Unlocking sunlight driven photocatalysis: synthesis, diversity, and applications of COF-based S-scheme heterojunctions

Abstract

Covalent organic frameworks (COFs) are highly porous and crystalline organic polymers with remarkable thermal and chemical stability. Their tunable structures and properties have enabled their applications in diverse fields. However, COFs suffer from significant drawbacks, including poor processability, strong self-stacking tendencies, limited electrical conductivity, restricted ion transport owing to pore blockage, and quick recombination of photogenerated electron–hole pairs. To address these limitations, construction of heterojunctions between COFs and other semiconductors has emerged as an effective approach. In particular, the S-scheme heterojunction design has recently attracted increasing interest owing to its ability to suppress charge carrier recombination while preserving strong redox capability, thereby enhancing photocatalytic efficiency. Despite these advantages, there is a scarcity of comprehensive reviews focusing on COF-based S-scheme heterojunctions. This review provides a detailed overview of the structural and functional properties of these materials in photocatalysis. It further discusses various synthesis strategies and charge-transfer mechanisms involved in constructing S-scheme heterojunctions by integrating COFs with different semiconductor materials. Additionally, recent advancements in COF-based S-scheme heterojunction photocatalysts are summarized, highlighting their various applications. Finally, the persistent challenges and potential future research directions in this domain are critically examined.

---

Introduction

Photocatalysis is a vital strategy for harnessing solar energy and converting it into chemical energy for various applications.¹ However, conventional inorganic semiconductor photocatalysts face several challenges, including wide band gaps, limited structural tunability, and a lack of abundant active sites.² These drawbacks contribute to the low efficiency of photocatalytic processes, thereby restricting their scalability for industrial applications.³ As a result, there is a growing need for advanced materials and innovative structural designs to improve photocatalytic performance. Recent years have witnessed significant progress in the development of porous materials with tailored structures for photocatalysis.³ Among them, metal–organic frameworks (MOFs), covalent organic frameworks (COFs), hydrogen-bonded organic frameworks (HOFs), and porous organic polymers (POPs) have attracted considerable interest.^4,5 MOFs, composed of metal clusters and organic ligands connected by coordination bonds, exhibit well-ordered structures, high porosity, and tunable properties.⁶ However, their practical application in photocatalysis is hindered by limited chemical stability.⁷ Similarly, HOFs, which rely on hydrogen bonding interactions to form defined structures with low densities, also suffer from poor chemical stability, limiting their functional versatility.⁸ POPs, in contrast, are highly stable due to their covalent bonding networks, but their irregular pores and undefined structures make it challenging to establish structure–property relationships.⁹

Among these porous materials, COFs stand out as a distinct class of crystalline, covalently linked polymers with well-defined structures.¹⁰ Their ability to integrate the advantageous properties of MOFs, HOFs, and POPs while addressing their limitations makes them attractive candidates for photocatalytic applications.¹¹ COFs exhibit semiconducting behavior and a tailorable electronic structure, which permits for specific control over light absorption and charge transfer processes.¹² The high surface area and porous architecture of COFs provide numerous active sites for catalysis while minimizing particle aggregation, thereby improving reaction efficiency.¹³ Furthermore, their chemical versatility enables fine-tuning of their properties through rational design and modification, allowing them to accomplish the specific prerequisites of different photocatalytic reactions.¹⁴ The structural flexibility of COFs offers additional advantages, as their framework composition can be engineered to optimize physical and electronic properties.¹⁵ By adjusting the building blocks, functional groups, and connectivity, COFs can be tailored for enhanced light absorption, charge separation, and catalytic activity.¹⁶ These properties make COFs promising platforms for a wide range of applications, such as photocatalysis, sensors, gas separation, and drug delivery.^17,18 In photocatalysis, their ability to facilitate electron transfer and provide efficient reaction sites plays a crucial role in improving reaction rates and selectivity.¹⁹ In spite of these benefits, the photocatalytic efficiency of COFs remains hindered by their relatively large band gaps and slow charge transfer kinetics.²⁰ These factors reduce their ability to utilize a broad range of solar radiation and hinder the overall reaction efficiency.²¹ Additionally, recombination of photogenerated electron–hole pairs further limits their large-scale application.²² To overcome these limitations, researchers have developed various strategies, including the construction of heterojunction composites, metal doping, and the incorporation of co-catalysts.^23–25 Among these, the design of heterojunction systems has proven to be particularly impressive in promoting charge separation and increasing photocatalytic efficiency.

Heterojunction photocatalysts function by forming an interface between different semiconductor materials to facilitate charge transfer and reduce recombination.²⁶ Traditional heterojunction designs, such as type-II and Z-scheme configurations, have demonstrated success in improving photocatalytic activity.²⁷ However, these systems have inherent drawbacks, including insufficient redox potential and inefficient charge separation.²⁸ Recently, an innovative approach, S-scheme heterojunctions, has emerged as a potential substitute for overcoming these limitations.²⁹ The S-scheme heterojunction is formed by coupling an oxidation photocatalyst (OP) with a reduction photocatalyst (RP). Upon integration, electrons from the RP migrate to the OP, leading to band bending at the interface. This charge migration results in the generation of an internal electric field directed from the RP to OP. Owing to coulombic interactions and the inherent electric field, non-contributory charge carriers recombine, while high-energy electrons and holes remain effectively separated. This unique charge transfer mechanism enhances photocatalytic performance by maintaining a strong redox potential while simultaneously reducing charge carrier recombination,^30–32 resulting in S-scheme heterojunctions exhibiting superior photocatalytic activity compared to traditional heterojunction designs.

Integrating COFs with S-scheme heterojunctions presents an outstanding approach for further increasing photocatalytic performance.³³ The combination of the COF's high surface area, tunable band structure, and stability with the efficient charge separation process of S-scheme heterojunctions leads to improved light absorption and enhanced photocatalytic efficiency.³⁴ By leveraging the synergistic effects of COFs and S-scheme heterojunctions, researchers have explored applications in diverse photocatalytic fields, including hydrogen production, CO₂ reduction, and H₂O₂ synthesis.^35,36 COF-based S-scheme heterojunctions capitalize on the advantages of both COFs and S-scheme charge transfer pathways, leading to increased reaction rates and improved photocatalytic efficiency.³⁷ The porous structure of COFs facilitates a beneficial environment for reactant adsorption and mass transport, while the heterojunction design ensures optimal charge carrier dynamics. This integration enables COFs to serve as efficient photocatalytic platforms for renewable energy conversion and environmental remediation.³⁸ Despite these advancements, research on COF-based S-scheme heterojunctions is still in its infancy, and plentiful barriers remain. For instance, the synthesis of COF-based S-scheme heterojunctions is limited by factors such as poor interfacial compatibility, difficulty in achieving precise structural alignment, and challenges in controlling crystallinity and porosity.^33–37 These issues can hinder effective charge transfer and overall material stability.

The rational design and precise synthesis of COFs with tailored electronic structures for optimal charge transfer require further exploration. In addition, the recyclability and scalability of these candidates require further investigation to enable practical applications. Addressing these challenges through advanced material design and improved synthetic methodologies is crucial for unlocking the full potential of COF-based S-scheme heterojunction photocatalysts. Currently, the existing literature lacks a comprehensive review that systematically examines recent progress in COF-based S-scheme heterojunction photocatalysts. Considering the accelerating interest in this domain, a detailed analysis of their synthesis strategies, charge transfer mechanisms, and photocatalytic applications is essential for guiding future research. This review provides a detailed overview of the latest status of research, addresses existing challenges, and proposes future directions for advancing this promising field. By consolidating recent findings and offering insights into their potential applications, this review seeks to contribute to the continued development of high-performance COF-based S-scheme heterojunction photocatalysts. This review provides an in-depth review of COF-based S-scheme heterojunction photocatalysts, covering their fundamental properties, fabrication methods, various types of COF-based S-scheme heterojunctions, and photocatalytic mechanisms. Additionally, this review highlights recent advancements in their applications in hydrogen production, CO₂ reduction, and H₂O₂ production. Finally, existing obstacles and future viewpoints are highlighted to provide further insights to promote this emerging research field.

Photocatalytic properties of COFs

COFs are largely structured organic polymers with remarkable porosities, offering significant potential for photocatalytic applications. The attractive properties of the COFs are discussed below.

Morphology

COFs can be synthesized in 2D and 3D forms, with each structure providing distinct advantages for different applications such as separation, sensing, and catalysis.^39,40 The most promising and emerging applications of COFs are photocatalytic solar fuel generation, where they effectively convert CO₂ into useful products, including CH₄, CO, CH₃OH, and HCOOH, reduce N₂ to NH₃, and split water into H₂ using solar energy.^41–43 COFs are prepared using techniques such as imine condensation, boronate ester formation, and Schiff base condensation, which links organic monomers to create extended networks with periodic pores. These monomers contain active centers, such as boronic acids, amines, and aldehydes, facilitating the fabrication of intricate 2D or 3D structures.³² In 2D COFs, the organic molecules are arranged in layers connected by π–π stacking interactions.⁴⁴ These interactions stabilize the structure and create channels for photocatalysis. The large surface area provided by these layers is crucial for photocatalytic applications, while π–π stacking also aids in charge transport, making 2D COFs suitable for photocatalysis.^45,46 Additionally, 2D COFs are adaptable, allowing for modifications in surface area, pore size, and electrical properties by choosing specific building blocks and adjusting the synthesis process.⁴⁷ On the other hand, 3D COFs have a more robust structure because of their extensive network of covalent bonds in three dimensions, offering high mechanical and thermal stability.⁴⁸ This makes them ideal for applications that require durability and long-term performance under harsh conditions. The interconnected pores of 3D COFs provide a large surface area for reactant adsorption, making them valuable in processes like carbon capture and water splitting.²⁹ While 2D COFs are more common, creating 3D COFs presents challenges due to the limited availability of stereoscopic building blocks and crystallization issues.⁴⁹ Nevertheless, top-down or bottom-up methods have been used to produce COF nanosheets, providing opportunities to study the relationship between structure and performance for catalytic improvements.⁵⁰

Charge separation and migration

Among the main photocatalytic parameters, COFs have the ability to produce photoinduced charge carriers upon light exposure, which is enhanced by the careful choice of building materials that promote excellent charge migration, enabling the transformation of sunlight into chemical energy.⁵¹ By optimizing the electronic properties of COFs, such as through the design of donor–acceptor pairs or improving π-conjugation, their photocatalytic efficiency can be further increased.⁵² The combination of efficient electron transfer and structured electron channels makes COFs promising candidates for CO₂ reduction and other photocatalytic applications.⁵³ By incorporating photosensitizers or designing COFs with alternating donor–acceptor units, researchers can optimize charge transfer properties and enhance photocatalytic performance, while adjusting the bandgap and electronic properties further improves their efficiency.⁵⁴ COFs' ability to form thin films is also advantageous for applications requiring fast mass and charge transport, though synthesizing highly crystalline and stable COF thin films can be challenging; however, techniques like layer-by-layer assembly, interfacial assembly, and support-assisted growth have made significant progress, demonstrating promise in heterogeneous photocatalysis, particularly for long-term stability and efficiency.⁵⁵ In recent years, strategies such as designing donor–acceptor COFs, incorporating photosensitizers, and creating COF heterojunctions have been explored to improve photocatalytic activity.⁵⁶ COFs can form heterojunctions with materials like inorganic semiconductors, MOFs, and carbon-based materials, which enhances their photocatalytic performance by providing the migration of photogenerated charge carriers, improving charge separation, and increasing efficiency for energy production and pollutant removal applications.⁵⁷ The efficiency of COFs in photocatalysis is largely influenced by their capability to enable charge separation and transport, as the strength of exciton binding energy dictates how readily excitons dissociate into individual charge carriers.⁵⁸ For optimal performance, excitons must migrate effectively through the COF network to reach active sites before recombining, making minimizing recombination and ensuring effective electron–hole separation key to enhancing the availability of free carriers for catalytic activity.⁵⁹ Charge transport mechanisms enable these carriers to reach reaction sites where they contribute to photocatalytic transformations, and various strategies can improve this process, including the incorporation of heterojunctions, protonation, or the induction of an internal electric field.⁶⁰ Heterojunctions establish an interface between materials with distinct band alignments, which aids in electron–hole separation, while protonation alters the electronic properties of COFs, leading to increased charge carrier mobility.⁶¹ Additionally, an internal electric field helps steer electrons and holes in opposite directions, thereby reducing their recombination rate and enhancing overall catalytic efficiency.⁶²

Surface area and porosity

Larger surface areas and well-defined pores provide optimal environments for substrates to interact with catalytic sites, facilitating efficient reactions.⁶³ The highly porous nature of COFs, with customizable surface areas and pore sizes, enhances their ability to adsorb reactants and convert them into valuable products.⁶⁴ Their stability, both chemically and physically, ensures durability and recyclability as photocatalysts, making them an ideal choice for photocatalytic applications.⁶⁵ Additionally, COFs are crystalline materials formed through covalent bonding between multifunctional monomers, resulting in well-defined 2D and 3D structures.⁶⁶ The geometry, pore distribution, and surface area of COFs might be carefully controlled through the picking of appropriate organic building blocks, enabling fine-tuning for specific applications.⁶⁷ For photocatalytic applications, the surface area and pore size of COFs are critical factors. Adjustments to the stacking arrangement of COF nanosheets can influence the pore sizes and catalytic performance, while incorporating metal nanoparticles further enhances their photocatalytic properties. The combination of precise pore control, large surface areas, and the ability to host catalytic sites makes COFs highly effective for photocatalytic applications.⁶⁸ COFs possess a modular design, allowing for precise control of their structural attributes, including topology, symmetry, pore size, and connectivity. This structural versatility is achieved by selecting organic precursors with distinct molecular lengths and symmetries.⁶⁹ Due to their highly crystalline nature, COFs exhibit a substantial surface area and a uniform distribution of pores, which play a crucial role in offering abundant photocatalytic sites, facilitating substrate accommodation, and ensuring efficient mass transport during catalytic reactions.⁷⁰ Additionally, their well-defined framework serves as an excellent matrix for anchoring molecular catalysts or photosensitizers, enhancing their applicability in heterogeneous photocatalysis.⁷¹ By carefully designing the pore architecture through the selection of specific monomeric units, it is possible to tailor both pore dimensions and morphology. Moreover, post-synthetic modifications offer additional avenues for fine-tuning the porous framework, as pore organization and surface area are closely tied to the crystallinity and packing arrangement of COFs. For example, highly crystalline COF-1 exhibits a notably greater specific surface area (2143 m² g⁻¹) in contrast to its amorphous counterpart, COP-1 (13.8 m² g⁻¹).⁷² The extensive surface area of COFs ensures accessible catalytic sites, while their porous network supports rapid charge migration toward the surface, optimizing their photocatalytic efficacy.

Versatility

The versatility of COF structures offers considerable potential for tailoring materials for specific photocatalytic applications.⁵¹ This can be achieved through two primary approaches: the bottom-up method, where monomers are designed to form the desired COF directly, and post-synthetic modifications, where functional groups are added to existing COF structures.⁵² These modifications include metalation, where metal ions are introduced into the COFs, and other techniques like click reactions that introduce new functional groups to enhance properties.⁵³ Chemical conversions, such as oxidation–reduction or cyclization, can also modulate the structure, providing additional means to fine-tune COF properties.⁵⁴ COFs' high surface area, lightweight elements, and ability to adjust pore size (ranging from micropores to mesopores) render them perfect for a variety of applications.⁶⁰ The pore size and shape can be customized through monomer design, although predicting these characteristics in 3D COFs is more challenging due to potential interpenetration of structures.⁶¹ Researchers control pore size using three strategies: adjusting the length of building blocks, modifying substituent groups, and altering the stacking arrangement of COF nanosheets.⁶² The pore size of COFs can also be customized. For example, halogens can be used to modify the pore environment of 2D COFs, enhancing their photocatalytic performance, while functionalized nanosheets can improve photocatalytic activity for fuel generation and pollutant removal.⁶³ COFs' low density and high surface area, along with their ability to host functional groups, make them attractive for a variety of applications.⁶⁴

Stability

Stability is crucial for the practical use of COFs, especially for post-synthetic modifications.⁷³ COFs generally exhibit effective thermal durability due to their strong covalent bonds, which enable them to maintain structural integrity at temperatures beyond 500 °C under inert conditions.⁷⁴ While early boron-derived COFs showed inadequate chemical durability in water or acidic/basic environments, advancements have significantly improved their stability.⁵⁴ Stronger linkages, such as hydrazine, ether, azine, and imine bonds, have enhanced the chemical stability of COFs, even under harsh conditions.⁵⁵ Post-synthetic strategies, such as converting a 2D imine-linked COF into a more stable thiazole-linked version, further improve their stability and suitability for practical applications.⁷⁵ COFs' thermal and chemical stability, due to the robust covalent bonds in their structure, allow them to maintain their properties in challenging environments.⁴³ COFs' stability plays a key role in their practical use for photocatalytic applications. Advances in linkage chemistry have greatly improved the stability of COFs, making them more robust in various environments. Some COFs are stabilized through processes like aza-Diels–Alder cycloaddition or sulfurization, increasing their resilience under harsh conditions.⁷⁶ A crucial aspect in the rational design of photocatalysts is ensuring both photostability and chemical durability. From the standpoint of reusability, it is vital that photocatalysts maintain long-term stability while withstanding different environmental factors such as acidic or basic conditions, oxidative species, co-catalysts, and sacrificial reagents.⁷⁷ COFs can achieve remarkable stability by carefully selecting molecular components and robust linkages, thereby sustaining their photocatalytic efficiency in diverse reactions.⁷⁸ Furthermore, COFs inherently resist photodegradation and photobleaching, enhancing their longevity under light exposure.⁴⁹ In certain cases, their photochemical robustness can be further improved through protonation or post-synthetic modifications.⁶⁵

Band position

The performance of photoreactions depends on the alignment of the CB and VB of the material. By designing COFs with suitable energy levels, researchers can optimize their ability to absorb visible light (400–800 nm), enhance charge separation, and improve photocatalytic performance.⁷⁹ Modifications, such as halogenation or the addition of functional groups like sulfone or porphyrin, can further tune COFs' energy levels and boost photocatalytic activity.⁸⁰ Improved electron transfer and charge carrier mobility contribute to more efficient photocatalytic processes.⁸¹ Additionally, the inhibition of electron–hole recombination through structural modifications, such as chlorination or fluorination, enhances the overall photocatalytic performance.⁸² For a photochemical reaction to occur, the absorbed photon must have energy ≥ the bandgap of the catalyst. COFs exhibit photocatalytic activity that is highly dependent on their optical band structure, which ideally should be around 1.80 eV to maximize solar energy utilization.⁸³ Typically, COFs possess optical band gaps in the range 1.50–3.0 eV, with a predominant value near 2.20 eV due to their extensive π-conjugation.³³ The energy levels of COFs can be tailored by incorporating extended aromatic moieties, donor–acceptor pairs, or heteroatoms into their molecular structure, as well as by modifying their linkages or applying post-synthetic alterations.⁴³ Beyond the band gap, the absolute energy levels of COFs are critical for photocatalytic processes. Functionalizing COFs or altering their linkage structures can significantly influence the band gap and band alignment, improving charge carrier dynamics and light absorption.⁷⁰ Upon irradiation, electrons migrate from the VB to the CB, producing electron–hole pairs that transfer to the surface and drive redox processes.⁶² The presence of π-conjugation in COFs plays a key role in reducing the band gap, thereby enhancing their ability to capture light.⁵¹ Additionally, incorporating specific structural components or heteroatoms can further optimize electronic properties. The introduction of functional groups such as imidazole, porphyrin, or sulfone into the COF backbone can enhance light absorption.⁷⁵ The inherent bandgap of COFs generally falls between 1.30 eV and 2.90 eV, allowing them to absorb light up to 800 nm.³⁴ As a result, COFs with well-matched energy levels and strong optical absorption can serve as efficient photocatalysts.

Crystallinity

COFs stand out from amorphous porous organic polymers (POPs) because of their ordered crystalline structures.⁸⁴ This well-defined arrangement ensures that π-conjugated units are periodically positioned, allowing precise control over both topology and stacking through molecular engineering.⁸⁵ The optical and electronic behavior of COFs is significantly influenced by the stacking arrangement and interlayer distances, as these factors regulate orbital interactions in both the ground and excited states.⁶¹ Unlike conventional organic polymers, COFs are synthesized using a bottom-up strategy that strategically organizes electron donors (D) and acceptors (A) in a repeating pattern.²⁸ This structural design enhances charge separation upon photoexcitation while mitigating charge recombination. Additionally, the extended periodicity in layered COFs facilitates the efficient transport of charge carriers or excitons across the material, reducing recombination losses.³⁹ Moreover, the crystalline structure of COFs contributes to increased charge carrier lifetime and kinetic energy, both of which are essential for efficient photocatalysis. Achieving high crystallinity in COFs requires careful selection of molecular precursors along with optimized reaction conditions, including solvent choice, catalyst type, reaction temperature, and synthesis duration.²⁵ The long-range order in COFs reflects the structured positioning of building units at both local and extended scales. Several factors play a critical role in the formation of highly crystalline COFs. Due to variations in reaction dynamics and thermodynamic stability, the selection of linkage chemistry and molecular components greatly impacts the final crystallinity.⁸⁶ Furthermore, reaction parameters such as temperature, pressure, monomer concentration, and synthesis duration are crucial in controlling nucleation and ensuring the formation of large, well-ordered COF crystals,⁶⁹ among which optimization of the monomer concentration and reaction temperature is particularly vital for promoting crystallization. Typically, solvothermal synthesis of highly crystalline COFs requires a reaction period of 3 to 7 days.⁸⁷ For instance, PI-COF-1, a polyimide-linked COF, was successfully synthesized over 5 to 7 days under solvothermal conditions.³⁸ In general, well-ordered COFs exhibit better photocatalytic efficiency in contrast to their disordered counterparts, as their high crystallinity reduces defect sites, suppresses charge carrier recombination, and enhances charge transport efficiency.⁸⁵ Despite their promising properties for photocatalysis, challenges remain in enhancing the conductivity of COFs. Solutions include developing hybrids with mobile semiconductors or increasing the crystallinity and conjugation of COFs.⁶⁸ The well-defined pore structures, high surface area, and wide light harvesting capabilities of COFs make them excellent candidates for heterogeneous photocatalysis, particularly in solar fuel generation under harsh conditions.⁶⁹ Their high crystallinity and periodic arrangements promote charge separation and migration, which is pivotal for photocatalytic efficiency.³⁵

Heterojunction formation of COFs

Most COFs possess a relatively large bandgap, which restricts their ability to efficiently absorb and exploit solar radiation, leading to low photocatalytic efficiency. This is partly due to the slow electron transfer rate and reduced quantum efficiency.^25,32 One potential solution is to combine COFs with semiconductors, as the high porosity and large surface area of COFs can enhance the dispersion of coupled materials. This improved dispersion support hinders cluster formation and ensures a more regular supply of reactive sites, thereby enhancing the photocatalytic process.^26,31 The formation of heterojunctions between COFs and semiconductor materials further improves the overall performance by enhancing light absorption, particularly in the visible range.³⁰ These heterojunctions suppress the recombination of charge carriers, and increasing the transfer efficiency of photogenerated electron–hole pairs. As a result, the separation and transfer of charges are enhanced, creating supportive conditions for photochemical reactions.²⁷ There are three main types of heterojunctions commonly employed in COF-semiconductor systems: type-II, Z-scheme, and S-scheme heterojunctions. The configuration of each heterojunction dictates the behavior of the photoinduced charge pairs at the interface between the two materials. In these configurations, the bandgaps of the semiconductors differ, and their CB and VB positions align to facilitate charge transfer between the materials. Herein, we discuss three different types of COF-based heterojunction photocatalysts, Fig. 1.

Fig. 1 Schematic illustration of COF-based (a) Type II heterojunction, (b) indirect Z-scheme heterojunction, (c) direct Z-scheme heterojunction, and (d) S-scheme heterojunction (notably, the band edge position of the COF and integrated semiconductor within these heterojunctions is exchangeable).

COF-based type II heterojunction

In a type-II heterojunction, when the semiconductor has a higher CB energy level and a lower VB energy level than the integrated COF, as shown in Fig. 1a, the photoinduced electrons from the CB of the semiconductor move to the CB of the COF. Meanwhile, photogenerated holes from the VB of the COF migrate to the VB of the integrated semiconductor.^28,44 This transfer results in the spatial separation of photoinduced charge carriers, diminishing the likelihood of recombination and improving the efficiency of the photocatalytic reaction.³³ However, type-II heterojunction systems have some limitations. A major issue is the repulsion between photoinduced electrons in the CBs of the COF and integrated semiconductor.²⁷ This repulsion can slow down or prevent the efficient migration of photoinduced electrons from the CB of the semiconductor to that of the COF.⁸⁵ Additionally, this repulsion hinders the migration of photoinduced holes from the VB of the COF to the VB of the integrated semiconductor.⁶⁹ From a thermodynamic standpoint, this reduces the overall redox capacity of the system, limiting its ability to drive photocatalytic reactions. Consequently, although type-II heterojunctions can be effective in some cases, this mechanistic limitation hinders their potential in traditional photocatalytic applications. For example, Yu et al. employed an in situ approach to couple TiO₂ with three different COFs (TTh-COF, Th-COF, and Ph-COF) to form type-II heterojunctions to study the photocatalytic activity for uranium reduction.⁸⁸

Compared to the pure COF and TiO₂, the XPS spectra of the TTh-COF-TiO₂ heterojunction showed that the binding energies of the COF orbitals shifted positively, while those of TiO₂ orbitals shifted negatively, as shown in Fig. 2a and b. This revealed not only the transport of electrons from TTh-COF to TiO₂ due to the generation of N–Ti covalent bonds but also confirmed the development of the heterojunction between TiO₂ and TTh-COF. Furthermore, the extended light absorption ability of TTh-COF-TiO₂ compared with that of pure TiO₂ (Fig. 2c) verified that heterojunction formation effectively promoted light absorption. Similarly, the smallest diameter in the EIS Nyquist plots and the highest photocurrent density of TTh-COF-TiO₂ validated the successful formation of the heterojunction. The optimized TTh-COF-TiO₂ photocatalyst displayed a maximum uranium reduction efficiency of 98% within 40 min of visible-light illumination. Additionally, six recycling experiments revealed only a small performance loss, from 98% to 95.3%, indicating a strong stability. Upon light excitation, the photogenerated electrons in the CB of TTh-COF shifted to the CB of TiO₂ through the N–Ti covalent bond, whereas holes migrated from the VB of TiO₂ to that of TTh-COF. This resulted in the enhanced separation of photogenerated charge pairs. Ultimately, the accumulated electrons in the CB of TiO₂ removed uranium, whereas holes in the VB of TTh-COF were depleted by methanol. Similarly, Wang et al. reported that the development of a type-II heterojunction between CdS and TpPa-1 COF provided enhanced separation and transfer of photogenerated charge pairs, leading to an NH₃ generation rate of 241 μmol g⁻¹ h⁻¹ from N₂ fixation, which was 3- and 17-fold superior to that of CdS and TpPa-1, respectively.⁹⁰

Fig. 2 High resolution XPS spectra of pure materials and TTh-COF-TiO₂ heterojunction (a) N 1s, (b) Ti 2p, and (c) absorption spectra of fabricated materials,⁸⁸ copyright 2024, Elsevier (d) SEM image, (e) PL spectra, (f) EIS Nyquist plots, and (g) ASS Z-scheme electron–hole transfer process of TzDa/Ag/AgBr,⁷¹ copyright 2022, Elsevier. (h) Absorption spectra and high-resolution XPS spectra of different materials: (i) Bi 4f, (j) Br 3d, (k) N 1s, and (l) S-scheme charge-migration mechanism of Bi₄O₅Br₂/COF,⁸⁹ copyright 2024, American Chemical Society.

COF-based Z-scheme heterojunction

In type-II heterojunctions, two semiconductors with different band structures are coupled, which helps suppress the recombination of photoinduced charge carriers by spatially separating them. However, this separation can reduce the photocatalyst's redox capability because electrons and holes are not available at the same site for oxidation and reduction reactions, resulting in a lower overall performance.⁸¹ Z-scheme heterojunctions were proposed to address the above limitation, mimicking the catalytic mechanism of photosynthesis. This structure retains a high redox potential while separating the electrons and holes, enhancing photocatalytic performance.⁸² Z-scheme heterojunctions can be categorized as indirect Z-scheme heterojunctions and direct Z-scheme heterojunctions depending on the existence of charge migration mediators, as shown in Fig. 1b and c.⁸³ Indirect Z-scheme systems use mediators, which can be liquid-phase electron donor/acceptor pairs or solid-state mediators like noble metals.⁵⁷ In conventional Z-scheme heterojunctions, two semiconductors are paired with electron donor/acceptor species in the liquid phase.²³ The component with a more negative reducibility preserves the photoinduced electrons, while the component with a superior oxidizability preserves the photoinduced holes, preventing recombination, and retaining a strong redox potential. All-solid-state (ASS) Z-scheme heterojunctions use precious metals as charge mediators, preventing reverse reactions typical in liquid-phase systems.²⁴ The solid-state mediator provides better stability and suppresses unwanted reactions that could degrade the photocatalyst.⁸⁸

Direct Z-scheme heterojunctions combine the advantages of both the indirect types. They preserve both excellent redox properties and efficient electron–hole separation, maintaining strong oxidative and reductive capabilities without relying on mediators. This results in higher photocatalytic efficiency.⁹⁰ For instance, Shi et al. utilized Ag as a charge mediator between AgBr and COF TzDa to construct the ASS Z-scheme heterojunction using the in situ technique for photocatalytic pollutant removal under simulated sunlight irradiation.⁹⁰ The microstructure analysis (Fig. 2d) revealed the simultaneous dispersion of AgBr and Ag nanoparticles onto COF TzDa nanosheets, forming strong interfacial contacts that facilitated rapid mass transport and extended light absorption, thereby enhancing the photoreaction. Additionally, the porous structure of COF TzDa/Ag/AgBr increases the surface area, promoting catalyst–light interactions, charge migration, and reactant transport through the pores. This was confirmed by the lower PL intensity and smaller EIS Nyquist arc radius for TzDa/Ag/AgBr compared to the pure materials, as shown in Fig. 2e and f. The optimized TzDa/Ag/AgBr photocatalyst demonstrated a maximum TC removal rate of 80% within 30 min, which was ascribed to the widened optical absorption, promoted electron–hole separation, and large surface area. Furthermore, TzDa/Ag/AgBr exhibited negligible performance loss after four consecutive recycling experiments, with no changes in the XRD results, indicating promising efficiency and structural stability of the photocatalyst. Under light irradiation, the photoinduced electrons in the CB of AgBr recombined with the photoinduced holes via the Ag mediator, leading to the increased separation of photoinduced electrons in the CB of COF TzDa and holes in the VB of AgBr, as illustrated in Fig. 2g. These preserved electrons reduce O₂ to ˙O²⁻ radicals, where holes react with H₂O to produce ˙OH radicals, which subsequently degrade pollutants.

Similarly, Yang et al. formed a direct Z-scheme WO₃/Tp-TAPB heterojunction for photocatalytic H₂O₂ production under visible light excitation.⁶⁶ The microstructure examination confirmed the effective integration of WO₃ nanocubes with COF nanosheets, resulting in extended light absorption and improved transfer and separation of photogenerated charge pairs. Additionally, compared to pure catalysts, the improved surface area of the heterojunction not only provided numerous reactive sites for redox reactions but also promoted light interaction and reactant diffusion, enhancing the photocatalytic performance. Consequently, the optimized WO₃/Tp-TAPB exhibited the highest H₂O₂ production rate of 1.49 mmol g⁻¹ h⁻¹, which was approximately 72 and 3 fold higher than those of WO₃ and Tp-TAPB, respectively. Furthermore, the optimized WO₃/Tp-TAPB catalyst retained both its performance and structural stability after four recycling experiments, confirming the robust stability of the photocatalyst. Upon light excitation, the photogenerated electrons in the CB of WO₃ directly perished with holes in the VB of Tp-TAPB, thus maintaining the photoinduced electrons and holes in the CB of Tp-TAPB and VB of WO₃, respectively. These retained electrons and holes then produce ˙O²⁻ and ˙OH radicals, leading to the generation of H₂O₂. COFs have attracted significant attention for their use in Z-scheme heterojunctions for photocatalysis, especially in liquid-phase systems. However, these systems face challenges such as unwanted charge migration from the CB of the coupled component to the CB of the COF, which leads to charge recombination and reverse reactions that reduce efficiency.⁹¹ Additionally, the liquid-phase nature of these systems limits their applications. COF-based ASS Z-scheme heterosystems rely on expensive precious metals as charge mediators to facilitate electron transfer.⁴⁷ While these metals can improve charge separation, they also increase the overall cost and interfere with the semiconductor's ability to absorb light, diminishing the system's overall efficiency.²⁷ Furthermore, the precise mechanisms of charge transfer in these systems are still not fully understood, leading to ongoing debates. A more recent development, the direct Z-scheme heterojunction, addresses these issues by eliminating the need for electron mediators. This system involves two semiconductors in direct contact, which lowers costs and retains strong redox capabilities.⁵³ However, despite its advantages, it shares similarities with older Z-scheme systems, and some still mistakenly associate it with the drawbacks of reverse electron transfer seen in those previous models.⁹² To overcome these limitations, Yu et al. documented the idea of the S-scheme heterojunction.⁹³

COF-based S-scheme heterojunction

The S-scheme heterojunction is designed to optimize charge separation and minimize reverse reactions by combining two semiconductors with different energy levels: the more closely integrated, the difference in their E_f levels is eliminated, resulting in the migration of free electrons from the reduced photocatalyst (RP) to the oxidized photocatalyst (OP).²⁸ This electron transfer facilitates a positively charged layer near the RP, bending its band structure upwards, while a negatively charged layer develops near the side of OP, bending its band structure downward, as shown in Fig. 1d.⁶⁸ The resulting IEF, which points from the RP to the OP, plays a crucial role in separating charge carriers and preventing recombination. It also aids in the recombination and perishing of photogenerated holes in the VB of the RP with electrons in the CB of the OP, consuming unnecessary charge carriers and improving the system's efficiency.⁵⁶ Additionally, this electric field and the Coulomb repulsion between electrons prevent the unwanted transport of photoinduced electrons from the CB of the RP to the CB of the OP, effectively inhibiting reverse reactions and maintaining a strong redox capacity in the system.⁹⁴ As a result, the S-scheme heterojunction is perceived as a more excellent and cost-effective alternative to conventional Z-scheme and type-II heterojunctions, addressing the limitations of older systems and enhancing photocatalytic performance.

For example, Yue et al. integrated TTD-COF strips with Bi₄O₅Br₂ spheres through a solvothermal process to create an S-scheme heterojunction, with the aim of enhancing the photocatalytic performance for H₂O₂ production.⁸⁹ Microstructure analysis demonstrated that TTD-COF was effectively wrapped around Bi₄O₅Br₂via strong interfacial interactions, resulting in increased surface area and enhanced light absorption (Fig. 2h). Additionally, the higher current density of the heterojunction, in contrast to bare semiconductors, further confirmed the enhanced separation and transport of photoinduced charge pairs, attributed to the generation of the S-scheme heterojunction between Bi₄O₅Br₂ and TTD-COF.

Consequently, the Bi₄O₅Br₂/COF photocatalyst exhibited the highest H₂O₂ production rate of 5221 μmol g⁻¹ h⁻¹, nearly double that of TTD-COF alone. XPS spectra showed that the binding energies of Bi₄O₅Br₂ orbitals exhibited a red shift, while those of TTD-COF displayed a blue shift, indicating the migration of free electrons from Bi₄O₅Br₂ to TTD-COF due to the formation of the heterojunction. Under light irradiation, however, the binding energies of the Bi₄O₅Br₂ and TTD-COF orbitals in the heterojunction exhibited a reverse shift compared to the dark state, as shown in Fig. 2i–k, confirming that photoinduced electrons migrated from TTD-COF to Bi₄O₅Br₂ in accordance with the S-scheme charge migration mechanism. The preserved electrons in the CB of Bi₄O₅Br₂ and holes in the VB of TTD-COF generated ˙O²⁻and ˙OH radicals, leading to H₂O₂ production. Similarly, An et al. employed a self-assembly technique to integrate TiO₂ with Tp-Tta COF via an S-scheme heterojunction and investigated the photocatalytic performance for CO₂ reduction.⁴⁹ Microstructure analysis confirmed a strong interface interaction between Tp-Tta COF and TiO₂ nanosheets, which resulted in an increased surface area that promoted mass transport and charge migration. Additionally, the XPS study confirmed the successful fabrication of the TiO₂/COF heterojunction, which extended the light absorption, providing additional photoinduced charge carriers to support the photoreaction. The optimized TiO₂/COF heterojunction exhibited the highest CO generation efficiency of 11.6 μmol h⁻¹, 14.5 fold greater compared to that of pure TiO₂. EPR spectra confirmed the development of the S-scheme charge transfer between TiO₂ and COF. Upon light excitation, the photogenerated electrons in the CB of TiO₂ perished with holes in the VB of COF, leading to the separation of electrons in the CB of COF and holes in the VB of TiO₂, thus facilitating CO₂ reduction. Although the formation of COF-based S-scheme heterojunctions has immense potential for photocatalytic applications, research in this field is still in its early stages. Consequently, significant research is expected in the near future to explore this emerging area.

Synthesis methods

The formation of COF-based S-scheme heterojunctions is a potential approach for developing highly efficient photocatalysts. Owing to their unique properties, integrating COFs with other semiconductors presents a challenging synthetic approach. Additionally, precise control over the crystallinity and porosity of COFs during heterojunction formation requires careful attention to factors such as the pressure, processing time, reaction temperature, and solvent combinations. These parameters play a critical role in maintaining the balance between the synthesis and crystallization of COF-based S-scheme heterojunctions. The common preparation methods for these photocatalysts are summarized in Fig. 3.

Fig. 3 Summary of various synthesis methods discussed to synthesize COF-based S-scheme heterojunctions with corresponding advantages and limitations.

Hydrothermal synthesis

The hydrothermal synthesis route is extensively adopted, especially in the preparation of COF-based photocatalysts because it is economical, as it relies primarily on water as the solvent, which is abundant and inexpensive.⁹⁵ Hydrothermal synthesis offers several advantages, including the ability to produce photocatalysts with good grain development, small particle sizes, uniform distribution, and well-controlled morphology.⁹⁶ These features are essential for photocatalytic reactions, as they improve the COF-based material's efficiency in applications, including hydrogen production, pollution degradation, and other energy-related reactions. One of the pivotal characteristics of the hydrothermal method is its ability to produce high-crystallinity COF-based materials with precise control over their morphology.⁹⁷ Notably, the precursors of the COF and selected semiconductor are mixed in a solvent, typically water, under stirring or ultrasonic treatment, and then placed in an autoclave. The process typically involves dissolving the sample in water and heating it to a specific temperature (80–200 °C), where it undergoes re-crystallization.⁹⁸ During this period, surfactants or templates may be added to guide the formation of the desired structure. These surfactants facilitate the self-assembly of particles, and as the crystals grow, they form the desired shape, whether they are fine particles, thin films, or more complex structures, such as hollow particles.⁹⁹ Overall, the hydrothermal method offers precise control over the crystallization process, making it ideal for synthesizing a wide range of photocatalytic materials with specific properties.¹⁰⁰ These materials are typically used for applications that require a high surface area, efficient electron transport, and excellent photocatalytic performance. The capability to regulate the size, shape, and distribution of the nanoparticles allows for the design of advanced materials with tailored properties, which is a significant advantage in fields like environmental remediation and renewable energy.¹⁰¹

For example, Zhang et al. employed a hydrothermal method to integrate BiOBr and TzDa COF to evaluate its photocatalytic performance in Cr(VI) reduction.¹⁰² The precursors of both materials were dissolved in ethylene glycol (EG) under stirring, and the resulting uniform solution was transferred to an autoclave for hydrothermal treatment at 120 °C for 6 h. The final product was collected, washed, and dried at 60 °C for 6 h. The synthetic process is illustrated in Fig. 4a. As shown in Fig. 4b, microstructure analysis revealed that the BiOBr nanoplates (∼100 nm) were effectively anchored with the porous TzDa COF, enhancing the surface area, light absorption, and charge separation, thereby improving the photocatalytic performance. Consequently, the optimized BiOBr/TzDa COF photocatalyst achieved a Cr(VI) reduction efficiency of 97% within 40 min under visible-light irradiation. The microwave-assisted hydrothermal synthesis, which uses microwaves to generate heat, has also been effectively employed to prepare COF-based S-scheme heterojunctions.¹⁰³ Replacing conventional heating with microwave heating accelerates the reaction rate, provides better control over the process, yields higher product purity, and enhances efficiency.⁸⁷ However, the use of polar solvents is essential for effective heating during the reaction.²⁸ For example, Bao et al. employed a microwave-assisted hydrothermal process to fabricate a TpBD COF/ZnIn₂S₄ S-scheme heterojunction to investigate its photocatalytic H₂ generation performance.⁴⁸ The precursors, in the required molar ratio, were mixed in water under ultrasonication before initiating the microwave-assisted hydrothermal reaction, as depicted in Fig. 4c. SEM analysis (Fig. 4d) revealed that the TpBD COF particles (2–7 μm) were effectively integrated with the flower-like ZnIn₂S₄ structures, providing abundant active sites and enhanced light absorption. Furthermore, the strong interfacial contact between TpBD COF and ZnIn₂S₄ promoted the separation and migration of photogenerated carriers, resulting in an effective H₂ generation efficiency of 2304 μmol g⁻¹ h⁻¹ under visible light. In addition, in situ hydrothermal synthesis was employed to fabricate COF-based S-scheme heterojunction photocatalysts. This approach ensures strong interfacial interactions, improves the separation and transport of photoinduced charge pairs, and allows precise control over the heterojunction's morphology.³³ As an example, Wang et al. utilized an in situ hydrothermal process to fabricate defective ZnIn₂S₄/TpPa-1 S-scheme photocatalysts for H₂ generation.¹⁰⁴ Microstructure analysis revealed that flower-like ZnIn₂S₄ nanosheets were intimately connected with block-like TpPa-1, resulting in an improved surface area, widened optical absorption, and enhanced charge dynamics. The optimized ZnIn₂S₄/TpPa-1 photocatalyst achieved an outstanding H₂ production efficiency of 2745 μmol g⁻¹ h⁻¹ under visible light. Despite the promising potential of in situ hydrothermal synthesis, its application in the fabrication of COF-based S-scheme heterojunction photocatalysts remains limited. However, it is anticipated that this method will attract increasing attention in the development of S-scheme photocatalysts with yields suitable for industrial applications.

Fig. 4 (a) Synthetic process and (b) SEM image of BiOBr/TzDa COF. Reproduced with permission from ref. 102, copyright 2021, Elsevier. (c) Synthetic scheme, and (d) SEM image of TpBD COF/ZnIn₂S₄. Reproduced with permission from ref. 48, copyright 2023, Elsevier. (e) Synthetic scheme, and (f) SEM image of COF/MOF. Reproduced with permission from ref. 81, copyright 2022, American Chemical Society.

The hydrothermal technique is effective in generating highly crystalline COFs with well-defined particle sizes and minimal aggregation.¹⁰³ However, its confined reaction conditions result in relatively low yields.³³ This method, though suitable for large-scale synthesis, is time-intensive, energy-demanding, and often requires specialized equipment.⁴⁸ It enables the formation of COFs under controlled conditions, yielding materials with minimal structural defects and uniform morphology.¹⁰² However, prolonged exposure to high temperatures and pressures poses challenges for industrial applications. Moreover, COFs with borate linkages struggle with stability in humid environments, limiting their usability in specific applications.¹⁰⁴

Solvothermal synthesis

The solvothermal preparation technique is performed within a sealed container under high pressure and temperature above the boiling point of the chosen solvent. This technique begins by dissolving precursors in a solvent (such as tetrahydrofuran, dimethyl sulfoxide, N,N-dimethylformamide or 1,4-dioxane).^105,106 Once dissolved, the mixture is kept in a stainless-steel Teflon-lined autoclave to withstand the high temperatures and pressures involved.¹⁰⁷ The autoclave is then heated in an oven, typically within a temperature range of 100 °C to 200 °C, to facilitate the reaction. The duration of the heating process generally ranges from 48 to 96 h, but can be extended by relying on the features of the precursors and the complexity of the desired product.^108,109 During this reaction, the process occurs heterogeneously, meaning that the different components in the mixture interact in various phases, which is key for the formation of COF-based heterojunctions. As the autoclave heats up, the pressure inside the container increases significantly. This elevated pressure accelerates the reaction rate, which in turn promotes the nucleation of particles and the gradual growth of COF crystals.¹¹⁰ After the reaction is complete, the final product is isolated using a centrifuge, a process that can be repeated in different stages to ensure complete separation of the crystalline products.⁸⁵ This is known as the solvothermal reaction and is highly effective in synthesizing COF-based heterojunctions, as it allows for a controlled environment that promotes crystal formation. For example, Niu et al. used N,N-dimethylformamide (DMF) to develop a COF/MOF S-scheme heterojunction via solvothermal synthesis, investigating its photocatalytic performance for CO₂ reduction.⁸¹ In this process (Fig. 4e), the required amounts of COF and MOF precursors were mixed in DMF under stirring, followed by a solvothermal treatment at 120 °C for approximately 12 hours. TEM analysis (Fig. 4f) revealed a robust interfacial contact between the COF and MOF, which facilitated the migration and separation of photoinduced charges. This was confirmed by the smallest EIS Nyquist arc diameter of the composite, in contrast to that of the pure materials. Consequently, the COF/MOF photocatalyst exhibited excellent CO generation performance, which was 5 and 4.4 fold superior to that of MOF and COF, respectively. Similarly, the precursors of g-C₃N₄ and COF were mixed in a DMF solution under stirring, followed by heat treatment at 120 °C for one day under vacuum.⁸⁵ The g-C₃N₄/COF S-scheme heterojunction was successfully developed. The COF films grown on g-C₃N₄ enhanced the surface area, widened the light-harvesting range, and promoted the separation and transfer of photogenerated electron–hole pairs, leading to a sevenfold increase in hydrogen generation performance in contrast to bare g-C₃N₄. The selectivity of solvents plays a crucial role during the solvothermal synthesis of COF-based S-scheme heterojunction photocatalysts.³⁰ In some cases, a combination of solvents has been shown to improve crystallinity and porosity, which in turn promotes surface area, light absorption, and charge separation.⁴⁷ However, despite its effectiveness, the solvothermal synthesis route has some drawbacks, particularly its long reaction times, significant energy consumption required to maintain high temperatures for extended periods, and the need for specialized, bulky equipment that can handle high-pressure conditions.

To overcome these challenges, microwave-assisted solvothermal synthesis has been introduced as a faster and more energy-efficient alternative. Microwaves, first discovered in 1946 by P. L. Spencers, are a form of electromagnetic radiation with wavelengths ranging from 1 m to 1 mm and frequencies ranging from 300 MHz to 300 GHz.¹¹¹ Local microwave ovens usually operate at a frequency of 2.45 GHz.⁴⁸ Microwave-assisted solvothermal synthesis takes advantage of these microwaves to accelerate chemical reactions, significantly reducing the time required for synthesis.⁵⁵ This method can carry out reactions within just 4 minutes to 4 hours, depending on the nature of the reaction and the materials involved. The process works using an oscillating electric field generated by microwaves, which causes the dipole molecules present in the reaction mixture to move rapidly. This motion generates heat, which further accelerates the chemical reactions.^112,113 The heat generated by this process improves the nucleation rate of COF-based materials, leading to faster crystal formation.¹¹⁴ One of the major advantages of microwave-assisted solvothermal synthesis is that it offers precise control over both temperature and pressure, which is vital for achieving high-quality products. This method is particularly beneficial for synthesizing materials with specific morphological characteristics, as it permits a better command over the shape and dimension of the final product.³⁸ Additionally, the microwave-assisted solvothermal method produces high crystallization rates and improved purity, making it ideal for applications that require materials with well-defined structures and properties. The shorter synthesis times and the ability to tailor product morphology make microwave-assisted synthesis a potential substitute to conventional methods such as solvothermal fabrication, which typically require much longer reaction times and more cumbersome equipment.⁵⁸ For instance, microwave-assisted solvothermal processes have demonstrated greater efficacy than the conventional solvothermal method in increasing the surface area and light absorption of the TpPa-1/BiVO₄ S-scheme heterojunction photocatalyst, leading to nearly 72-fold higher H₂O₂ production compared to pure BiVO₄ under visible light.⁵⁵ Notably, the solvothermal method is a simple and efficient technique for synthesizing COF-based S-scheme heterojunctions, enabling faster reaction rates and larger yields, thereby facilitating the development of high-performance photocatalysts. The in situ solvothermal growth of COF onto a semiconductor forms a heterojunction characterized by robust interfacial interactions between the constituent semiconductors, enhanced porosity and surface area, and strong chemical stability.¹¹⁴ Furthermore, this direct integration of materials optimizes band alignment, reduces interfacial charge migration resistance, and consequently improves electron–hole separation and photocatalytic efficiency.²⁸ As an example, Sun et al. utilized an in situ solvothermal growth process to fabricate a WO₃/TpPa-1 S-scheme heterojunction with improved surface area, distributed light harvesting, and enhanced charge separation, achieving nearly fivefold higher H₂ production compared to pure TpPa-1 under visible light.⁸⁶ The solvothermal approach is widely utilized for synthesizing COF-based materials due to its straightforward process and ability to produce high-quality crystals.⁸⁵ Despite these advantages, it demands extended reaction times, organic solvents, and strict reaction conditions, making it challenging to control.¹⁰⁸ This method typically employs high-dielectric, proton-donating solvents, while pH adjustments through acids or bases help in achieving desired material properties.¹⁰⁶ The reaction occurs at temperatures exceeding the solvent's boiling point, ensuring higher yields with controlled morphology, size distribution, and crystallinity.⁸¹ However, solvent removal from the COF pores is a complex process, often requiring vacuum drying or solvent-based washing with ethanol or methanol.⁸⁵

Electrostatic self-assembly synthesis

Self-assembly is a process in which small components spontaneously organize themselves into larger ordered structures. This phenomenon is seen in nature, such as when molecules form cell membranes, viruses, or biopolymer fibers.⁴¹ Recently, scientists and engineers have adapted this process to create nanoscale materials in the laboratory using a bottom-up approach. In this approach, the individual components arrange themselves into specific, pre-designed structures to form larger systems, minimizing the overall free energy of the system.⁴⁶

One common method of self-assembly involves electrostatic interactions, which are especially effective in producing heterojunctions.⁴⁹ Electrostatic self-assembly relies on electrostatic interactions between nanoparticles of opposite charges, causing them to naturally align and organize in a medium. This method is gentle and allows for specific regulation of the shape and size of the assembled particles.⁵⁸ For example, when using electrostatic interactions, COFs and semiconductors with opposite charges can attract each other, resulting in their assembly into the desired structure.⁶⁵ Since the self-assembly process occurs under mild conditions, it preserves the original properties of COFs and semiconductors, leading to a heterojunction that retains the characteristics of both materials.⁹² Additionally, the self-assembly process prevents these materials from aggregating.⁹⁶ As an example, Wang et al. employed a self-assembly method to couple Tp-Tta COF with g-C₃N₄, forming an S-scheme g-C₃N₄/Tp-Tta COF heterojunction for photocatalytic CO₂ reduction under visible light.⁴⁶ The precursors of both semiconductors were mixed in water, ultrasonicated, and heated at 80 °C until the solvent evaporated to obtain the heterojunction. The synthesis scheme is shown in Fig. 5a. XRD analysis confirmed that the original phases of both materials were preserved in the heterojunction, with no impurities or transformation phases observed, demonstrating the effectiveness of the self-assembly synthesis. The TEM image (Fig. 5b) revealed that the self-assembly process effectively distributed Tp-Tta COF nanosheets onto g-C₃N₄, creating an interfacial contact that facilitated the separation and migration of charge carriers while extending light absorption up to 700 nm, as validated by DRS and PL analyses. Consequently, the optimized g-C₃N₄/Tp-Tta COF system exhibited the maximized CO generation rate of 11.25 μmol h⁻¹, which was 45 fold greater compared to that of pure g-C₃N₄.

Fig. 5 (a) Synthetic process and (b) TEM image of g-C₃N₄/Tp-Tta COF. Reproduced with permission from ref. 46, copyright 2022, Elsevier. (c) Synthetic process, and (d) TEM image of ZnO/TpPa-Cl. Reproduced with permission from ref. 50, copyright 2022, Elsevier. (e) Synthesis process and (f) TEM image of Rh-COF@COF. Reproduced with permission from ref. 105, copyright 2023, Elsevier.

Similarly, Zhang et al. utilized a self-assembly process to integrate positively charged ZnO nanoparticles with negatively charged TpPa-Cl, creating an S-scheme heterojunction for photocatalytic H₂O₂ production.⁵⁰ Initially, the authors employed Schiff-base condensation to synthesize TpPa-Cl, followed by calcination of ZnO precursors at 700 °C to obtain ZnO, as depicted in Fig. 5c. Subsequently, ZnO nanoparticles were anchored onto TpPa-Cl. XRD patterns confirmed that no phase transformation occurred in either ZnO or TpPa-Cl. Moreover, TEM analysis (Fig. 5d) demonstrated that the ZnO nanoparticles were intimately anchored onto TpPa-Cl, enhancing the porosity and surface area, thereby facilitating reactant adsorption and increasing the density of active sites. Additionally, the robust interfacial interaction between ZnO and TpPa-Cl significantly improved the separation and transport of photoinduced charge carriers, resulting in a maximum H₂O₂ generation efficiency of 2443 μmol g⁻¹ h⁻¹ under simulated sunlight, which was nearly nine times higher than that of pure TpPa-Cl. However, the limitations of electrostatic forces often constrain the complexity and diversity of the structures that can be developed, thereby hindering further advancements in these COF-based heterojunctions. This method ensures precise control over the interaction between COFs and semiconductors, leading to improved crystallinity and tunable functional properties.⁴¹ It enables synthesis under mild conditions, preserving structural integrity and enhancing material stability.⁵⁰ However, achieving a uniform interface remains difficult, and controlling COF thickness is a challenge.⁴⁶ Additionally, defects and interface irregularities formed during the assembly process may hinder the photocatalytic efficiency of the resulting material.⁹⁶

In situ growth

The in situ growth synthesis of COF-based S-scheme heterojunctions is a versatile method that hinges on the intrinsic characteristics of the materials involved. This approach takes advantage of the functional groups and surface properties of both the COFs and semiconductors, enabling precise interactions essential for seamless integration.¹¹⁵ Without the need for additional post-treatment processes, this strategy simplifies the fabrication while ensuring robust interfacial contact. Such a contact is pivotal in creating band-aligned heterojunctions that enhance charge carrier dynamics, leading to faster separation and migration of photogenerated charge carriers.¹¹⁶ Moreover, this synthesis method ensures the uniform deposition of COFs onto the semiconductor surface, resulting in a well-ordered, porous structure. This uniformity not only enhances the material's adsorption ability for reactants but also enhances their diffusion efficiency within the photocatalyst framework.¹¹⁷ The maximized porosity further contributes to a higher density of active sites, thereby supporting prolonged carrier lifetimes and improving the overall photocatalytic efficiency.¹¹⁷ Additionally, the covalent interactions that develop during this process impart remarkable thermal stability to the material. This stability is critical for maintaining the integrity and performance of the photocatalyst under extreme operating conditions, making it highly durable for long-term applications.¹¹⁸ The synergy between the structural, interfacial, and electronic properties enabled by this method paves the way for the development of advanced photocatalytic systems with superior sunlight conversion capabilities.¹¹ For instance, Zhao et al. utilized an in situ growth approach to prepare a hierarchical core–shell S-scheme heterojunction by integrating hierarchical Rh-COF with a hollow COF, where Rh-COF functioned as the shell.¹⁰⁵Fig. 5e illustrates the schematic preparation of the hierarchical Rh-COF@COF core–shell structure. TEM analysis (Fig. 5f) revealed that the hollow COF core was encapsulated with a hierarchical Rh-COF shell, resulting in an increased surface area that enhanced the catalyst–light interaction and facilitated mass transport. This structural optimization significantly improved the separation and transport of the photoinduced charge carriers, thereby supporting the photocatalytic reaction. The photocatalytic performance of the S-scheme Rh-COF@COF photocatalyst was 3.7 times higher than that of pure Rh-COF under visible-light irradiation.

Similarly, Qiu et al. employed an in situ growth process to grow In₂S₃ nanosheets onto COF substrates, aiming to evaluate their photocatalytic performance for H₂O₂ generation.⁶⁹ The authors initially synthesized COF through Schiff base condensation of TAPB and DMTP, which was subsequently dissolved with InCl₃ in water. Indium ions coordinated with the active sites on the COF, while TAA decomposed to release sulfur ions, leading to the formation of In₂S₃ nanosheets robustly anchored to the COF. Microstructural analysis revealed that the core–shell nanostructure of COF/In₂S₃ developed through the network growth of In₂S₃ nanosheets on the COF. This integration increased the surface area from 87 to 267 m² g⁻¹, extended the light absorption edge to 700 nm, and enhanced the separation and transfer of photogenerated charge pairs, as confirmed by various characterization techniques. The optimized COF/In₂S₃ photocatalyst achieved a remarkable H₂O₂ generation efficiency of 5713.2 μmol g⁻¹ h⁻¹ in the absence of a sacrificial agent. Xia et al. adopted a similar strategy, initially preparing COF via Schiff base condensation between TG and DHPA.¹⁰⁶ The synthesized COF was then mixed with bismuth nitrate, allowing Bi ions to adsorb onto the COF. These ions were subsequently reduced to Bi₂O₃ nanoparticles, which were strongly anchored to the COF via robust interfacial interactions. The resulting heterojunction displayed increased porosity and surface area, which facilitated mass transport and reactant diffusion. This design yielded a 5.6-fold enhancement in H₂O₂ production compared that with of pure Bi₂O₃, which was attributed to the improved separation and transfer of photogenerated charge pairs. Hence, the in situ growth of COF-based S-scheme heterojunctions offers effective control over morphology, promotes band alignment between coupled semiconductors, and enhances the separation and transport of electron–hole pairs, collectively boosting photocatalytic performance. This technique enhances charge separation, minimizes electron–hole recombination, and improves overall material stability.¹¹ The strong interfacial bonding promotes effective reactant diffusion and increased catalytic performance.¹¹⁵ Furthermore, the uniform porosity contributes to better adsorption capacity and selectivity for organic pollutants.¹¹⁸ Enhanced conductivity and reduced interfacial resistance further improve long-term stability and ion transport efficiency.¹¹⁶ Additionally, the robust bonding between COFs and 2D materials offers excellent resistance to corrosion. However, controlling the morphology during synthesis can be difficult, sometimes leading to structural imperfections.¹⁰⁵

Other synthesis methods

The sonochemical synthesis process has gained attention in photocatalyst fabrication owing to its advantages such as low power consumption, rapid reaction rates, and cost-effectiveness. The generation and collapse of acoustic cavitation bubbles create localized hot spots, leading to extreme temperatures (>5000 K) and pressures (>1000 atm) within short durations.¹¹⁹ These dynamic conditions enable rapid heating and cooling cycles, promoting COF-based S-scheme photocatalyst crystallization.¹²⁰ This process enhances crystallinity while optimizing properties such as size, morphology, and defect structure, which are vital for photocatalytic efficiency. Sonochemically prepared S-scheme photocatalysts feature large surface areas, which increase the number of active sites for photocatalytic reactions.¹²¹ This also improves mass transport and reactant diffusion, ensuring effective interactions with active sites. This method achieves high reaction rates, as seen in its short synthesis duration (∼60 min), making it efficient and scalable.²⁸ Moreover, it enables the formation of heterostructures and S-scheme interfaces with improved light absorption and charge separation, enhancing sunlight-driven photocatalysis.⁴² Tailoring parameters such as ultrasonic power, frequency, solvent, and precursors offers precise control over structural and functional properties, making the sonochemical method versatile and valuable for synthesis of advanced materials.⁶⁰

For example, Zhang et al. utilized a sonochemical method to integrate BiOBr nanosheets onto TpPaCl₂, forming an S-scheme TpPaCl₂/BiOBr system for the photocatalytic generation of H₂O₂.⁹⁴ The authors initially synthesized TpPaCl₂via Schiff-base polymerization of Tp and PaCl₂. Subsequently, TpPaCl₂ was mixed with KBr and bismuth nitrate, which served as BiOBr precursors, in a water–glycerol mixture under sonication for 30 min to fabricate the TpPaCl₂/BiOBr heterojunction. Microstructural analysis revealed that the TpPaCl₂ microspheres were partially covered with BiOBr nanoparticles, resulting in an increased surface area and porosity. These structural improvements facilitated the formation of additional active sites, enhanced mass transport, and reduced charge recombination rate. The band position alignment between the two materials established an S-scheme heterojunction, achieving a remarkable H₂O₂ generation rate of 3749 μmol g⁻¹ h⁻¹ under simulated sunlight, which is 27 times higher than that of bare BiOBr. Thus, the sonochemical process effectively enhances the surface area and porosity, highlighting its potential for the industrial-scale synthesis of COF-based S-scheme heterojunctions. Thermal annealing involves heating the constituent precursors to high temperatures, which enables the formation of specific microstructures with enhanced crystallinity, composition, and interface properties. This process increases surface area and improves charge separation, thereby supporting photocatalytic activity.⁶⁸ For example, Niu et al. employed thermal annealing to develop an S-scheme CuWO₄/COF heterojunction with improved photocatalytic performance for CO₂ reduction under visible light.³⁴ The COF was fabricated by dispersing specific amounts of TMT and TFPT in a mixture of 1-butanol and o-dichlorobenzene, followed by a solvothermal reaction at 120 °C for 72 hours. CuWO₄ was prepared by dissolving copper nitrate and sodium tungstate dihydrate in water, followed by co-precipitation and calcination. To construct the heterojunction, the required amount of COF dissolved in methanol was ultrasonically mixed with CuWO₄ and heated at 80 °C, followed by calcination at 250 °C for 1 h. Microstructural analysis showed that CuWO₄ nanoparticles were distributed along the COF nanofibers, resulting in an improved surface area, extended light harvesting, and promoted electron–hole separation. This led to a CO generation level of 7.17 μmol g⁻¹ h⁻¹, highly surpassing that of pure COF (1.6 μmol g⁻¹ h⁻¹). Despite its potential, thermal annealing remains underexplored for the synthesis of COF-based S-scheme heterojunctions.

Wet chemical synthesis is a highly versatile method that provides precise control over the crystallinity, particle size, and morphology of COF-based S-scheme heterojunctions by carefully selecting precursors, shape-controlling agents, and reducing agents.¹²² Surfactants play a crucial role by modulating the surface energy, driving crystallization, and promoting stable crystal structures, which enhance charge separation and migration.¹²³

This approach also offers scalability and cost-effectiveness, operating under mild conditions while enabling uniform COF deposition on substrates for improved photocatalytic performance.¹²⁴ However, challenges include maintaining stoichiometry and homogeneity and the potential for impurities from additives while ensuring consistency.¹²⁵ Despite these limitations, wet chemical synthesis remains a promising strategy, with future efforts needed to optimize reaction conditions, develop sustainable surfactants, and improve scalability for industrial applications. For instance, Zhang et al. employed a wet chemistry technique to anchor MgIn₂S₄ nanosheets onto a hollow COF, forming an S-scheme heterojunction for visible-light-activated photocatalytic H₂O₂ generation.⁶⁴ The authors initially synthesized COF via Schiff-based polymerization and crystallization between TPT-CHO and TAPB, followed by the application of wet chemistry to grow MgIn₂S₄ nanosheets on the COF surface. Microstructure analysis revealed robust interfacial interactions between the MgIn₂S₄ nanosheets and COF, leading to an increased surface area, widened light harvesting, and enhanced mass transport to favor the photocatalytic reaction. As a result, the optimized MgIn₂S₄/COF heterojunction achieved a remarkable H₂O₂ generation level of 4.52 mmol g⁻¹ h⁻¹, which was 9.4 fold superior to that of bare COF, highlighting the effectiveness of wet chemistry synthesis in developing heterojunctions with superior photocatalytic properties. However, material aggregation during synthesis reduces the surface area and limits the carrier transfer rates, partially compromising the photocatalytic performance.³⁹ However, the method remains cost-effective and suitable for industrial applications.

Various kinds of COF-based S-scheme heterojunctions

The porosity and high surface area of COFs play a pivotal role in the prevention of semiconductor aggregation when coupled with suitable semiconductors. This attribute ensures the accessibility of a larger density of photocatalytic reactive sites.²⁸ Moreover, the framework and porous structure of COFs are instrumental in reducing carrier transport distances when combined with other semiconductors, thereby accelerating the transfer of photoinduced carriers.³⁰ Importantly, compared to traditional heterojunctions like type II or Z-scheme configurations, the S-scheme heterojunctions exhibit a notable extension in the photoinduced carrier lifetime while significantly decreasing the resistance of the composite material. This results in an enhanced redox capability, underlining the superior performance of these composites.⁴⁰ In the context of classification, COF-based S-scheme heterojunctions can be categorized by their heterojunction structure or COF modification strategies. From a structural perspective, these heterojunctions are divided into single-and dual S-scheme heterojunctions. However, from the standpoint of modification strategies, the focus shifts to COF–COF heterojunctions, MOF–COF heterojunctions, and S-scheme heterojunctions involving co-catalysts coupled with Schottky junctions. Although these methodologies have gained recognition, they remain at various stages of development, with ongoing research aimed at addressing challenges and fostering innovation.

Structural level

Single S-scheme heterojunctions

At the structural level, the integration of semiconductor materials with COFs to construct single S-scheme heterojunctions has gained significant attention. The selection of materials in this context is influenced by differences in the E_f levels, CB, and VB of the involved components. Materials with higher E_f levels and more negative CB values are classified as RP, whereas those with lower E_f levels and more positive VB values are categorized as OP.^45,58 While earlier studies primarily emphasized on n-type semiconductors for this construction, recent advancements have expanded the scope to include p-type semiconductors, broadening the versatility of the S-scheme heterojunction framework, as shown in Fig. 6.

Fig. 6 Photoinduced electron–hole migration mechanism of four kinds of single S-scheme heterojunction systems under light excitation: n–n heterojunction (n with a more negative CB potential is RP while the other is OP), p–n heterojunction (n is OP while p is RP), n–p heterojunction (n is RP while p is OP), and p–p heterojunction (p with a more negative CB potential is RP while the other is OP).

Zhao et al. employed an in situ encapsulation method to construct an S-scheme W₁₈O₄₉@COF heterojunction, achieving enhanced photocatalytic performance for aerobic oxidation under visible light.³⁵ SEM analysis (Fig. 7a) confirmed the effective wrapping of W₁₈O₄₉ nanobundles with TpPa-H COF through robust interfacial interactions, resulting in a high surface area and abundant pore channels that facilitated mass transport and suppressed recombination of photoinduced electron–hole pairs. This was further validated by photoluminescence (PL) and electrochemical impedance spectroscopy (EIS) Nyquist studies (Fig. 7b and c), which demonstrated the lowest PL intensity and smallest EIS Nyquist arc radius for the W₁₈O₄₉/TpPa-H heterojunction compared to its individual components. Additionally, the absorption spectra corroborated the effectiveness of the S-scheme heterojunction by exhibiting enhanced light absorption, which provided abundant charges for supporting photocatalytic reactions. Consequently, the W₁₈O₄₉/TpPa-H photocatalyst achieved 100% benzylamine conversion within 4 h, which is approximately three times higher than that of pure TpPa-H. This improvement was ascribed to the smaller work function of TpPa-H (5.62 eV) than that of W₁₈O₄₉ (8.05 eV), which induced spontaneous electron migration from W₁₈O₄₉ to TpPa-H upon integration. This migration results in band bending and the generation of interfacial BIEF, facilitating charge separation and migration. Under light irradiation, the photoinduced electrons in the CB of W₁₈O₄₉ recombined with holes in the VB of TpPa-H, effectively separating the high-potential electrons in the CB of TpPa-H and the holes in the VB of W₁₈O₄₉ to drive the aerobic oxidation reaction (Fig. 7d).

Fig. 7 (a) SEM image, (b) PL spectra, (c) EIS Nyquist plots, and (d) charge transfer mechanism of W₁₈O₄₉@COF. Reproduced with permission from ref. 35, copyright 2024, Elsevier. XPS analysis (e) C 1s, (f) N 1s, (g) S 2p, (h) In 3d, and (i) charge-transfer mechanism of TpMa-H/In₂S₃. Reproduced with permission from ref. 84, copyright 2024, Elsevier.

Similarly, Chen et al. developed a single S-scheme heterojunction by integrating In₂S₃ nanosheets with TpMa nanotubes to enhance photocatalytic performance for H₂O₂ production.⁸⁴ TEM analysis confirmed the successful integration of In₂S₃ nanosheets with TpMa nanotubes via strong interfacial interactions. The absorption spectra revealed that the generation of the TpMa-H/In₂S₃ system significantly broadened light harvesting, increasing the excitation and availability of photoinduced charge carriers. The improved separation and migration of these charges were validated by the lowest EIS Nyquist arc radius observed for the optimized TpMa-H/In₂S₃ heterojunction compared with its counterparts. Consequently, the TpMa-H/In₂S₃ heterojunction exhibited a maximum H₂O₂ generation level of 311.07 μmol L⁻¹. The development of an S-scheme system between TpMa-H and In₂S₃ was further confirmed by in situ XPS analysis. As shown in Fig. 7e–h, the binding energies of the TpMa-H orbitals exhibited a negative shift, whereas those of In₂S₃ showed a positive shift after heterojunction formation, indicating electron migration from In₂S₃ to TpMa-H. Under light irradiation, the binding energies of the component orbitals reversed their shifts in contrast to the dark state. This indicates that the photogenerated electrons and holes in the CB of TpMa-H and holes in the VB of In₂S₃ recombine, enhancing the separation of electrons in the CB of TpMa-H and holes in the VB of In₂S₃. These separated charge carriers facilitate the redox reactions necessary for H₂O₂ production (Fig. 7i).

Qi et al. reported improved photocatalytic performance for SMT degradation by forming an S-scheme heterosystem between COF and g-C₃N₄ using an in situ polymerization process.⁸⁷ Microstructural analysis demonstrated that the COF nanosheets were intimately coated onto the g-C₃N₄ substrate via strong π–π interactions, resulting in an enhanced surface area with abundant charge transport channels and numerous catalytic sites to support photoreactions. The formation of the heterojunction also promoted charge carrier separation and migration, as verified by the diminished PL intensity and smallest EIS Nyquist arc diameter. Consequently, the optimized COF/g-C₃N₄ photocatalyst exhibited a 6.6-fold improvement in SMT degradation performance compared to pure COF. Upon light irradiation, the photogenerated electrons in the CB of g-C₃N₄ recombined and perished with holes in the VB of COF, enhancing the separation of electrons in the CB of COF and holes in the VB of g-C₃N₄. This process generates superoxide and hydroxyl radicals, which actively degrade SMT. Although these studies achieved significant performance improvements, further analysis is required to establish the recyclability and scalability of these materials in various operational environments, such as real wastewater matrices or continuous flow reactors. In addition, the development of COF-based S-scheme heterojunctions introduces a new dimension of π–π interactions, which warrants a deeper investigation into how these interactions influence the electronic band structures and active site accessibility. Future research should also address the eco-friendliness and cost-effectiveness of synthetic processes to align these advancements with sustainable development goals and ensure practical viability beyond laboratory conditions. Integrating computational studies, such as DFT simulations, can further direct the formation of S-scheme heterojunctions by predicting the optimal material combinations and their electronic properties for specific applications.

Dual S-scheme heterojunctions

The formation of single S-scheme heterojunctions involving COFs as one of the components has demonstrated a significant improvement in photocatalytic efficiency by enhancing the separation of high-potential photoinduced charge carriers.⁴⁸ However, their photocatalytic efficiency remains insufficient to meet the demands of large-scale applications. Recently, the concept of dual S-scheme heterojunctions has garnered considerable attention owing to their superior ability to facilitate charge-carrier separation on a larger scale. This enhancement is attributed to the increased driving force provided by the synergistic effects of two BIEFs formed at the interfaces of the two junctions.³⁶ The formation of dual S-scheme heterojunctions requires specific prerequisites, including proper energy band alignment, appropriate band edge positions, and a sufficient dissimilarity in E_f levels between the component materials.²⁹ These factors collectively ensure the effective formation of dual S-scheme heterojunctions, resulting in optimal charge transfer and separation. For researchers interested in the principles governing charge transfer and separation, the conditions necessary for heterojunction formation, and the precise roles of band edge positions and E_f level differences, we recommend our recently published review paper on dual S-scheme heterojunction photocatalysts. This paper provides an in-depth exploration of all aspects related to these dual S-scheme heterojunctions.⁹⁴ In this section, we specifically emphasize on the enhanced performance exhibited by COF-based dual S-scheme heterojunctions, emphasizing their potential for advancing photocatalytic applications. For instance, Fan et al. transitioned the S-scheme CdS@COF core–shell heterojunction, which exhibited poor photocatalytic performance, into a high-performance dual S-scheme CdS/BiVO₄@COF heterojunction by introducing BiVO₄ at the interface of CdS@COF.¹⁰⁹ Notably, the role of the COF shell shifted from an OP to an RP in the dual S-scheme heterojunction because of the significant matching of the band structures and the E_f level differences. The authors first anchored BiVO₄ nanoparticles onto the exterior of CdS, which was subsequently enveloped by a COF shell. The resulting heterostructure exhibited an enhanced surface area due to the combined effect of the three components, thereby extending light absorption, improving mass transport, and promoting electron–hole separation. The CdS/BiVO₄@COF photocatalyst achieved the maximum syngas and O₂ yields of 453.6 and 230.5 μmol g⁻¹ h⁻¹, respectively. These improvements were ascribed to the increased separation of high-potential charges driven by the generation of two BIEFs at the interfaces. The enhanced charge separation was validated by the highest current density and lowest PL intensity of CdS/BiVO₄@COF compared to those of CdS@COF and pure materials. KPFM analysis was employed to authenticate the S-scheme charge-migration mechanism, and the results are presented in Fig. 8. For the CdS@COF heterojunction, a high surface potential difference of 60.6 mV between CdS and COF indicated the formation of an interfacial BIEF owing to the charge redistribution process in the dark. Under light irradiation, the surface potential of CdS decreased, whereas that of COF increased, lowering the potential difference between CdS and COF to 49.42 mV, indicating the transfer of photogenerated electrons from the COF to CdS. In contrast, the CdS/BiVO₄@COF heterojunction demonstrated a higher potential in the outermost region and a reduced potential in the middle region, suggesting the flow of free carriers from CdS and COF to BiVO₄. Additionally, in contrast with CdS@COF, the potential difference of CdS/BiVO₄@COF in the dark increased to 80.15 mV, which then decreased to 36.9 mV upon light exposure. This significant potential difference under dark and light conditions confirmed that the photogenerated electrons in the CB of CdS and COF perished with photoinduced holes in the VB of BiVO₄, thereby confirming the dual S-scheme charge migration mechanism in CdS/BiVO₄@COF.

Fig. 8 KPFM analyses of (a) CdS/COF and (b) CdS/BiVO₄@COF: AFM studies (left) and KPFM analyses (right) under dark and light irradiation; (bottom) line-scanning surface potential curves under dark and light irradiation. Reproduced with permission from ref. 109, copyright 2024, Elsevier.

Similarly, Lu et al. demonstrated that the formation of a dual S-scheme α-Fe₂O₃/g-C₃N₄/COF heterojunction effectively enhanced the visible-light-driven photoactivity for TC degradation.¹¹⁵ Microstructural analysis revealed that the net structure of the COF, featuring wide pore routes, was effectively integrated with lamellar-shaped g-C₃N₄ and hexagonal α-Fe₂O₃. This integration facilitates a large surface area and high porosity, promoting the dispersion of active sites, offering additional carrier transport channels, and reducing the charge diffusion distance, thereby impeding the recombination of photogenerated charge pairs. Furthermore, the unique heterostructure broadens light harvesting, generating abundant charges for the redox processes. Consequently, the TC degradation rate reached 0.00561 min⁻¹, which was 2.5 and 2 fold superior to those of the COF and α-Fe₂O₃/g-C₃N₄, respectively. Additionally, the optimal candidate exhibited robust stability over five continuous cycles without structural degradation, highlighting its potential for large-scale applications. Upon light excitation, the photogenerated electrons in the CB of α-Fe₂O₃ recombine with holes in the VB of g-C₃N₄ and COF, enhancing the separation of electrons in the CBs of g-C₃N₄ and COF and holes in the VB of α-Fe₂O₃. This process generates superoxide and hydroxyl radicals, which effectively degrade the TC. Despite the remarkable ability of COF-based dual S-scheme heterojunctions to enhance the photocatalytic performance, few studies have explored this strategy. However, given the exceptional properties of COFs and the charge separation potential of dual S-scheme heterojunctions, this approach is anticipated to witness a significant surge in research interest in the near future.

COF level

COF–MOF-based S-scheme heterojunctions

The effective preparation of porous MOFs enhances the surface area, offering abundant active sites and providing shortened charge transport channels, which collectively increase the adsorption of reactants, elevate reaction activity, and accelerate the separation and transfer of photogenerated electron–hole pairs.⁴⁵ These features significantly improve the photocatalytic efficiency by addressing key restrictions like recombination of electron–hole pairs and limited interaction between the catalyst and reactants.⁶⁰ Furthermore, the nanostructure and composition of MOFs can be modulated to achieve extended light absorption, a critical factor in exciting and generating sufficient charge carriers to facilitate redox reactions under various light conditions.¹¹ This structural tunability, combined with their versatile properties, has made MOFs a prominent subject of attention in the domain of photocatalysis. In addition to these advantages, the development of a COF-MOF S-scheme system synergistically integrates the strengths of both materials. This unique heterojunction not only enhances the surface area and extends light harvesting capacity but also significantly improves charge separation efficiency, which is vital for sustaining high photocatalytic activity.⁷⁵ The spatially directed charge transfer within the S-scheme heterojunction ensures effective exploitation of the generated charges, thereby boosting photocatalytic reactions.³⁹ As an example, the MOF-808-NH₂ was coupled with TpTt-COF through a hydrothermal method to form an S-scheme heterojunction for photocatalytic AP removal under visible light.⁸³ Microstructural analysis revealed a core–shell structure, where rod-like TpTt-COF grew onto the surface of rope-like MOF-808-NH₂, increasing the surface area to 54.95 m² g⁻¹ and creating wide pore channels to enhance mass transport and light absorption. The formation of C–N bonds facilitates charge transport, improving the separation and transport of photoinduced charge pairs. Consequently, the optimal MOF-808-NH₂/TpTt-COF heterojunction achieved an AP removal rate of 98% within 2 h with negligible performance loss and structural degradation after four recycling experiments, confirming the potential of S-scheme heterojunctions. Niu et al. employed a solvothermal process to develop S-scheme heterojunctions by integrating COF nanofibers with MOF octahedra in varying molar ratios for photocatalytic CO₂ reduction under visible light.⁸¹ SEM analysis (Fig. 9a) demonstrated that MOF was effectively embedded onto COF with strong interfacial interactions, enhancing carrier separation and transport. The higher surface area of the COF/MOF heterojunction (900 m² g⁻¹) compared to that of the individual components facilitated CO₂ adsorption and activation, provided abundant active sites, and shortened the charge transport distance. This was validated by the diminished PL intensity and the smallest EIS Nyquist arc diameter of the optimized heterojunction, as shown in Fig. 9b and c. TAS analysis (Fig. 9d and e) further confirmed prolonged charge carrier lifetimes, with the average relaxation lifespan of the heterojunction (242.1 ps) significantly exceeding that of the MOF (119 ps) within the 550–800 nm wavelength range. Consequently, the optimized COF/MOF heterojunction achieved a CO generation rate of 6.56 μmol g⁻¹ h⁻¹, 5- and 4.4-fold superior to those of the MOF and COF, respectively. XPS spectra provided evidence of the S-scheme charge transport route. As Fig. 9f–i show, the binding energies of N 1s and C 1s in COF exhibited a red shift within the heterojunction under dark conditions, whereas Zr 3d and O 1s in MOF showed a blue shift, indicating electron migration from COF to MOF. Under light irradiation, the binding energies of the constituent orbitals were reversed, verifying photoinduced electron migration from the MOF to COF. This separation of electrons in the CB of the COF and holes in the VB of the MOF enabled CO₂ and H₂O to react, generating CO, H⁺, and O₂. Similarly, Hu et al. utilized ultrasonic treatment to integrate H₂N-Cu-MOF nanosheets with flower-like TpPa-1-COF, developing an S-scheme heterosystem to enhance H₂ generation performance.⁶⁷ The resulting nanostructure improved the surface area, widened light harvesting, and reduced charge diffusion distances, promoting efficient charge separation and migration. The optimized H₂N-Cu-MOF/TpPa-1-COF photocatalyst exhibited a hydrogen generation efficiency of 4.19 mmol g⁻¹ h⁻¹, surpassing TpPa-1-COF and H₂N-Cu-MOF by factors of 22.1 and 46.5, respectively. Stability analysis over four cycles revealed a negligible performance loss, confirming the robustness of the photocatalyst. ISIXPS analysis verified the S-scheme charge migration route, where electrons in the CB of TpPa-1-COF reduced H⁺ to H₂ and holes in the VB of H₂N-Cu-MOF were scavenged by ascorbic acid. While these studies show that the S-scheme heterojunction effectively facilitates spatial charge separation, there is scope to explore the impact of external factors, such as solution pH, temperature, and competing ions, on the stability and robustness of these heterojunctions. Moreover, leveraging in situ techniques, including XPS, fs-TAS, and KPFM, could offer additional insights into the transient behavior of charge carriers and their interaction with reactants during the photocatalytic process. Overall, while these studies exemplify the potential of MOF/COF S-scheme heterojunctions, future research should prioritize scalable synthesis methods, enhanced mechanistic understanding, and long-term stability in complex environments to transition these materials to industrial-scale photocatalytic applications.

Fig. 9 (a) SEM image, (b) PL spectra, (c) EIS Nyquist plots, (d and e) TAS analysis, ex situ/in situ XPS analysis, (f) C 1s, (g) N 1s, (h) O 1s, and (i) Zr 3d of COF/MOF. Reproduced with permission from ref. 81, copyright 2022, American Chemical Society.

COF–COF-based S-scheme heterojunctions

This novel approach leverages the distinctive features of individual COFs, including their tunable porosity, high crystallinity, and large surface areas, while overcoming their intrinsic limitations, such as inefficient charge separation and low light absorption.³⁷ By grafting one COF onto another, the resulting heterojunctions not only optimize charge carrier dynamics but also enhance light harvesting capabilities through synergistic interactions.²⁷ Furthermore, the integration ensures more efficient utilization of the visible light and improves the redox potential of the photocatalyst.⁴² These atomic-level charge transfer pathways, facilitated by the strong interfacial coupling between COF units, reduce the recombination rates of photogenerated carriers, and prolong their lifetimes.⁶¹ This, in turn, significantly boosts the efficiency of photocatalytic reactions, particularly for applications such as hydrogen production, CO₂ reduction, organic transformation, and pollutant removal.

For example, Luan et al. employed an in situ process to decorate Pt nanoparticles and COF-316 nanosheets into TbBpy-COF nanosheets, thereby fabricating an S-scheme COF-316/Pt@TbBpy-COF heterojunction for evaluating the efficiency of overall water splitting under visible light.⁶⁸ TEM analysis (Fig. 10a) confirmed the intimate integration of COF-316 nanosheets with nanoribbons of TbBpy-COF, along with the effective decoration of Pt particles within the pores of COF. The formation of the COF-316/Pt@TbBpy-COF heterostructure via π–π interactions effectively enhanced the surface area, promoting the availability of catalytic centers and facilitating catalyst–light interactions to generate abundant charge carriers. Moreover, COF-316/Pt@TbBpy-COF exhibited the highest current density and the lowest PL intensity compared to its counterparts, verifying that the heterojunction significantly increased the separation and migration of photoinduced electron–hole pairs. Consequently, this heterostructure acquired a maximum H₂ generation of 220.4 μmol g⁻¹ h⁻¹ with no performance loss over five consecutive recycling experiments, and an AQY of 5.75% at 500 nm. EPR analysis further verified the S-scheme charge-separation route. As shown in Fig. 10b, the DMPO-˙OH signal for COF-316/TbBpy-COF was considerably larger than those for pure COF-316 and TbBpy-COF, confirming the accumulation of photogenerated charge pairs with a strong redox capacity, a hallmark of the S-scheme heterojunction. Under light irradiation, the photogenerated electrons and holes in the CB of COF-316 and the VB of TbBpy-COF recombined, enhancing the separation of electrons and holes in the CB of TbBpy-COF and the VB of COF-316, respectively. These separated charge carriers subsequently drive water splitting, producing H₂ and O₂.

Fig. 10 (a) TEM image and (b) DMPO-˙OH spectra of COF-316/Pt@TbBpy-COF. Reproduced with permission from ref. 68, copyright 2024, Wiley. (c) TAS analysis, and (d) TRPL analysis of Az-COF/rGO/Hz-COF. Reproduced with permission from ref. 82, copyright 2024, American Chemical Society.

Niu et al. synthesized an S-scheme heterostructured system by integrating Az-COF and Hz-COF using a solvothermal process, with rGO serving as a mediator for overall water splitting.⁸² UPS analysis revealed that the work function of Az-COF (3.72 eV) was inferior to that of Hz-COF (4.39 eV), representing charge migration from Az-COF to Hz-COF upon integration. This migration created an interfacial IEF, which served as a dynamic force for the improved separation of photogenerated electrons in Az-COF and holes in Hz-COF under light irradiation. TAS analysis (Fig. 10c) verified that the intensity of the excited absorption peaks in the Az-COF/rGO/Hz-COF heterojunction was higher than that in the pure COFs, signifying the enhanced separation of photoinduced electron–hole pairs due to the S-scheme heterojunction. TRPL analysis (Fig. 10d) further supported this observation, showing the shortest charge recombination lifetime for Az-COF/rGO/Hz-COF, indicating accelerated charge separation and migration across the interface. The Az-COF/rGO/Hz-COF photocatalyst achieved the highest H₂ and O₂ levels of 120 and 58 μmol g⁻¹ h⁻¹, respectively. Hence, the modularity and structural diversity of COFs provide further opportunities to fine-tune the electronic properties of these heterojunctions, enabling precise control of their photocatalytic behavior. These studies revealed that controlling the interfacial coupling between COFs is essential for optimizing charge–transfer pathways, as challenges such as mismatched energy levels and insufficient contact can hinder the formation of efficient heterojunctions. Advanced synthesis techniques, including template growth and molecular pre-design, can overcome these limitations and enhance photocatalytic performance. Notably, these photocatalysts highlight their potential; however, scalability remains a concern because of the challenges of maintaining crystallinity and structural integrity in large-scale synthesis methods, such as in situ decoration and solvothermal processes. Furthermore, integrating computational studies such as DFT calculations with experimental analyses such as EPR, UPS, TAS, and TRPL could provide deeper insights into interfacial charge dynamics and band alignment, offering a direction for the rational formation of next-generation COF-based heterojunctions.

COF-QD-based S-scheme heterojunctions

Quantum dots (QDs) exhibit several matchless characteristics that render them highly appropriate for enhancing the photocatalytic performance, especially when integrated into COF-based S-scheme heterojunction photocatalysts. Their ability to absorb light in the visible region allows for efficient utilization of solar energy, which is a crucial factor in promoting photocatalytic reactions.¹²⁶ Furthermore, the quantum confinement effect in QDs facilitates multiple exciton generation, enabling the emission of up to three electrons per absorbed photon, which is a significant improvement compared to the single-electron emission typical of bulk semiconductors.¹²⁷ This characteristic is particularly advantageous for boosting the charge carrier density in photocatalytic systems. Additionally, QDs possess excellent charge separation and transport capabilities because of their nanoscale dimensions and large intrinsic dipole moments, which align well with the charge transport pathways in S-scheme heterojunctions.¹²⁸ The high exciton coefficient of QDs in the visible region further enhances their light absorption efficiency, ensuring effective solar spectrum utilization.¹²⁹ Moreover, their luminescence properties contribute to reducing charge carrier recombination, thereby improving photocatalytic efficiency.³³ When integrated with COF semiconductors, QDs extend the visible-light harvesting extent of the resulting heterojunction, ensuring complementary light harvesting.¹³⁰ These synergistic properties make QDs an exceptional component in COF-based S-scheme heterojunction photocatalysts, where efficient charge separation, enhanced light absorption, and optimized charge transport are vital for driving advanced photocatalytic applications.

For instance, Zhao et al. used a self-assembly process to integrate a flower-like COF with ZnSe QDs, thereby developing an S-scheme heterojunction for photocatalytic H₂O₂ production.⁵⁸ Microstructure analysis confirmed the uniform decoration of ZnSe QDs onto the flower-like COF, with a robust interfacial interaction. The absorption spectra revealed that the absorption shoulder of the COF extended from 690 nm to 810 nm upon integration with ZnSe QDs (Fig. 11a), resulting in increased excitation and the creation of more photoinduced electron–hole pairs. Furthermore, TPRL analysis indicated that the average lifetime of ZnSe/COF was shorter than those of bare COF and ZnSe, verifying the reduced recombination of high-potential photoinduced charge carriers. This was also supported by the smallest EIS Nyquist arc radius of ZnSe/COF compared with its competitors. The optimized ZnSe/COF exhibited the highest H₂O₂ generation level (1895 μmol g⁻¹ h⁻¹) under light irradiation. XPS analysis not only confirmed the formation of the heterojunction but also authenticated the S-scheme charge transfer pathway between the COF and ZnSe. As Fig. 11b–e show, without light irradiation, the binding energies of COF orbitals in ZnSe/COF exhibited a negative shift compared to those of pure COF orbitals, while the binding energies of ZnSe in ZnSe/COF were higher than those of pure ZnSe orbitals. This variation in the binding energy confirmed the transfer of free carriers from ZnSe to COF upon the creation of the heterojunction. Conversely, upon light exposure, the binding energies of the constituent orbitals of both materials within the heterojunction exhibited a reverse shift, validating the performance of photogenerated electrons in the LUMO of COF with holes in the VB of ZnSe, which is consistent with the S-scheme mechanism (Fig. 11f). The separated electrons in the CB of ZnSe reduced O₂ and H⁺ to generate H₂O₂, whereas holes in the HOMO of COF were oxidized through ethanol.

Fig. 11 (a) Light absorption spectra, ex situ/in situ XPS analysis (b) O 1s, (c) N 1s, (d) Zn 2p, (e) Se 3d, and (f) charge transfer mechanism of ZnSe/COF. Reproduced with permission from ref. 58. Copyright 2024, Elsevier. TAS analysis of (g and h) COF, (i and j) ZnCdS, and (k and l) COF/ZnCdS. Reproduced with permission from ref. 97, copyright 2024, Wiley.

The presence of high-energy surface atoms in QDs can lead to particle agglomeration during photocatalytic reactions, reducing the surface area and charge transport channels. However, the high surface area and plentiful active sites of COF can effectively reduce QD agglomeration. For example, Ma et al. reported that the creation of an S-scheme heterojunction between ZnCdS QDs and COF effectively mitigated agglomeration and promoted the uniform dispersion of ZnCdS QDs.⁹⁷ This resulted in a substantial enhancement in the surface area of COF/ZnCdS, providing abundant active sites and charge transport channels, which reduced recombination and increased the generation of charge carriers, as evidenced by the maximized current density and extended light absorption. Consequently, the optimized COF/ZnCdS exhibited the highest H₂O₂ generation efficiency of 5171 μmol g⁻¹ h⁻¹ under visible light, which is double that of the pure COF. The enhanced performance was ascribed to the retention of high-potential photoinduced charge pairs without perishing owing to the formation of an S-scheme heterojunction between ZnCdS and COF. The S-scheme charge-transfer pathway was verified using fs-TAS analysis (Fig. 11g–l). Under light irradiation, pure COF showed maximum absorption at 480 nm, whereas pure ZnCdS exhibited absorption peaks at 430 and 485 nm. Compared with a single exciton, the second exciton showed a red shift for ZnCdS, representing a decrease in the energy required to introduce the second exciton. Similarly, the COF/ZnCdS composite exhibited two absorption peaks at 420 and 495 nm, showing a red shift in the second exciton aligned with ZnCdS. However, TAS analysis of COF/ZnCdS showed an additional negative absorption at 700 nm, attributed to the migration of electrons from ZnCdS to COF, aligned with the S-scheme carrier transport route. Similarly, He et al. anchored CsPbBr₃ QDs onto TPA-COF to construct an S-scheme heterojunction, leading to enhanced photocatalytic efficiency for CO₂ reduction.¹¹⁶ Microstructure analysis confirmed the uniform dispersion of CsPbBr₃ QDs onto the TPA-COF surface, leading to extended light harvesting and improved electron–hole separation, thus supporting the photocatalytic reaction. The optimized CsPbBr₃/TPA-COF exhibited maximum CH₄ and CO generation rates of 13.7 and 41.2 μmol g⁻¹ h⁻¹, respectively. fs-TAS analysis confirmed the S-scheme charge migration pathway. Compared with pure materials, the formation of an additional signal and the shortening of lifetimes (τ₁ and τ₂) in TPA-COF/CsPbBr₃ relative to pure TPA-COF and CsPbBr₃ verified the formation of new charge transport channels and the migration of charge from TPA-COF to CsPbBr₃, aligned with the S-scheme route. These results emphasize the potential of COF-based materials for overcoming the intrinsic limitations of QD photocatalysts. However, to fully exploit the capabilities of S-scheme heterojunctions, further research should focus on elucidating the interplay between the structural stability, electronic interactions, and charge dynamics under realistic conditions. Such studies will pave the way for the rational design of next-generation photocatalysts with enhanced performance and durability.

Applications

COF-based S-scheme heterojunction photocatalysts have demonstrated significant potential for various applications. Table 1 lists several applications of these advanced materials. Although their current use primarily focuses on photocatalysis, their unique properties also suggest their potential in areas such as gas storage, chemical sensing, biomedicine, and beyond. COF-based S-scheme heterojunctions are particularly notable for photocatalytic applications, such as hydrogen generation, CO₂ conversion, degradation of environmental pollutants, and organic transformation, as shown in Fig. 12.

Table 1 Applications of COF-based S-scheme heterojunction photocatalysts

Catalyst	Synthesis method	Light source/λ (nm)	Application	Performance	Ref.
COF/CdS	Self-assembly	300 W Xe/>420	H2 generation	8.67 mmol h^−1 g^−1	41
COF/ZnIn2S4	Condensation reaction	300 W Xe/>400		0.70 mmol h^−1 g^−1	42
COFs-Ph@CdS	Hydrothermal	300 W Xe		0.73 mmol h^−1 g^−1	44
TpBD COF/ZnIn2S4	Hydrothermal	300 W Xe/>420		2.30 mmol h^−1 g^−1	48
ZnIn2S4/TpPa-1COF	Hydrothermal	300 W Xe/>420		2.75 mmol h^−1 g^−1	104
WO3/TpPa-1COF	Hydrothermal	300 W Xe/>420		19.89 mmol h^−1 g^−1	86
BiOBr/COF	Solvothermal	300 W Xe/>420		7.79 mmol h^−1 g^−1	85
TiO2/COF	—	300 W Xe/>420		15.33 mmol h^−1 g^−1	51
CdS/COF	Self-assembly	300 W Xe/>420		15.1 mmol h^−1 g^−1	52
COF/g-C3N4	Chemical approach	300 W Xe/>420		1.15 mmol h^−1 g^−1	56
COF/PUP	In situ growth	350 W Xe/>420		94.5 mmol h^−1 g^−1	57
BiOBr/TpBD COF	In situ transformation	300 W Xe/>420		16.17 mmol h^−1 g^−1	62
COF/ZnxCd1−xS	Self-assembly	300 W Xe/>400		8.41 mmol h^−1 g^−1	65
MOF/TpPa-1 COF	Ultrasonic treatment	300 W Xe/320–780		4.19 mmol h^−1 g^−1	67
COF/CeO2	In situ growth	300 W Xe/>420	CO2 reduction	CO: 66.2 μmol g^−1	40
g-C3N4/COF	Self-assembly	300 W Xe/>400		CO: 11.25 μmol h^−1	46
TiO2/COF	In situ growth	300 W Xe		CO: 7.51 μmol h^−1	47
TiO2/COF	Self-assembly	300 W Xe		CO: 11.6 μmol h^−1	49
TiO2/COF	Co-precipitation	300 W Xe		CO: 50.5 μmol h^−1 g^−1	59
COF/ZnIn2S4	In situ growth	300 W Xe/>420		CO: 1117 μmol h^−1 g^−1	66
Zr-MOF/Pro-COF	In situ growth	300 W Xe		CO: 106.35 μmol g^−1	11
COF/MOF	Solvothermal	300 W Xe/>420		CO: 6.56 μmol g^−1 h^−1	81
ZnO/COF	Interfacial schiff-base reaction	300 W Xe		CO: 11.07 μmol g^−1; CH4: 4.86 μmol g^−1	91
TiO2/BTTA COF	In situ growth	300 W Xe/350–780	H2O2 production	0.74 mmol L^−1 h^−1	37
COF/In2S3	Hydrothermal	300 W Xe/>420		0.31 mmol L^−1 h^−1	84
BiOBr/COF	In situ growth	300 W Xe		3.75 mmol g^−1 h^−1	94
ZnO/COF	Self-assembly	300 W Xe		2.44 mmol g^−1 h^−1	50
TpPa-1COF/BIVO4	Solvothermal	LED/420		0.72 mmol g^−1 h^−1	55
ZnSe/COF	Self-assembly	300 W Xe		1.90 mmol g^−1 h^−1	58
MgIn2S4/COF	Wet chemistry	300 W Xe/>420		4.52 mmol g^−1h^−1	64
COF/In2S3	In situ growth	360 W Xe		5.71 mmol g^−1 h^−1	69
CdS/COF	Hydrothermal	300 W Xe/>420		0.61 mmol L^−1 h^−1	80
ZnCdS/COF	—	300 W Xe/>420		5.17.1 mmol g^−1 h^−1	97
Bi4O5Br2/COF	Hydrothermal	White LED/400–700		5.22 mmol g^−1 h^−1	89
SnS2/COF	Hydrothermal	300 W Xe/420–800		1.47 mmol g^−1 h^−1	117

---

Fig. 12 Various applications of COF-based S-scheme heterojunctions.

Hydrogen production

The environmental challenges posed by fossil fuels necessitate a shift toward renewable and eco-friendly energy sources. Among these, hydrogen is emerging as a promising energy carrier owing to its large energy density and clean-burning properties.⁴¹ Hydrogen, with an energy content of 120–142 MJ kg⁻¹, is expected to dominate global energy generation by 2080, and current annual production exceeds 44.5 million tons.⁴⁴ However, the majority of hydrogen is still derived from steam methane reforming, which accounts for approximately 95% of global production.⁴² To make hydrogen production greener and more sustainable, photocatalytic water splitting offers an efficient method for converting solar energy into hydrogen fuel.⁸⁴ Since Fujishima and Honda's breakthrough in 1972, significant advancements have been made in photocatalytic hydrogen production.⁹⁴ However, many semiconductor photocatalysts capable of water splitting exhibit wide bandgaps and are only activated under UV light, which represents a small fraction of the solar spectrum. This limitation emphasizes the need for visible-light-responsive photocatalysts that are efficient, stable, environmentally safe, and cost-effective for large-scale hydrogen production.⁵⁰ To achieve efficient water splitting, the photocatalyst must satisfy several criteria. It must absorb light effectively to generate electron–hole pairs, minimize charge recombination, and prevent photocorrosion or toxic byproduct formation during the reaction. Additionally, the bandgap energy should ideally range between 1.23 eV and 3.0 eV to balance thermodynamic feasibility and visible-light activation.⁵¹ For overall water splitting, the CB must be more negative than the H⁺/H₂ potential (0 V), while the VB must be more positive than the O₂/H₂O potential (1.23 V). These requirements ensure that the photocatalyst can drive the redox reactions necessary for splitting water into hydrogen and oxygen.⁵⁶ COFs are particularly advantageous for this application because of their high surface area and porous frameworks that provide abundant active sites and minimize particle aggregation. The integration of COFs with S-scheme heterojunctions enhances charge carrier separation while maintaining high redox capability, resulting in synergistic performance improvements.¹³¹

For example, an in situ growth process was employed to decorate positively charged ZnIn₂S₄ onto negatively charged COF, forming an S-scheme COF/ZnIn₂S₄ core–shell heterojunction for photocatalytic H₂ generation.⁴² Microstructural analysis (Fig. 13a) revealed that flower-like ZnIn₂S₄, serving as the shell, enveloped the spherical COF nanoparticles, creating a robust interface between the core and shell. This design enhanced the surface area and porosity, promoted efficient mass transport, and reduced the charge-diffusion distance. These improvements were corroborated by the EIS Nyquist spectra (Fig. 13b), which demonstrated that the heterojunction significantly reduced the interfacial charge transfer resistance and accelerated charge migration and separation. Additionally, the substantial changes in the binding energies of the C 1s, N 1s, Zn 2p, In 3d, and S 2p orbitals in the COF/ZnIn₂S₄ system compared to those of pure ZnIn₂S₄ and COF confirmed the successful formation of the heterojunction. The COF/ZnIn₂S₄ heterojunction attained an outstanding H₂ production rate of 695 μmol g⁻¹ h⁻¹, which was further enhanced to 2711 μmol g⁻¹ h⁻¹ after the introduction of Pt as a co-catalyst. Moreover, this heterojunction displayed an AQY of 2.45% at 400 nm and a 98.9% increase in lifetime, in contrast to the bare COF. Upon light irradiation, the photogenerated electrons in the LUMO of the COF perished with holes in the VB of ZnIn₂S₄, effectively retaining high-potential electrons in the CB of ZnIn₂S₄ and holes in the HOMO of the COF, as illustrated in Fig. 13c. The preserved electrons react with H⁺ to generate H₂, whereas the holes are utilized by Na₂S/Na₂SO₃. The S-scheme charge-migration mechanism was further validated by EPR analysis (Fig. 13d). Pure COF did not exhibit any DMPO-˙O₂ signals owing to its insufficient reduction potential (0.12 V vs. NHE) compared to the standard ˙O²⁻ potential (−0.33 V vs. NHE). Conversely, ZnIn₂S₄ displayed a strong DMPO-˙O₂ signal due to its adequate CB potential (−1.14 V vs. NHE). Notably, the COF/ZnIn₂S₄ system showed a significantly enhanced DMPO-˙O₂ signal, indicating the pile-up of high-potential electrons in the CB of ZnIn₂S₄ rather than their migration to the low-potential COF, which aligned well with the S-scheme charge transport route. This study underscores that in situ anchoring of the ZnIn₂S₄ shell around the COF core not only enhances charge separation but also facilitates numerous reactive sites, facilitating efficient photocatalytic H₂ generation.

Fig. 13 (a) SEM image, (b) EIS Nyquist plots, (c) charge transfer mechanism, and (d) DMPO analysis of COF/ZnIn₂S₄. Reproduced with permission from ref. 42. Copyright 2024, Elsevier. (e) SEM image, (f) photocurrent response, (g) charge transfer mechanism, ex situ/in situ XPS analysis, (h) S 2p, (i) In 3d, (j) Zn 2p, (k) C 1s, and (l) N 1s of SVs-ZnIn₂S₄/TpPa-1. Ref. 104, copyright 2024, Elsevier.

Defect formation is critical for optimizing the photocatalytic properties of COF-based materials, particularly S-scheme heterojunction systems. Defects introduce active sites, enhance reactant–catalyst interactions, and create mid-gap states that expand visible-light absorption. These structural irregularities act as charge trapping centers, reducing electron–hole recombination and complementing the spatial charge separation of S-scheme heterojunctions.¹³² Additionally, defects improve interfacial bonding within the heterojunction, boosting charge transfer efficiency.¹³³ As a self-modification strategy, defect engineering maintains the chemical stability and durability of COF-based materials, making it an effective approach to enhance photocatalytic performance without compromising stability.

For example, Wang et al. used a hydrothermal method to introduce sulfur vacancies (SVs) in ZnIn₂S₄, subsequently integrating SVs-ZnIn₂S₄ onto TpPa-1 at varying mass ratios to construct S-scheme heterojunctions for photocatalytic H₂ generation.¹⁰⁴ TEM analysis (Fig. 13e) confirmed that flower-like spheres of SVs-ZnIn₂S₄, composed of nanosheets, were uniformly distributed on block-like TpPa-1, establishing a robust interfacial interaction. This configuration enhances the surface area and increases the density of the catalytic centers. The successful creation of SVs in ZnIn₂S₄ was validated through EPR analysis, which demonstrated their role as charge-trapping sites, facilitating the separation of photoinduced electron–hole pairs. This was also supported by the highest photocurrent density recorded for the SVs-ZnIn₂S₄/TpPa-1 heterojunction compared with its pure counterparts, as shown in Fig. 13f. Moreover, the formation of SVs introduced impurity levels below the CB of ZnIn₂S₄, effectively reducing its bandgap and extending its light harvesting range. This extension facilitated the generation of additional charges for the redox processes. Consequently, the optimal SVs-ZnIn₂S₄/TpPa-1 achieved an impressive hydrogen production level of 2745 μmol g⁻¹ h⁻¹, with an AQY of 4.67% at 420 nm. Furthermore, the photocatalyst demonstrated exceptional stability and maintained its performance and structural integrity after multiple consecutive cycles. Upon light irradiation, the photogenerated electrons in the CB of TpPa-1 perished with holes in the VB of SVs-ZnIn₂S₄ (Fig. 13g), enhancing the separation of the high-potential electrons in the CB of SVs-ZnIn₂S₄, which subsequently reduced H⁺ to H₂. The S-scheme charge migration route was verified using ISIXPS analysis (Fig. 13h–l). The binding energies of the Zn 2p, In 3d, and S 2p orbitals of SVs-ZnIn₂S₄, as well as C 1s and N 1s in TpPa-1, exhibited red and blue shifts in the dark state within the heterojunction compared to those of the pure materials, demonstrating the migration of free electrons from SVs-ZnIn₂S₄ to TpPa-1. Upon light irradiation, the binding energies shifted in the reverse direction, confirming that photoinduced electrons from TpPa-1 migrated to SVs-ZnIn₂S₄, which matched well with the S-scheme charge transfer route. Thus, the formation of defects has a profound effect on the photocatalytic efficiency of S-scheme heterojunctions. In COF-based S-scheme heterojunctions, defects not only modulate the electronic properties and energy levels of the integrated component but also enhance reaction kinetics by introducing additional active sites and charge trap locations.³³

Furthermore, defect engineering extends light absorption and manipulates carrier density, supporting the separation and migration of photoinduced charges.⁴⁷ It is thus established that defect formation induces a red shift in light harvesting due to the production of intermediate levels. However, it is important to note that this extension in light absorption occurs at the cost of the redox potential of the photoinduced electron–hole pairs. A redox potential that is too low cannot overcome the potential barrier for hydrogen generation, adversely impacting the photocatalytic performances.⁴⁴ Additionally, defects can act as charge recombination centers, which increase the recombination level of photogenerated electron–hole pairs, further reducing the photocatalytic H₂ generation efficiency.⁵⁸ Therefore, future studies should focus on optimizing the defect density and distribution within the heterojunction by selecting advanced synthetic methods to mitigate the negative effects on light absorption and charge redox potential, ultimately enhancing H₂ generation performance.

Microstructure optimization is another crucial approach for improving the H₂ generation performance of COF-based S-scheme photocatalysts.¹³² Various specific COF nanostructures, such as 0D (e.g., nanoparticles and quantum dots), 1D (e.g., nanorods, nanotubes, and nanowires), 2D (e.g., nanosheets and nanoplates), and 3D (e.g., nanonetwork structures and nanoflowers), each with different physiochemical properties, have been used to construct S-scheme heterojunctions. These nanostructures enable the controllable formation and utilization of photogenerated charge carriers.^134,135 Among these, 2D COF-based S-scheme heterojunctions stand out because of their ability to increase the surface area, provide abundant active sites, shorten charge diffusion distances, and offer multiple charge transport channels. These advantages help decrease the recombination rate of photogenerated charge carriers and extend the light absorption. For example, Sun et al. adopted an in situ process to decorate WO₃ nanoplates onto the surface of flower-like TpPa-1-COF to develop an S-scheme WO₃/TpPa-1-COF heterojunction for photocatalytic H₂ production.⁸⁶ TEM analysis confirmed a strong interfacial interaction between TpPa-1-COF and WO₃, facilitating an increased surface area that improved mass transportation and provided frequent active sites to favor the photoreaction. The maximized current density and lowest EIS Nyquist arc diameter of the optimized WO₃/TpPa-1-COF heterojunction, compared to its counterparts, demonstrated that the creation of the S-scheme heterojunction effectively improved the separation and transfer of photoinduced charge pairs, leading to a higher density of charges available for redox processes. Consequently, the optimized WO₃/TpPa-1-COF heterojunction exhibited a maximum hydrogen generation efficiency of 19.89 mmol g⁻¹ h⁻¹, which was five times higher than that of the pure TpPa-1-COF. Moreover, the photocatalyst showed an AQE of 12.4% at 420 nm with negligible performance and structural loss, even after four recycling experiments, demonstrating its strong stability. Under light irradiation, the photogenerated electrons in the CB of WO₃ perished with holes in the VB of TpPa-1-COF, enhancing the separation of the high-potential electrons in the CB of TpPa-1-COF and holes in the VB of WO₃. The retained electrons in the CB of TpPa-1-COF migrated to Pt, where they reacted with H⁺ to produce hydrogen, whereas the holes in the VB of WO₃ were captured by ascorbic acid. This work revealed that controlling the microstructure can enhance the surface area and the separation and transfer of photogenerated charge carriers, thus promoting photocatalytic H₂ generation. Other examples of COF-based S-scheme heterojunctions that are considered excellent photocatalysts for H₂ generation are listed in Table 1. In comparison to binary COF-based heterojunctions, dual S-scheme heterojunctions offer enhanced charge transport and separation capabilities. However, to the best of our knowledge, no COF-based dual S-scheme heterojunctions for photocatalytic H₂ generation have been reported to date. Future studies should explore this approach, as the combined effects of the three semiconductors could significantly extend light absorption and promote charge separation, minimize electron–hole recombination, and ultimately improve photocatalytic performance.

CO₂ reduction

Photocatalytic CO₂ reduction, alongside hydrogen evolution, holds significant promise for addressing environmental and energy challenges.³⁴ Converting CO₂ into fuels such as methane and methanol not only mitigates atmospheric CO₂ levels but also offers alternative energy sources.⁴⁰ However, CO₂ reduction poses considerable difficulties due to its chemical inertness and the need for high over-potentials to drive multi-electron transfer processes.⁴⁶ This complexity makes the reaction less straightforward compared to hydrogen generation and often results in poor product selectivity. Despite these hurdles, photocatalytic CO₂ reduction under ambient conditions has emerged as a potential approach capable of lowering greenhouse gas emissions while generating beneficial compound feedstocks.⁴⁷ Since the groundbreaking research by Inoue and colleagues, numerous studies have highlighted the potential of this method for generating products such as carbon monoxide, methanol, and methane.⁴⁹ A key area of recent research has focused on the application of COF-based S-scheme heterojunctions for photocatalytic CO₂ capture and conversion. These materials, composed of porous networks formed from organic linkers, boast high surface areas and tailorable pore structures, rendering them efficient outlets for CO₂ adsorption.⁵⁹ Advancements in COF-based S-scheme heterojunction design have enhanced both the adsorption capacity and selectivity for CO₂. Such modifications are particularly important for separating CO₂ from mixed gas streams, which is a common requirement for industrial processes and power plant emissions. The ability of COFs to selectively target CO₂, even in the presence of other gases, is a critical advantage in practical applications. Additionally, the reusability of these materials further enhances their appeal as sustainable tools for carbon capture and conversion technologies.³² The implications of the above extend beyond theoretical interest, offer tangible solutions for lowering greenhouse gas emissions, and promote sustainable energy applications.

For instance, Wang et al. reported an enhanced photocatalytic CO₂ conversion efficiency utilizing an S-scheme g-C₃N₄ (NH)/Tp-Tta COF heterojunction synthesized through a self-assembly process.⁴⁶ TEM analysis revealed that Tp-Tta COF nanosheets were uniformly distributed over the surface of g-C₃N₄ (NH) nanosheets through strong interfacial interactions. This homogeneous scattering increases the surface area, providing abundant active sites, reducing charge diffusion distances, and creating multiple charge transport channels. These factors collectively facilitated mass transport and broadened light absorption, both of which are crucial for supporting the photocatalytic reaction. The optical properties were validated using absorption spectra and PL analysis (Fig. 14a and b). The heterojunction exhibited the most pronounced extension in light harvesting and the lowest PL intensity compared with its individual components. These results indicate that the strong interfacial bonding and π–π interactions between g-C₃N₄ (NH) and Tp-Tta COF effectively enhanced the light absorption and charge separation. Consequently, the optimized photocatalyst achieved a CO generation level of 11.25 μmol h⁻¹, which is 3.2 fold superior to that of pure g-C₃N₄ (NH). Moreover, the photocatalyst maintained stable CO generation over a continuous 7-hours period. The S-scheme charge-transfer route within the g-C₃N₄ (NH)/Tp-Tta COF heterojunction was confirmed by EPR analysis (Fig. 14c). Pure Tp-Tta COF exhibited no DMPO-˙O₂⁻ signal due to its CB potential being less negative than the O₂/˙O₂⁻ redox potential (−0.33 V vs. NHE). Conversely, g-C₃N₄ (NH) showed a distinct DMPO-˙O₂ signal, confirming that its photogenerated electrons had sufficient reduction potential. The g-C₃N₄ (NH)/Tp-Tta COF heterojunction displayed a significantly stronger DMPO-˙O₂⁻ signal, indicating the accumulation of high-energy photoinduced electrons in the CB of g-C₃N₄ (NH) owing to the S-scheme charge-migration mechanism. In the absence of this mechanism, electrons would migrate from the high-potential g-C₃N₄ (NH) to the lower-potential Tp-Tta COF, resulting in a diminished DMPO-˙O₂⁻ signal. Upon light excitation, the preserved electrons in the CB of g-C₃N₄ (NH) converted CO₂ to CO, whereas the holes in the VB of Tp-Tta COF oxidized TEOA. Thus, the intimate interaction and creation of the S-scheme heterojunction between g-C₃N₄ (NH) and Tp-Tta COF are key factors in the enhanced CO₂ reduction performance.

Fig. 14 (a) PL analysis, (b) absorption spectra, and (c) EPR analysis of g-C₃N₄ (NH)/Tp-Tta COF. Reproduced with permission from ref. 46, copyright 2022, Elsevier. (d) SEM image, (e) photocurrent response, (f) EIS Nyquist plots, in situ XPS spectra (g) C 1s, and (h) Zn 2p of N–ZnO/TAPT-DMTP COF. Reproduced with permission from ref. 91, copyright 2025, Elsevier. (i) TEM image, (j) DMPO-˙O²⁻ signal, (k) DMPO-˙OH signal, and (l) charge-transfer mechanism of N-COF/TiO₂. Reproduced with permission from ref. 47, copyright 2022, Elsevier.

The energy band structure design plays a critical role in controlling the charge transfer behavior and the redox capabilities of COF-based S-scheme heterojunctions.¹³⁶ Carefully tuning the band positions not only governs the directionality and efficiency of photogenerated electron–hole pair separation but also aligns the thermodynamics to favor CO₂ reduction processes.¹³⁷ Therefore, selecting semiconductors with appropriate band structures and rationally adjusting their electronic properties are essential strategies to optimize charge transfer and separation in COF-based S-scheme heterojunctions. As an example, Mu et al. employed an imine-bridged strategy to couple a D–A COF, specifically TAPT-DMTP COF, onto imine-modified ZnO (N–ZnO) with matching energy bands, forming an S-scheme N-ZnO/TAPT-DMTP COF heterojunction for CO₂ reduction.⁹¹ SEM analysis (Fig. 14d) confirmed the successful integration of flower-like N–ZnO onto the coral-like surface of TAPT-DMTP COF, resulting in a porous morphology with an increased surface area. This structure provides abundant active sites and shortens the charge diffusion distance, facilitating the separation and transfer of photogenerated electron–hole pairs. The electrochemical measurements further supported these results. The N–ZnO/TAPT-DMTP COF heterojunction displayed the highest photocurrent density and lowest EIS Nyquist arc diameter among its counterparts (Fig. 14e and f). This behavior suggests that the amine groups in N–ZnO interact strongly with DMTP, creating robust interfacial coupling and enhancing charge-transport channels for rapid charge migration. XPS analysis confirmed the formation of the N–ZnO/TAPT-DMTP COF heterojunction. A significant negative shift in the C 1s binding energy and a positive shift in the Zn 2p binding energy, compared to the pure materials, indicated electron transfer from N–ZnO to TAPT-DMTP COF. Under light irradiation, these shifts reversed (Fig. 14g and h), indicating that the migration of photoinduced electrons from the TAPT-DMTP COF to N–ZnO—aligned well with the S-scheme charge transfer route. The optimized N–ZnO/TAPT-DMTP COF photocatalyst demonstrated superior CO and CH₄ generation levels of 11.07 and 4.86 μmol g⁻¹, respectively, significantly outperforming the individual components. Additionally, the optimized N–ZnO/TAPT-DMTP COF photocatalyst exhibited insignificant efficiency and structural loss after five consecutive runs, indicating its high stability. Under illumination, photogenerated electrons in the CB of N–ZnO perished with holes in the VB of the TAPT-DMTP COF owing to the synergistic effects of BIEF, band bending, and electrostatic attraction. This process enhances the separation of high-energy electrons in the CB of the TAPT-DMTP COF and holes in the VB of N–ZnO. The retained electrons in the CB of the TAPT-DMTP COF facilitated the reduction of CO₂ into CO and CH₄, whereas the holes in the VB of N–ZnO oxidized water to produce ˙OH radicals. More research efforts are expected to deepen the understanding of the matching energy band structure to realize charge transfer dynamics through advanced techniques, quantifying the role of BIEF using DFT to guide more effective interface engineering while optimizing the morphology (e.g., nanosheets or hollow structures) to improve the surface area and active site exposure. Expanding light harvesting through plasmonic enhancements or co-catalyst incorporation could further improve performance, while exploring alternative reaction conditions and scaling synthesis methods for real-world applications would advance the practical potential of photocatalysts.

The interface quality of COF-based S-scheme heterojunctions plays a pivotal role in governing the migration and separation efficiency of photogenerated charge carriers.²⁹ Enhancing the photocatalytic performance hinges on improving the interfacial contact between integrated materials, as the interface is central to facilitating efficient charge carrier migration and separation.⁴⁴ Strengthening the interface bonding is crucial for reducing interfacial charge transfer resistance. One of the most efficient processes to improve the interface bonding strength is in situ growth for constructing S-scheme heterojunctions. For example, Wang et al. employed an in situ anchoring approach to integrate defective TiO₂ onto the surface of T-COF, forming an S-scheme heterojunction with improved interfacial bonding strength for photocatalytic CO₂ reduction.⁴⁷ TEM analysis (Fig. 14i) showed that TiO₂ nanoparticles were uniformly distributed across the surface of T-COF spheres, indicating successful integration with robust interfacial contact. The resulting porous morphology contributed to an increased surface area and larger pore size, facilitating CO₂ migration to the exposed active sites for redox reactions. EXAFS analysis confirmed that TiO₂ strongly riveted onto the T-COF network via coordination with N atoms, validating the formation of a stable heterojunction with robust interfacial interactions. The w.f. of TiO₂ (5.10 eV) exceeded that of T-COF (4.85 eV), indicating the transport of free electrons from T-COF to TiO₂ and the generation of an interfacial BIEF directed from T-COF to TiO₂. EPR study verified the S-scheme carrier transfer between T-COF and TiO₂, as shown in Fig. 14j and k. Defective TiO₂ exhibited no DMPO-˙O²⁻ signal but showed a significant DMPO-˙OH signal, suggesting that its electrons lacked sufficient reduction potential, whereas its holes possessed a strong oxidation potential. Conversely, the lack of a DMPO-˙OH signal and the presence of a pronounced DMPO-˙O²⁻ signal in the EPR spectra of T-COF highlight the effectiveness of the photoinduced electrons. In contrast, the EPR spectra of N-COF/TiO₂ exhibited strong intensities for both the DMPO-˙O²⁻ and DMPO-˙OH signals, indicating the accumulation of high-potential electrons in the conduction band of T-COF and holes in the valence band of TiO₂, which matched well with the S-scheme transfer route (Fig. 14l). The optimized N-COF/TiO₂ photocatalyst demonstrated excellent CO₂ reduction performance, achieving an AQY of 6.81% at 365 nm, which was 8- and 9-fold superior to that of T-COF and TiO₂, respectively. Furthermore, this photocatalyst exhibited robust stability over five successive rounds, with an insignificant efficiency reduction. These results indicate the significance of precise interface engineering in COF-based heterojunctions. In situ growth methods have demonstrated potential for enhancing interfacial bonding and photocatalytic efficiency, although the underlying atomic-level mechanisms require further investigation. Comparative studies with other semiconductor materials can also provide a deeper understanding of the advantages and limitations of COF-based S-scheme heterojunctions.

H₂O₂ production

Hydrogen peroxide (H₂O₂), a green reactive oxygen species (ROS) with strong oxidative stability, is ideal for antibacterial applications.¹ It is widely used in sanitation and environmental remediation. Semiconductor photocatalysis offers a sustainable, renewable, and cost-effective method for H₂O₂ synthesis, eliminating the need for expensive and hazardous precious-metal catalysts. This solar-driven process enables on-site H₂O₂ production, thereby reducing storage and transportation risks.¹³⁸ Recent advances in semiconductor photocatalysis have focused on enhancing H₂O₂ production efficiency and selectivity.¹³⁹ S-scheme heterojunctions have proven effective in addressing conventional photocatalytic limitations by combining semiconductors with different bandgaps and electronic properties. This structure promotes superior charge separation and transfer: electrons in the CB of one semiconductor migrate to another with a more negative potential, while holes move to the VB of the other semiconductor with a more positive potential.^140,141 This reduces electron–hole recombination and enhances photocatalytic performance. For H₂O₂ synthesis, this mechanism allows electrons to reduce oxygen on the photocatalyst surface, forming H₂O₂, while holes oxidize various substrates.¹⁴² Acidic conditions enhance proton availability, aiding hydroxyl radical generation, which further reacts to form H₂O₂, completing the cycle.¹⁴³ H₂O₂ photocatalytic production involves the oxygen reduction reaction (ORR) and water oxidation reaction (WOR). The ORR follows a two-electron pathway to produce superoxide anions (O²⁻), which convert to H₂O₂, or a four-electron pathway reducing oxygen directly to water.¹⁴⁴ WOR oxidizes water to generate hydroxyl radicals, contributing to H₂O₂ formation.¹⁴⁵ The photocatalyst's selectivity for the two-electron ORR, requiring a CB potential more negative than 0.68 V, is crucial for high H₂O₂ yields.¹⁴⁶ COFs offer tunable porosity, large surface area, and structural flexibility, optimizing light absorption and charge dynamics essential for redox reactions.¹⁴⁷ COFs' chemical stability and functionalizable surfaces make them suitable for diverse environmental conditions.

For example, Xu et al. employed an in situ growth approach to decorate SnS₂ nanoplates onto TaTp COF via a hydrothermal process, forming an S-scheme SnS₂/TaTp COF heterojunction to enhance photocatalytic H₂O₂ production.¹¹⁷ Microstructural analysis (Fig. 15a) revealed that the nanoplates, composed of agglomerated nanoparticles, were effectively grown on the TaTp COF surface, ensuring intimate interfacial contact. This structure resulted in an increased surface area, providing abundant active sites and efficient charge transfer routes to facilitate the photoreaction. The formation of the SnS₂/TaTp COF heterojunction not only extended the light absorption, generating additional charge carriers, but also improved the charge separation and transport owing to the excellent conductivity of SnS₂. This effectively suppressed electron–hole recombination, as evidenced by the smallest arc diameter in the EIS Nyquist plot and the highest photocurrent density observed for SnS₂/TaTp COF compared to its counterparts. The optimized heterojunction exhibited H₂O₂ production efficiencies 4.6- and 2.1-fold superior to that of bare SnS₂ and TaTp COF, respectively. Moreover, stability tests over four consecutive cycles demonstrated no loss in performance or structural integrity, confirming the high stability of the SnS₂/TaTp COF photocatalyst. To verify the S-scheme charge transfer route, in situ XPS analysis was conducted (Fig. 15b–f). Under dark conditions, the binding energies of the C 1s, N 1s, and O 1s orbitals in the SnS₂/TaTp COF heterojunction exhibited a blue shift, whereas the Sn 3d and S 2p orbitals indicated a positive shift, indicating the migration of free electrons from SnS₂ to TaTp COF. Upon light irradiation, the binding energies of C 1s, O 1s, Sn 3d, and S 2p reversed, confirming the transfer of photogenerated electrons from TaTp COF to SnS₂, in alignment with the S-scheme charge transfer route. During photocatalysis, the retained electrons in the CB of SnS₂ reduce O₂ to ˙O²⁻, which consequently reduces H⁺ to produce H₂O₂. Simultaneously, the photoinduced holes in the VB of TaTp COF oxidize H₂O, producing O₂ and H⁺. These findings emphasize the potential of engineered heterojunctions in solar-driven H₂O₂ production, while underscoring the need for further optimization in interface engineering, electronic structure modulation, and long-term stability to advance scalable photocatalytic applications.

Fig. 15 (a) SEM image and XPS analysis (b) C 1s, (c) N 1s, (d) O 1s, (e) Sn 3d, and (f) S 2p of SnS₂/TaTp COF. Reproduced with permission from ref. 117, copyright 2025, Elsevier. XPS analysis of (g) Ti 2p and (h) N 1s of TiO₂/COF. Reproduced with permission from ref. 118. Copyright 2025, Elsevier. XPS analysis (i) C 1s, (j) N 1s, (k) Ti 2p, and (l) O 1s of TiO₂/BTTA (reproduced with permission from ref. 37 copyright 2023, Elsevier).

For heterojunction photocatalysts, the semiconductor surface serves as a pivotal site for redox reactions, whereas the interface plays a critical role in controlling the migration and separation of photogenerated charge pairs. Consequently, the microstructure of the heterojunction and its interface significantly influences charge migration and the extent of redox processes, thus affecting the overall photocatalytic efficiency. To optimize the structure of S-scheme heterojunctions, controlling the morphology is essential to enhance light transmission depth and expose additional active sites, fostering better interaction between active sites and reactants, thus boosting photocatalytic efficiency. For instance, Zhao et al. assembled TiO₂ nanoparticles onto a flower-like COF to construct an S-scheme TiO₂/COF heterojunction with robust interfacial contact, resulting in enhanced photocatalytic performance for H₂O₂ production.¹¹⁸ TEM examination indicated that 5 nm TiO₂ nanoparticles were uniformly distributed across the surface of the flower-like COF, indicating successful integration. This integration leads to extended light absorption, promoting increased charge generation. Furthermore, the fs-TA decay kinetics demonstrated a reduced average carrier lifetime for TiO₂/COF compared to pure TiO₂, while exhibiting a longer lifetime than pure COF. This indicates enhanced separation and transport of photoinduced charge pairs due to the generation of an S-scheme heterojunction between TiO₂ and COF. The optimized TiO₂/COF photocatalyst achieved a maximum H₂O₂ generation level of 1326 μmol g⁻¹ h⁻¹, maintaining excellent structural integrity and performance stability over five successive recycling experiments. The superior w.f. of TiO₂ (6.73 eV) in contrast to that of COF (5.04 eV) facilitated the transfer of free electrons from COF to TiO₂, leading to the formation of an interfacial IEF. This phenomenon was corroborated by the XPS analysis. As depicted in Fig. 15g and h, the binding energies of the Ti 2p and N 1s orbitals exhibited negative and positive shifts within the TiO₂/COF heterojunction under dark conditions compared to pure TiO₂ and COF, respectively, indicating electron migration from COF to TiO₂ upon integration. Upon light excitation, the binding energies of the Ti 2p and N 1s orbitals moved in the opposite direction, signifying the transport of photoinduced electrons from the CB of TiO₂ to the HOMO of COF, which matched well with the S-scheme charge transfer route. The retained electrons in the LUMO of the COF reacted with O₂ to produce ˙O²⁻ radicals, which subsequently reacted with H⁺ to yield H₂O₂. Simultaneously, the retained holes in the VB of TiO₂ oxidize water to generate ˙OH radicals.

Optimized interfacial contact areas facilitate abundant charge transport channels, ensuring swift transfer and effective separation of photogenerated electron–hole pairs.²⁹ Heterojunctions formed through mere physical mixing of semiconductors provide limited interfacial contact, resulting in insufficient charge carrier separation and migration, thereby diminishing photocatalytic performance.⁴⁸ Conversely, the formation of core–shell heterojunctions offers extensive active surfaces, enabling effective charge carrier migration and utilization.²² The robust interfacial interaction between the core and shell not only provides ample charge diffusion and transport pathways but also enhances light absorption and utilization due to multiple light reflections within the cavity of the core–shell heterojunction. For instance, Yang et al. reported enhanced photocatalytic performance for H₂O₂ production using an S-scheme TiO₂/BTTA core–shell heterojunction.³⁷ Microstructural analysis revealed a rough surface on the TiO₂/BTTA composite; attributed to the in situ anchoring of BTTA nanoparticles onto TiO₂ nanofibers which produce a porous shell, promoting the contact of reactants to the reactive centers on the TiO₂ core. This structural modification led to an increased surface area and abundant active sites, thereby promoting mass transport and enhancing redox reaction capacity. The creation of the heterojunction was verified by Raman spectroscopy, which displayed characteristic peaks corresponding to both TiO₂ and BTTA. XPS analysis (Fig. 15i–l) demonstrated shifts in the binding energies, providing deeper insight into the charge-transfer behavior. The N 1s and C 1s orbitals in the TiO₂/BTTA heterojunction exhibited a positive shift compared to pure BTTA, whereas the Ti 2p and O 1s orbitals showed a negative shift relative to pure TiO₂ under dark conditions. These observations indicate electron migration from BTTA to TiO₂.

Under light irradiation, the binding energies shifted in the opposite direction, suggesting photoinduced electron transfer from TiO₂ to BTTA, which matched well with the S-scheme charge transfer route. The optimized TiO₂/BTTA photocatalyst achieved a maximum H₂O₂ production rate of 740 μmol L⁻¹ h⁻¹ and a furoic acid transformation efficiency of 96%. Moreover, the photocatalyst exhibited negligible performance degradation and no structural changes after four consecutive cycles, confirming its high stability. Upon light exposure, the photoexcited electrons in the CB of TiO₂ perish with holes in the HOMO of BTTA. This route enhances the separation of high-energy electrons in the LUMO of BTTA and holes in the VB of TiO₂. The retained electrons reduce O₂ to superoxide radicals, leading to H₂O₂ formation, while the holes oxidize ethanol and water to produce ˙C₂H₅OH and ˙OH radicals, which subsequently undergo dehydrogenation to form furoic acid. However, despite these promising results, future studies should explore the scalability of such heterojunctions, recyclability under real-world conditions, and the potential impact of different reaction environments on charge separation ability. In addition, a deeper insight into the interplay between the structural properties and charge dynamics could pave the way for further optimization, potentially extending the application of such heterojunctions to broader photocatalytic processes, including water splitting and CO₂ reduction.

Organic transformation

Photocatalytic organic synthesis provides a green and sustainable strategy for producing valuable chemicals by utilizing solar energy for power reactions while reducing harmful byproducts.¹⁴⁸ Traditional photocatalysts, such as metal complexes and organic dyes, often suffer from drawbacks such as instability, photodegradation, and high costs.¹⁴⁹ To overcome these issues, researchers are actively exploring novel photocatalysts that offer enhanced efficiency, long-term stability, and cost-effectiveness, particularly those free from precious metals.¹⁵⁰ COFs have attracted attention as potential candidates owing to their high surface area, robust stability, and customizable structures. Consequently, the application of COFs in photocatalytic organic synthesis has gained significant traction, leading to notable discoveries and improvements. However, a major challenge in COF-based photocatalysis is the rapid recombination of photoinduced charges, which restricts their performance.¹⁵¹ One effective strategy to address this concern is the development of S-scheme heterojunctions, which facilitate charge separation by selectively directing electrons and holes, thereby improving the photocatalytic activity. For example, Zhao et al. employed an in situ encapsulation strategy to synthesize a series of S-scheme heterojunctions incorporating W₁₈O₄₉ nanobundles and TpPa-H COF, aiming to explore their potential in the photocatalytic transformation of benzylamine into imine.¹⁵² SEM analysis (Fig. 16a) confirmed that the W₁₈O₄₉ nanobundles were effectively encapsulated within TpPa-H, forming a robust interfacial contact that enhanced mass transport and light absorption. This strong interface facilitates the availability of numerous active sites and a high density of charge carriers, thereby promoting the photocatalytic reaction. Furthermore, XPS analysis demonstrated that the binding energies of the W 4f orbital in W₁₈O₄₉ (Fig. 16c) exhibited a blue shift in the heterojunction compared with pristine W₁₈O₄₉. Conversely, the C 1s orbital in TpPa-H (Fig. 16b) displays a red shift relative to that of pure TpPa-H. These shifts provide compelling evidence of electron migration from TpPa-H to W₁₈O₄₉, driven by the formation of a well-defined heterojunction with strong interfacial interactions. Additionally, the optimized W₁₈O₄₉/TpPa-H heterojunction exhibited the smallest EIS Nyquist arc diameter and lowest PL intensity, indicating that heterojunction formation significantly reduced the interfacial charge migration resistance. This enhancement facilitated efficient charge separation, thereby promoting redox reactions.

Fig. 16 (a) SEM image, XPS spectra (b) C 1s, and (c) W 4f of pure materials and W₁₈O₄₉/TpPa-H heterojunction. Ref. 152, copyright 2024, Elsevier, (d) AFM image, surface potential distribution of the W₁₈O₄₉/TpPa-H heterojunction; (e) dark state; (f) light irradiation; (g) photoexcitation of KPFM; line scanning potential difference (h) dark state; and (i) light irradiation. Ref. 153, copyright 2023, Elsevier.

As a result, the optimized heterojunction achieved an outstanding benzylamine conversion efficiency of 99% within 240 min, which was approximately three and ten times higher than those of TpPa-H and W₁₈O₄₉, respectively. Furthermore, recycling experiments confirmed the exceptional stability of the developed heterojunction, as no noticeable variation in the benzylamine conversion rate or changes in the XRD patterns were observed, indicating the preservation of both performance and structural integrity. The w.f. of W₁₈O₄₉ (8.05 eV) was considerably larger than that of TpPa-H (5.62 eV), leading to the generation of an interfacial BIEF from TpPa-H to W₁₈O₄₉. This results in band bending owing to electron migration upon integration. Within this system, the photogenerated electrons in the CB of W₁₈O₄₉ perished with holes in TpPa-H, thereby enhancing the charge separation by accumulating electrons in TpPa-H and holes in W₁₈O₄₉. These electrons subsequently react with O₂ to generate superoxide radicals, whereas the holes react with benzylamine to form intermediate ions, which subsequently react with superoxide radicals to yield the final imine product. Future research should focus on tailoring the heterojunction interface through defect engineering, doping, or ligand functionalization to further modulate charge transfer dynamics and enhance selectivity.

The morphology of the component materials in an S-scheme heterojunction plays a crucial role in determining the interface quality and interfacial charge transport pathways.¹⁵⁴ Compared to nanoparticles or larger blocks, the fabrication of nanosheets and tiny blocks not only strengthens the interfacial interaction between the constituent materials but also provides abundant active sites and short charge transport channels. This, in turn, enhanced the charge separation efficiency and accelerated redox reactions, ultimately leading to superior photocatalytic performance.¹⁵⁵ For example, Sun et al. documented the development of an S-scheme CdS/COF heterojunction to evaluate its efficiency in hydrogen production alongside an organic transformation process.¹⁵³ TEM imaging confirmed the successful integration of TpBD onto the CdS nanosheets, forming a robust interface that extended light absorption (500–600 nm), thereby generating a large number of charges for redox processes. Furthermore, the CdS/TpBD heterojunction exhibited a higher current density and a lower EIS Nyquist arc diameter than its pure components, validating the improved separation and migration of the photogenerated charge pairs. This improvement was attributed to the presence of abundant charge-transport pathways and reduced charge-transport impedance. Consequently, the S-scheme photocatalyst achieved an excellent hydrogen production rate of 15.1 mmol g ⁻¹ h ⁻¹ while simultaneously converting ascorbic acid to 2,3-diketo-L-gulonic acid with high efficiency. To further substantiate the S-scheme charge transport route between CdS and TpBD, a KPFM analysis was conducted. AFM imaging (Fig. 16d) confirmed the anchoring of TpBD onto CdS nanosheets. As illustrated in Fig. 16e, the surface potential of point A (CdS) was superior to that of point B (TpBD) in the absence of light, generating a potential difference of 62 mV and establishing a BIEF from TpBD to CdS. Upon light irradiation, the surface potential of both materials increased because of their n-type conductivity, leading to a relatively higher surface potential difference of 95 mV between points A and B (Fig. 16f–i). This shift indicated the transport of photoinduced electrons from CdS to TpBD under the influence of BIEF (Fig. 16g). Consequently, the photoinduced holes in the VB of TpBD recombine with electrons in the CB of CdS, following the S-scheme charge migration mechanism. The remaining electrons in the CB of TpBD participated in the reduction of H⁺ to generate hydrogen, whereas the holes in the VB of CdS facilitated the oxidation of ascorbic acid into C₆H₈O₇. However, the scalability and real-world applicability of CdS/COF S-scheme heterojunctions remain unclear. Investigating its performance under natural sunlight conditions and in flow-reactor systems could bridge the gap between laboratory-scale research and industrial applications. Addressing these gaps is essential for advancing S-scheme heterojunctions beyond fundamental research for practical and sustainable photocatalytic applications.

The construction of an S-scheme heterojunction by combining organic COFs with an appropriate inorganic material featuring a suitable band structure effectively merges the inherent advantages of porous COFs with the charge-separation capabilities of S-scheme systems. This synergy enhances the charge-carrier diffusion and transport, thereby facilitating photocatalytic reactions. For example, an in situ growth technique was applied to prepare ZnIn₂S₄ nanosheets on Tp-Tta nanoplates, forming an S-scheme heterojunction to investigate their cooperative photocatalytic performance for H₂ production and organic transformation.¹⁵⁶ Microstructural analysis confirmed that ZnIn₂S₄ nanosheets were vertically aligned on Tp-Tta nanoplates with strong interfacial contact, leading to enhanced light scattering, improved absorption, and more effective transfer and separation of photoinduced charges to drive the reaction. The XPS analysis provided compelling evidence for heterojunction formation. The constituent orbitals of ZnIn₂S₄ exhibited a positive binding energy shift within the ZnIn₂S₄/Tp-Tta system compared to that of pure ZnIn₂S₄, whereas Tp-Tta showed a negative shift relative to its pristine form. This binding energy shift not only confirmed the charge transfer between the two materials but also highlighted the presence of BIEF and band bending. Under light irradiation, the reversal of these orbital shifts further authenticated the S-scheme charge-transport route. As a result, the ZnIn₂S₄/Tp-Tta photocatalyst demonstrated outstanding performance, achieving maximum H₂ production and furfural generation levels of 9.73 and 12.1 mmol g⁻¹ h⁻¹, respectively, while maintaining excellent structural stability. Mechanistically, the photogenerated electrons in the CB of ZnIn₂S₄ actively reacted with H⁺ to generate hydrogen, whereas the holes in the VB of Tp-Tta oxidized furfuryl alcohol to generate furfuryl. Future research should focus on defect engineering and interface modifications to enhance charge separation efficiency and product selectivity for broader photocatalytic applications.

Summary and future perspectives

In summary, COF-based S-scheme heterojunctions have emerged as a potential strategy to address the inherent limitations of COFs, such as insufficient light absorption, rapid charge recombination, and restricted redox potential. By integrating S-scheme heterojunctions, these materials exhibit enhanced photocatalytic properties, benefiting from improved charge-separation efficiency, extended light absorption, and superior redox capabilities. Despite rapid advancements in this field, a comprehensive review of COF-based S-scheme heterojunction photocatalysts is still lacking. This work, therefore, provides an in-depth discussion of synthesis approaches, structural configurations, and charge transfer mechanisms, particularly distinguishing S-scheme heterojunctions from other heterojunction categories, such as type-II and Z-schemes. Furthermore, various preparation strategies and recent developments in COF-based S-scheme heterojunctions are systematically summarized, followed by an exploration of their application in diverse photocatalytic processes. Although these photocatalysts have demonstrated significant potential, the field remains in its early phases, facing various constraints that need to be addressed for broader practical applications. Key aspects, such as structural stability, charge transport efficiency, and large-scale synthesis, require further research and optimization. This review focuses on facilitating a valuable understanding of the current advancements and inspires future advancements, ultimately promoting the growth of COF-based S-scheme heterojunction photocatalysts for a wider range of applications. Although noteworthy progress has been made in this promising domain, there are still a number of challenges for further exploration.

(I) To fully harness the advantages of COFs in S-scheme heterojunctions, a deeper insight into the electronic connections between different materials is essential. This includes clarifying how these interactions influence charge carrier behavior, particularly in terms of separation efficiency and transfer dynamics. Moreover, the impact of the surrounding reaction conditions on the photocatalytic efficiency, such as variations in temperature, humidity, and chemical composition, requires further exploration to optimize operational efficiency.

(II) Despite their promising properties, the synthesis of COF-based S-scheme heterojunction photocatalysts typically relies on energy-intensive methods that involve high temperature and pressure. These approaches not only increase production costs, but also pose environmental concerns.¹⁵⁷ The development of scalable and sustainable fabrication techniques that minimize energy consumption and eliminate hazardous reagents is a crucial step toward broader adoption in industrial applications. To address these challenges, researchers are actively exploring alternative synthetic pathways, including low-temperature in situ growth, green solvent approaches, and self-assembly methods.

(III) Accurate characterization is fundamental to understanding the structural and functional properties of these photocatalysts. However, techniques such as scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), and Raman spectroscopy have been used. Advancements in in situ and operando characterization techniques, such as real-time X-ray diffraction (XRD) and transmission electron microscopy (TEM), will enable the direct observation of structural and electronic changes during photocatalytic reactions. Additionally, transient absorption spectroscopy (TAS) offers insights into charge carrier dynamics, whereas in situ X-ray absorption spectroscopy (XAS) can provide information on electronic state evolution, enhancing the mechanistic understanding of these materials.

(IV) From a morphological perspective, most COF-based photocatalysts feature predominantly microporous architecture, which can limit the mass transport and active site accessibility. Engineering hierarchical porosity by integrating mesopores and macropores or designing more complex architectures, such as hollow or open-framework structures, could enhance diffusion properties and catalytic performance. Techniques such as template-assisted synthesis, defect engineering, and selective etching are viable routes for achieving these structural modifications.

(V) Another constraint is the limited range of semiconducting materials currently integrated into COF-based S-scheme heterojunctions. Most studies have focused on various oxides, sulfides, and carbonaceous materials. The inclusion of other potential photocatalysts, including metal nitrides, bismuth oxyhalides, and metal selenides, can significantly enhance the efficiency and versatility of these heterojunctions. Moreover, although COF/non-COF heterojunctions have been extensively studied, the development of COF-on-COF heterojunctions remains relatively unexplored. Such all-COF structures offer unique advantages in terms of tunability, electronic coupling, and interfacial charge transport.

(VI) A deeper exploration of the correlation between the material structure and performance is necessary to fully realize the potential of COF-based S-scheme heterojunctions in photocatalysis. In situ spectroscopic techniques, including infrared and Raman spectroscopy, can provide real-time monitoring of the structural and compositional changes during photocatalytic reactions.¹⁵⁸ Additionally, maximizing the concurrent consumption of photoinduced electrons and holes in redox reactions could enable high-value chemical transformation. Because the porous nature of COFs influences reactant diffusion, investigating the interfacial pore connectivity is critical for optimizing the mass transfer properties.¹⁵⁹ Finally, translating these materials from laboratory-scale powders to macroscale configurations such as gels, films, or membranes would greatly enhance their practical applicability in various photocatalytic processes.

(VII) The formation of heterojunctions inherently involves lattice mismatch owing to differences in crystal structures, leading to strain-induced defects, increased charge recombination, and reduced photocatalytic performance.¹⁶⁰ This issue is particularly significant in COF-based S-scheme heterojunctions, where interfacial defects hinder the charge transport. To overcome this challenge, strategies such as epitaxial growth, strain modulation, and interfacial passivation can minimize the mismatch-induced defects and enhance charge separation. van der Waals epitaxy offers a promising route by reducing the strain at the heterointerface, whereas advanced buffer layers with tailored mechanical and electronic properties can alleviate structural disparities without impeding charge dynamics. Additionally, linker engineering, molecular-level design, and π–π stacking optimization can improve interfacial compatibility and charge transport. Future research should focus on rationally designing charge-transport materials with tunable lattice parameters, integrating computational approaches to optimize heterojunction architecture, and developing scalable fabrication methods to enhance the stability and efficiency of COF-based S-scheme photocatalysts for practical solar energy conversion applications.

By advancing these key research directions, the development and application of COF-based S-scheme heterojunction photocatalysts can be accelerated significantly, opening new possibilities for their integration into sustainable energy and environmental technologies.

Data availability

Data availability is not applicable to this article as no new data were created or analyzed in this study.

Author contributions

I. A. and G. L. designed the concept of this work. I. A., Z. U. D. B., A. A.-O. and S. B. A. wrote the first manuscript. Y. Z., G. L. and S. B. A. got the fund and wrote the final manuscript. All the authors discussed the results and commented on the manuscript.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Gao Li acknowledge the financial support by the National Natural Science Foundation of China (22172167) and Open fund of State Key Laboratory of Catalysis (DICP 2024SKL-A-006). The authors extend their appreciation to the Deanship of Research and Graduate Studies at King Khalid University for funding this work through the Large Research Project under grant number RGP2/376/46.

References

1. J. Z. Hassan, A. Raza, U. Qumar and G. Li, Sustainable Mater. Technol., 2022, 33, e00478.
2. Q. Shi, Z. Qin, C. Yu, A. Waheed, H. Xu, Y. Gao, H. Abroshan and G. Li, Nano Res., 2020, 13, 939–946.
3. F. Xu and B. Weng, J. Mater. Chem. A, 2023, 11, 4473–4486.
4. Q. Li, Z.-Q. Wu, D. Li, T.-H. Liu, H.-y. Yin, X.-B. Cai, W. Zhu, Z.-L. Fan and R.-Z. Li, J. Mater. Chem. A, 2023, 11, 2957–2968.
5. M. S. Khalid, Z. Li, Y. Miao, X. Gu, Y. Zhang and G. Li, ACS Appl. Nano Mater., 2025, 8, 6201–6209.
6. G. Li, C. Zhu, X. Xi and Y. Cui, Chem. Commun., 2009, 2118–2120.
7. S. Ghosh, H. Küçükkeçeci, R. P. Paitandi, V. Weigelt, V. Dippold, S. Seki and A. Thomas, J. Mater. Chem. A, 2024, 12, 247–255.
8. Z. Zhao, W. Chen, G. Zhang and Y. Chen, J. Mater. Chem. A, 2023, 11, 26052–26062.
9. Y. Liu, X. Jiang, L. Chen, Y. Cui, Q.-Y. Li, X. Zhao, X. Han, Y.-C. Zheng and X.-J. Wang, J. Mater. Chem. A, 2023, 11, 1208–1215.
10. F. Huang, Y. Wang, X. Dong and X. Lang, J. Mater. Chem. A, 2024, 12, 7036–7046.
11. J. Wang, Z. Dai, L. Wang, D. Zhang, Y. Wang, J. Li, F. Zhou and J. Huang, J. Mater. Chem. A, 2023, 11, 2023–2030.
12. C. Yang, H. Hu, C. Qian and Y. Liao, J. Mater. Chem. A, 2023, 11, 25899–25909.
13. I. M. A. Mekhemer, M. M. Elsenety, A. M. Elewa, K. D. G. Huynh, M. M. Samy, M. G. Mohamed, D. M. Dorrah, D. C. K. Hoang, A. F. Musa, S.-W. Kuo and H.-H. Chou, J. Mater. Chem. A, 2024, 12, 10790–10798.
14. Z. Li, T. Deng, S. Ma, Z. Zhang, G. Wu, J. Wang, Q. Li, H. Xia, S.-W. Yang and X. Liu, J. Am. Chem. Soc., 2023, 145, 8364–8374.
15. H. He, R. Shen, P. Zhang, G. Liang and X. Li, J. Mater. Chem. A, 2024, 12, 227–232.
16. 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.
17. X. Yu, K. Gong, S. Tian, G. Gao, J. Xie and X.-H. Jin, J. Mater. Chem. A, 2023, 11, 5627–5635.
18. C. Liu, D.-L. Ma, P.-J. Tian, C. Jia, Q.-Y. Qi, G.-F. Jiang and X. Zhao, J. Mater. Chem. A, 2024, 12, 16063–16069.
19. T. Wang, R. Zhang, P. Zhai, M. Li, X. Liu and C. Li, J. Mater. Chem. A, 2024, 12, 1292–1299.
20. W. Dong, Y. Xiao, Z. Qin, B. Qiao and L. Li, J. Mater. Chem. A, 2023, 11, 14760–14767.
21. I. Ahmad, Y. Zhang, S. A. AlFaify, G. Li and I. M. Ashraf, J. Alloys Compd., 2024, 178309.
22. Z. Li, L. Xu, Z. Babar, A. Raza, Y. Zhang, X. Gu, Y. Miao, Z. Zhao and G. Li, Nano Res., 2024, 17, 4729–4736.
23. Q. Li, Y. Zhang, L. Xu, H. Chen, L. Zhang, M. S. Khalid, Z. Li, Z. Li and G. Li, Appl. Catal., B, 2025, 361, 124706.
24. Y. Sun, Q. Shi, X. Gu, B. Wang, B. Lumbers and G. Li, J. Colloid Interface Sci., 2024, 662, 76–86.
25. Q. Shi, X. Zhang, Z. Li, A. Raza and G. Li, ACS Appl. Mater. Interfaces, 2023, 15, 30161–30169.
26. H. Shi, Q. Shi, S. Piracha and G. Li, J. Mater. Chem. A, 2024, 12, 20270–20277.
27. Q. Shi, G. Ping, X. Wang, H. Xu, J. Li, J. Cui, H. Abroshan, H. Ding and G. Li, J. Mater. Chem. A, 2019, 7, 2253–2260.
28. H. Wang, J. Fan, J. Zou, Y. Zheng, D. Wang and J. Jiang, J. Mater. Chem. A, 2024, 12, 30429–30441.
29. A. Raza, Z. Qin, S. Ahmad, M. Ikram and G. Li, J. Environ. Chem. Eng., 2021, 9, 106569.
30. R. Xiong, X. Ke, W. Jia, Y. Xiao, B. Cheng and S. Lei, J. Mater. Chem. A, 2023, 11, 2178–2190.
31. S. Zhang, X. Gong, Q. Shi, G. Ping, H. Xu, A. Waleed and G. Li, ACS Omega, 2020, 5, 15942–15948.
32. P. Sanwal, X. Gu, Y. Zhang and G. Li, Precis. Chem., 2025 DOI:10.1021/prechem.4c00087.
33. Y. Pan, Z. Huang, D. Zheng and C. Yang, J. Colloid Interface Sci., 2023, 644, 19–28.
34. Q. Niu, Q. Chen, G. Huang, L. Li, Y. He and J. Bi, J. Colloid Interface Sci., 2023, 643, 102–114.
35. X. Zhao, A. Li, D. Yang, T.-Y. Qiu, Z. Zhao, S.-L. Wang, X. Mu and H.-Q. Tan, J. Colloid Interface Sci., 2024, 653, 67–76.
36. J. Bi, Z. Zhang, J. Tian and G. Huang, J. Colloid Interface Sci., 2024, 661, 761–771.
37. Y. Yang, J. Liu, M. Gu, B. Cheng, L. Wang and J. Yu, Appl. Catal., B, 2023, 333, 122780.
38. Y. Wen, Z. Huang, M. Zhao and L. Zhao, Appl. Surf. Sci., 2022, 596, 153485.
39. M.-L. Xu, J.-R. Li, X.-M. Wu, T. Yu, G.-Y. Qin, F.-J. Wang, L.-N. Zhang, K. Li and X. Cheng, Appl. Surf. Sci., 2022, 602, 154371.
40. S. Li, H. Yu, Y. Wang, S. Wang, L. Zhang, P. Zhu, C. Gao and J. Yu, Appl. Surf. Sci., 2023, 615, 156335.
41. F. Wang, B. Da, Y. Jin, P. Sanwal, L. Cui, S. Chen, J. Xu, B. Xue and G. Li, J. Environ. Sci., 2025, 155, 37–47.
42. Y. Chen, Y. Li, W. Chen, W. Xu, Z. Han, A. Waheed, Z. Ye, G. Li and A. Baiker, Nano Res., 2022, 15, 1366–1374.
43. A. Raza, M. Ikram, S. Guo, A. Baiker and G. Li, Adv. Sustainable Syst., 2022, 6, 2200087.
44. D. You, Z. Pan and Q. Cheng, J. Alloys Compd., 2023, 930, 167069.
45. S. He, Q. Rong, H. Niu and Y. Cai, Appl. Catal., B, 2019, 247, 49–56.
46. J. Wang, Y. Yu, J. Cui, X. Li, Y. Zhang, C. Wang, X. Yu and J. Ye, Appl. Catal., B, 2022, 301, 120814.
47. Y. Wang, Z. Hu, W. Wang, Y. Li, H. He, L. Deng, Y. Zhang, J. Huang, N. Zhao, G. Yu and Y.-N. Liu, Appl. Catal., B, 2023, 327, 122419.
48. S. Bao, Q. Tan, S. Wang, J. Guo, K. Lv, S. A. C. Carabineiro and L. Wen, Appl. Catal., B, 2023, 330, 122624.
49. X. An, J. Bian, K. Zhu, R. Liu, H. Liu and J. Qu, Chem. Eng. J., 2022, 442, 135279.
50. Y. Zhang, J. Qiu, B. Zhu, M. V. Fedin, B. Cheng, J. Yu and L. Zhang, Chem. Eng. J., 2022, 444, 136584.
51. Y.-P. Zhang, W. Han, Y. Yang, H.-Y. Zhang, Y. Wang, L. Wang, X.-J. Sun and F.-M. Zhang, Chem. Eng. J., 2022, 446, 137213.
52. G. Sun, J. Zhang, B. Cheng, H. Yu, J. Yu and J. Xu, Chem. Eng. J., 2023, 476, 146818.
53. Z. Hu, Y. Wang, L. Wang, Q. Wang, Q. Zhang, F. Cui and G. Jiang, Chem. Eng. J., 2024, 479, 147534.
54. M. Zhang, H. Cao, M. Wu, Y. Zhang, J. Zhu, J. Xu, J. Deng and T. Zhou, Chem. Eng. J., 2024, 487, 150684.
55. Z. Tong, D.-K. Ma, J. Wu, X. Hu, N. Jiang, X. Yang and T. Peng, Chem. Eng. J., 2024, 496, 154262.
56. P. Dong, A. Zhang, T. Cheng, J. Pan, J. Song, L. Zhang, R. Guan, X. Xi and J. Zhang, Chin. J. Catal., 2022, 43, 2592–2605.
57. Z. Liang, R. Shen, P. Zhang, Y. Li, N. Li and X. Li, Chin. J. Catal., 2022, 43, 2581–2591.
58. Y. Zhao, C. Yang, S. Zhang, G. Sun, B. Zhu, L. Wang and J. Zhang, Chin. J. Catal., 2024, 63, 258–269.
59. L. Wang, G. Huang, L. Zhang, R. Lian, J. Huang, H. She, C. Liu and Q. Wang, J. Energy Chem., 2022, 64, 85–92.
60. H.-X. Zhang, S.-H. Ma, H.-X. Wang, S.-C. Li, H.-Y. Shen, D.-M. Kong, F.-Y. Wang and L.-N. Zhu, J. Environ. Chem. Eng., 2024, 12, 111899.
61. Z. Huang, C. Zhang, W. Zhang and H. Liu, J. Environ. Chem. Eng., 2024, 12, 114501.
62. H. Ran, X. Liu, J. Fan, Y. Yang, L. Zhang, Q. Guo, B. Zhu and Q. Xu, J. Materiomics, 2025, 11, 100918.
63. J. Sun, H. Liu, S. Wang, Y. Zhang, C. Bie and L. Zhang, J. Materiomics, 2025, 11, 100975.
64. Y. Zhang, Y. Wang, Y. Liu, S. Zhang, Y. Zhao and J. Zhang, J. Materiomics, 2025, 11, 100985.
65. Z. Fan, X. Guo, F. Liu, Y. Li, L. Zhang and Z. Jin, Appl. Mater. Today, 2022, 29, 101637.
66. S. Yuan, Y. Feng, S. Liang, Y. Zhang, J. Zhao, Y. Wang, P. Yang and Q. Gu, Mater. Today Energy, 2025, 48, 101804.
67. H. Hu, X. Zhang, K. Zhang, Y. Ma, H. Wang, H. Li, H. Huang, X. Sun and T. Ma, Adv. Energy Mater., 2024, 14, 2303638.
68. B.-B. Luan, X. Chu, Y. Wang, X. Qiao, Y. Jiang and F.-M. Zhang, Adv. Mater., 2024, 36, 2412653.
69. J. Qiu, K. Meng, Y. Zhang, B. Cheng, J. Zhang, L. Wang and J. Yu, Adv. Mater., 2024, 36, 2400288.
70. R. Gao, R. Shen, C. Huang, K. Huang, G. Liang, P. Zhang and X. Li, Angew. Chem., Int. Ed., 2025, 64, e202414229.
71. Z. Shi, Z. Chen, Y. Zhang, X. Wang, T. Lu, Q. Wang, Z. Zhan and P. Zhang, Sep. Purif. Technol., 2022, 288, 120717.
72. Y.-P. Zhang, H.-L. Tang, H. Dong, M.-Y. Gao, C.-C. Li, X.-J. Sun, J.-Z. Wei, Y. Qu, Z.-J. Li and F.-M. Zhang, J. Mater. Chem. A, 2020, 8, 4334–4340.
73. B. Dong, Y. Wan, Q. Cheng, H. Zhou and Z. Pan, Environ. Sci.: Nano, 2022, 9, 2799–2814.
74. L. Sun, W. Wang, T. Kong, H. Jiang, H. Tang and Q. Liu, J. Mater. Chem. A, 2022, 10, 22531–22539.
75. J. Pang, W. Chen, J. Hu, J. Cheng, M. Tang, Z. Liu and R. Tan, Catal. Sci. Technol., 2023, 13, 6198–6210.
76. H. Shi, Q. Shi, X. Gu, B. Wang, B. Lumbers and G. Li, J. Colloid Interface Sci., 2024, 673, 527–536.
77. Y. Tang, Y. Wang, X. Cheng and H. Zhang, J. Mater. Chem. A, 2025, 13, 3350–3358.
78. Q. Xiong, Q. Shi, X. Gu, X. Sheng, Y. Sun, H. Shi, L. Xu and G. Li, J. Colloid Interface Sci., 2024, 654, 1365–1377.
79. P. Dong, T. Cheng, J.-l. Zhang, J. Jiang, L. Zhang, X. Xi and J. Zhang, ACS Appl. Energy Mater., 2023, 6, 1103–1115.
80. J. Z. Hassan, A. Zaheer, A. Raza and G. Li, Sustainable Mater. Technol., 2023, 36, e00609.
81. Q. Niu, S. Dong, J. Tian, G. Huang, J. Bi and L. Wu, ACS Appl. Mater. Interfaces, 2022, 14, 24299–24308.
82. Q. Niu, W. Chen, G. Pan, L. Li, Y. Yu, J. Bi, L. Wu and J. C. Yu, ACS Mater. Lett., 2024, 6, 1411–1417.
83. P. Chen, M. Yan, J. Shi, M. Feng, Q. Li, H. Chen, Y. Xie, J. Lei, M. Wakeel, B. Hu and X. Wang, Chem.–Eur. J., 2025, e202500305.
84. H. Chen, S. Gao, G. Huang, Q. Chen, Y. Gao and J. Bi, Appl. Catal., B, 2024, 343, 123545.
85. J. Zhang, F. B. Busari, Y. Zhang, S. Guo, Y. Zhao, B. Wang, Q. Zeng, Z. Zhao and G. Li, Nano Res. Energy, 2025, 4, e9120140.
86. L. Sun, L. Li, J. Fan, Q. Xu and D. Ma, J. Mater. Sci. Technol., 2022, 123, 41–48.
87. L. Qi, C. Xiao, W. Lu, H. Zhang, Y. Zhou, J. Qi, Y. Yang, Z. Zhu and J. Li, Sep. Purif. Technol., 2025, 354, 128758.
88. S. Yu, C. Li, Y. Lin, J. Zhang, Y. Liu and F. Yu, Sep. Purif. Technol., 2024, 341, 126888.
89. J.-Y. Yue, Z.-X. Pan, P. Yang and B. Tang, ACS Mater. Lett., 2024, 6, 3932–3940.
90. C. Wang, S. Shi, F. Duan, S. Lu, H. Zhu, M. Du, X. Chen and M. Chen, J. Mater. Chem. A, 2022, 10, 16524–16532.
91. M. Mu, L. Meng, S. Ma, W. Chen, X. Yin and G. Bai, J. Colloid Interface Sci., 2025, 686, 63–74.
92. R. Gao, R. Shen, C. Huang, K. Huang, G. Liang, P. Zhang and X. Li, Angew. Chem., Int. Ed., 2025, 64, e202414229.
93. L. Zhang, J. Zhang, H. Yu and J. Yu, Adv. Mater., 2022, 34, 2107668.
94. H. Zhang, J. Liu, Y. Zhang, B. Cheng, B. Zhu and L. Wang, J. Mater. Sci. Technol., 2023, 166, 241–249.
95. L. Wang, J. Mao, G. Huang, Y. Zhang, J. Huang, H. She, C. Liu, H. Liu and Q. Wang, Chem. Eng. J., 2022, 446, 137011.
96. Y. Wang, H. Li, Q. Lin, J. Zhao, X. Fang, N. Wen, Z. Zhang, Z. Ding, R. Yuan, X.-H. Huang and J. Long, ACS Catal., 2023, 13, 15493–15504.
97. X. Ma, S. Li, Y. Gao, N. Li, Y. Han, H. Pan, Y. Bian and J. Jiang, Adv. Funct. Mater., 2024, 34, 2409913.
98. Y. Xu, Z. Sun, S. Fan, X. Han, L. Li, Z. Gao and C. Wang, J. Mater. Chem. A, 2024, 12, 27180–27205.
99. G. Zhao, C. Yang, W. Meng and X. Huang, J. Mater. Chem. A, 2024, 12, 3209–3229.
100. T. Mazuelo, T. Naranjo, M. Gomez-Mendoza, A. Herrero Pizarro, L. Collado, M. Barawi, F. Gándara, M. Liras and V. A. de la Peña O'Shea, J. Mater. Chem. A, 2024, 12, 1476–1487.
101. C. Huang, T. Huang, X. L. Li, W. Zhou and M. Ding, J. Mater. Chem. A, 2024, 12, 8734–8746.
102. Y. Zhang, Z. Chen, Z. Shi, T.-t. Lu, D. Chen, Q. Wang and Z. Zhan, Sep. Purif. Technol., 2021, 275, 119216.
103. N. Arif, M. N. Zafar, M. Batool, M. Humayun, M. A. Iqbal, M. Younis, L. Li, K. Li and Y.-J. Zeng, J. Mater. Chem. C, 2024, 12, 12653–12691.
104. S.-D. Wang, L.-Y. Huang, L.-J. Xue, Q. Kang, L.-L. Wen and K.-L. Lv, Appl. Catal., B, 2024, 358, 124366.
105. H. Zhao, L. Wang, G. Liu, Y. Liu, S. Zhang, L. Wang, X. Zheng, L. Zhou, J. Gao, J. Shi and Y. Jiang, ACS Catal., 2023, 13, 6619–6629.
106. Y. Xia, K. Zhang, H. Yang, L. Shi and Q. Yi, Acta Phys.-Chim. Sin., 2024, 40, 2407012.
107. H. Hu, X. Sun, H. Li, H. Pan, Y. Ma, H. Huang and T. Ma, Small, 2025, 21, 2407117.
108. Y.-N. Gong, X. Guan and H.-L. Jiang, Coord. Chem. Rev., 2023, 475, 214889.
109. Z. Fan, Y. Cai, Z. Yang, X. Zhang, R. Shao, S. Zhong, L. Zhao, D. Liu and S. Bai, Appl. Catal., B, 2024, 351, 123979.
110. H. He, S. Chen, W. Bi, X. Wen, S. Sun, P. Zhang, R. Shen and X. Li, Sol. RRL, 2023, 7, 2300532.
111. Q. Shi, Z. Qin, S. Sharma and G. Li, Chem. Rec., 2021, 21, 879–892.
112. P. Mou, G. Wan, L. Wu, D. Liu and G. Wang, J. Mater. Chem. A, 2023, 11, 4345–4354.
113. J. Xu, W. Ma, P. He, Y. Zhou, X. Liu, Y. Chen, P. Zuo and Q. Zhuang, J. Mater. Chem. A, 2023, 11, 330–340.
114. Q. Xia, Z. Han, Z. Zhang, Z. Huang, X. Wang, J. Chang, Q. Chen and M. Chen, J. Mater. Chem. C, 2023, 11, 4552–4569.
115. W. Lu, T. Qin, B. Wang, J. Li, H. Zhu, G. Bai, Y. Li, L. Chen and X. Yan, ChemPhotoChem, 2023, 7, e202200230.
116. Y. He, P. Hu, J. Zhang, G. Liang, J. Yu and F. Xu, ACS Catal., 2024, 14, 1951–1961.
117. D. X. Xu, S. Dong, J. Lv, G. Huang, Q. Chen and J. Bi, Appl. Surf. Sci., 2025, 689, 162489.
118. Y. Zhao, Y. Zhang, H. Tan, C. Ai and J. Zhang, J. Materiomics, 2025, 11, 100970.
119. J. Zhang, H. Wang, Y. Zhang, Z. Li, D. Yang, D. Zhang, T. Tsukuda and G. Li, J. Phys. Chem. Lett., 2023, 14, 4179–4184.
120. G. Sahoo, T. G. Senthamaraikannan, H. S. Jeong, P. Bandyopadhyay, D.-H. Lim, S. K. Nayak and S. M. Jeong, J. Mater. Chem. A, 2024, 12, 15019–15035.
121. Z. Qin, D. Zhao, L. Zhao, Q. Xiao, T. Wu, J. Zhang, C. Wan and G. Li, Nanoscale Adv., 2019, 1, 2529–2536.
122. F. K. Busari, Z. Babar, A. Raza and G. Li, Sustainable Mater. Technol., 2024, 40, e00958.
123. J. Ding, Z. Peng, Z. Wang, C. Zeng, Y. Feng, M. Yang, G. Hu, J. Luo and X. Liu, J. Mater. Chem. A, 2024, 12, 28023–28031.
124. X. Li, L. Gao, M. Wang, D. Lv, P. He, Y. Xie, X. Zhan, J. Li and Z. Lin, J. Mater. Chem. A, 2023, 11, 6772–6795.
125. A. Raza, A. A. Rafi, J. Z. Hassan, A. Rafiq and G. Li, Appl. Surf. Sci. Adv., 2023, 15, 100402.
126. H.-L. Wu, M.-Y. Qi, Z.-R. Tang and Y.-J. Xu, J. Mater. Chem. A, 2023, 11, 3262–3280.
127. W. Song, G. Qi and B. Liu, J. Mater. Chem. A, 2023, 11, 12482–12498.
128. Z. Yang, H. Li, T. Xu, M. She, J. Chen, X. Jia, P. Liu, X. Liu and J. Li, J. Mater. Chem. A, 2023, 11, 2679–2689.
129. M. Yu, P. Li, R. Huang, C. Xu, S. Zhang, Y. Wang, X. Gong and X. Xing, J. Mater. Chem. B, 2023, 11, 734–754.
130. A. Tiwari, S. Walia, S. Sharma, S. Chauhan, M. Kumar, T. Gadly and J. K. Randhawa, J. Mater. Chem. B, 2023, 11, 1029–1043.
131. J. Zhang, W. Zhou, J. Zhao, L. Xu, X. Jiang, Z. Li, Y. Peng and G. Li, Small, 2025, 21, 2408324.
132. X. Ren, M. Li, L. Qiu, X. Guo, F. Tian, G. Han, W. Yang and Y. Yu, J. Mater. Chem. A, 2023, 11, 5754–5765.
133. L. Chen, G. Chen, C. Gong, Y. Zhang, Z. Xing, J. Li, G. Xu, G. Li and Y. Peng, Nat. Commun., 2024, 15, 10501.
134. X. Liu, Y. Peng, Z. Liang, L. Wang, S. Han, Z. Dou, X. Lu, J. Tong, J. Liu and J. Li, J. Mater. Chem. C, 2023, 11, 2871–2879.
135. J. Prakash, P. Kumar, N. Saxena, Z. Pu, Z. Chen, A. Tyagi, G. Zhang and S. Sun, J. Mater. Chem. A, 2023, 11, 10015–10064.
136. Z. Ali, J. Ma, M. Hong and R. Sun, J. Mater. Chem. A, 2023, 11, 3297–3314.
137. K. Zheng, J. Zhang, D. Zhao, Y. Yang, Z. Li and G. Li, Nano Res., 2019, 12, 501–507.
138. S. Guo, S. Zhang, Q. Fang, H. Abroshan, H. J. Kim, M. Haruta and G. Li, ACS Appl. Mater. Interfaces, 2018, 10, 40599–40607.
139. Z. Li, Y. Xie, J. Gao, X. Zhang, J. Zhang, Y. Liu and G. Li, J. Mater. Chem. A, 2023, 11, 26573–26579.
140. J. Shen, X. Qiu and Y. Zhu, J. Mater. Chem. A, 2023, 11, 11704–11711.
141. Q. Shi, A. Raza, L. Xu and G. Li, J. Colloid Interface Sci., 2022, 625, 750–760.
142. N. Wang, Y.-S. Zhang, D.-D. Wei, H.-M. Duan, L.-M. Mo and H.-Y. Wang, J. Mater. Chem. A, 2023, 11, 4162–4169.
143. M. Tamtaji, S. Cai, W. Wu, T. Liu, Z. Li, H.-Y. Chang, P. R. Galligan, S.-i. Iida, X. Li, F. Rehman, K. Amine, W. A. Goddard and Z. Luo, J. Mater. Chem. A, 2023, 11, 7513–7525.
144. J. Zhang, Y. Xie, Q. Jiang, S. Guo, J. Huang, L. Xu, Y. Wang and G. Li, J. Mater. Chem. A, 2022, 10, 16920–16927.
145. Z. Li, H. Wang, Y. Xie, Y. Zhang, J. Gao, Y. Xu, L. Chen, L. Xu, E. Wang and G. Li, J. Colloid Interface Sci., 2025, 681, 16–24.
146. Y. Yao, H. Wang, K. Dong, H. Li, J. Liang, R. Li, S. Sun, Z. Cai, X. He, D. Zheng, Y. Luo, S. Alfaifi, D. Ma, W. Hu and X. Sun, J. Mater. Chem. A, 2023, 11, 22154–22160.
147. W. Yu, Z. Zhu, C. Hu, S. Lin, Y. Wang, C. Wang, N. Tian, Y. Zhang and H. Huang, J. Mater. Chem. A, 2023, 11, 6384–6393.
148. H. Chen, H. S. Jena, X. Feng, K. Leus and P. Van Der Voort, Angew. Chem., Int. Ed., 2022, 61, e202204938.
149. Z. Wang, Z. Qi, S. Wang, J. Du, W. Dai, F. Lu and Q. Deng, J. Colloid Interface Sci., 2025, 683, 736–745.
150. M. Deng, J. Chakraborty, G. Wang, K. S. Rawat, L. Bourda, J. Sun, I. Nath, Y. Ji, P. Geiregat, V. Van Speybroeck and X. Feng, J. Am. Chem. Soc., 2025, 147, 10219–10230.
151. M. Lei, Y. Xue, L. Gan, B. Li, Y. Lu, Q. Yang, W. Wang, S. Wang, Q. Su and Q. Wu, J. Mol. Catal., 2025, 577, 114976.
152. X. Zhao, A. Li, D. Yang, T. Y. Qiu, Z. Zhao, S. L. Wang, X. Mu and H. Q. Tan, J. Colloid Interface Sci., 2024, 653, 67–76.
153. G. Sun, J. Zhang, B. Cheng, H. Yu, J. Yu and J. Xu, Chem. Eng. J., 2023, 476, 146818.
154. S. Ghasemi, A. Parastesh, M. Padervand, H. Ren, X. Li, A. Labidi, M. Signoretto, E. A. Dawi, T. Hamzehlouyan, E. Lichtfouse and C. Wang, Curr. Opin. Chem. Eng., 2025, 47, 101059.
155. M. Bilal, M. Q. Alfaifi, S. B. Ahmed, M. M. Abduljawad, Y. I. Alrashed, E. S. Aldurahim and Y. A. Alassmy, Mater. Sci. Semicond. Process., 2025, 186, 109051.
156. Y. Wang, H. Li, Q. Lin, J. Zhao, X. Fang, N. Wen, Z. Zhang, Z. Ding, R. Yuan, X. H. Huang and J. Long, ACS Catal., 2023, 13(23), 15493–15504.
157. D. Golberg, Y. Bando, C. C. Tang and C. Y. Zhi, Adv. Mater., 2007, 19, 2413–2432.
158. J. Yang, Z. Chen, L. Zhang and Q. Zhang, ACS Nano, 2024, 18, 21804–21835.
159. L. Yao, A. M. Pütz, H. Vignolo-González and B. V. Lotsch, J. Am. Chem. Soc., 2024, 146, 9479–9492.
160. S.-P. Qi, R.-T. Guo, Z.-X. Bi, Z.-R. Zhang, C.-F. Li and W.-G. Pan, Small, 2023, 19, 2303632.

---