoPc/g-C₃N₄-P_1.5 hybrid catalyst. No product was detected under an argon atmosphere, in the dark, in the absence of the CoPc/g-C₃N₄-P_x hybrid catalyst, or with only CoPc (without a photosensitizer). Only trace CH₄ was detected with only g-C₃N₄-P_x (without CoPc) under visible-light irradiation. These results confirm that the CO originates from the photoreduction of CO₂ catalysed by the CoPc/g-C₃N₄-P_x hybrid catalyst, and CoPc itself (without a sensitizer) has no activity for CO₂ reduction under visible-light irradiation.

Under visible-light illumination (420 nm < λ < 780 nm), the prepared CoPc/g-C₃N₄-P_x hybrid catalysts show high photocatalytic activity and selectivity for CO₂ reduction to CO, attributed to the intrinsic CO selectivity of CoPc. Only CO was detected, no CH₄, H₂, C₂ gas, or liquid products in CO₂ reduction based on the CoPc/g-C₃N₄-P_x hybrid photocatalyst were detected. The order of performance under visible light was as follows: CoPc/g-C₃N₄-P_1.5 (14.95 mmol g⁻¹) > CoPc/g-C₃N₄-P_2.0 (13.85 mmol g⁻¹) > CoPc/g-C₃N₄-P_1.0 (12.67 mmol g⁻¹) > CoPc/g-C₃N₄-P_3.0 (8.73 mmol g⁻¹) > CoPc/g-C₃N₄-P_0.5 (5.54 mmol g⁻¹) > CoPc/g-C₃N₄ (2.47 mmol g⁻¹) after 6 hours. The highest CO yield of 14.95 mmol g⁻¹ was achieved with the CoPc/g-C₃N₄-P_1.5 catalyst, which is approximately 6.1 times that of the unmodified CoPc/g-C₃N₄. When pyridine was grafted onto the g-C₃N₄ framework, the CO yield initially increased significantly with increasing pyridine content. Further increasing the pyridine grafting amount resulted in a decrease in the photocatalytic activity. Table S2 lists the recent advances in the photocatalytic reduction of CO₂ based on molecular catalyst/semiconductor hybrid photocatalysts, and the CoPc/g-C₃N₄-P_1.5 hybrid catalyst exhibits competitive performance, especially in terms of CO yield based on non-noble metal molecular/g-C₃N₄ hybrid catalysts under visible-light irradiation.

To evaluate the superior stability of the CoPc/g-C₃N₄-P_1.5 hybrid catalyst compared to the unmodified CoPc/g-C₃N₄, a continuous 12-hour CO₂ photoreduction test was conducted. As shown in Fig. S6, the time-dependent CO evolution was monitored using two catalytic systems: 50 mg of g-C₃N₄-P_1.5 hybrid with 1.0 mg of CoPc, and 50 mg of g-C₃N₄ hybrid with 1.0 mg of CoPc, both dispersed in an acetonitrile/water/triethanolamine (CH₃CN/H₂O/TEOA) mixed solution (3 : 1 : 1, 100 mL) under visible-light irradiation. The CO yield from the CoPc/g-C₃N₄-P_1.5 catalyst increased almost linearly over 12 hours, demonstrating significantly higher efficiency and durability than CoPc/g-C₃N₄. These results confirm the excellent stability of the CoPc/g-C₃N₄-P_1.5 hybrid under visible-light conditions. Notably, the dynamic coordination bond in the CoPc/g-C₃N₄-P_1.5 hybrid interface confers a certain self-repairing capability, thereby enhancing the overall durability of the catalytic system.

2.4 Mechanism of the enhanced photocatalytic performance

2.4.1 EIS Nyquist plots and transient photocurrent response. To elucidate the enhanced photocatalytic performance, electrochemical impedance spectroscopy (EIS) Nyquist plots and transient photocurrent–time response (I–t) curves were recorded using g-C₃N₄ and g-C₃N₄-P_x films as working electrodes. EIS reflects the charge transfer resistance at the electrode/electrolyte interface, while the I–t curves provide further insight into the synergistic effect of light absorption and photo-generated charge separation. The impedance parameters can be derived from the spectra using an equivalent circuit shown in the inset in Fig. 5a. In the equivalent circuit, R_ct corresponds to the charge transfer impedance. The R_ct values of g-C₃N₄, g-C₃N₄-P_0.5, g-C₃N₄-P_1.0, g-C₃N₄-P_1.5, g-C₃N₄-P_2.0 and g-C₃N₄-P_3.0 were fitted to be 2394, 1738, 761, 759, 768, and 1377 Ω, respectively. The R_ct value decreases significantly from g-C₃N₄ to g-C₃N₄-P_1.5, and then gradually increases from g-C₃N₄-P_2.0 to g-C₃N₄-P_3.0. The smallest radius is observed for g-C₃N₄-P_1.5, and accordingly, g-C₃N₄-P_1.5 shows the lowest R_ct value and, therefore, the lowest charge transfer resistance and most efficient charge separation, which correlates well with the CO₂ photoreduction activity shown in Fig. 4d and is further supported by the transient photocurrent results below. These findings suggest that appropriate incorporation of pyridine significantly facilitates charge separation and transfer.

Fig. 5 (a) EIS Nyquist plots (the inset shows the equivalent circuit), and (b) the transient photocurrent response (I–t) of the pristine g-C₃N₄ and g-C₃N₄-P_x (x = 0.5–3.0) samples.

Moreover, under visible-light irradiation, all pyridine-modified samples (g-C₃N₄-P_x) exhibit substantially higher photocurrents than pristine g-C₃N₄ (Fig. 5b). As indicated in Fig. 4a, the enhanced visible-light absorption resulting from pyridine grafting contributes to the improved photocurrent response. The photocurrent increases progressively from g-C₃N₄ to g-C₃N₄-P_1.5, and then decreases from g-C₃N₄-P_1.5 to g-C₃N₄-P_3.0. The trend in photocurrent intensity aligns closely with the photocatalytic CO production performance (Fig. 4d), with minor discrepancies observed for g-C₃N₄-P_1.0 and g-C₃N₄-P_3.0. Among all the samples, g-C₃N₄-P_1.5 demonstrates the highest photocurrent, implying optimal synergy between light harvesting and charge separation. Therefore, appropriate edge-grafting of pyridine is an effective strategy for enhancing the photoelectrochemical properties of g-C₃N₄.

2.4.2 Steady-PL and TR-PL kinetics decay. Steady-state and time-resolved photoluminescence (steady-PL and TR-PL) decay spectra were acquired with excitation at 370 nm and emission monitored at 460 nm to probe the charge carrier dynamics within the materials. As presented in Fig. 6a, both pristine g-C₃N₄ and pyridine-edge-functionalized g-C₃N₄-P_x samples exhibit a broad fluorescence emission band. The pristine g-C₃N₄ demonstrates the highest emission intensity, indicative of strong radiative charge recombination. A notable quenching of the steady-PL intensity is observed across all pyridine-modified samples, particularly from g-C₃N₄-P_1.0 onward. This suppression implies significantly reduced radiative recombination, which we attribute to facilitated electron transfer via non-radiative pathways and enhanced charge separation resulting from pyridine grafting. The introduction of pyridine moieties at the framework edges likely provides charge delocalization channels and promotes interfacial electron transfer, thereby suppressing recombination.

Fig. 6 (a) Steady-PL and (b) TR-PL kinetics decay spectra probed at 460 nm for the pristine g-C₃N₄ and as-synthesized g-C₃N₄-P_x (x = 0.5–3.0) samples at 370 nm excitation.

Concurrently, a systematic red shift in the emission peak from 453 nm (g-C₃N₄) to 570 nm (g-C₃N₄-P_3.0) is observed with increasing pyridine content. This red-shift reflects a narrowing of the effective bandgap due to the formation of an intramolecular donor–acceptor (D–A) structure, wherein the electron-donating triazine units interact with the electron-accepting pyridine on the edge of g-C₃N₄. The variation trend of the red shift in the PL emission peak aligns well with the red shift of the absorption onset shown in Fig. 4a, corroborating the electronic modulation of the material through functionalization. These PL results collectively affirm that pyridine edge-grafting effectively alters the electronic structure of g-C₃N₄, facilitating charge separation and transfer—key factors underlying the enhanced photocatalytic performance.

The TR-PL decay spectra monitored at 460 nm (Fig. 6b) provide further insight into the charge transfer dynamics. All samples exhibit typical exponential decay behaviour, with the pyridine-modified g-C₃N₄-P_x samples demonstrating markedly accelerated decay kinetics compared to pristine g-C₃N₄. Moreover, a progressive acceleration in PL decay is observed from g-C₃N₄-P_0.5 to g-C₃N₄-P_3.0. The fluorescence lifetimes were quantitatively analysed using a tri-exponential decay model, and the fitted results are summarized in Table 1. Each lifetime component (τ₁, τ₂ and τ₃) exhibits a consistent decreasing trend with increasing pyridine content. The average fluorescence lifetime (τ_av) decreases sequentially from 6.11 ns (pristine g-C₃N₄) to 1.84 ns (g-C₃N₄-P_3.0), confirming enhanced non-radiative decay pathways.

Table 1 The fitted parameters for the fluorescence lifetime of the as-synthesized g-C₃N₄-P_x (x = 0–3.0) sample

	g-C3N4	g-C3N4-P0.5	g-C3N4-P1.0	g-C3N4-P1.5	g-C3N4-P2.0	g-C3N4-P3.0
τ1 (ns)	4.03	3.70	3.40	2.66	2.58	2.07
Rel.%	52.03%	44.61%	43.25%	39.82%	36.82%	31.59%
τ2 (ns)	16.67	14.51	12.90	11.1	11.19	8.81
Rel.%	22.33%	25.27%	24.72%	15.86%	12.58%	11.07%
τ3 (ns)	1.14	0.82	0.71	0.51	0.48	0.36
Rel.%	25.64%	30.12%	32.04%	44.32%	50.60%	57.34%
τ (ns)	6.11	5.56	4.89	3.05	2.60	1.84

---

This systematic reduction in lifetime reflects more efficient electron transfer from the g-C₃N₄ matrix to the pyridine functional groups at the material's edge. The introduced pyridine rings serve as electron-accepting sites, facilitating rapid separation and extraction of photo-generated electrons, thereby suppressing radiative recombination. Together, the steady-state and time-resolved PL results conclusively demonstrate that pyridine edge functionalization promotes the separation of photoinduced electron–hole pairs and inhibits their recombination.

2.4.3 Theoretical calculation. To gain deeper mechanistic insights into how pyridine edge-grafting enhances the photocatalytic performance and charge separation in g-C₃N₄-P_x, density functional theory (DFT) calculations were performed to evaluate the molecular dipole moment and the electron density distribution of the highest occupied molecular orbitals (HOMOs) and the lowest unoccupied molecular orbitals (LUMOs). The molecular model of g-C₃N₄ was constructed using three connected 3-s-triazine units, while the g-C₃N₄-P_x model was built by grafting one pyridine ring onto the terminal triazine unit. As shown in Fig. 7, the optimized structures, computed dipole moments, and spatial distributions of the HOMO and LUMO are compared for g-C₃N₄ and g-C₃N₄-P_x. The dipole moment increases from 4.75 debye in g-C₃N₄ to 5.47 debye in g-C₃N₄-P_x, indicating enhanced molecular polarity induced by pyridine functionalization. This greater polarity facilitates improved charge separation and promotes more efficient charge transport within the material.

Fig. 7 Optimized structures, calculated dipole moments, and isodensity surface plots of the HOMO and LUMO for g-C₃N₄ and g-C₃N₄-P_x model structures.

Orbital analysis in Fig. 7 reveals that in pristine g-C₃N₄, the HOMO is delocalized evenly across all three triazine units, whereas the LUMO is predominantly localized on two of the triazine units. In contrast, for g-C₃N₄-P_x, the HOMO is mainly distributed over the two triazine units distal to the pyridine group, with minimal contribution from the pyridine-linked triazine and none from the pyridine itself. The LUMO, however, is primarily located on the pyridine-anchored triazine unit, with additional density extending into the pyridine ring and the remaining triazine motifs. The frontier molecular orbitals, the HOMO and LUMO, correspond to bonding and anti-bonding orbitals, respectively, and dictate the spatial distribution of holes and electrons upon photoexcitation. Compared with pristine g-C₃N₄, the isodensity surface plots of g-C₃N₄-P_x reveal a more pronounced and directional shift of electron density from the 3-s-triazine core toward the edge-anchored pyridine ring under excitation. This redistribution supports efficient photoinduced electron transfer from the central region to the peripheral functional groups, thereby facilitating charge separation and reducing the likelihood of electron–hole recombination. The DFT simulation results align consistently with the fluorescence spectra discussed above.

2.4.4 Mechanism of the optimum photocatalytic performance at P_1.5. Thermodynamically, all g-C₃N₄-P_x samples can transfer electrons to CoPc [E_CB for g-C₃N₄-P_x = −1.19 to −1.26 V vs. NHE; E_red(CoPc) ≈ −0.90 V vs. NHE,⁵⁰ satisfying the requirement for CO₂ reduction (E_(CO2/CO) = −0.52 V vs. NHE)]. The slight variation in E_CB (0.02–0.07 eV) among g-C₃N₄-P_x samples implies negligible differences in driving force for electron transfer, indicating that the enhanced activity does not originate from thermodynamic factors. A charge transfer diagram for CO₂ photoreduction over CoPc/g-C₃N₄-P_x hybrid catalysts is proposed, as shown in Scheme 1. First, photoexcitation of g-C₃N₄-P_x generates electron–hole pairs; second, electrons transfer efficiently to CoPc via pyridine bridges, while holes migrate to the g-C₃N₄-P_x surface, competing with recombination; finally, CO₂ is reduced to CO at CoPc sites, while the holes are scavenged by the sacrificial agent TEOA. As shown in Fig. 4d, the photocatalytic activity of CoPc/g-C₃N₄-P_x depends strongly on the pyridine grafting level (x). The CO₂ reduction performance increases with pyridine loading up to an optimum level at x = 1.5, beyond which it declines. This trend can be explained by the interplay of several key factors.

Scheme 1 Charge transfer diagram for CO₂ photoreduction over CoPc/g-C₃N₄-P_x hybrid photocatalysts under visible-light illumination.

The optimum performance at P_1.5 reflects a kinetic spot between enhanced charge separation and competing recombination pathways. At lower loadings (x < 1.5), increasing pyridine content improves visible-light absorption, facilitates in-plane charge separation within g-C₃N₄, and promotes directional electron transfer from g-C₃N₄-P_x to CoPc. This is supported by the lowest charge-transfer resistance (Fig. 5a) and highest photocurrent response (Fig. 5b) observed at P_1.5. However, further increasing the pyridine loading beyond this point introduces excess surface defects or states that act as charge recombination centres, as indicated by the increased charge-transfer resistance and decreased photocurrent. These defects promote electron–hole recombination before electrons can reach the CoPc active sites. In addition, excess pyridine functionalization leads to an overly elevated valence band maximum (E_VB in Fig. 4c), which reduces the hole oxidation power and weakens the oxidative half-reaction, thereby impairing catalyst regeneration. Moreover, while the optimal pyridine content enables efficient coordination and electron transfer to CoPc, an excess may cause non-productive coordination or steric hindrance, blocking active sites and leading to local electron accumulation. This imbalance between pyridine and CoPc promotes recombination instead of catalysis, ultimately reducing the CO₂ reduction efficiency.^51,52

3 Conclusions

In conclusion, we successfully synthesized pyridine edge-functionalized g-C₃N₄-P_x coupled with CoPc for efficient photocatalytic CO₂ reduction to CO under visible light. In this hybrid system, g-C₃N₄-P_x serves as the light-harvesting unit, while CoPc acts as the catalytic site for CO₂ reduction. The grafting of electron-withdrawing pyridine groups enhances visible-light absorption and promotes directional in-plane charge separation within g-C₃N₄. Unlike conventional CoPc/g-C₃N₄ systems, which primarily rely on physical adsorption, static interactions, or direct covalent linking, our work introduces polarization engineering and coordination via pyridine edge functionalization. This approach not only promotes in-plane charge separation within g-C₃N₄, for improved light absorption, but also enables directional electron transfer and reversible coordination between g-C₃N₄ and CoPc, significantly increasing charge separation and CO₂ reduction performance. The optimal catalyst, CoPc/g-C₃N₄-P_1.5, achieves a CO yield of 14.95 mmol g⁻¹ after 6 h of visible-light irradiation, which is 6.1 times that of the non-functionalized counterpart. The improvement in the photocatalytic activity is mainly due to the increase in the electron velocity between the photosensitizer and catalyst, which emphasizes the superiority of directional electron transfer and the construction of a dynamic coordination interaction. The one-step thermal polymerization used for g-C₃N₄-P_x is simple, scalable, and based on low-cost precursors, and the coordination with CoPc is also straightforward and conducted under mild conditions, making the CoPc/g-C₃N₄-P_x hybrid suitable for large-scale production. This work offers strategic guidelines for designing efficient semiconductor–molecular hybrid photocatalysts for CO₂ conversion into solar fuel.

4 Experimental

4.1 Regents

All the reagents and solvents were chromatographic or analytical grade. Urea, 4-aminopyridine, TEOA, CH₃CN, and N,N-dimethylformamide (DMF) were purchased from Shanghai Aladdin Bio-chem Technology. Tetraethylammonium tetrafluoroborate (Et₄NBF₄) was obtained from Sigma-Aldrich. CO₂ with a purity of 99.999% was bought from Shanxi Beiwen Gases Company. Ultrapure water (Millipore Milli-Q grade, 18.2 MΩ cm) was used in all the experiments.

4.2 Synthesis

Pristine g-C₃N₄ was synthesized by direct thermal polymerization of urea at 560 °C for 4 h with a heating rate of 2.5 °C min⁻¹ in a lidded alumina crucible in a muffle furnace under an ambient atmosphere. The pyridine edge-functionalized g-C₃N₄-P_x samples were prepared through thermal polymerization of urea with different amounts of 4-aminopyridine, as for the synthetic process of g-C₃N₄. The edge-grafted amount of pyridine in the final g-C₃N₄-P_x composites was controlled by varying the quantity of 4-aminopyridine used in the thermal polymerization. The g-C₃N₄-P_x samples synthesized with 4-aminopyridine to urea weight percentages of 0.5%, 1.0%, 1.5%, 2.0%, and 3.0% were denoted as g-C₃N₄-P_0.5, g-C₃N₄-P_1.0, g-C₃N₄-P_1.5, g-C₃N₄-P_2.0, and g-C₃N₄-P_3.0, respectively.

CoPc was synthesized and characterized as previously reported.⁵³ The CoPc/g-C₃N₄-P_x hybrid catalysts were prepared by mixing 1 mg of CoPc with 50 mg of g-C₃N₄-P_x under stirring overnight in a 100 mL solution of CH₃CN/H₂O/TEOA (3 : 1 : 1, v : v : v). FT-IR spectra were recorded to verify the CoPc/g-C₃N₄-P_1.5 heterogeneous catalyst (Fig. S5).

4.3 Characterization

The morphology of the g-C₃N₄-P_x samples was observed on a Hitachi S4800 field emission SEM and FEI Tecnai G2 T20 TEM. XRD was measured on a Bruker D8 diffractometer with Cu Kα radiation. The BET specific surface area was measured by a surface area and porosity analyser (Micromeritics, TriStar II 3020). XPS was carried out on an ESCALAB 250Xi X-ray photoelectron spectrometer. UV-vis DRS was recorded on a Lambda 750 UV/vis/NIR spectrophotometer (Perkin-Elmer, USA) using BaSO₄ as the background. Steady-state PL and TR-PL decay spectra with excitation at 370 nm were collected on a NanoLOG-TCSPC spectrophotometer (Horiba Jobin Yvon, USA). Transient I–t curves and Mott–Schottky plots were tested with a frequency of 1 kHz. EIS measurements were performed under an AC amplitude of 10 mV with a frequency range of 10⁻¹ Hz to 10⁵ Hz. All of the photoelectrochemical measurements were performed in a three-electrode cell on a CHI 660D potentiostat (Shanghai Chenhua) in DMF solution using 0.1 M Et₄NBF₄ as the electrolyte. g-C₃N₄-P_x films on FTO glass, Pt sheet and Ag/AgCl were used as working, counter and reference electrodes, respectively.

4.4 Photocatalytic CO₂ reduction

Photoreduction of CO₂ was performed over CoPc/g-C₃N₄-P_x hybrid photocatalysts in a CH₃CN/H₂O/TEOA (3 : 1 : 1, 100 mL) mixed solution in a photoreaction system (Labsolar-IIIAG, Beijing Perfectlight Technology Co., Ltd.) saturated with 28 kPa of CO₂ at 15 °C, as reported.¹¹ The volume of the reactor was 250 mL, and the volume of the gas circulation pipeline was 150 mL. The photoreaction system was vacuumed and purged with high-purity CO₂ gas for 1 h before illumination. The reactor was illuminated under visible light (420 nm < λ < 780 nm) using a 300 W Xe lamp (Microsolar 300, Beijing Perfectlight Technology Co., Ltd.) with a 420 nm cut-off filter and a 780 nm IR-cut filter. The spot diameter is 60 mm, and the irradiation height is 10 cm. The light intensity was calibrated using a silicon photodiode power meter, with 280 mW cm⁻². The total photon flux was calculated to be approximately 1.26 × 10⁻⁶ photons per s m⁻². The spectrum of Microsolar 300 Xe light (420 nm < λ < 780 nm) is shown in Fig. S7. The gas reduction products were analysed by using an Agilent 7890A gas chromatograph equipped with a hydrogen flame ionization detector (FID) and a thermal conductivity detector (TCD) using He as the carrier gas. The FID was fitted with a methanizer to detect CO. Typical chromatographic curves of CH₄ and CO with residence times and GC calibration curves are shown in Fig. S8 and S9, respectively. A series of control experiments (i.e., purging Ar instead of CO₂, in the dark without illumination, with only CoPc and without the photosensitizer, with only g-C₃N₄-P_x and without CoPc, and without the hybrid photocatalyst) were carried out.

4.5 Theoretical calculation

Theoretical calculation was carried out using Gaussian software using the DFT method.⁵⁴ Becke's three-parameter hybrid functional with the LYP correlation functional (B3LYP) together with the 6-31G(d) basis set was employed. Geometry optimization and electronic property analysis of the g-C₃N₄ and g-C₃N₄-P_x models with the minimum energy conformation were carried out without any symmetry constraint. The molecular dipole moment and electron distribution of the HOMO and LUMO were obtained.

Author contributions

Xuehua Zhang: resources, methodology, formal analysis, data curation, writing – original draft. Chunlei Su: resources, formal analysis, data curation. Rui Shi: data curation. Mingming Li: methodology. Lili Fu: formal analysis. Rongji Liu: methodology, investigation. Yong Chen: writing – review & editing, visualization, validation. Bin Li: formal analysis, review & editing, funding acquisition. Guangjin Zhang: formal analysis, writing – review & editing, supervision, funding acquisition and project administration.

Conflicts of interest

The authors declare no conflict of interest.

Data availability

The data supporting this article have been included as part of the supplementary information (SI).

Supplementary information: the model structure unit of CoPc/g-C₃N₄-P_x hybrid photocatalysts; the possible synthetic scheme of g-C₃N₄-P_x; XPS survey spectra of g-C₃N₄, g-C₃N₄-P_x, CoPc and CoPc/g-C₃N₄-P_1.5 hybrid, XPS spectra of C1s and N1s of g-C₃N₄ and g-C₃N₄-P_x; Tauc plots and photographs of g-C₃N₄ and g-C₃N₄-P_x; IR spectra of g-C₃N₄-P_1.5, CoPc and CoPc/g-C₃N₄-P_1.5 hybrid; time dependence of CO evolution based on CoPc/g-C₃N₄ and CoPc/g-C₃N₄-P_x in 12 h; the out spectrum of Microsolar 300 Xe light; typical chromatographic curves of CH₄ and CO and GC calibration curves; results of the corresponding control experiments for CO₂ photoreduction and recent advances of CO₂ photoreduction based on molecular catalyst/semiconductor hybrid photocatalysts. See DOI: https://doi.org/10.1039/d5im00296f.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (22378402; 22178361), the International Partnership Project of CAS (039GJHZ2022029GC), Fundamental research project of China National Tobacco Corporation (No. 110202403002) and the Major Project of the Innovation Platform at the Key Laboratory of Tobacco Processing Technology, State Tobacco Monopoly Administration (CNTC, 202025AWCX01).

References

1. T. Inoue, A. Fujishima, S. Konishi and K. Honda, Photoelectrocatalytic reduction of carbon dioxide in aqueous suspensions of semiconductor powders, Nature, 1979, 277, 637–638.
2. W. Tu, Y. Zhou and Z. Zou, Photocatalytic conversion of CO₂ into renewable hydrocarbon fuels: State-of-the-art accomplishment, challenges, and prospects, Adv. Mater., 2014, 26, 4607–4626.
3. X. Li, J. Yu, M. Jaroniec and X. Chen, Cocatalysts for selective photoreduction of CO₂ into solar fuels, Chem. Rev., 2019, 119, 3962–4179.
4. Y. Wang, E. Chen and J. Tang, Insight on reaction pathways of photocatalytic CO₂ conversion, ACS Catal., 2022, 12, 7300–7316.
5. C. Huang, X. Zhang, D. Li, M. Wang and Q. Wu, The influence of the precursor molar ratio on the structure of the CdS catalyst during synthesis and visible-light driven CO₂ reduction into solar fuel, New J. Chem., 2022, 46, 10339–10346.
6. D. Zhang and H. Wang, Polyoxometalate-based nanostructures for electrocatalytic and photocatalytic CO₂ reduction, Polyoxometalates, 2022, 1, 9140006.
7. T. Morikawa, S. Sato, K. Sekizawa, T. Suzuki and T. Arai, Solar-driven CO₂ reduction using a semiconductor/molecule hybrid photosystem: From photocatalysts to a monolithic artificial leaf, Acc. Chem. Res., 2022, 55, 933–943.
8. S. Zhang, R. Liu, C. Streb and G. Zhang, Design and synthesis of novel polyoxometalate-based binary and ternary nanohybrids for energy conversion and storage, Polyoxometalates, 2023, 2, 9140037.
9. 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.
10. G. Zhao, H. Pang, G. Liu, P. Li, H. Liu, H. Zhang, L. Shi and J. Ye, Co-porphyrin/carbon nitride hybrids for improved photocatalytic CO₂ reduction under visible light, Appl. Catal., B, 2017, 200, 141–149.
11. L. Lin, C. Hou, X. Zhang, Y. Wang, Y. Chen and T. He, Highly efficient visible-light driven photocatalytic reduction of CO₂ over g-C₃N₄ nanosheets/tetra(4-carboxyphenyl)porphyrin iron(III) chloride heterogeneous catalysts, Appl. Catal., B, 2018, 221, 312–319.
12. X. Zhang, L. Lin, D. Qu, J. Yang, Y. Weng, Z. Wang, Z. Sun, Y. Chen and T. He, Boosting visible-light driven solar-fuel production over g-C₃N₄/tetra(4-carboxyphenyl)porphyrin iron(III) chloride hybrid photocatalyst via incorporation with carbon dots, Appl. Catal., B, 2020, 265, 118595.
13. 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.
14. Y. Wei, L. Chen, H. Chen, L. Cai, G. Tan, Y. Qiu, Q. Xiang, G. Chen, T.-C. Lau and M. Robert, Highly efficient photocatalytic reduction of CO₂ to CO by in situ formation of a hybrid catalytic system based on molecular iron quaterpyridine covalently linked to carbon nitride, Angew. Chem., Int. Ed., 2022, 61, e202116832.
15. P. Li, C. Hou, X. Zhang, Y. Chen and T. He, Ethylenediamine-functionalized CdS/tetra(4-carboxyphenyl)porphyrin iron (III) chloride hybrid system for enhanced CO₂ photoreduction, Appl. Surf. Sci., 2018, 459, 292–299.
16. P. Li, X. Zhang, C. Hou, L. Lin, Y. Chen and T. He, Visible-light-driven CO₂ photoreduction over Zn_xCd_1-xS solid solution coupling with tetra(4-carboxyphenyl)porphyrin iron(III) chloride, Phys. Chem. Chem. Phys., 2018, 20, 16985–16991.
17. P. Li, X. Zhang, C. Hou, Y. Chen and T. He, Highly efficient visible-light driven solar-fuel production over tetra(4-carboxyphenyl)porphyrin iron(III) chloride using CdS/Bi₂S₃ heterostructure as photosensitizer, Appl. Catal., B, 2018, 238, 656–663.
18. A. Perazio, G. Lowe, R. Gobetto, J. Bonin and M. Robert, Light-driven catalytic conversion of CO₂ with heterogenized molecular catalysts based on fourth period transition metals, Coord. Chem. Rev., 2021, 443, 214018.
19. F. Arcudi, L. Đorđević, B. Nagasing, S. I. Stupp and E. A. Weiss, Quantum dot-sensitized photoreduction of CO₂ in water with turnover number >80,000, J. Am. Chem. Soc., 2021, 143, 18131–18138.
20. J. Wang, L. Jiang, H. Huang, Z. Han and G. Ouyang, Rapid electron transfer via dynamic coordinative interaction boosts quantum efficiency for photocatalytic CO₂ reduction, Nat. Commun., 2021, 12, 4276.
21. J. Bian, J. Feng, Z. Zhang, Z. Li, Y. Zhang, Y. Liu, S. Ali, Y. Qu, L. Bai, J. Xie, D. Tang, X. Li, F. Bai, J. Tang and L. Jing, Dimension-matched zinc phthalocyanine/BiVO₄ ultrathin nanocomposites for CO₂ reduction as efficient wide-visible-light-driven photocatalysts via a cascade charge transfer, Angew. Chem., Int. Ed., 2019, 58, 10873–10878.
22. J. Sun, J. Bian, J. Li, Z. Zhang, Z. Li, Y. Qu, L. Bai, Z. Yang and L. Jing, Efficiently photocatalytic conversion of CO₂ on ultrathin metal phthalocyanine/g-C₃N₄ heterojunctions by promoting charge transfer and CO₂ activation, Appl. Catal., B, 2020, 277, 119199.
23. S. Xu, X. Li, S. Li, H. Rao, J.-S. Qin, P. She, W.-C. Cheong and L. Jing, Recent advances of photocatalytic CO₂ reduction based on hybrid molecular catalyst/semiconductor photocatalysts: A review, Small, 2025, 21, 2408075.
24. S. McGuigan, S. J. Tereniak, C. L. Donley, A. Smith, S. Jeon, F. Zhao, R. N. Sampaio, M. Pauly, L. Keller, L. Collins, G. N. Parsons, T. Lian, E. A. Stach and P. A. Maggard, Discovery of a hybrid system for photocatalytic CO₂ reduction via attachment of a molecular cobalt-quaterpyridine complex to a crystalline carbon nitride, ACS Appl. Energy Mater., 2023, 6, 10542–10555.
25. Y.-H. Li, Y. Chen, J.-Y. Guo, R. Wang, S.-N. Zhao, G. Li and S.-Q. Zang, Engineering coordination microenvironments of polypyridine Ni catalysts embedded in covalent organic frameworks for efficient CO₂ photoreduction, Chin. J. Catal., 2025, 74, 155–166.
26. J.-W. Wang, F. Zhao, H.-H. Huang, Z. Han and G. Ouyang, Molecular catalyst coordinatively bonded to organic semiconductors for selective light-driven CO₂ reduction in water, Nat. Commun., 2024, 15, 9779.
27. M. Schulz, M. Karnahl, M. Schwalbe and J. Vos, The role of the bridging ligand in photocatalytic supramolecular assemblies for the reduction of protons and carbon dioxide, Coord. Chem. Rev., 2012, 256, 1682–1705.
28. X. Wang, K. Maeda, A. Thomas, K. Takanabe, 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.
29. W. Ong, L. Tan, Y. Ng, S. Yong and S. Chai, Graphitic carbon nitride (g-C₃N₄)-based photocatalysts for artificial photosynthesis and environmental remediation: Are we a step closer to achieving sustainability?, Chem. Rev., 2016, 116, 7159–7329.
30. J. Wen, J. Xie, X. Chen and X. Li, A review on g-C₃N₄-based photocatalysts, Appl. Surf. Sci., 2017, 391, 72–123.
31. R. Umapathi, C. V. Raju, S. M. Ghoreishian, G. M. Rani, K. Kumar, M.-H. Oh, J. P. Park and Y. S. Huh, Recent advances in the use of graphitic carbon nitride-based composites for the electrochemical detection of hazardous contaminants, Coord. Chem. Rev., 2022, 470, 214708.
32. S. M. Ghoreishian, K. S. Ranjith, M. Ghasemi, B. Park, S.-K. Hwang, N. Irannejad, M. Norouzi, S. Y. Park, R. Behjatmanesh-Ardakani, S. M. Pourmortazavi, S. Mirsadeghi, Y.-K. Hanb and Y. S. Huh, Engineering the photocatalytic performance of B-C₃N₄@Bi₂S₃ hybrid heterostructures for full-spectrum-driven Cr(VI) reduction and in-situ H₂O₂ generation: Experimental and DFT studies, Chem. Eng. J., 2023, 452, 139435.
33. Y. Li, T. Kong and S. Shen, Artificial photosynthesis with polymeric carbon nitride: When meeting metal nanoparticles, single atoms, and molecular complexes, Small, 2019, 15, 1900772.
34. Q. Chen, S. Li, H. Xu, G. Wang, Y. Qu, P. Zhu and D. Wang, Co-MOF as an electron donor for promoting visible-light photoactivities of g-C₃N₄ nanosheets for CO₂ reduction, Chin. J. Catal., 2020, 41, 514–523.
35. Q. Chen, G. Gao, Y. Zhang, Y. Li, H. Zhu, P. Zhu, Y. Qu, G. Wang and W. Qin, Dual functions of CO₂ molecular activation and 4f levels as electron transport bridges in erbium single atom composite photocatalysts therefore enhancing visible-light photoactivities, J. Mater. Chem. A, 2021, 9, 15820–15826.
36. C. Li, H. Wu, D. Zhu, T. Zhou, M. Yan, G. Chen, J. Sun, G. Dai, F. Ge and H. Dong, High-efficient charge separation driven directionally by pyridine rings grafted on carbon nitride edge for boosting photocatalytic hydrogen evolution, Appl. Catal., B, 2021, 297, 120433.
37. Z. Ma, X. Zong, Q. Hong, L. Niu, T. Yang, W. Jiang, D. Qu, L. An, X. Wang, Z. Kang and Z. Sun, Electrostatic potential of the incorporated asymmetry molecules induced high charge separation efficiency of the modified carbon nitride copolymers, Appl. Catal., B, 2022, 319, 121922.
38. Q. Zhang, J. Chen, H. Che, P. Wang, B. Liu and Y. Ao, Recent advances in g-C₃N₄-based donor–acceptor photocatalysts for photocatalytic hydrogen evolution: An exquisite molecular structure engineering, ACS Mater. Lett., 2022, 4, 2166–2186.
39. Y. Wu, Z. Jiang, X. Lu, Y. Liang and H. Wang, Domino electroreduction of CO₂ to methanol on a molecular catalyst, Nature, 2019, 575, 639–642.
40. S. Ren, D. Joulié, D. Salvatore, K. Torbensen, M. Wang, M. Robert and C. Berlinguette, Molecular electrocatalysts can mediate fast, selective CO₂ reduction in a flow cell, Science, 2019, 365, 367–369.
41. W. W. Kramer and C. C. L. McCrory, Polymer coordination promotes selective CO₂ reduction by cobalt phthalocyanine, Chem. Sci., 2016, 7, 2506–2515.
42. K. E. R. Cruz, Y. Liu, T. L. Soucy, P. M. Zimmerman and C. C. L. McCrory, Increasing the CO₂ reduction activity of cobalt phthalocyanine by modulating the σ-donor strength of axially coordinating ligands, ACS Catal., 2021, 11, 13203–13216.
43. T. L. Soucy, W. S. Dean, J. Zhou, K. E. R. Cruz and C. C. L. McCrory, Considering the influence of polymer−catalyst interactions on the chemical microenvironment of electrocatalysts for the CO₂ reduction reaction, Acc. Chem. Res., 2022, 55, 252–261.
44. H. Niu, Y. Liu, B. Mao, N. Xin, H. Jia and W. Shi, In-situ embedding MOFs-derived copper sulfide polyhedrons in carbon nanotube networks for hybrid supercapacitor with superior energy density, Electrochim. Acta, 2020, 329, 135130.
45. Z. Lin and X. Wang, Nanostructure engineering and doping of conjugated carbon nitride semiconductors for hydrogen photosynthesis, Angew. Chem., Int. Ed., 2013, 52, 1735–1738.
46. S. Morrison, Electrochemistry at semiconductor and cxidized metal electrodes, Plenum Press, New York, 1980.
47. A. Ishikawa, T. Takata, J. Kondo, M. Hara, H. Kobayashi and K. Domen, Oxysulfide Sm₂Ti₂S₂O₅ as a stable photocatalyst for water oxidation and reduction under visible light irradiation (λ ≤ 650 nm), J. Am. Chem. Soc., 2002, 124, 13547–13553.
48. Y. Matsumoto, Energy positions of oxide semiconductors and photocatalysis with iron complex oxides, J. Solid State Chem., 1996, 126, 227–234.
49. Y. Matsumoto, K. Omae, I. Watanabe and E. Sato, Photoelectrochemical properties of the Zn-Ti-Fe spinel oxides, J. Electrochem. Soc., 1986, 133, 711–715.
50. S. Roy and E. Reisner, Visible-light-driven CO₂ reduction by mesoporous carbon nitride modified with polymeric cobalt phthalocyanine, Angew. Chem., Int. Ed., 2019, 58, 12180–12184.
51. S. M. Ghoreishian, K. S. Ranjith, H. Lee, H. Ju, S. Z. Nikoo, Y.-K. Han and Y. S. Huh, Hierarchical N-doped TiO₂@Bi₂W_xMo_1-xO₆ core-shell nanofibers for boosting visible–light–driven photocatalytic and photoelectrochemical activities, J. Hazard. Mater., 2020, 391, 122249.
52. S. M. Ghoreishian, K. S. Ranjith, B. Park, S.-K. Hwang, R. Hosseini, R. Behjatmanesh-Ardakani, S. M. Pourmortazavi, H. U. Lee, B. Son, S. Mirsadeghi, Y.-K. Han and Y. S. Huh, Full-spectrum-responsive Bi₂S₃@CdS S-scheme heterostructure with intimated ultrathin RGO toward photocatalytic Cr(VI) reduction and H₂O₂ production: Experimental and DFT studies, Chem. Eng. J., 2021, 419, 129530.
53. F. Moser and A. Thomas, The Phthalocyanines, Vol. II: Properties, CRC Press, Boca Raton, 1983.
54. M. Frisch, et al., Gaussian 09, Revision D.01, Gaussian, Inc., Wallingford CT, 2016.

---