Integrating β-ketoenamine linkages into covalent organic frameworks toward efficient overall photocatalytic hydrogen peroxide production

Abstract

Covalent organic frameworks (COFs) are promising photocatalysts for hydrogen peroxide (H₂O₂) production from water and oxygen. However, the H₂O₂ generation from a dual-channel pathway, the oxygen reduction reaction (ORR), and especially the water oxidation reaction (WOR), is still challenging for COF photocatalysts. Here, a series of COFs were constructed by introducing different amounts of hydroxy groups into the COF skeleton to adjust the redox potentials, exciton dissociation and active center. Overall, the β-ketoenamine-linked COF, Tz–THBZ, exhibited a more positive oxidation potential and lower exciton binding energy, leading to efficient photo-excited electron and hole separation ability, and shows an overall photocatalytic H₂O₂ production yielding a rate of 4688 μmol h⁻¹ g⁻¹ in pure water. Density functional theory calculations further demonstrate the keto-formed benzene and triazine rings as the photooxidation and reduction centers.

---

Introduction

Hydrogen peroxide (H₂O₂) ranks among the top 100 most important chemicals globally with a demand exceeding 5 million tons annually,¹ and shows extensive applications in industries such as textile, pulp and paper bleaching.² Nowadays, the dominance of the anthraquinone oxidation method accounts for over 95% of industrial H₂O₂ production, while its high energy consumption and generation of hazardous by-products limit its environmental sustainability.³ The emergence of photocatalytic H₂O₂ production is a promising technology that offers a sustainable approach to convert solar energy to chemical energy.⁴ The photocatalytic production of H₂O₂ typically involves two key half-reactions: the oxygen reduction reaction (ORR) and the water oxidation reaction (WOR).⁵ Currently, most established photocatalysts produce H₂O₂ through the ORR pathway, often relying on sacrificial reagents (such as ethanol or benzyl alcohol) to quench photogenerated holes to achieve significant production efficiency.^6–8 To promote solar energy utilization efficiency, optimizing the ORR pathway and adequately leveraging the WOR pathway to directly generate H₂O₂ from H₂O are crucial. However, challenges arise due to the high oxidation potential in thermodynamic terms (e.g., the conversion of H₂O to H₂O₂ requires 1.78 V vs. NHE at pH = 0) and the presence of competitive reactions such as hydroxyl radical or oxygen generation, limiting the overall performance of the WOR pathway.^9,10 Furthermore, the microenvironment of the reaction sites plays a critical role in which the hydrophilic environment can facilitate the water adsorption that results in the acceleration of the WOR process.¹¹ Therefore, developing dual-channel H₂O₂ production that can synergistically coordinate the ORR and WOR pathways to achieve maximum solar energy and 100% atomic utilization remains a significant challenge in advancing photocatalytic H₂O₂ production.

Organic photocatalysts, such as graphitic carbon nitride (g-C₃N₄),¹² conjugated microporous polymers (CMPs),¹³ covalent triazine frameworks (CTFs),¹⁴ and conjugated linear polymers (CLPs),¹⁵ have emerged as promising candidates for photocatalytic H₂O₂ production. These polymers show large surface areas, tunable electronic and optical properties, and functional diversity, offering significant potential in this field.¹⁶ Among them, covalent organic frameworks (COFs) have emerged as promising candidates for H₂O₂ photosynthesis due to their strong covalent bonds, crystalline structures, and ordered pore arrangements,^7,9,17 which allow COFs to provide highly accessible active centers with reactants. The structural versatility also allows COFs to tune the photoelectronic properties at the molecular level. Various strategies have been employed to boost photocatalytic H₂O₂ production with COFs. Van Der Voort and co-workers constructed diarylamine units into COFs to promote strong reduction properties to efficiently reduce oxygen to H₂O₂;⁷ Han's group introduced oxygen functional groups as a strong electron/proton extractor in COFs to enhance the photocatalytic H₂O₂ production;¹⁸ Thomas and co-workers using a three-component Doebner reaction synthesized multicomponent COFs with acid–base bifunctionality for efficient photocatalytic H₂O₂ evolution;¹⁹ Xu and co-workers introduced acetylene or diacetylene moieties into CTFs to enhance photocatalytic H₂O₂ production.¹⁴ Despite these advancements, only a limited number of studies exhibited a successful H₂O₂ generation using COF-based photocatalysts through both the WOR and ORR pathways. For instance, the work by Tang and co-workers has reported thiophene-containing COFs by adjusting the N-heterocycle modules to achieve dual ORR and WOR performance.²⁰ Challenges including insufficient electron–hole separation and limited oxidizing ability often hinder the exciton dissociation and photoexcited hole utilization, resulting in poor WOR ability. Moreover, the availability of catalytic sites is also important, necessitating stronger interactions with reactants (e.g. O₂ and H₂O).

In this study, a series of COFs, namely Tz–HBZ, Tz–DHBZ, and Tz–THBZ, were synthesized by incorporating varying numbers of hydroxy groups. The presence of electron-donating hydroxy groups allows the formation of a β-ketoenamine structure for the modulation of oxidation ability and the planar structure in the COFs, facilitating efficient charge transfer and creating a viable WOR pathway for enhanced H₂O₂ photosynthesis.^21,22 Among the three COFs, Tz–THBZ, featuring irreversible β-ketoenamine linkages, exhibited the highest H₂O₂ production yield of 4688 μmol h⁻¹ g⁻¹ under visible light (λ > 420 nm) irradiation compared to Tz–HBZ and Tz–DHBZ. Additionally, Tz–THBZ achieved a solar-to-chemical conversion (SCC) efficiency of 0.36% and demonstrated excellent stability over six cycles (equivalent to 60 hours) with minimal changes in its chemical structure. Control experiments confirmed that all three COFs produced H₂O₂ through both the ORR and WOR dual-channel pathways, underscoring the strategic utilization of the WOR route to enhance the photocatalytic efficiency of H₂O₂ production.

Results and discussion

Synthesis and structural characterization

Three ordered COFs, Tz–HBZ, Tz–DHBZ and Tz–THBZ, were constructed under solvothermal conditions via the Schiff-base condensation reaction. As illustrated in Fig. 1, the 4,4′,4′′-(1,3,5-triazine-2,4,6-triyl)trianiline (Tz) as an amine monomer was reacted with different aldehyde precursors, 2-hydroxy-1,3,5-benzenetricarboxaldehyde (HBZ), 2,4-dihydroxy-1,3,5-benzenetricarboxaldehyde (DHBZ) and 1,3,5-triformylphloroglucinol (THBZ), which contain varying numbers of hydroxy groups (Fig. S1–S3†). Tz–THBZ exhibited an irreversible keto–enol tautomerization towards the keto form, while Tz–HBZ and Tz–DHBZ displayed tautomerization reversibility. All the COFs exhibited a similar 2D layer structure with hexagonal topology, displaying long-range ordered characteristics. The crystal structures of these COFs were analyzed using powder X-ray diffraction (PXRD). The most intense peak at 6.1° was attributed to the (100) planes, while additional weak reflections and a broad reflection around 9.9°, 11.8°, and 25.4° corresponded to the (110), (200), and (001) planes, respectively (Fig. 1a, b and c). By comparing the experimental PXRD with calculated patterns, it can be confirmed that the structural arrangement of the three COFs matched well with the simulated eclipsed (AA) stacked models through Pawley refinements (Fig. S5†).

Fig. 1 The synthetic routes and chemical structures of the COFs (top). C: grey; N: blue and O: red. Pawley refinements of the PXRD patterns of (a) Tz–HBZ; (b) Tz–DHBZ and (c) Tz–THBZ. Experimental diffraction patterns (green, orange and blue for Tz–HBZ, Tz–DHBZ and Tz–THBZ), Pawley refinement profile (red), simulated structural model pattern (black) and residual (grey).

Solid-state ¹³C CP-MAS NMR spectroscopy and Fourier transform infrared (FT-IR) spectroscopy were carried out to investigate the chemical structure of the synthesized COFs. As depicted in Fig. 2a, all three COFs exhibited peaks around 170 ppm and 140–106 ppm which were attributed to the carbon atoms in the triazine and aromatic rings, respectively.^14,23 Tz–THBZ displayed a prominent peak at approximately 183 ppm, corresponding to the carbonyl carbon (C=O) group and indicating the prevalence of the keto form.^24,25 In contrast, the absence of a peak around 180 ppm in Tz–HBZ, along with the appearance of a peak at 157 ppm for the imine bond (C=N),^25,26 suggested the presence of the enol form. Owing to the reversible enol-to-keto tautomerization, Tz–DHBZ exhibited both signals at around 183 ppm and 157 ppm, related to the C=O and C=N group.²⁷ Furthermore, all three COFs exhibited characteristic peaks in the FT-IR spectra (Fig. 2b and S6†), notably at 1628 cm⁻¹ and 1580–1509 cm⁻¹, corresponding to the stretching vibration of the C=N and C=C bonds.²⁸ Additionally, an increased peak at 1287 cm⁻¹ in Tz–DHBZ and Tz–THBZ was attributed to the stretching vibration of the C–N bond.²²

Fig. 2 (a) Solid-state ¹³C CP-MAS NMR spectra of the COFs. (b) Transmission FT-IR spectra of the COFs. (c) Nitrogen adsorption/desorption isotherms for the COFs. (d) The water contact angle of the COFs.

The permanent porosity of the synthesized COFs was confirmed through nitrogen adsorption–desorption experiments. The apparent Brunauer–Emmett–Teller (BET) surface areas of the three COFs were calculated to be 1334.6, 973.5 and 1032.7 m² g⁻¹ for Tz–HBZ, Tz–DHBZ and Tz–THBZ, respectively (Fig. 2c). The pore size distribution analysis using the nonlocal density functional theory (NLDFT) and cylindrical pore model revealed that the pore size of all the COFs was predominantly centered around 1.2 nm and also showed some other distributions which may attributed to the stacking faults in the COF structure (Fig. S7†). Scanning electron microscopy (SEM) and high-resolution transmission electron microscopy (HRTEM) images depicted nanorod and nanoparticle structures for all three COFs (Fig. S8 and S9†) and HRTEM images also observed clear lattice fringes for all COFs (Fig. S9†). The water contact angles decreased from 57.4° to 35.5° with the increase of hydroxy groups from Tz–HBZ to Tz–THBZ, indicating that Tz–THBZ shows stronger interactions with water (Fig. 2d). Thermal stability for all COFs was determined using thermogravimetric analysis (TGA) which showed that Tz–THBZ exhibited exceptional stability up to 400 °C under a nitrogen atmosphere (Fig. S10†). Tz–THBZ maintained its crystallinity even after immersion in a 1 M H₂O₂ solution for 48 hours in contrast to Tz–HBZ and Tz–DHBZ (Fig. S11†). The more stable chemical structure belongs to the more hydrogen bonding interactions within the molecule, which is highly desired for photocatalytic H₂O₂ production.

Photophysical and photochemical property study

The ultraviolet-visible diffuse reflection spectroscopy (UV/vis DRS) spectra revealed strong absorption of all the COFs in the ultraviolet and visible regions (Fig. 3a). The optical band gaps (E_g) were calculated to be 1.85 eV for Tz–HBZ, 1.96 eV for Tz–DHBZ, and 2.25 eV for Tz–THBZ using Tauc plots (Fig. S12†). Mott–Schottky plots showed positive slopes for all COFs, indicating n-type semiconductor behavior (Fig. S13†). The conduction band (CB) potentials were estimated to be −0.46, −0.41, and −0.39 V (vs. NHE, pH = 7) for Tz–HBZ, Tz–DHBZ, and Tz–THBZ, respectively, based on the Nernst equation. Combining the band gaps and CB potentials, the valence band (VB) potentials were also calculated to be 1.39, 1.55, and 1.85 V (vs. NHE, pH = 7) using the equation (E_VB − E_CB = E_g). The band positions, as shown in Fig. 3c, indicated that these COFs are thermodynamically feasible for overall photocatalytic H₂O₂ production via both the ORR and WOR pathways. In particular, Tz–THBZ showed much more positive potential, implying a stronger oxidizing ability. Density functional theory (DFT) simulations also demonstrated that Tz–THBZ exhibited a more extensive distribution of the highest occupied molecular orbitals (HOMO) compared to the other two COFs (Fig. S14†). In particular, Tz–THBZ showed a more positive VB potential than the others, indicating a favorable HOMO energy level beneficial for photoinduced oxidation reactions.²²

Fig. 3 (a) Solid-state UV/vis absorption spectra of the COFs. (b) Time-resolved PL spectra of the COFs. (c) Energy levels of the COFs. Temperature-dependent PL spectra (inset) and corresponding fitting curves of (d) Tz–HBZ, (e) Tz–DHBZ and (f) Tz–THBZ.

Based on the feasible thermodynamic properties of these COFs for redox reactions, an in-depth analysis of the photochemical properties was conducted to gain a comprehensive understanding of the photocatalysis process. In Fig. S15,† all COFs exhibited significantly enhanced light response compared to the dark conditions, while Tz–THBZ demonstrated superior photocurrent response capability to Tz–HBZ and Tz–DHBZ. Electrochemical impedance spectroscopy (EIS) curves (Fig. S16†) further illustrated that Tz–THBZ displayed a reduced arc radius, indicative of the smallest charge transfer resistance within the system. The time-resolved photoluminescence (TR-PL) spectra revealed that Tz–THBZ boasted a fluorescence lifetime of 1.13 ns, surpassing the 0.93 ns values for both Tz–HBZ and 1.02 for Tz–DHBZ (Fig. 3b). This longer fluorescence lifetime suggests that Tz–THBZ possesses an extended carrier diffusion time and superior charge transfer capability. To delve deeper into the separation behaviour of photoexcited electron–hole pairs, temperature-dependent photoluminescence (TD-PL) spectra were employed to estimate the exciton binding energy for these COFs. As depicted in Fig. 3d–f (insets), the PL intensity of all three COFs decreased monotonically as the temperature rose from 80 K to 298 K. The exciton binding energy (E_b) was then estimated by combining PL intensity and temperature using the Arrhenius equation.²⁹ By utilizing the Arrhenius equation, the E_b was determined to be 57.8 meV for Tz–HBZ, 59.0 meV for Tz–DHBZ, and notably lower at 25.6 meV for Tz–THBZ. The lower E_b value for Tz–THBZ suggests that it is more favourable for exciton dissociation and charge delocalization during the photocatalytic process,³⁰ aligning with its superior photochemical behaviour discussed previously.

Photocatalytic H₂O₂ production performances

The photocatalytic H₂O₂ production experiments were initially conducted to assess the performance of the three COF-based catalysts under visible light (λ > 420 nm) irradiation in an O₂-saturated aqueous solution without using any sacrificial agents. As illustrated in Fig. 4a, all three COFs exhibited steady H₂O₂ production as the reaction time extended to 5 hours. All the COFs showed a much higher performance than their ingredients and physical mixtures, indicating the necessity of the COF structure (Fig. S17†). The H₂O₂ production yields were measured to be 1880 μmol h⁻¹ g⁻¹ for Tz–HBZ, 1250 μmol h⁻¹ g⁻¹ for Tz–DHBZ, and 2580 μmol h⁻¹ g⁻¹ for Tz–THBZ. The higher performance of Tz–THBZ can be attributed to its enhanced hydrophilicity, stronger oxidation potential and more favourable exciton dissociation. Furthermore, by optimizing the dosage to prevent catalyst aggregation and ensuring efficient light utilization, the performance of Tz–THBZ was further enhanced to reach 4688 μmol h⁻¹ g⁻¹ (Fig. S18†).³¹ The reproducibility of the photocatalytic performance of Tz–THBZ was confirmed by conducting photocatalysis experiments using three different batches of the material under the same conditions. The results revealed that all three batches of the as-prepared Tz–THBZ exhibited similar photocatalytic performance (Fig. S19†). The wavelength-dependent external quantum efficiency (EQE) of Tz–THBZ was found to align closely with the UV/vis DRS spectra, confirming that the reaction was initiated by the absorption of incident photons (Fig. 4b). Tz–THBZ achieved the highest EQE of 5.5% at 420 nm. Furthermore, the solar-to-chemical efficiency of Tz–THBZ, evaluated under simulated sunlight (AM 1.5 G, 100 mW cm⁻²), was calculated to be 0.36%, significantly surpassing the solar-to-biomass efficiency of plants (∼0.1%).³²

Fig. 4 (a) Photocatalytic H₂O₂ production performances of Tz–HBZ, Tz–DHBZ and Tz–THBZ (10 mg catalyst in 50 mL water under visible light (λ > 420 nm)). (b) EQE of Tz–THBZ at various incident light wavelengths. (c) Photocatalytic cycle performance of Tz–THBZ (50 mg catalyst in 50 mL water under visible light, λ > 420 nm, 10 hours for each cycle). (d) H₂O₂ decomposition measurement of the COFs (10 mg catalyst in 50 mL 0.1 M H₂O₂ solution in a N₂ atmosphere under visible light (λ > 420 nm)).

The photo-stability of a photocatalyst is a crucial parameter for its potential applications (Fig. 4c). The photocatalytic performance of Tz–THBZ demonstrated excellent stability over 6 cycles with a total reaction time of 60 hours. Only a slight decrease was observed in the last cycle, likely attributed to inevitable catalyst losses during the recycling process (Fig. S20†).³³ The stability of Tz–THBZ surpassed that of most reported organic materials (Table S1†). In contrast to Tz–HBZ and Tz–DHBZ, where aromatic signals were detected in the ¹H-NMR spectra of the photocatalytic solution after 50 hours of reaction (Fig. S21 and S22†), Tz–THBZ showed negligible changes, at least at the NMR level, in its chemical structure for the recycled catalyst (Fig. S21 and S22†). Following the reaction, it is important to prevent the generated H₂O₂ from decomposing in a light-driven process.³⁴ As depicted in Fig. 4d, even in the absence of any catalyst, the H₂O₂ concentration gradually decreased after 5 hours of irradiation. Notably, Tz–HBZ and Tz–DHBZ were found to facilitate this decomposition process further, which might be the promoted H₂O₂ adsorption of the imine groups in the COF skeleton.³⁵ In contrast, the presence of β-ketoenamine-linked Tz–THBZ inhibited the H₂O₂ decomposition, indicating its superior oxidizing ability to utilize holes for H₂O₂ generation through the WOR rather than decomposition.

Photocatalytic mechanism

The photocatalytic reactions were further investigated under different gas atmospheres to understand the performance of the COF-based catalysts. In Fig. 5a, it is observed that the performance of all COFs improved as the gas atmosphere changed from nitrogen to air to oxygen, indicating that the photocatalytic H₂O₂ production reaction was primarily driven by oxygen-involved reactions. In the nitrogen atmosphere, Tz–HBZ, Tz–DHBZ, and Tz–THBZ show the H₂O₂ yields of 390, 470, and 940 μmol h⁻¹ g⁻¹, respectively (Fig. S23†). This production showed a linear increase over 5 hours of irradiation, highlighting the ability of these COFs to drive the WOR through H₂O to H₂O₂ conversion. These data suggest that all COFs possess dual-channel photocatalytic H₂O₂ production capabilities through WOR and ORR pathways.

Fig. 5 (a) Photocatalytic reaction under different atmospheres (10 mg catalyst in 50 mL water under visible light (λ > 420 nm) for 1 hour). (b) The photocatalytic reaction using different quenching agents under N₂ (b) and (c) O₂ atmospheres. (d and e) In situ ATR-SEIRAS spectra vs. illumination time for the photocatalytic system of Tz–THBZ. (f) Electrostatic surface potential maps of Tz–THBZ. DFT-calculated adsorption configuration of (g) O₂ and (h) H₂O on keto-formed Tz–THBZ.

Further investigations were conducted to elucidate the photocatalytic pathway. Under dark conditions, no H₂O₂ was detected. The introduction of quenching agents such as potassium bromate (KBrO₃), isopropanol (IPA), and p-benzoquinone (p-BQ) revealed insights into the involvement of electrons, holes, and intermediates during the photocatalytic reaction. In a nitrogen atmosphere, the presence of methanol (hole quenching agent) inhibited performance compared to the blank condition (pure water). The presence of KBrO₃ (1 mM) can quench electrons that increase H₂O₂ production yields for all COFs, which demonstrates the occurrence of the WOR (Fig. 5b and S24†). Moreover, under an oxygen atmosphere, the introduction of IPA as a hole-quenching agent significantly boosted performance, indicating the dominance of the ORR. In contrast, the presence of p-BQ inhibited H₂O₂ production yields, underscoring the crucial role of ˙O₂⁻ and the two-step single-electron ORR pathway (Fig. 5c). Electron paramagnetic resonance (EPR) measurements were conducted using 5,5-dimethyl-1-pyrroline N-oxide (DMPO) as a spin-trap agent to detect radicals generated during the photocatalytic reaction. In Fig. S25,† a comparison between the dark conditions and after-lighting revealed the presence of typical six-line characteristic peaks corresponding to DMPO–˙O₂⁻ for all COFs.³⁶ This observation provides further evidence supporting the involvement of the indirect 2e⁻ ORR pathway in the photocatalytic process. Tz–THBZ exhibited a stronger EPR signal compared to the other COFs, indicating its superior ability to generate active ˙O₂⁻ radicals.

In situ attenuated total reflection surface-enhanced infrared absorption spectroscopy (ATR-SEIRAS) was employed to gain insights into the intermediates formed during the photocatalytic process of Tz–THBZ under an O₂ atmosphere. In Fig. 5d, it is observed that once Tz–THBZ reaches equilibrium with water in the dark, the peak at around 949 cm⁻¹ corresponding to O–O bonding shows a continuous increase upon light irradiation. Additionally, vibrational peaks at approximately 1100 cm⁻¹ and 1283 cm⁻¹ gradually increase, indicating the presence of ˙O₂⁻ and endoperoxide intermediates, supporting the two-step single-electron ORR pathway.²³ Furthermore, a negative peak around 3190 cm⁻¹ is observed during the ongoing photocatalytic process, suggesting the consumption of adsorbed water and indicating water involvement in the WOR (Fig. 5e).³³

The electrostatic surface potential (ESP) map and DFT calculations also provide crucial insights into the charge distribution and adsorption energy within Tz–THBZ. As shown in Fig. 5f, the keto-formed Tz–THBZ, resulting from the irreversible keto-enamine form, exhibits a distinct charge distribution compared to the enol-formed structure. The positive charge, denoted by the blue region, is predominantly localized on the triazine unit, establishing it as a reduction center within the structure. Conversely, the negative charge, depicted by the red region, is specifically situated on the carbonyl group of the keto-Tp unit, suggesting its role as an oxidation center. Furthermore, DFT calculations have elucidated the adsorption energy between the redox center and reactants in the enol/keto forms of Tz–THBZ. In the enol-formed structure, the O₂ tended to interact with the –OH group of the enol-Tp unit (ΔE_O2 = −3.47 kJ mol⁻¹) (Fig. S27a†), while H₂O preferred binding to the oxygen and nitrogen atom of the –OH and C=N bonds (ΔE_H2O = −13.7 kJ mol⁻¹) (Fig. S27b†). Conversely, in the keto-formed structure, the O₂ and H₂O favoured interaction with the oxygen atom of the keto-Tp unit (ΔE_O2 = −6.61 kJ mol⁻¹) (Fig. 5g) and the carbon and nitrogen atom of the triazine unit (ΔE_H2O = −24.91 kJ mol⁻¹) (Fig. 5h). As a result, the β-keto-enamine linked structure demonstrates a more favourable interaction with reactants (e.g. O₂ and H₂O), exhibiting higher adsorption energies compared to the enol-formed structure.

DFT simulations also demonstrated that the HOMO were mainly concentrated on the hydroxy group containing-unit, while the LUMO were concentrated on the Tz unit, which was consistent with their ESP results. Moreover, Tz–THBZ showed a more separated HOMO and LUMO distribution, indicating an efficient separation of light-induced electrons and holes.³¹ Based on the above results, a probable mechanism for photocatalytic H₂O₂ production is proposed. When exposed to visible light, the COF absorbs photons, leading to the generation of electron–hole pairs. These charge carriers then migrate to the surface of the photocatalyst. The photogenerated electrons engage in a reaction with the adsorbed oxygen on the COF surface, ultimately producing H₂O₂ through an indirect sequential two-step single-electron ORR pathway. Meanwhile, the holes are involved in the WOR process through a direct 2e⁻ reaction.

Conclusions

In summary, a series of COF photocatalysts were synthesized by incorporating varying amounts of hydroxy groups to enable visible light-induced hydrogen peroxide production. Both experimental and theoretical analyses suggest that Tz–THBZ, featuring a β-ketoenamine linkage, exhibits a higher HOMO energy and positively charged potentials, enhancing its WOR capability. Time-resolved PL and temperature-dependent PL studies demonstrate that Tz–THBZ displays prolonged charge carrier lifetimes and improved exciton dissociation, thereby boosting photocatalysis performance. Control experiments confirm that the β-ketoenamine-linked COF, Tz–THBZ, can generate H₂O₂ through both the ORR and WOR pathways, achieving a peak photocatalytic activity of 4688 μmol h⁻¹ g⁻¹ in pure water under visible light irradiation. DFT calculations further reveal that the triazine unit acts as the photoreduction center, while the keto-formed benzene ring serves as the oxidation site. This study underscores how the linkages within COFs impact the HOMO energy, excited-state properties, and electronic structure, significantly influencing oxidation capabilities and active centers in the photocatalytic process.

Data availability

The data supporting this article have been included as part of the ESI.†

Author contributions

Xiaoyan Wang and Bien Tan conceived the study and supervised this work. Chang Shu and Peixuan Xie performed the experiments and analysis and contributed equally to this work. Xiaoju Yang and Xuan Yang performed and conceived the ATR-SEIRAS analysis. Hui Gao carried out PXRD analyses. All authors interpreted the data and contributed to the preparation of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Grant No. 52203259; Grant No. 21975086; Grant. No. 22204054); Natural Science Foundation of Hubei Province (Grant No. 2022CFB720); Fundamental Research Funds for the Central Universities (Grant No. 2024JYCXJJ041).

Notes and references

1. R. L. Myers, The 100 Most Important Chemical Compounds: a Reference Guide, Bloomsbury Publishing USA, 2007.
2. Y. Wu, J. Wu, F. Yang, C. Tang and Q. Huang, Polymers, 2019, 11, 776.
3. G. Goor, in Catalytic Oxidations with Hydrogen Peroxide as Oxidant, ed. G. Strukul, Springer Netherlands, Dordrecht, 1992, pp. 13–43.
4. H. Hou, X. Zeng and X. Zhang, Angew. Chem., Int. Ed., 2020, 59, 17356–17376.
5. S. Qu, H. Wu and Y. H. Ng, Adv. Energy Mater., 2023, 13, 2301047.
6. H. Wang, C. Yang, F. Chen, G. Zheng and Q. Han, Angew. Chem., Int. Ed., 2022, 61, e202202328.
7. C. Krishnaraj, H. Sekhar Jena, L. Bourda, A. Laemont, P. Pachfule, J. Roeser, C. V. Chandran, S. Borgmans, S. M. J. Rogge, K. Leus, C. V. Stevens, J. A. Martens, V. Van Speybroeck, E. Breynaert, A. Thomas and P. Van Der Voort, J. Am. Chem. Soc., 2020, 142, 20107–20116.
8. L. Chen, C. Chen, Z. Yang, S. Li, C. Chu and B. Chen, Adv. Funct. Mater., 2021, 31, 2105731.
9. J. Sun, J. Chakraborty, M. Deng, A. Laemont, X. Feng, Y.-Y. Liu and P. Van Der Voort, J. Mater. Chem. A, 2023, 11, 21516–21540.
10. S. Wu and X. Quan, ACS ES&T Eng., 2022, 2, 1068–1079.
11. W. Yu, C. Hu, L. Bai, N. Tian, Y. Zhang and H. Huang, Nano Energy, 2022, 104, 107906.
12. Y. Shiraishi, S. Kanazawa, Y. Kofuji, H. Sakamoto, S. Ichikawa, S. Tanaka and T. Hirai, Angew. Chem., Int. Ed., 2014, 53, 13454–13459.
13. Y.-X. Ye, J. Pan, Y. Shen, M. Shen, H. Yan, J. He, X. Yang, F. Zhu, J. Xu, J. He and G. Ouyang, Proc. Natl. Acad. Sci. U.S.A., 2021, 118, e2115666118.
14. L. Chen, L. Wang, Y. Wan, Y. Zhang, Z. Qi, X. Wu and H. Xu, Adv. Mater., 2020, 32, 1904433.
15. L. Liu, M.-Y. Gao, H. Yang, X. Wang, X. Li and A. I. Cooper, J. Am. Chem. Soc., 2021, 143, 19287–19293.
16. X. Xu, Y. Sui, W. Chen, W. Huang, X. Li, Y. Li, D. Liu, S. Gao, W. Wu, C. Pan, H. Zhong, H.-R. Wen and M. Wen, Appl. Catal., B, 2024, 341, 123271.
17. Z. Yong and T. Ma, Angew. Chem., Int. Ed., 2023, 62, e202308980.
18. L. Li, X. Lv, Y. Xue, H. Shao, G. Zheng and Q. Han, Angew. Chem., Int. Ed., 2024, 63, e202320218.
19. P. Das, G. Chakraborty, J. Roeser, S. Vogl, J. Rabeah and A. Thomas, J. Am. Chem. Soc., 2023, 145, 2975–2984.
20. J.-Y. Yue, L.-P. Song, Y.-F. Fan, Z.-X. Pan, P. Yang, Y. Ma, Q. Xu and B. Tang, Angew. Chem., Int. Ed., 2023, 62, e202309624.
21. L.-J. Wang, P.-Y. Dong, G. Zhang and F.-M. Zhang, Energy Fuels, 2023, 37, 6323–6347.
22. H. Wang, C. Qian, J. Liu, Y. Zeng, D. Wang, W. Zhou, L. Gu, H. Wu, G. Liu and Y. Zhao, J. Am. Chem. Soc., 2020, 142, 4862–4871.
23. D. Chen, W. Chen, Y. Wu, L. Wang, X. Wu, H. Xu and L. Chen, Angew. Chem., Int. Ed., 2023, 62, e202217479.
24. W.-R. Cui, C.-R. Zhang, R.-H. Xu, X.-R. Chen, R.-H. Yan, W. Jiang, R.-P. Liang and J.-D. Qiu, Small, 2021, 17, 2006882.
25. S. Haldar, D. Chakraborty, B. Roy, G. Banappanavar, K. Rinku, D. Mullangi, P. Hazra, D. Kabra and R. Vaidhyanathan, J. Am. Chem. Soc., 2018, 140, 13367–13374.
26. L. Zhang, S.-L. Wang, G.-H. Zhang, N. Shen, H. Chen, G. Tao, G.-H. Tao, F. Yong, J. Fu, Q.-H. Zhu and L. He, Cell Rep. Phys. Sci., 2022, 3, 101114.
27. M. Traxler, S. Gisbertz, P. Pachfule, J. Schmidt, J. Roeser, S. Reischauer, J. Rabeah, B. Pieber and A. Thomas, Angew. Chem., Int. Ed., 2022, 61, e202117738.
28. S. Chandra, D. Roy Chowdhury, M. Addicoat, T. Heine, A. Paul and R. Banerjee, Chem. Mater., 2017, 29, 2074–2080.
29. X. Li, Y. Wu, S. Zhang, B. Cai, Y. Gu, J. Song and H. Zeng, Adv. Funct. Mater., 2016, 26, 2435–2445.
30. G. Fu, D. Yang, S. Xu, S. Li, Y. Zhao, H. Yang, D. Wu, P. S. Petkov, Z.-A. Lan, X. Wang and T. Zhang, J. Am. Chem. Soc., 2024, 146, 1318–1325.
31. C. Shu, C. Han, X. Yang, C. Zhang, Y. Chen, S. Ren, F. Wang, F. Huang and J.-X. Jiang, Adv. Mater., 2021, 33, 2008498.
32. X.-G. Zhu, S. P. Long and D. R. Ort, Curr. Opin. Biotechnol., 2008, 19, 153–159.
33. C. Shu, X. Yang, L. Liu, X. Hu, R. Sun, X. Yang, A. I. Cooper, B. Tan and X. Wang, Angew. Chem., Int. Ed., 2024, 63, e202403926.
34. Y. Nosaka and A. Y. Nosaka, Chem. Rev., 2017, 117, 11302–11336.
35. J. Sun, H. Sekhar Jena, C. Krishnaraj, K. Singh Rawat, S. Abednatanzi, J. Chakraborty, A. Laemont, W. Liu, H. Chen, Y.-Y. Liu, K. Leus, H. Vrielinck, V. Van Speybroeck and P. Van Der Voort, Angew. Chem., 2023, 135, e202216719.
36. G. R. Buettner, Free Radical Biol. Med., 1987, 3, 259–303.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ta03950e

‡ These authors contributed equally to this work.

---