Strategic design of covalent organic frameworks (COFs) for photocatalytic hydrogen generation

Abstract

Covalent organic frameworks (COFs) are an emerging class of crystalline materials that are attracting increasing attention due to their high porosity, crystallinity, and tunable properties. Consequently, the strategic design of COF-based photocatalysts for various applications, including energy and environmental remediation, has attracted considerable interest. In particular, the sustainable production of clean fuel – hydrogen (H₂) – by water splitting is a promising means to meet the global energy demand and to address the atmospheric CO₂ concentration caused by the excessive use of fossil fuels. In this regard, COFs offer potential advantages due to their modular nature, which facilitates their rational design from suitable organic building blocks to achieve optimal properties of visible light harvesting properties and easy charge transport. As a result, extensive research has been devoted to the design of photoresponsive COFs for efficient water splitting to generate hydrogen. Here, we provide a comprehensive review of recent developments in the strategic design of COF-based photocatalysts for solar fuel production via water splitting. The various organic linkers used in the construction of photocatalytic COFs and their structure–property correlations are discussed in detail. The role of bandgap engineering in tuning the hydrogen evolution efficiency of COFs is also discussed. Furthermore, the current challenges and future perspectives of COF-based solid catalysts for green and sustainable clean fuel production are presented. Indeed, this review demonstrates the importance of COF-based photocatalysts for the visible-light-driven hydrogen evolution reaction (HER) and can be beneficial for the future design of photocatalytic systems.

---

Kamal Prakash

Dr Kamal Prakash received his PhD degree in Chemistry from the Indian Institute of Technology (IIT) Roorkee, India in May 2018. Following his PhD, he worked as a Postdoctoral Fellow at the University of Houston, Texas, USA in 2018–2019. Later, he joined the Indian Institute of Technology Ropar as an Institute Postdoctoral Fellow in 2021. His current research interests focus on the design and synthesis of novel covalent organic frameworks (COFs) and their photocatalytic application in hydrogen evolution and CO2 reduction/conversion.

Bikash Mishra

Bikash Mishra is currently pursuing his doctor of philosophy at the department of chemical and biological sciences at S. N. Bose National Centre for Basic Sciences, Kolkata, India. He was awarded his Bachelor of sciences in chemistry from Sidho-Kanho-Birsha University, Purulia. He pursued his master of science in organic chemistry from Banaras Hindu University, Varanasi. Currently, his research focusses on water splitting by covalent organic frameworks (COF) for hydrogen production.

David Díaz Díaz

Dr David Díaz Díaz received his PhD in Chemistry from the University of La Laguna (ULL) (2002). Then he joined Prof. Finn's group at TSRI, USA. Since 2006, he has held positions in academia and industry (Ramón y Cajal, UAM, Spain, 2006; Dow, Switzerland, 2007; CSIC, Spain, 2009; University of Regensburg, Germany, Alexander von Humboldt Researcher (2010), Heisenberg Professor (2013), and Privatdozent (since 2018)). In 2020, he was appointed as Distinguished Researcher (ULL). His main research interest focuses on the development of new functional materials for biomedical, environmental, and energy applications.

C. M. Nagaraja

Dr C. M. Nagaraja is an associate professor at the Indian Institute of Technology Ropar. He received his PhD from the Indian Institute of Science, Bangalore in 2007. Subsequently, he carried out postdoctoral research at Brandeis University, USA and Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR), Bangalore. Prior to joining the Indian Institute of Technology Ropar in 2012, he worked as an Assistant Professor at IIT Jodhpur. His research mainly focuses on the design of framework materials for utilization of carbon dioxide and inorganic nanostructured materials for generation of solar fuels.

Pradip Pachfule

Dr Pradip Pachfule studied chemistry at Solapur University, India and graduated in 2008. He received his PhD at CSIR-National Chemical Laboratory, Pune, India under the supervision of Prof. Rahul Banerjee in 2014. Later, he worked as a JSPS postdoctoral research fellow in the laboratory of Prof. Qiang Xu at AIST, Kansai, Japan. This was followed by working in the group of Prof. Arne Thomas as an Alexander von Humboldt postdoctoral fellow and a postdoctoral research fellow at the Technische Universität Berlin, Germany (2017–2021). He is currently working as an assistant professor at S. N. Bose National Centre for Basic Sciences, Kolkata, India. His research is focused on covalent organic frameworks and their applications in photocatalytic water splitting and CO2 reduction.

---

1. Introduction

Since the beginning of industrialization, fossil fuels such as natural gases, petroleum, and coal have been the primary source of energy.¹ As a result, the consumption of fossil fuels has rapidly increased over the past few decades, resulting in excessive utilization of these non-renewables. Moreover, this process produces CO₂ in large amounts, leading to various environmental issues, such as the greenhouse effect, global warming, and climate change.² To reduce our dependence on traditional energy sources and to keep the environment safe, it is necessary to decarbonize energy sources and develop carbon-free renewable energy sources such as solar, wind, tidal, biomass, etc. However, most of these energy sources are nondependable.^3–7 In these circumstances, hydrogen (H₂) has been considered an alternative clean fuel with the potential to be an eco-friendly and cost-effective renewable energy source for future generations.⁸ Notably, hydrogen is an essential feedstock for many industrial applications, such as petroleum refining, synthesis of fertilizers, and pharmaceuticals. Hydrogen can be accumulated and transported in power plants to electrify gas turbines for electricity generation (Fig. 1).⁹

Fig. 1 The applications and sources of hydrogen. (a) The applications of hydrogen in different fields. (Source: web). (b) Various methods applied for hydrogen generation.

Notably, clean and sustainable hydrogen will be crucial to utilize as synthetic fuel for transportation to prevent greenhouse emissions from fossil fuel-based vehicles. Furthermore, large-scale hydrogen generation can be beneficial because it can be compressed, liquified, and stored for future energy requirements. In addition, green hydrogen is also an important feedstock in the fertilizer and pharmaceutical industries for synthesizing ammonia and other valuable products. Thus, hydrogen is a vital feedstock and a promising environmentally friendly alternative to fossil fuels with zero carbon emissions. However, the large-scale generation, storage, and transportation of hydrogen continues to be a significant challenge that needs to be addressed to realize a complete transition to a hydrogen economy.^10–12

Hydrogen can be produced from various resources, such as fossil fuels, biomass, water, agricultural waste, and sewer sludge.^13–15 The traditional methods of hydrogen generation from fossil fuels, such as coal and natural gases, are the most dominant, costly, and environmentally hazardous, accounting for the highest CO₂ emissions to the atmosphere. However, hydrogen generation from biomass and agricultural waste is a controlled process involving heat, steam, and oxygen without combustion, and as a result, these methods contribute to low-level CO₂.^16,17 Considering the abundance of natural sunlight and water, the cleanest way to produce hydrogen is through the water-splitting process, in which water molecules are split into hydrogen and oxygen using photocatalysts. This process can be accomplished via photocatalysis using solar energy or photoelectrocatalysis by applying electricity in the presence of solar energy. In a pioneering work, Fujishima and Honda demonstrated hydrogen generation by water splitting using a TiO₂ semiconductor as a photocatalyst for the first time.¹⁸ Motivated by this work, many metal oxide/sulfide-based semiconductor materials have been studied as photocatalysts for hydrogen generation.^19–21 Although metal oxide-based semiconductor nanomaterials such as TiO₂, SrTiO₃, ZnO, ZrO₂, and Fe₂O₃ have shown great potential due to their high chemical and photostability, most of them are active mainly in the UV region of the solar spectrum, limiting their application under visible light. However, it is worth noting that recently, Domen and coworkers immobilized SrTiO₃:Al on a silica support and utilized it for photocatalytic water splitting. This system has shown a solar-to-hydrogen (STH) conversion efficiency of 0.76% per 1 m² panel, representing the largest demonstration of solar hydrogen generation.²² In addition to metal oxides, metal nitrides and metal sulfide-based semiconductor nanomaterials have also been employed as visible light active photocatalysts for H₂ generation.^23–31

In addition, layered two-dimensional materials such as graphene oxide (GO), reduced graphene oxide (rGO), and graphitic carbon nitride (g-C₃N₄) are commonly employed in photocatalysis. Their layered structure promotes facile separation and migration of photoinduced charge carriers, resulting in enhanced photocatalytic efficiency.³² In particular, g-C₃N₄ is widely studied as a metal-free visible-light-active photocatalyst due to its favorable band gap (∼2.7 eV) and high chemical/thermal stability.³³ However, further functionalization of these carbon-based materials and their porosity enhancement are still challenging.³⁴ Furthermore, in contemporary research, porous organic polymers (POPs) have emerged as promising candidates as photocatalysts for hydrogen generation owing to their unique properties of visible light absorption, tailored pore size, and functionalities.^35–38 Hence, researchers have targeted various types of porous organic materials, such as hypercross-linked polymers (HCPs),³⁹ polymers of intrinsic microporosity (PIMs),⁴⁰ porous aromatic frameworks (PAFs),⁴¹ covalent triazine frameworks (CTFs),⁴² and covalent organic frameworks (COFs).^43,44 For instance, Yoshino and coworkers investigated the application of poly(p-phenylene) as a photocatalyst for the HER in the presence of trimethylamine (TMA) as a sacrificial electron donor (SED).⁴⁵ Similarly, Jiang's group designed a conjugated organic polymer encapsulated in a dendritic shell for photocatalytic hydrogen evolution from water.⁴⁶ Furthermore, the application of a fully conjugated three-dimensional poly(azomethine) for HER has been reported.⁴⁷ Among these materials, COFs are the most prominent class of porous organic materials with permanent porosity, good thermochemical stability and well-defined crystalline structure.^48–53 Owing to their unique properties, COFs offer potential applications in the fields of catalysis, sensing, optoelectronics, energy, drug delivery, and so on.^54–56 After the first landmark synthesis of two-dimensional COFs by Yaghi and coworkers⁴³ in 2005, many two-(2D) and three-dimensional (3D) ordered COFs with permanent porosities, large surface areas, and chemical stabilities have been synthesized through covalent bonding between organic precursors.^57–59 Additionally, COFs can be structurally predesigned and functionalized by varying their organic synthons as per requirements.⁶⁰ The chemical and thermal stability of COFs is an added advantage in inhibiting photocorrosion during photocatalysis. Thus, COFs are applied for many photocatalytic transformations, including water splitting to hydrogen and oxygen evolution, CO₂ reduction, and organic transformations.^61,62

In particular, the rational design of COFs with optimal properties suitable for visible light-promoted water splitting for hydrogen generation has gained significant interest from researchers worldwide. In this direction, intensive research efforts are being made to design COFs, which can promote good visible light absorption, rapid diffusion of charge carriers, and an enhanced lifetime of excited species. The ordered porous network structure creates a good interface with reactants, electrolytes, sacrificial donors, and cocatalysts to render efficient photocatalytic activity. In addition, the extended π-conjugation of organic building blocks can be organized in particular stacking patterns to achieve facile transport of charge carriers, rendering a longer lifetime of excited states. Owing to these unique features, COFs have been identified as highly promising materials for efficient and durable photocatalysts for light-driven water splitting to hydrogen evolution.^63,64

2. Basic principle of photocatalytic water splitting

The overall water splitting is a thermodynamically energy-demanding process with a free energy requirement of ΔG = +237.2 kJ mol⁻¹ and comprises simultaneous red-ox processes, as shown below:⁶⁵

The process offers an energy barrier of 1.23 eV that can be overcome by utilizing photocatalysts possessing an optimum band gap that lies in the range of 1.5–3.2 eV.⁶⁶ For an efficient water-splitting reaction, the conduction band (CB) of the photocatalyst should be more negative than the reduction potential of H⁺/H₂ [NHE (normal hydrogen electrode) at pH = 0], and the valence band (VB) should be more positive than the oxidation potential of O₂/H₂O (1.23 V vs. NHE at pH = 0).

The basic steps involved in the photocatalytic overall water splitting reaction are depicted in Fig. 2a. The photocatalyst absorbs light with energy higher than the bandgap to generate charge carriers, i.e., electrons and holes at the conduction band (CB) and valence band (VB), respectively. Then, the photogenerated electrons and holes migrate to the catalyst's surface and are utilized for the hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) in the presence of reduction and oxidation cocatalysts, respectively. It is important to note that the photogenerated electrons and holes can undergo recombination before being utilized for the redox process, resulting in the loss of charge carriers. Overall water splitting is practically difficult to achieve because it suffers from slow kinetics and unfavorable thermodynamics.^67–69 In addition, separating the hydrogen and oxygen generated is another major concern of the overall water-splitting process. Hence, the water-splitting reaction is generally performed as two half-reactions, wherein at one half, the reduction of water to hydrogen evolution and in another half, the oxidation of water to O₂ generation is being carried out. Consequently, during hydrogen generation, the oxidation half-reaction is generally replaced by using sacrificial electron donors (SEDs) such as triethanolamine,⁷⁰ triethylamine,⁷¹ ascorbic acid,⁷² ethylenediaminetetraacetic acid (EDTA),⁷³ and lactic acid (Fig. 2b).⁷⁴

Fig. 2 Mechanisms involved in the water splitting process. (a) Schematic representation of photocatalytic overall water splitting. (b) Photocatalytic hydrogen evolution. (c and d) Type-II and Z-scheme heterojunction systems.

Furthermore, it is important to overcome electron–hole recombination for effective hydrogen generation by water reduction, which is achieved by introducing cocatalyst and heterojunction strategies. Generally, Pt is used as a cocatalyst for COF-based catalysts, which effectively utilize photogenerated electrons to reduce water to hydrogen due to its large work functions and electron-transport properties. Furthermore, heterojunction photocatalysts formed by combining COF with other materials have shown enhanced photocatalytic hydrogen generation aided by facile separation and migration of charge carriers to the interface. Among the various heterojunctions studied, type-II heterojunctions and Z-scheme-based systems are known to be effective in achieving improved HER activity.^75,76 As shown in Fig. 2c, in type-II heterojunctions, the CB of photosystem-I (PS-I) is more negative than that of PS-II. This arrangement favors the facile transfer of a photogenerated electron from the CB of PS-I to that of PS-II. However, the holes migrate from the VB of PS-II to PS-I, resulting in effective separation and utilization of electrons for reducing water to hydrogen generation. In the case of Z-scheme heterojunctions, the photoinduced electrons in the CB of PS-II recombine with the holes in the VB of PS-I, resulting in the accumulation of photogenerated electrons on PS-I with a negative CB potential and holes on PS-II with a positive VB potential (Fig. 2d). Thus, electron–hole recombination is prevented, leading to enhanced photocatalytic activity. Overall, the cocatalyst and heterojunction strategies effectively promote efficient separation and migration of the photogenerated charge carriers for improved hydrogen generation efficiency.

3. Construction of covalent organic frameworks for hydrogen generation

COFs – exclusively covalently bonded in a thermodynamically stable ordered form – are constructed using a bottom-up approach from organic linkers through reversible reactions (Fig. 3a). The application of COFs as heterogeneous catalysts for hydrogen generation is growing steadily with time, which is evident from the increased number of articles published over the past few years (Fig. 3b). Notably, in the last few years, there has been an upsurge in research on the design of COF-based photocatalysts for hydrogen evolution, which indicates that COFs are preferred as potential photocatalytic materials for HER owing to their unique properties (Table 1).

Fig. 3 (a) Pictorial representation of the general synthesis of COFs (C₃ + C₂). (b) Year wise number of publications of COF for photocatalytic hydrogen evolution (Source: Clarivate).

Table 1 Early reports of COFs for photocatalytic water splitting

Photocatalyst	Light source	Cocatalyst	Sacrificial agent	Hydrogen evolution rate (μmol h^−1 g^−1)	AQY (%)	Reference
Early reported covalent organic framework for hydrogen evolution
TFPT-COF	Visible light	Pt	Triethanolamine	1970	2.2 (420 nm)	Chem. Sci., 2014, 5, 2789
Early modified covalent organic frameworks for hydrogen evolution
CdS-COF (inorganic)	300 W Xe lamp	Pt	Lactic acid	3678	4.2 (420 nm)	Chem.–Eur. J., 2014, 20, 15961
CN-COF (organic)	300 W Xe lamp	Pt	Triethanolamine	10 100	20.7 (425 nm)	Chem. Commun., 2019, 55, 5829
Highest hydrogen evolution activity for covalent organic framework
CYANO-CON	300 W Xe lamp	Pt	Ascorbic acid	134 200	82.6 (450 nm)	Nat. Commun., 2022, 13, 2357
Higher apparent quantum efficiency for covalent organic framework
CYANO-CON	300 W Xe lamp	Pt	Ascorbic acid	134 200	82.6 (450 nm)	Nat. Commun., 2022, 13, 2357
Photocatalytic hydrogen evolution activity without cocatalyst
CYANO-CON	300 W Xe lamp	—	Ascorbic acid	17.4	2.72 (450 nm)	Nat. Commun., 2022, 13, 2357
PhBp-CTF-Ir	Visible light	—	Triethanolamine	242	2.36 (420 nm)	ACS Appl. Energy Mater., 2022, 5, 7473
BT-TAPT COF	Visible light	—	Ascorbic acid	9.45	—	Chem. Commun., 2020, 56, 12612
Py-ClTP-BT-COF	Visible light	—	Ascorbic acid	2200	—	Angew. Chem. Int. Ed., 2020, 59, 16902
TpPa-COF-(CH3)2	300 W Xe lamp	—	Sodium ascorbate	72.10	—	ChemCatChem, 2019,10, 1002
Cu-sulphen-HDCOF-NS	Visible light	—	Triethanolamine	36 990	5.77 (420 nm)	J. Mater. Chem. A,2020, 8, 25094
Photocatalytic hydrogen evolution activity with molecular cocatalyst
N2-COF	300 W Xe lamp	Chloro(pyridine)cobaloxime(Co-I)	Triethanolamine	782	0.16 (400 nm)	J. Am. Chem. Soc., 2017, 139, 16228
15-[Ni(pymt)2]n/CTF-HC2	300 W Xe lamp	[Ni(pymt)2]	Triethanolamine	3472	—	ChemCatChem, 2021, 13, 958
TpDTz-COF	300 W Xe lamp	NiME	Triethanolamine	941	0.20 (400 nm)	J. Am. Chem. Soc., 2019, 141, 11082
TpPa-1-COF	300 W Xe lamp	Ni12P5	L-Ascorbic acid	31.6		Small, 2022, 2, 201340
Covalent organic frameworks for simultaneous hydrogen and oxygen evolution
r-CTF NSs	300 W Xe lamp	Pt	Triethanolamine	HER: 10 200	11.3 (420 nm)	Angew. Chem. Int. Ed., 2021, 60, 25381
	300 W Xe lamp	Co	AgNO3	OER: 247	5.6 (420 nm)
sp^2-COF	300 W Xe lamp	Pt	Triethanolamine	HER:1360	—	Chem, 2019, 5, 1632
	300 W Xe lamp	Co	AgNO3	OER: 22	—
CTF-HUST-A1-^tBuOK	Visible light	NiPx + Pt	—	HER: 25.4	0.8 (420 nm)	Angew. Chem. Int. Ed., 2020, 59, 6007
	Visible light	NiPx + Pt	—	OER: 12.9
Early azine-based covalent organic framework for hydrogen evolution
N3-COF	300 W Xe lamp	Pt	Triethanolamine	1703	0.44 (450 nm)	Nat. Commun., 2015, 6, 8508
Early imine-based covalent organic framework for hydrogen evolution
CN-COF	300 W Xe lamp	Pt	Triethanolamine	10 100	20.7 (425 nm)	Chem. Commun., 2019, 55, 5829
Early olefin-linked covalent organic framework for hydrogen evolution
sp^2-COF	300 W Xe lamp	Pt	Triethanolamine	1360	—	Chem, 2019, 5, 1632
sp^2-COFERDN				2120	0.4 (495 nm)
Early hydrazone-based covalent organic framework for hydrogen evolution
TFPT-COF	Visible light	Pt	Triethanolamine	1970	2.2 (420 nm)	Chem. Sci., 2014, 5, 2789
Early β-ketoenamine-based covalent organic framework for hydrogen evolution
TP-BDDA-COF	300 W Xe lamp	Pt	Triethanolamine	324	1.8 (520 nm)	J. Am. Chem. Soc., 2018, 140, 1423
Early imide-based covalent organic framework for hydrogen evolution
COF-imide	Visible light	Pt	Triethanolamine	34	—	Adv. Sci., 2020, 1, 902988

---

A snapshot of the important COFs developed over the past few years with their photocatalytic HER performance is shown in Fig. 4. In a pioneering work, Lotsch and coworkers demonstrated the application of TFPT-COF as a visible light photocatalyst for hydrogen generation from water using Pt as a proton reduction catalyst.⁷⁷ The COF was synthesized from 1,3,5-tris-(4-formyl-phenyl)triazine (TFPT) and 2,5-diethoxy terephthalohydrazide (DETH) linkers (Fig. 4). The triazine-based TFPT COF featuring mesopores of 3.8 nm in diameter and a high surface area of 1603 m² g⁻¹ showed a hydrogen evolution rate of 1970 μmol h⁻¹ g⁻¹ (see below). Since then, researchers worldwide have been making extensive research efforts on the rational design of COF materials for hydrogen generation applications.

Fig. 4 Representative examples of COFs with their hydrogen evolution performances showcasing the advancement in the hydrogen evolution activity with time.

In an important demonstration, Lotsch and coworkers have shown that the electronic and steric variations in the precursor molecules are transferred to the resulting frameworks, thus leading to progressively enhanced hydrogen generation with an increase in the number of nitrogen atoms in the frameworks.⁷⁸ Furthermore, the same group demonstrated single-site hydrogen evolution by applying azine-linked N₂-COF as a photosensitizer and chloro(pyridine)cobaloxime as a cocatalyst. This is a promising step toward designing “COF-molecular cocatalyst”-based photocatalytic systems free of noble metals (e.g., Pt).⁷⁹ In 2018, Cooper and coworkers demonstrated the design of sulfone-based COFs named S-COF (S = sulfone) and FS-COF (FS = fused sulfone). As a result of high crystallinity, strong visible light absorption, and the presence of wettable and hydrophilic mesopores in the COF matrix, S-COF and FS-COF showed hydrogen evolution with steady rates of 4.44 and 10.1 mmol h⁻¹ g⁻¹ under visible light (see below).⁸⁰ Similarly, using halogenated (TaPa-Cl₂)⁸¹ and donor–acceptor (BT-TAPT)⁸² type COFs, enhanced photocatalytic hydrogen evolution has been achieved. These results highlight the influence of the electronic and photophysical properties of the COFs on the HER performance. To study the application of fully conjugated 2D COFs, Jiang and coworkers developed sp²-carbon conjugated COFs for water splitting (Fig. 4).¹⁴⁰ Owing to its excellent chemical (water, acid, and base) stability and extended conjugation, sp²c-COF showed good HER activity with a rate of 1360 μmol h⁻¹ g⁻¹ under visible light (≥420 nm).

Recently, Thomas and coworkers identified the effect of protonation as a promising means to improve the photocatalytic hydrogen evolution performance of a 2D imine-linked COF.⁸³ The protonated TtaTfa COF showed hydrogen generation at a rate of 20.7 mmol g⁻¹ h⁻¹ under visible light in the presence of ascorbic acid as a sacrificial electron donor (see below). In a similar attempt to improve the photocatalytic hydrogen evolution performance of COFs, incorporation of a β-ketene-cyano donor–acceptor pair in a cyano-functionalized COF nanosheet (CYANO-CON) was achieved by Yang and coworkers.⁸⁴ These nanosheets have shown an apparent quantum yield (AQY) of up to 82.6% (450 nm) and very high hydrogen generation performance. Inspired by the promising performance of COFs, several photocatalytic COFs have been rationally prepared, and their optical and catalytic properties have been engineered to achieve improved hydrogen evolution. However, further advancement in the design of COFs would be desired to accomplish enhanced hydrogen evolution performance with high durability. To improve the HER activity of COFs, various strategies, such as bandgap engineering, cocatalysts, and heterojunction formation, have been employed, which are discussed in the following sections.

3.1 Favorable properties of COFs as photocatalysts

A material that acts as an efficient photocatalyst should possess the essential requirements of an optimal bandgap, improved light harvesting, high porosity, and crystallinity.⁸⁵ The crystalline nature of photocatalysts with fewer defect sites is highly desired to achieve enhanced catalytic activity by eliminating recombination of photogenerated charge carriers. The modular nature of these materials allows the strategic design of COFs with tunable pore structure and functionality by the judicious choice of organic building blocks. Thus, one can rationally construct COF-based photocatalysts incorporating multiple functionalities to achieve desired properties such as visible light absorption, facile charge transport, high chemical and photostability along with high porosity and crystallinity (Fig. 5). As a result, COFs offer potential advantages as photocatalysts for hydrogen generation from water, which is also reflected in the rapid increase in the application of COF-based photocatalysts for water splitting (Table 2).

Fig. 5 Schematic representation showing essential characteristics of an ideal photocatalyst.

Table 2 The photocatalytic hydrogen evolution properties of functionalized COFs

Name of COF	Linker	Light intensity	Conc. of cocatalyst	Sacrificial donor	AQY (%)	Hydrogen evolution rate (μmol h^−1 g^−1)	Reference
Triazine-based COFs
TFTP-COF	2,5-Diethoxy-terephthalohydrazide (DETH)	300 W Xe lamp, λ ≥ 420 nm	2.4 μL of 8 wt% Pt in H2O	1 mL TEOA	2.2	1970	Chem. Sci., 2014, 5, 2789
N0-COF	N0-aldehyde	300 W Xe lamp, λ ≥ 420 nm	2.5 μL of 8.0 wt% Pt in H2O	1 μL TEOA	0.001–0.44	23	Nat. Comm., 2015, 6, 8508
N1-COF	N1-aldehyde	300 W Xe lamp, λ ≥ 420 nm	2.5 μL of 8.0 wt% Pt in H2O	1 μL TEOA	0.001–0.44	90	Nat. Comm., 2015, 6, 8508
N2-COF	N2-aldehyde	300 W Xe lamp, λ ≥ 420 nm	2.5 μL of 8.0 wt% Pt in H2O	1 μL TEOA	0.001–0.44	438	Nat. Comm., 2015, 6, 8508
N3-COF	N3-aldehyde	300 W Xe lamp, λ ≥ 420 nm	2.5 μL of 8.0 wt% Pt in H2O	1 μL TEOA	0.001–0.44	1703	Nat. Comm., 2015, 6, 8508
BT-TAPT-COF	4,40-(Benzothiadiazole-4,7-diyl)dibenzaldehyde (BT)	300 W Xe lamp, λ ≥ 420 nm	10 μL of 8.0 wt% Pt in H2O	0.1 M ascorbic acid	0.19	949	Chem. Commun., 2020, 56, 12612
BDF-TAPT-COF	4,4′-(Benzo[1,2-b:4,5-b′]difuran-4,8-diyl)dibenzaldehyde (BDF-CHO)	300 W Xe lamp, AM 1.5 G	10 μL of 8.0 wt% Pt in H2O	0.1 M ascorbic acid	7.8	1390	Chem. Commun., 2021, 57, 4464
TA-COF	Terephthalaldehyde	300 W Xe lamp, full wavelength	3.0 wt% Pt	10 vol% TEOA	—	10	Appl. Surface Sci., 2021, 537, 148082
TFA-COF	2,3,5,6-Tetrafluoroterephthaldehyde	300 W Xe lamp, full wavelength	3.0 wt% Pt	10 vol% TEOA	—	80	Appl. Surface Sci., 2021, 537, 148082
TtaTfa_AC	1,3,5-Tris(4-formylphenyl)benzene	300 W Xe lamp, λ ≥ 420 nm	8.0 wt% Pt	Ascorbic acid	1.43	20 700	Angew. Chem. Int. Ed., 2021, 60, 19797
TpaTfa_AC	Tris(4-formylphenyl)amine (Tfa)	300 W Xe lamp, λ ≥ 420 nm	8.0 wt% Pt	Ascorbic acid	—	14 900	Angew. Chem. Int. Ed., 2021, 60, 19797
TtaTpa_AC	1,3,5-Tris(4-formylphenyl)benzene (Tpa-CHO) and 1,3,5-tris(4-aminophenyl)benzene	300 W Xe lamp, λ ≥ 420 nm	8.0 wt% Pt	Ascorbic acid	—	10 800	Angew. Chem. Int. Ed., 2021, 60, 19797
TTI-COF	2,4,6-Tris(4-aminophenyl)-1,3,5-triazine	300 W Xe lamp, λ > 420 nm	3.0 wt% Pt	0.1 M ascorbic acid	—	460	ACS Catal., 2022, 12, 10718–10726
TTV-COF	2,4,6-Tris[4-(diethyl-phosphono-methyl)phenyl]-1,3,5-triazine	300 W Xe lamp, λ > 420 nm	3.0 wt% Pt	0.1 M ascorbic acid	—	5500	ACS Catal., 2022, 12, 10718–10726
TTAN-COF	2,4,6-Tris(4-cyano-methyl-phenyl)-1,3,5-triazine	300 W Xe lamp, λ > 420 nm	3.0 wt% Pt	0.1 M ascorbic acid	—	11 940	ACS Catal., 2022, 12, 10718–10726
NKCOF-112 M	[2,2′-Bipyridine]-5,5′-dicarbaldehyde	300 W Xe lamp, λ > 420 nm	50 mM H2PtCl6	TEOA	—	2800	Adv. Sci., 2022, 2, 203832
NKCOF-113 M	5,5′-Bis-(cyanomethyl)-2,2′-bipyridine	300 W Xe lamp, λ > 420 nm	50 mM H2PtCl6	TEOA	56	13 100	Adv. Sci., 2022, 2203832
NKCOF-114 M	5,5′-Bis-(cyanomethyl)-2,2′-bipyridine	300 W Xe lamp, λ > 420 nm	50 mM H2PtCl6	TEOA	—	1160	Adv. Sci., 2022, 2, 203832
BTH-1	Benzobisthiazole	300 W Xe lamp, λ > 420 nm	8.0 wt% Pt	0.1 M ascorbic acid	1.925	10 500	Nat. Commun., 2022, 13, 100
BTH-2	Benzobisthiazole	300 W Xe lamp, λ > 420 nm	8.0 wt% Pt	0.1 M ascorbic acid	0.241	1200	Nat. Commun., 2022, 13, 100
BTH-3	Benzobisthiazole	300 W Xe lamp, λ > 420 nm	8.0 wt% Pt	0.1 M ascorbic acid	1.256	15 100	Nat. Commun., 2022, 13, 100
COF-JLU100	1,4-Phenylene-diacetonitrile	300 W Xe lamp, λ > 420 nm	12.0 wt% Pt	TEOA	5.13	107 380	Angew. Chem. Int. Ed., 2022, 61, e202208919
TThB-TZ-COF	Trithienyl-benzene	300 W Xe lamp, λ > 420 nm	8.0 wt% Pt	0.1 M ascorbic acid	0.31	1030	ACS Appl. Mater. Interfaces, 2023, 15, 16794–16800
BTTh-TZ-COF	Benzotri-thiophenes	300 W Xe lamp, λ > 420 nm	8.0 wt% Pt	0.1 M ascorbic acid	1.38	5220	ACS Appl. Mater. Interfaces, 2023, 15, 16794–16800
COF-JLU35	Benzo[1,2-b:3,4-b′:5,6-b′′]trithiophene-2,5,8-tricarbaldehyde and 1,4-phenyldiacetoni rile	300 W Xe lamp, λ ≥ 420 nm	1.0 wt% Pt	Ascorbic acid	3.21	70 800	J. Am. Chem. Soc., 2023, 145, 8364–8374
COF-JLU36	1,4-Phenylene-diamine	300 W Xe lamp, 420 < λ < 780 nm	1.0 wt% Pt	0.1 M ascorbic acid	0.37	23 600	J. Am. Chem. Soc., 2023, 145, 8364–8374
COF-OH-0	1,3,5-Tri-formylbenzene	300 W Xe lamp, λ ≥ 420 nm	1.0 wt% Pt	Ascorbic acid	—	110	J. Mater. Chem. A, 2022, 10, 24620–24627
COF-OH-1	2,4,6-Triformyl-phenol	300 W Xe lamp, λ ≥ 420 nm	1.0 wt% Pt	Ascorbic acid	—	—	J. Mater. Chem. A, 2022, 10, 24620–24627
COF-OH-2	2,4,6-Triformyl-resorcinol	300 W Xe lamp, λ ≥ 420 nm	1.0 wt% Pt	Ascorbic acid	—	2910	J. Mater. Chem. A, 2022, 10, 24620–24627
COF-OH-3	2,4,6-Triformyl-phloroglucinol	300 W Xe lamp, λ ≥ 420 nm	1.0 wt% Pt	Ascorbic acid	0.15	9890	J. Mater. Chem. A, 2022, 10, 24620–24627
Pyrene-based COFs
A-TEBPY-COF	Hydrazine	300 W Xe lamp, AM 1.5 G	6 μL of 8 wt% Pt in H2O	1 mL TEOA	—	98	Adv. Energy Mater., 2018, 8, 1703278
A-TENPY-COF	Hydrazine	300 W Xe lamp, AM 1.5 G	6 μL of 8 wt% Pt in H2O	1 mL TEOA	—	22	Adv. Energy Mater., 2018, 8, 1703278
A-TEPPY-COF	Hydrazine	300 W Xe lamp, AM 1.5 G	6 μL of 8 wt% Pt in H2O	1 mL TEOA	—	6	Adv. Energy Mater., 2018, 8, 1703278
sp^2c-COFEDRN	p-Phenylenediacetonitrile (PDAN)	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	10 vol% TEOA	—	2120	Chem, 2019, 5, 1632
sp^2c-COF	p-Phenylenediacetonitrile (PDAN)	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	10 vol% TEOA	0.46	1360	Chem, 2019, 5, 1632
Tz-COF-4	Naphthalene-2,6-diamine	300 W Xe lamp, λ ≥ 420 nm	8.0 wt% Pt	0.1 M ascorbic acid	1.3	4296	J. Am. Chem. Soc., 2020, 142, 11131
IL-COF-2	Naphthalene-2,6-diamine	300 W Xe lamp, λ ≥ 420 nm	8.0 wt% Pt	0.1 M ascorbic acid	—	1644	J. Am. Chem. Soc., 2020, 142, 11131
Py-ClTP-BT COF	4,7-Dibromo-5,6-dichlorobenzo[c][1,2,5]thiadiazole	300 W Xe lamp, λ ≥ 420 nm	5.0 wt% Pt	0.1 M ascorbic acid	8.45	8875	Angew. Chem. Int. Ed., 2020, 59, 16902
Py-FTP-BT COF	4,7-Dibromo-5,6-difluorobenzo[c][1,2,5]thiadiazole	300 W Xe lamp, λ ≥ 420 nm	5.0 wt% Pt	0.1 M ascorbic acid	—	2875	Angew. Chem. Int. Ed., 2020, 59, 16902
Py-HTP-BT COF	4,7-Dibromo-benzo[c][1,2,5]thiadiazole	300 W Xe lamp, λ ≥ 420 nm	5.0 wt% Pt	0.1 M ascorbic acid	—	1078	Angew. Chem. Int. Ed., 2020, 59, 16902
NK-COF-108	Benzothiadiazole moieties	300 W Xe lamp, λ ≥ 420 nm	52 μL of 50 mM Pt in H2O	0.1 M ascorbic acid	2.96	11 600	ACS Catal., 2021, 11, 2098
PyTz-COF	4,4′-(Thiazolo[5,4-d]thiazole-2,5-diyl)dibenzaldehyde (TzDA)	300 W Xe lamp, AM 1.5 G	3.0 wt% Pt	0.1 M ascorbic acid	—	2072.4	Angew. Chem. Int. Ed., 2021, 60, 1869
PY-DHBD-COF	1,4-Dihydroxybenzidine	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Ascorbic acid	8.4	71 160	Nat. Comm., 2022, 13, 1355
Ni-Bn-COF	Bis(1,2-diphenyl-glyoximato)-nickel(II)(Ni(dpg)2)	300 W Xe lamp, λ > 420 nm	Pt	Ascorbic acid	—	1805	Angew. Chem. Int. Ed., 2022, 61, e202204326
Ni-Py-COF	Bis(1,2-diphenyl-glyoximato)nickel(II)	300 W Xe lamp, λ ≥ 420 nm	Pt	Ascorbic acid	4.28	13 231	Angew. Chem. Int. Ed., 2022, 61, e202204326
DABT-Py-COF	4,4′,4′′,4′′′-(Benzo[c][1,2,5]thiadiazole-4,7-diylbis(9,9-dimethyl-9,10-dihydroacridine-10,2,7-triyl))tetrabenzaldehyde	300 W Xe lamp, AM 1.5 G	Pt	0.1 M ascorbic acid	—	5458	J. Mater. Chem. A, 2023, 11, 4007–4012
Co/Zn-Salen-COF	Ethylenediamine	300 W Xe lamp, λ ≥ 420 nm	CoN2O2	Ascorbic acid	—	1378	Angew. Chem. Int. Ed., 2023, 62, e202214143
TAPFy-PhI COF	Phthalimide	300 W Xe lamp, λ > 420 nm	1.0 wt% Pt	0.1 M ascorbic acid	3.34	2718	ACS Appl. Mater. Interfaces, 2023, 15, 20310–20316
Porphyrin-based COFs
H2Por-DETH-COF	2,5-Diethoxy-terephthalohydrazide (DETH)	300 W Xe lamp, λ ≥ 400 nm	2.5 μL of 8 wt% Pt in H2O	50 μL TEOA	—	80	Nat. Comm., 2021, 12, 1354
CoPor-DETH-COF	Co + 2,5-diethoxy-terephthalohydrazide (DETH)	300 W Xe lamp, λ ≥ 400 nm	2.5 μL of 8 wt% Pt in H2O	50 μL TEOA	—	25	Nat. Commun., 2021, 12, 1354
NiPor-DETH-COF	Ni + 2,5-diethoxy-terephthalohydrazide (DETH)	300 W Xe lamp, λ ≥ 400 nm	2.5 μL of 8 wt% Pt in H2O	50 μL TEOA	—	211	Nat. Commun., 2021, 12, 1354
ZnPor-DETH-COF	Zn + 2,5-diethoxy-terephthalohydrazide (DETH)	300 W Xe lamp, λ ≥ 400 nm	2.5 μL of 8 wt% Pt in H2O	50 μL TEOA	0.32	413	Nat. Commun., 2021, 12, 1354
[Mo3S13]^2−@ZnP-Pz-DHTP-COF	Pyrazine-2,5-dialdehyde and 2,5-dihydroxy-terephthalaldehyde	300 W Xe lamp, λ > 420 nm	[Mo3S13]^2−	Lactic acid	—	4700	Nat. Commun., 2023, 14, 329
[Mo3S13]^2−@ZnP-Pz-PEO-COF	Pyrazine-2,5-dialdehyde and 2,5-dihydroxy-terephthalaldehyde	300 W Xe lamp, λ > 420 nm	[Mo3S13]^2−	Lactic acid	5.7	11 000	Nat. Commun., 2023, 14, 329
[Mo3S13]^2−@ZnP-Pz-COF	Pyrazine-2,5-dialdehyde	300 W Xe lamp, λ > 420 nm	[Mo3S13]^2−	Lactic acid	—	10.1	Nat. Commun., 2023, 14, 329
TP-COF	1,1,2,2-Tetra(4-formyl-(1,1′-biphenyl)ethene	300 W Xe lamp, λ > 420 nm	5.0 wt% Pt	20 vol% TEOA	—	58.4	Applied Surface Science, 2023, 613, 155966
1,3,5-Triformylphloroglucinol-containing COFs
TP-DTP	4,4′′-Diamino-p-terphenyl	300 W Xe lamp, λ ≥ 395 nm	3.0 wt% Pt	4 mL (10% V) TEOA	—	20	J. Am. Chem. Soc., 2018, 140, 1423
TP-EDDA	4,4′-(Ethyne-1,2-diyl)dianiline (EDDA)	300 W Xe lamp, λ ≥ 395 nm	3.0 wt% Pt	4 mL (10% V) TEOA	—	30	J. Am. Chem. Soc., 2018, 140, 1423
TP-BDDA	4,4′-(buta-1,3-diyne-1,4-diyl)dianiline (BDDA)	300 W Xe lamp, λ ≥ 395 nm	3.0 wt% Pt	4 mL (10% V) TEOA	1.8	324	J. Am. Chem. Soc., 2018, 140, 1423
TP-COF	4,4′′-Diamino-p-terphenyl	300 W Xe lamp, λ ≥ 395 nm	8.0 wt% Pt	0.1 M ascorbic acid	—	1600	Nat. Chem., 2018, 10, 1180
S–COF	3,7-Diaminodibenzo-[b,d]thiophene sulfone (SA)	300 W Xe lamp, λ ≥ 395 nm	8.0 wt% Pt	0.1 M ascorbic acid	—	4440	Nat. Chem., 2018, 10, 1180
FS-COF	3,9-Diamino-benzo[1,2-b:4,5-b′]-bis[1]benzothiophene sulfone (FSA)	300 W Xe lamp, λ ≥ 395 nm	8.0 wt% Pt	0.1 M ascorbic acid	3.2	10 100	Nat. Chem., 2018, 10, 1180
TpDTz	4,4′-(Thiazolo[5,4-d]thiazole-2,5-diyl)dianiline	300 W Xe lamp, AM 1.5 G	Ni-thiolate hexameric cluster	10 vol% TEOA	0.044	941	J. Am. Chem. Soc., 2019, 141, 11082
TpDTP	4,4′′-Diamino-p-terphenyl	300 W Xe lamp, AM 1.5 G	Ni-thiolate hexameric cluster	10 vol% TEOA	—	160	J. Am. Chem. Soc., 2019, 141, 11082
Tp-PDA	p-Phenylenediamine	300 W Xe lamp, λ ≥ 420 nm	5.0 wt% Pt	Sodium ascorbate	0.76	600	Small, 2021, 17, 2101017
Tp-DBN	2,5-Diaminobenzonitrile	300 W Xe lamp, λ ≥ 420 nm	5.0 wt% Pt	Sodium ascorbate	2.12	1800	Small, 2021, 17, 2101017
TpPa-H	p-Phenylenediamine (Pa–H)	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	—	11.13	Chem. Eng. J., 2021, 419, 129984
TpPa-Cl2	2,5-Dichloro-1,4-phenylene-diamine (Pa-Cl2)	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	17	11.73	Chem. Eng. J., 2021, 419, 129984
TpPa-(CH3)2	2,5-Dimethyl-1,4-phenylenedia (Pa-(CH3)2)	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	—	4.44	Chem. Eng. J., 2021, 419, 129984
TpPa-SO3H	2,5-Diaminobenzenesulphonic acid (Pa-SO3H)	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	—	3.62	Chem. Eng. J., 2021, 419, 129984
Tp-BD	Benzidine	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	—	7.19	Chem. Eng. J., 2021, 419, 129984
Tp-DTP	4,4′′-Diamino-p-terphenyl	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	—	4.76	Chem. Eng. J., 2021, 419, 129984
Tp(BT0.05TP0.95)-COF	4,4′-(Benzo-2,1,3-thiadiazole-4,7-diyl)dianiline (BT) and 4,4′-diamino-p-terphenyl (TP)	300 W Xe lamp, λ ≥ 420 nm	5.0 wt% Pt	0.1 M ascorbic acid	2.34	9839	Polym. Chem., 2021, 12, 3250
TAB-TFP-COF	440,400-Boranetriyltris(3,5-dimethylaniline)	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Ascorbic acid	0.69	666.4	J. Mater. Chem. A, 2022, 10, 17691–17698
Tp-2C/BPy^2+-COF (19.10%)	2,2′-Bipyridine	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	0.1 M ascorbic acid	6.93	34 600	Angew. Chem. Int. Ed., 2021, 60, 9642–9649
Tta-Tp	4,4′,4′′-(1,3,5-Triazine-2,4,6-triyl) trianiline	300 W Xe lamp, λ > 420 nm	2.2 wt% Pt	0.1 M ascorbic acid	0.64	6286	ChemSusChem, 2022, 15, e202101510
Tapb-Tp	1,3,5-Tris(4-aminophenyl)-benzene	300 W Xe lamp, λ > 420 nm	2.2 wt% Pt	0.1 M ascorbic acid	0.21	1828	ChemSusChem, 2022, 15, e202101510
Ttba-Tp	4,4′,4′′-(1,3,5-Triazine-2,4,6-triyl)-tris(1,1′-biphenyl)trianiline)	300 W Xe lamp, λ > 420 nm	2.2 wt% Pt	0.1 M ascorbic acid	0.44	3386	ChemSusChem, 2022, 15, e202101510
Tta-Tb	4,4′,4′′-(1,3,5-Triazine-2,4,6-triyl) trianiline	300 W Xe lamp, λ > 420 nm	2.2 wt% Pt	0.1 M ascorbic acid	0.04	152	ChemSusChem, 2022, 15, e202101510
Tz-COF-1	2,6-Diaminobenzo [1,2-d:4,5-d]-bisthiazole	300 W Xe lamp, λ > 420 nm	3.0 wt% Pt	0.8 M ascorbic acid	—	1100	ACS Catal., 2022, 12, 9494–9502
Tz-COF-2	2,6-Diaminobenzo [1,2-d:4,5-d]-bisthiazole	300 W Xe lamp, λ > 420 nm	3.0 wt% Pt	0.8 M ascorbic acid	—	7100	ACS Catal., 2022, 12, 9494–9502
Tz-COF-3	2,6-Diaminobenzo [1,2-d:4,5-d]-bisthiazole	300 W Xe lamp, λ > 420 nm	3.0 wt% Pt	0.8 M ascorbic acid	6.9	43 200	ACS Catal., 2022, 12, 9494–9502
TpPa(Δ)-Cu(II)–COF	(S)- or (R)-1-Benzenemethanamine	300 W Xe lamp, λ ≥ 420 nm	—	L-Cysteine	0.78	14 720	Nat. Chem., 2022, 13, 5768
Pt@TpBpy-NS	2,2′-Bipyridine-5,5′-diamine	300 W Xe lamp, λ > 420 nm	—	Sodium ascorbate	—	22 666	Nat. Commun., 2023, 14, 593
Pt@TpBpy-2-NS	3,3′-Bipyridine-6,6′-diamine	300 W Xe lamp, λ > 420 nm	—	Sodium ascorbate	—	9523	Nat. Commun., 2023, 14, 593
Plm-COF2	1H-Phenanthro-[9,10-d] imidazole-5,10-diamine	300 W Xe lamp, λ ≥ 420 nm	H2PtCl6	0.1 M ascorbic acid	2.55	7417.5	ACS Appl. Energy Mater., 2023, 6, 1126–1133
Plm-COF1	1H-Phenanthro-[9,10-d] imidazole-5,10-diamine	300 W Xe lamp, λ ≥ 420 nm	H2PtCl6	0.1 M ascorbic acid	0.64	528.5	ACS Appl. Energy Mater., 2023, 6, 1126–1133
Tp-DB-H2-COF	4,4′-Diamino-biphenyl benzidine	300 W Xe lamp, λ > 400 nm	3.0 wt% Pt	Sodium ascorbate	—	600	Molecular Catalysis, 2023, 535, 112807
Tp-DB-(CH3)2-COF	3,3′-Dimethyl-benzidine	300 W Xe lamp, λ > 400 nm	3.0 wt% Pt	Sodium ascorbate	—	810	Molecular Catalysis, 2023, 535, 112807
Tp-DB-(OCH3)2-COF	3,3′-Dimethoxy benzidine	300 W Xe lamp, λ > 400 nm	3.0 wt% Pt	Sodium ascorbate	—	1230	Molecular Catalysis, 2023, 535, 112807
Tp-DB-(NO2)2–COF	3,3′-Dinitro-benzidine	300 W Xe lamp, λ > 400 nm	3.0 wt% Pt	Sodium ascorbate	—	15	Molecular Catalysis, 2023, 535, 112807
FOO-COF	2,7-Diamino-9 H-fluoren-9-one	300 W Xe lamp, λ ≥ 420 nm	Pt	Ascorbic acid	20.5	119 100	Appl. Catal., B., 2023, 330, 122581
FO-COF	2,7-Diamino-fluorene	350 W Xe lamp, λ > 400 nm	Pt	Ascorbic acid	—	12 300	Appl. Catal., B., 2023, 330, 122581
Bipyridine-based COFs
TPCBP B-COF	1,4-Dibromobutane	300 W Xe lamp, λ ≥ 420 nm	—	TEOA	79.69	1029	ACS Appl. Mater. Interfaces, 2023, 15, 18836–18844
BTT-BPy-PCOF(AC)	Benzotrithiophene-2,5,8-tricarbaldehyde	300 W Xe lamp, AM 1.5 G	3.32 wt% Pt	0.1 M ascorbic acid	—	21 200	Angew. Chem. Int. Ed., 2023, 62, e202300224
Others COFs
ODA-COF	5,5′-(2,5-Diethoxy-1,4-phenylene)-bis(2-phenyl-1,3,4 oxadiazole)	300 W Xe lamp, λ > 420 nm	9.0 wt% Pt	TEOA	0.42	2615	Angew. Chem. Int. Ed., 2022, 61, e202115655
H–COF	N 1,N4-Di((E)–benzylidene)-2,5-diethoxytere-phthalohydrazide	300 W Xe lamp, λ > 420 nm	9.0 wt% Pt	TEOA	—	609	Angew. Chem. Int. Ed., 2022, 61, e202115655
RuCOF-ETTA	Ru(bpy-oEt)3(PF6)2	300 W Xe lamp, λ > 420 nm	H2PtCl6	0.01 M ascorbic acid	6.95	6429	Angew. Chem. Int. Ed., 2022, 61, e202208791
RuCOF-TPB	Ru(bpy-oEt)3(PF6)2	300 W Xe lamp, λ > 420 nm	H2PtCl6	0.01 M ascorbic acid	—	20 308	Angew. Chem. Int. Ed., 2022, 61, e202208791
RuCOF-ETTBA	Ru(bpy-oEt)3(PF6)2	300 W Xe lamp, λ > 420 nm	H2PtCl6	0.01 M ascorbic acid	—	14 745	Angew. Chem. Int. Ed., 2022, 61, e202208791
Zn@H-TpPa	p-Phenylene-diamine	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	—	28 000	ChemCatChem, 2022, 14, e202101800
v-2d-COF-NO1	2,6-Dimethylbenzo-[1,2-b:4,5-b′]-bisoxazole	250 W Xe lamp, λ > 400 nm	8.0 wt% Pt	DIPEA	—	1970	J. Am. Chem. Soc., 2022, 144, 13953–13960
V-2d-COF-NO2	2,6-Dimethylbenzo-[1,2-b:5,4-b′]-bisoxazole	250 W Xe lamp, λ > 400 nm	8.0 wt% Pt	DIPEA	—	863	J. Am. Chem. Soc., 2022, 144, 13953–13960
v-2d-COF-NS1	2,6-Dimethyl-[1,3]thiazolo-[5,4-f][1,3]benzo-thiazole	100 mW cm^−2, λ > 420 nm	8.0 wt%	TEA	—	4400	ACS Catal., 2023, 13, 1089–1096
BTT-PDA	Benzene-1,4-diamine	150 W Xe lamp, λ > 420 nm	3.0 wt% Pt	0.1 M ascorbic acid	—	2040	Angew. Chem. Int. Ed., 2023, 62, e202217416
BTT-NDA	Naphthalene-2,6-diamine	150 W Xe lamp, λ > 420 nm	3.0 wt% Pt	0.1 M ascorbic acid	5.42	5220	Angew. Chem. Int. Ed., 2023, 62, e202217416
BTT-AnthDA	Anthracene-2,6-diamine	150 W Xe lamp, λ > 420 nm	3.0 wt% Pt	0.1 M ascorbic acid	—	4230	Angew. Chem. Int. Ed., 2023, 62, e202217416
BTT-BPhDA	[1,1′-Biphenyl]-4,4′-diamine	150 W Xe lamp, λ > 420 nm	3.0 wt% Pt	0.1 M ascorbic acid	—	3270	Angew. Chem. Int. Ed., 2023, 62, e202217416

---

3.2 Choice of synthetic methods for COF preparation

Several synthetic methods have been utilized for preparing crystalline COFs in high-phase purity.⁸⁶ The solvothermal route is the most commonly employed synthetic technique to prepare crystalline COFs. It involves a condensation reaction between the organic building blocks in a vacuum-sealed tube at a specific temperature under solvent-generated pressure. The first landmark synthesis of COF-1 and COF-5 was achieved by Yaghi and coworkers by self-condensation of 1,4-benzene diboronic acid (BDBA) and its cocondensation with 2,3,6,7,10,11-hexahydroxytriphenylene (HHTP), respectively, under solvothermal conditions.⁴³ Herein, the solvothermal reaction carried out in a sealed tube promotes a condensation reaction between the precursors with controlled dehydration and diffusion of the molecules into the solution, facilitating the nucleation of crystalline material. The use of a closed reaction system sustains the availability of water for maintaining reversible conditions conducive to the growth of COF crystallites. Although the solvothermal method yields highly crystalline and porous COFs, it can be performed only at small scales with harsh reaction conditions for longer reaction times. Furthermore, many new approaches, such as microwave, mechanochemical, and room temperature solution methods, have been developed for COF synthesis.⁸⁷

In particular, the microwave-assisted solvothermal route was found to be beneficial for the rapid synthesis of porous COFs in good yield and crystallinity. In this regard, Wang and coworkers demonstrated microwave-assisted solvothermal synthesis of crystalline TpPa-COF from 1,3,5-triformylphloroglucinol (Tp) and p-phenylenediamine (Pa) precursors in a mesitylene/1,4-dioxane/acetic acid (3 : 3 : 1) solvent mixture at 100 °C within 60 min.⁸⁸ The ionothermal synthetic route has been employed to rapidly prepare 3D COFs by utilizing ionic liquid (IL) as a green solvent under ambient temperature and pressure conditions.⁸⁹ In this direction, Fang and coworkers reported the synthesis of 3D-IL-COFs using 1-butyl-3-methylimidazolium bis((trifluoromethyl)sulfonyl)imide ionic liquid, which acts as solvent as well as catalyst.⁸⁹ The Schiff base reaction of tetrakis(4-formylphenyl)methane (TFPM) with C₂ symmetric linear linkers with increasing chain length, p-phenylenediamine (6.1 Å), 4,4′-diaminobiphenyl (10.7 Å) and 4,4′′-diamino-p-terphenyl (15.3 Å) afforded 3D interpenetrated COFs named 3D-IL-COF-1–3 at room temperature within 12 h. The microwave-assisted and ionothermal methods offer an eco-friendly route for rapid and scalable synthesis of COFs over the traditional solvothermal process, which takes several days. However, the general applicability of these methods is lower than that of solvothermal methods.

In addition, room temperature synthesis of an imine-based COF (RT-COF-1) was achieved by a Schiff base reaction between 1,3,5-tris(4-aminophenyl)benzene (TAPB) and 1,3,5-benzenetricarbaldehyde (BTCA) using acetic acid along with m-cresol or DMSO.⁹⁰ Furthermore, the fabrication of RT-COF on various surfaces to obtain submicron patterns of COF by using ink-jet printing and wet lithography techniques has also been demonstrated. In another instance, Dichtel and coworkers reported rapid low-temperature synthesis of imine-linked COFs named TAPB-PDA by Sc(OTf)₂-catalyzed reaction of 1,3,5-tris(4-aminophenyl)benzene (TAPB) with terephtalaldehyde (PDA) in 1,4-dioxane/mesitylene (4 : 1, v/v) at RT within 30 min.⁹¹ This method was also extended to synthesize several heteroatom-containing COFs, demonstrating the general applicability of the synthetic process. To develop a rapid, eco-friendly, and solvent-free synthetic route, Banerjee and coworkers employed a mechanochemical route for synthesizing chemically and thermally stable COFs (TpPa-1/2) by manually grinding the precursors using a mortar and pestle within 45 min.⁹² This method yielded COFs with moderate crystallinity, marking a promising route for synthesizing COFs under environmentally friendly mild conditions. Banerjee's group further extended this synthetic strategy for the rapid synthesis of highly crystalline and ultra-porous COFs in seconds by utilizing a novel salt-mediated (p-toluenesulfonic acid) crystallization approach.⁹³ In addition, continuous synthesis of COFs was achieved using a twin-screw extruder, and the resulting COFs were processed into different shapes mimicking the ancient terracotta process. Overall, from the aforementioned discussions, it is evident that one can achieve a rapid synthesis of COFs under eco-friendly, mild, solvent-free conditions. These promising reports pave the way for scaling COF synthesis for large-scale production.

However, most of these rapid synthesis methods yield COFs with reduced crystallinity and surface area, leading to lower photocatalytic performance. To summarize, COFs synthesized by the solvothermal route possess high crystallinity and are reported to show enhanced hydrogen evolution performance over their amorphous counterparts (see below).⁷⁷

3.3 Role of organic moieties as building units

As discussed in the previous sections, COFs are formed by strong covalent bonding between the organic building blocks comprising light elements (B, C, H, N, O, S, etc.) through a reversible condensation reaction. The self-healing of COFs and their thermodynamically controlled covalent bonding results in long-range-ordered crystalline structures.^54,75,94 Thus, the network structure and properties of the resulting COFs will depend on the type of linkers used. The various organic linkers utilized in constructing photocatalytic COFs for hydrogen generation are shown in Fig. 6. Most COFs are prepared by the Schiff-base condensation reaction between aromatic amine and aldehyde ligands, both simple and substituted. Most of the linkers employed for COF synthesis have aromatic conjugation. The extent of aromaticity increases from simple phenyl linkers to di- and poly-phenyl-based linkers, further enhancing the visible light absorption properties of the material. In addition, it has been well established in the literature that an increase in the linker's aromatic conjugation results in enhanced COF visible-light photocatalytic activity.⁹⁵ Thus, highly π-conjugated systems act as strong electron donors and possess a narrow bandgap that promotes the effective separation and migration of photogenerated electrons and holes to the surface under visible light irradiation.⁵⁶ Consequently, COFs designed from linkers with extended conjugation have shown efficient hydrogen evolution over their lower conjugated counterparts.⁹⁶

Fig. 6 Representative linkers employed for the design of photocatalytic COFs and CTFs for water splitting.

It has been well demonstrated that COFs constructed from a judicious choice of organic linkers consisting of electron-donor and electron-acceptor moieties (donor–acceptor COFs) are beneficial in promoting facile transport of photogenerated electrons with longer lifetimes.⁹⁷ Consequently, the donor–acceptor COFs formed by amine-based donor-type linkers such as pyrene, porphyrins, triphenyl, etc., and aldehyde-based acceptor-type ligands having heteroatoms or electron-withdrawing substituents (–Cl, –F, –CN) are reported to exhibit enhanced HER activity (Fig. 6).⁹⁸ Herein, by virtue of donor–acceptor combination, the migration of photogenerated electrons from the donor to the acceptor ligand is favored by a push–pull effect.⁶⁰

3.4 Linkage chemistry for stable COFs

The chemical stability and durability of the synthesized COFs depend on the nature of the covalent linkage formed between the organic precursors (linkers and knots). It also affects the aromaticity and photostability of the COFs and their photocatalytic performance.^99,100 Notably, several covalent linkages, such as boroxine, boronic ester, imine, azine, hydrazine, β-ketoenamine, amide, olefin, dioxin, and ester, are employed for synthesizing COFs (Fig. 7). Among them, boroxine and boronic ester-linked COFs are reported to exhibit lower chemical stability, offering limited applications for photocatalytic water splitting.¹⁰¹ Hence, these COFs have not been utilized for photocatalytic water splitting applications because they are susceptible to water attack and can easily breakdown into individual precursor molecules. Owing to their excellent chemical stability and extended conjugation, imine-, azine-, hydrazone-, β-ketoenamine-, and olefin-linked COFs are preferred for photocatalytic HER applications.³⁶ The various types of linkages employed for the synthesis of COFs and their porosity, chemical stability, photocatalytic performance, and durability are shown in Fig. 7. In addition, covalent triazine frameworks (CTFs) have also been applied for efficient and durable water-splitting performance (see below).^42,102

Fig. 7 Various linkages, their properties and representative examples of the COFs employed for the synthesis of photocatalytically active COFs.

3.5 Functionalization of COFs for enhanced photocatalytic hydrogen generation

To enhance photocatalytic HER performance, pristine COFs are often functionalized by immobilizing catalytically active metal ions, complexes, or semiconductor nanomaterials (Fig. 8). In this context, COFs functionalized with -azo, -azine, and bipyridine are suitable for anchoring catalytically active species. For instance, bipyridine-functionalized COFs are incorporated with reactive metal ions by postsynthetic modification (PSM) to obtain hybrid COFs with enhanced hydrogen generation activity.^79,103,104 A similar observation of enhanced photocatalytic water oxidation activity has been reported for a cobalt-embedded bipyridine-based COF under visible light irradiation.¹⁰⁵ In another example, a β-ketoenamine-linked COF composed of a 2,2′-bipyridine linker was quaternized with dibromo alkanes into cyclic diquats, and its photocatalytic activity was studied. The resultant COF showed outstanding hydrogen generation activity (rate of 34 600 μmol h⁻¹ g⁻¹) compared to the parent COF.¹⁰⁶ In addition, bipyridine-functionalized COFs have been utilized for covalent anchoring of noble metal (Ru/Rh) complexes for photocatalytic applications. In addition, anchoring of a cadmium sulfide (CdS) semiconductor photocatalyst in TpPa COF to obtain a CdS/TpPa hybrid with enhanced hydrogen evolution activity over the parent COF and CdS has been demonstrated.¹⁰⁷ The crystalline, rigid network of COFs supports the formation of highly exposed catalytic sites in the large 1D channels – facilitating facile diffusion of the reactants and products.

Fig. 8 Various strategies employed for the design of COF-based hybrid photocatalysts for photocatalytic hydrogen evolution.

On the other hand, the integration of COFs with carbon-based 2D materials such as graphene, graphene oxide (GO), and graphitic carbon nitride (g-C₃N₄) has been found to be beneficial in enhancing the photocatalytic performance of COFs (Table 3)^108,109 The layered structure of both COFs and carbon materials facilitates facile charge transport by suppressing the charge recombination associated with individual materials, leading to enhanced hydrogen generation efficiency. In this regard, an imine-linked COF formed from 4,4′,4′′-(1,3,5-triazine-2,4,6-triyl)trianiline (TTA) and 1,3,5-triformylphloroglucinol (TP) was functionalized in situ with g-C₃N₄ (CN) to obtain a CN-COF hybrid. The hybrid obtained showed excellent photocatalytic hydrogen generation activity with a steady rate of 10.1 mmol g⁻¹ h⁻¹ under visible light irradiation (425 nm).¹¹⁰ The enhanced HER performance of the hybrid has been attributed to broader absorption of visible light and facile separation/migration of photogenerated electron–hole pairs aided by the heterojunction formation between COF and g-C₃N₄ with optimal band edge positions.

Table 3 The hydrogen evolution properties of COF-hybrid materials

Name of COF-hybrid photocatalyst	Name of COF	Dopant used	Light intensity	Conc. of cocatalyst	Sacrificial donor	AQY (%)	Hydrogen evolution rate (μmol h^−1 g^−1)	Reference
CdS-COF	COF(TpPa-2)	CdS	400 W Xe lamp, λ ≥ 420 nm	0.5 wt% Pt	1 mL lactic acid	4.2	3678	Chem.–Eur. J., 2014, 20, 15961–15965
N2-COF/Co1	N2-COF	Chloro(pyridine)-cobaloxime (Co1)	λ ≥ 420 nm	8.0 wt% Pt	1 μL TEOA		782	J. Am. Chem. Soc., 2017, 139, 16228
NH2-UiO-66/TpPa-1-COF	TpPa-1-COF	NH2-UiO-66 MOF	λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	—	23 410	Angew. Chem. Int. Ed., 2018, 130, 12282
NH2-UiO-66/TAPT-TP-COF	TAPT-TP-COF	NH2-UiO-66 MOF	AM 1.5 G	3.0 wt% Pt	10% TEOA	5.7	8440	ACS Appl. Mater. Interfaces, 2021, 3, 29916
MOF-808@TpPa-1-COF	TpPa-1-COF	NH2-UiO-808 MOF	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	10.23	11 880	J. Mater. Chem. A, 2021, 9, 16743
MoS2-TpPa-1-COF	TpPa-1-COF	MoS2	300 W Xe lamp, λ ≥ 420 nm	—	Ascorbic acid	0.76	5585	J. Mater. Chem. A, 2019, 7, 20193
Ni(OH)2/COF-TpPa-2	COF-TpPa-2	Ni(OH)2	300 W Xe lamp, λ ≥ 420 nm	—	Sodium ascorbate	—	1890	Chem. Eng. J., 2020, 379, 122342
α-Fe2O3/TpPa-2-COF	TpPa-2-COF	α-Fe2O3	300 W Xe lamp, λ ≥ 420 nm	—	Sodium ascorbate	0.137	3770	J. Mater. Chem. A, 2020, 8, 4334
TiO2-TpPa-1-COF	TpPa-1-COF	TiO2	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	7.6	11 190	Appl. Catal., B, 2020, 266, 118586
COF-CN	TpBD	g-C3N4	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	15.09	12 800	ACS Appl. Mater. Interfaces, 2020, 12, 51555
rGO-TpPa-1-COF	TpPa-1-COF	rGO (reduced graphene oxide)	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	7.53	11 980	J. Mater. Chem. A, 2020, 8, 8949
TBTA/g-C3N4	TBTA	g-C3N4	300 W Xe lamp, λ ≥ 420 nm	2.08 wt% Pt	Ascorbic acid	—	26 040	Catal. Sci. Technol., 2021, 11, 2616
PEG@BT-COF	BT-COF	PEG (polyethylene glycol)	300 W Xe lamp, λ ≥ 420 nm	—	Ascorbic acid	11.2	11 140	Nature Comm., 2021, 12, 3934
PdTCPP⊂PCN415(NH2)/TpPa (composite 2)	TpPa-COF	PdTCPP⊂PCN415(NH2)	300 W Xe lamp, λ ≥ 420 nm	1.2 wt% Pt	Ascorbic acid	5.9	13 980	Angew. Chem. Int. Ed., 2022, 61, e202114071
CN/TMP	TMP COF	g-C3N4	300 W Xe lamp, λ ≥ 420 nm	Pt	TEOA	4.07	2057	Appl. Catal., B, 2022, 315, 121568
TiO2-x/TpPa-1-COF (6 : 4)	TpPa-1-COF	TiO2	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Sodium ascorbate	6.7	15 330	Chem. Eng. J., 2022, 446, 137213
Pt/rGO(20%)/TpPa-1-COF	TpPa-1-COF diamine	rGO (reduced graphene oxide)	300 W Xe lamp, λ > 420 nm	3.0 wt% Pt	Sodium ascorbate	13.82	19 590	J. Colloid interface Sci., 2022, 608, 2613–2622
MS-c@TpPa-1	TpPa-1-COF	[Mo3S13]^2− nanoclusters	300 W Xe lamp, λ > 420 nm	[Mo3S13]^2− nanoclusters	Sodium ascorbate	0.54	528	Chem. Eng. J, 2022, 446, 136883
TpPa-1-COF/ZnIn2S4	TpPa-1-COF	ZnIn2S4	300 W Xe lamp, λ ≥ 420 nm	—	0.35 M Na2S + 0.25 M Na2SO3	2.08	853	ACS Appl. Energy Mater., 2023, 6, 1103–1115
Ni-COF-SCAU-1	COF-SCAU-1	Ni(acac)2	300 W Xe lamp, λ ≥ 420 nm	3.0 wt% Pt	Ascorbic acid	43.2	197 460	Adv. Energy Mater., 2023, 13, 2203695

---

Recently, Yan and coworkers demonstrated the hybridization of TpPa-1-COF with a metal–organic framework (NH₂-UiO-66) to obtain an NH₂-UiO-66/TpPa-1-COF hybrid with improved stability and photocatalytic activity.¹¹¹ The hybrid was fabricated via the Schiff base reaction of the COF precursors 4,4′,4′′-(1,3,5-triazine-2,4,6-triyl)trianiline (TAPT) and 2,4,6-triformylphloroglucinol (TP) in the presence of NH₂-UiO-66. Interestingly, the hybrid material exhibited improved stability in an alkaline environment and a high photocatalytic hydrogen generation rate (8.44 mmol h⁻¹ g⁻¹) under visible light (400 nm). Notably, the hybrid's hydrogen generation activity was higher than that of pure MOF and COF materials (see below). Hence, hybridizing COFs with appropriate materials has proven to be a promising strategy for generating hybrid materials with improved chemical/photostability, light-harvesting ability, and photocatalytic activity for sustainable hydrogen generation.

4. Photocatalytic hydrogen generation activity of COFs

The following section discusses the major types of COFs known for the photocatalytic hydrogen evolution reaction (HER) from water.

4.1 Triazine-based COFs and CTFs

In addition to triazine-containing COFs, covalent triazine frameworks (CTFs) have been vastly used for water splitting, which is discussed in the following sections.

4.1.1 Triazine-based COFs. Notably, COFs composed of pyrazine, triazine, tetrazine, pyrimidine, and heptazine-based nitrogen-rich linkers exhibit modest HOMO–LUMO separation, resulting in improved visible light harvesting and HER performance. Theoretical studies have shown that photoexcited triazine molecules can abstract a hydrogen atom from water via a hole transfer mechanism involving triazinyl–hydroxyl biradical formation. Thus, in a triazine system, the hydrogen-bonded water molecule decomposes into hydrogen and hydroxyl radicals in a biphotonic photochemical reaction facilitating water splitting.

To take advantage of these properties of triazine, Lotsch and coworkers demonstrated the application of a triazine COF for photocatalytic water hydrogen generation.⁷⁷ The TFPT-COF was designed by using 1,3,5-tris-(4-formylphenyl)triazine (TFPT) and 2,5-diethoxy-terephthalohydrazide (DETH) building blocks (Fig. 9a). The coupling of the triazine precursor with DETH resulted in a hydrazone-linked COF featuring good chemical stability, crystallinity and visible light harvesting properties (band gap of 2.8 eV). The TFPT-COF showed a Brunauer–Emmett–Teller (BET) surface area of 1603 m² g⁻¹ and pore volume of 1.03 cm³ g⁻¹. Indeed, TFPT-COF exhibited good HER activity with a moderate hydrogen evolution rate of 230 μmol h⁻¹ g⁻¹ with Pt as the cocatalyst and sodium ascorbate as the SED (Fig. 9b). Interestingly, the use of triethanolamine as SED led to an increase in the HER rate to 1970 μmol h⁻¹ g⁻¹ (Fig. 9b). Further studies on the effect of incident light on the HER revealed the best H₂ generation activity at 400 nm light with the highest apparent quantum efficiency (Fig. 9c). The recyclability studies showed partial loss of crystallinity of COF after using it for 95 h due to exfoliation of COF in water.

Fig. 9 Synthesis and photocatalytic applications of TFPT-COF. (a) Scheme of the synthesis of TFPT-COF. (b) Hydrogen evolution performance of TFPT-COF using sodium ascorbate and triethanolamine as sacrificial electron donors. (c) Wavelength-specific hydrogen generation of Pt-immobilized TFPT-COF. Reproduced with permission from ref. 77 Copyright© 2014, Royal Society of Chemistry.

In subsequent pioneering efforts to explore the structure–property relationship, Lotsch and coworkers studied the effect of tailoring the structure of COF at the molecular level by varying the number of nitrogen atoms in the precursor on the hydrogen evolution performance.⁷⁸ They prepared a series of N_X-COFs (X = 0–3) by condensation reaction of hydrazine with triphenylaryl aldehydes by introducing N atoms in the central phenyl ring, N = 0 (phenyl), N = 1 (pyridyl), N = 2 (pyrimidyl) and N = 3 (triazine) (Fig. 10a). The substitution of the C–H moiety by N resulted in loss of planarity of the molecule due to a change in the dihedral angle between the central aryl ring and peripheral phenyl rings. The COFs exhibited high crystallinity (Fig. 10b) and were applied for photocatalytic water splitting. All the N_X-COFs showed photocatalytic water splitting activity with hydrogen generation rates of 23, 90, 438, and 1703 μmol h⁻¹ g⁻¹ for N₀, N₁, N₂, and N₃-COFs, respectively. Interestingly, the hydrogen evolution rate increases with an increase in the number of nitrogen substitutions with a fourfold increase per nitrogen atom addition, and N₃-COF showed the best activity (Fig. 10c). The increase in the nitrogen content led to lowering of the HOMO energy, facilitating effective separation of photoinduced charge carriers and higher hydrogen evolution performance.

Fig. 10 Synthesis and applications of N_x-COFs for water splitting. (a) Scheme of the synthesis of N_x-COFs from N_x-aldehydes and hydrazine. PXRD patterns (b) and hydrogen evolution performance (c) of N_x-COFs. Reproduced with permission from ref. 78 Copyright© 2015, Springer Nature.

In another significant contribution, Lotsch and coworkers reported the application of a chloro(pyridine)cobaloxime (Co1) molecular complex as a cocatalyst to boost the photocatalytic HER performance of an azine-linked N₂-COF, replacing the traditional Pt-based cocatalyst.⁷⁹ Among the various metal complex-based cocatalysts known, cobaloximes are most efficient, offer a low overpotential for hydrogen generation and can be covalently anchored with COFs. The photocatalytic water splitting study of N₂-COF in the presence of cobaloxime cocatalyst showed a hydrogen evolution rate of 160 μmol h⁻¹ g⁻¹. Since the complex is unstable and undergoes ligand exchange, adding a dimethylglyoxime (dmgH₂) ligand during catalysis enhanced the hydrogen evolution activity. Consequently, the hydrogen evolution rate of N₂-COF was enhanced to 782 μmol h⁻¹ g⁻¹ with the addition of 4.69 mM dmgH₂ during the water splitting reaction, demonstrating noble metal (Pt)-free cocatalyst-supported COF-based H₂ generation. Lotsch's group extended the concept for immobilizing a chloro(pyridine)cobaloxime complex in hydrazone-based COF-42 via PSM to achieve improved and sustained photocatalytic activity.¹¹² The covalent anchoring of the complex cocatalyst was facilitated by coordination to the pyridine group in the COF, resulting in hybrid COFs (Co1a and Co1b). The photocatalytic HER studies of the hybrids revealed enhanced hydrogen evolution compared to pristine COF. The higher HER performance of the hybrid has been attributed to the close contact between the COF and cobaloxime complex in the pore walls of the COF, facilitating facile charge transfer between the complex cocatalyst and COF.

In conjugated polymeric materials, donor–acceptor systems show efficient light absorption and charge transport, essential for an effective photocatalytic water-splitting application. Utilizing this strategy, Dong and coworkers designed two benzothiadiazole-based donor–acceptor COFs, BT-TAPT-COF and BDF-TAPT-COF, featuring good crystallinity, chemical stability, and light harvesting properties.^82,113 The COFs were prepared by imine condensation of tris-(4-aminophenyl)triazine (TAPT) with two different aldehydes, 4,4′-(benzothiadiazole-4,7-diyl)dibenzaldehyde (BT-CHO) and 4,4′-(benzo[1,2-b:4,5-b′]difuran-4,8-diyl)dibenzaldehyde (BDF-CHO). In BDF-TAPT-COF, the triazine molecule acts as an electron acceptor, and the benzothiadiazole or benzodifurane molecule acts as an electron donor (Fig. 11a). The water-splitting experiments revealed higher hydrogen generation by BDF-TAPT-COF (1390 μmol h⁻¹ g⁻¹) than by BT-TAPT-COF (949 μmol h⁻¹ g⁻¹) (Fig. 11c). Additionally, UV-Vis diffuse reflectance spectrum also supports the enhanced light-harvesting property of BDF-TAPT-COF (Fig. 11b). This has been attributed to the higher electron-donating ability of the benzothiadiazole moiety rendering efficient charge separation and transfer for the reduction of water to hydrogen evolution.

Fig. 11 Synthesis and applications of BDF-TAPT-COF for photocatalytic hydrogen generation. (a) Synthesis of BDF-TAPT-COF following the solvothermal method. (b) UV-Vis spectra of BDF-TAPT-COF, TAPT and BDF-CHO linkers. Inset: Tauc plots for the determination of the optical band gaps. (c) Hydrogen evolution studies using BDF-TAPT COF. Reproduced with permission from ref. 113 Copyright 2021, Royal Society of Chemistry.

In a similar concept, Yan and coworkers reported the construction of fluorine-functionalized TFA-COF by the Schiff-base condensation reaction of 2,3,5,6-tetrafluoroterephthaldehyde (TFA-CHO) and 2,4,6-tris(4-aminophenyl)-1,3,5-triazine.¹¹⁴ The COF showed a high surface area (2490 m² g⁻¹) and optimum band gap (2.40–2.82 eV) owing to the electron-withdrawing effect of the fluorinated aldehyde. Interestingly, the TFA-COF showed HER with a rate of 80 μmol g⁻¹ h⁻¹, which is higher than that of nonfluorinated COF (TA-COF) having a terephthaldehyde (TA) linker with a H₂ generation rate of 10 μmol g⁻¹ h⁻¹. The higher catalytic activity of TFA-COF has been ascribed to a reduced band gap resulting in enhanced charge separation and migration aided by electron-accepting fluorine atoms. This work highlights the importance of the choice of linkers in tuning the photocatalytic performance of COFs.

Recently, Thomas and coworkers demonstrated the effect of protonation as a promising means to improve the photocatalytic hydrogen generation performance of an imine-based COF.⁸³ They prepared three donor–acceptor type imine-linked COFs named TtaTfa, TpaTfa, and TtaTpa from 2,4,6-tris(4-aminophenyl)triazine (Tta), tris(4-formylphenyl)amine (Tfa), 1,3,5-tris(4-formylphenyl)benzene (Tpa-CHO) and 1,3,5-tris(4-aminophenyl)benzene (Tpa-NH₂) building blocks (Fig. 12a). The COFs exhibited high crystallinity and a porous structure with well-defined 2D hexagonal honeycomb-like pore channels. The protonated COFs obtained by treatment with ascorbic acid showed enhanced hydrogen generation activity under visible light (>420 nm). For instance, the protonated COF, TtaTfa, showed the best hydrogen evolution activity with a rate of 20.7 mmol g⁻¹ h⁻¹. However, TpaTfa and TtaTpa showed hydrogen generation rates of 14.9 and 10.8 mmol g⁻¹ h⁻¹, respectively (Fig. 12b). Interestingly, the HER performance of the COFs was negligible when TEOA was used as a SED, highlighting the role of protonation on the hydrogen evolution activity (Fig. 12c). Here, protonation of COF led to broader absorption of light and better charge migration aided by planarity of the molecule, resulting in higher hydrogen generation activity over the nonprotonated counterpart (Fig. 12d).

Fig. 12 (a) Synthesis scheme for protonated 2D imine-linked COFs. (b and c) Comparison of the HER performance of protonated COFs carried out using AC and TEOA as SED and (d) mechanism of protonation of pristine COFs in L-AC. Reproduced with permission from ref. 83 Copyright© 2021, Wiley-VCH.

Donor–acceptor (D–A) type porous and conjugated materials are widely used in water splitting applications as the optoelectronic properties of donor–acceptor materials can be easily modulated by controlling the intramolecular charge transfer (ICT) from the electron-donating unit (D) to the electron-accepting unit (A). In this regard, Liu and co-workers have demonstrated the structural advantages of three-component donor–π–acceptor (TCDA) COFs, where three-component COFs have shown higher activity compared to analogous two-component COFs consisting of a node and linker, respectively.¹¹⁵ Following the multi-component strategy for the synthesis of two isomorphic TCDA COFs, the sp²-carbon-linked COF-JLU35 was synthesised by Knoevenagel condensation between electron-rich benzo[1,2-b:3,4-b′:5,6-b′′]trithiophene-2,5,8-tricarbaldehyde (BTT), electron-poor 1,3,5-tris-(4-formyl-phenyl)triazine (TFPT) and 1,4-phenyldiacetonitrile (PDAN) in the presence of a catalytic amount of Cs₂CO₃. On the other hand, imine-linked COF-JLU36 was prepared by Schiff base condensation of benzo[1,2-b:3,4-b′:5,6-b′′]trithiophene-2,5,8-tricarbaldehyde (BTT), 1,3,5-tris-(4-formylphenyl)triazine (TFPT) and 1,4-phenylenediamine (PDA) in the presence of acetic acid catalyst (Fig. 13a). A sp²-carbon-linked COF-JLU35 (λ = 616 nm, τ = 2.68 ns) exhibited a red-shifted absorption spectrum and longer excited state lifetime compared to COF-JLU36 (λ = 548 nm, τ = 0.73 ns), indicating enhanced π-electron efficiency and improved charge separation in the fully conjugated sp²-carbon bond (Fig. 13b and d). Photocatalytic hydrogen evolution was therefore carried out under visible light irradiation (Fig. 13c). Surprisingly, COF-JLU35 exhibited an impressive HER of up to 70 800 μmol h⁻¹ g⁻¹ compared to COF-JLU36 (23 600 μmol h⁻¹ g⁻¹) (Fig. 13e). This increase in activity can be attributed to the exceptional charge separation, broader light absorption and enhanced photocatalytic activity exhibited by the tri-component donor–π–acceptor COFs.

Fig. 13 Synthesis and applications of multicomponent COF-JLU35 and COF-JLU36 for photocatalytic hydrogen generation. (a) Synthesis of COF-JLU35 and COF-JLU36 following the solvothermal method. (b) UV-Vis spectra of COF-JLU35 and COF-JLU36. (c) Study of AQY for different specific wavelengths using COF-JLU35. (d) Photoluminescence decay plot of COF-JLU35 (red) and COF-JLU36 (blue). (e) Hydrogen evolution performances of TCDA COFs. Reproduced with permission from ref. 115 Copyright© 2023, American Chemical Society.

4.1.2 Triazine-based CTFs. Covalent triazine frameworks (CTFs) are a subclass of COFs and have gained special interest owing to their advantages, such as high nitrogen content, chemical tunability, high surface area, permanent microporosity and thermal/chemical stability. The CTFs mainly comprise triazine moieties linked together to form conjugated structures. They are synthesized by acid-catalyzed trimerization of aromatic nitriles or condensation reaction of benzylamine, aromatic aldehydes, or amides. Owing to the extended conjugation and light absorption in the visible region, the application of CTFs for hydrogen generation from water has been studied. The following section discusses selected examples of CTFs known for HER activity.

Early studies on the application of CTF-1 as a photocatalyst for hydrogen evolution were conducted by Wu and coworkers.¹¹⁶ The trifluoromethanesulfonic acid (TFMSA)-catalyzed trimerization of 1,4-dicyanobenzene at RT resulted in CTF-1. It showed absorption at 350–420 nm with an optical band gap of 2.94 eV. The photocatalytic studies showed a hydrogen evolution rate of 200 μmol h⁻¹ g⁻¹ under visible light irradiation (λ ≥ 420 nm) in the presence of Pt as a cocatalyst and triethanolamine as the SED.

In a major development, Thomas and coworkers studied the effect of shortening the reaction time from 40 h to as little as 30 min on the photocatalytic hydrogen evolution performance of CTF-1_X (X = 2.5, 5, 10, 15, 20, and 30 min) by carrying out the synthesis in molten ZnCl₂ at 400 °C (Fig. 14a).¹¹⁷ It was found that decreasing the reaction time led to reduced carbonization of the pre-CTF, rendering CTF-1 with improved crystallinity and catalytic activity. Indeed, CTF-1 prepared with just 10 min of reaction time showed impressive hydrogen generation activity with a rate of 1072 μmol h⁻¹ g⁻¹ (AQY of 9.2%) under visible light (>420 nm) in the presence of the Pt cocatalyst and TEOA as the sacrificial agent (Fig. 14b and c). The photocatalytic activity was found to be consistent over a period of 80 hours, proving the stability of the catalyst under water-splitting reaction conditions.

Fig. 14 Synthesis and applications of CTFs for water splitting. (a) Two-step synthesis of CTF in short reaction times. (b) Photocatalytic hydrogen evolution performance of pre-CTF and CTF-1_X min materials under visible light irradiation. (c) Comparison of the photocatalytic hydrogen evolution performance of the pre-CTF and CTF-1_X min catalysts under visible light irradiation. Reproduced with permission from ref. 117 Copyright 2017, Royal Society of Chemistry.

In another important demonstration of the effect of functionalization on the water-splitting activity of CTFs, Wu and coworkers synthesized phosphorus (P)-incorporated CTF by thermal treatment of CTF-1 with red phosphorus.¹¹⁸ The doping of phosphorus atoms in the CTF framework was found to alter the electronic and optical properties of CTF-1, promoting facile separation and migration of photoinduced electron–hole pairs, thereby enhancing the HER performance, which was found to be approximately 4.5 times higher than that of pure CTF. In similar attempts, Li and coworkers developed CTF-based heterostructures (CTF-BT/Th-x) by incorporating electron-withdrawing benzothiadiazole (BT) and electron-donating thiophene (Th) moieties in CTF to accomplish enhanced photocatalytic hydrogen evolution activity.¹¹⁹ The heterostructures were prepared by a sequential polymerization reaction of nitrile precursors catalyzed by trifluoromethanesulfonic acid (TfOH) (Fig. 15a). The integration of the electron-withdrawing (BT) and electron-donating (Th) groups was found to alter the HOMO and LUMO positions in CTF-BT/Th-x (where x represents the benzothiadiazole to thiophene molar ratio) in comparison. The resulting band structure at the heterojunction facilitated facile migration of photogenerated electrons and holes, leading to enhanced photocatalytic hydrogen evolution with a rate of 6600 μmol h⁻¹ g⁻¹ under visible light (420 nm) in the presence of triethanolamine (TEOA) as the sacrificial electron donor and Pt (3 wt%) nanoparticles as cocatalysts (Fig. 15b and c). These reports demonstrate the importance of doping heteroatoms in the CTF backbone to achieve efficient photocatalytic hydrogen generation.

Fig. 15 (a) Sequential polymerization strategy for the synthesis of covalently connected CTF-BT/Th, (b) illustration of the band alignment at the heterostructure, and (c) time-dependent photocatalytic hydrogen generation. Reproduced with permission from ref. 119 Copyright© 2019, Wiley-VCH.

For the preparation of highly active COF-based photocatalysts for solar energy harvesting and conversion, the donor–acceptor (D–A) strategy is an effective synthetic approach. Following a similar strategy, Jin and coworkers reported the construction of a series of D–A₁–A₂ system-based CTFs for photocatalytic hydrogen evolution.¹²⁰ The frameworks were prepared through copolymerization of 4,7-bis(4-formylphenyl)-2,1,3-benzothiadiazole (M-BT) and 3,6-dicarbaldehyde-N-ethylcarbazole (M-CBZ) in designated ratios with terephthalimidamide dihydrochloride. Thus, a series of ter-CTF-x (x = percentage of M-CBZ unit in CTF system varying from 10 to 90%) were prepared. Of these materials, ter-CTF-0.7 showed the best photocatalytic performance with a hydrogen evolution rate of 966 μmol h⁻¹ (AQY of 22.8%) at 420 nm. Due to the suppressed recombination of the electrons and holes, more delocalized structure for charge migration, and highest charge-transfer/separation efficiency, ter-CTF-0.7 showed efficient hydrogen evolution performance, suggesting the critical role of the organization mode and the distance of the D and A units. Similarly, using a series of bipyridine-based CTFs synthesized through polycondensation, Wisser and coworkers demonstrated a direct correlation of the bipyridine content in CTF with the surface area, band gap, charge separation and surface wettability with water and subsequently on the photocatalytic hydrogen evolution performance.¹²¹ Among the series of bipyridine-based materials, CTF-02-Bpy_0.66 showed the best performance, reaching the highest hydrogen evolution rate of 7.24 mmol h⁻¹ g⁻¹ under visible light irradiation in the presence of a cocatalyst and a sacrificial electron donor due to the optimum balance between optoelectronic properties and highest hydrophilicity. This report exemplifies that the balance between optoelectronic and surface properties and hydrophilicity determines the photocatalytic activity.

In another major development, Cooper and coworkers demonstrated the synthesis of a structurally diverse family of 39 CTFs by using the Suzuki–Miyaura coupling reaction of 2,4,6-tris[4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenyl]-1,3,5-triazine with various dibromo linkers,¹²² which includes five-membered aromatic heterocycles, substituted phenylenes, bipyridyl heterofluorenes, azoles, quinoxalines, etc. The photocatalytic hydrogen generation performance of these COFs was correlated to their ionization potential, electron affinity, optical gap, and transmittance. Out of all CTF-xx samples (where xx represents the number of dibromo monomers utilized during the synthesis), CTF-15 was found to show the best hydrogen evolution rate of 2946 μmol h⁻¹ g⁻¹ under visible light (>420 nm) irradiation over 50 hours. The electron affinity, ionization potential, optical band gap, and dispersibility of the CTF particles in solution are essential factors in determining the photocatalytic activity of the materials. These results demonstrated a powerful method to discover new photocatalysts and provided valuable insights into the structure–property relationships across a diverse range of chemical functionalities in the material backbones.

The presence of aromatic triazine linkages and fully conjugated structures enriched with aromatic nitrogen atoms make CTFs appealing in photocatalysis applications. Although many synthetic methods have been reported, most of the CTFs synthesized from those methods are amorphous. To develop crystalline, porous and hydrophilic CTFs using a wide array of building blocks, a polycondensation approach was employed for the construction of CTFs under mild conditions. Typically, the condensation reaction of an aldehyde and an amidine dihydrochloride involving Schiff base formation followed by Michael addition was employed for the construction of CTF-HUSTs in the presence of cesium carbonate in a dimethylsulfoxide-water solvent mixture heated at 60 °C for 12 h, then heated at 80 °C, and then 100 °C for 12 h, separately, before being heated at 120 °C for 3 days.¹²³ The crystalline and porous CTF-HUST-1 showed hydrogen evolution with a rate as high as 1460 μmol h⁻¹ g⁻¹, which is higher than that of CTF-1 synthesized by an ionothermal method (1072 μmol h⁻¹ g⁻¹). Similarly, using in situ formation of aldehyde monomers through the controlled oxidation of alcohols to aldehydes, Tan and coworkers achieved the synthesis of CTF-HUST-C1 by reticulating them with terephthalamidine – demonstrating that the improvement in crystallization lies in decreasing nucleation rates and generating lower concentrations of nuclei.¹²⁴ Using CTF-HUST-C1, which has much-improved crystallinity, higher thermal stability and layered structures, highly efficient photocatalytic hydrogen evolution performance (5100 μmol h⁻¹ g⁻¹) than the low crystalline CTF-HUST-1 (1460 μmol h⁻¹ g⁻¹) or amorphous CTFs (1072 μmol h⁻¹ g⁻¹) was achieved due to the easier electronic transport and better light absorption.

Furthermore, Jin and coworkers synthesized CTF-HUST-A1 by utilizing a benzylamine-functionalized monomer that initially formed an imine bond, which was further transformed into a triazine moiety.¹²⁵ Herein, the basicity of the base reagent employed was found to be very crucial in the generation of crystalline COF. CTF-HUST-A1-^tBuOK showed superior hydrogen evolution activity with a rate of 9200 μmol h⁻¹ g⁻¹ and AQE of 7.4% under visible light (>420 nm) irradiation along with good stability and recyclability. Tan and coworkers demonstrated the effect of noble metal cocatalysts on hydrogen generation activity by using two isostructural highly crystalline CTFs, named CTF-HC2 and CTF-HC6, with different pyridinic nitrogen contents.¹²⁶ The CTFs were obtained by following a controlled feeding rate method¹²⁷ in which the aldehyde solution [4,4′-biphenyl dicarboxaldehyde (2,2′-bipyridine)-5,5′-dicarbaldehyde] was slowly dropped into the amidine (terephthalamidine dihydrochloride) solution. The CTFs were treated with Pd(OAc)₂ and H₂PtCl₆ to generate Pd@CTFs and Pt@CTFs, respectively. Interestingly, the Pd@CTFs were found to exhibit higher HER performance than the Pt@CTFs, which has been attributed to the smaller loading of Pd nanoparticles facilitating better separation of photogenerated electron–hole pairs in comparison to the larger Pt nanoparticle-loaded sample. Hydrogen generation was found to be durable and recyclable for multiple cycles.

Most photocatalytic water splitting studies using COFs are usually performed using materials in the form of suspended powders. Owing to their convenient separation and recycling potential, preparations of COF and CTF films are desired. However, existing synthetic approaches for COF and CTF preparation mainly result in insoluble and unprocessable powders, making their water-splitting applications challenging. To address these issues, Tan and coworkers recently reported an aliphatic amine-assisted interfacial polymerization technique to obtain free-standing and semicrystalline CTF films (Fig. 16a).¹²⁸ In particular, the dimethylsulfoxide soluble aldehyde monomer was transformed into a dimethylsulfoxide-insoluble imine precursor by reacting with n-hexylamine (Fig. 16b). The initial arrangement of aldehyde monomers is guided via the imine precursor distributed at the DMSO surface, generating a DMSO/air interface at the same time, which is helpful for the preparation of CTF thin films. Using this technique, CTF films with sizes as high as 250 cm² and average thicknesses up to 30 to 500 nm were prepared (Fig. 16c and d). These films immobilized on a glass support were used for photocatalytic hydrogen generation, which exhibited good photocatalytic hydrogen evolution performance (5.4 mmol h⁻¹ m⁻²) in the presence of Pt cocatalysts (Fig. 16e–h). The considerably high hydrogen evolution performance was ascribed to the good light absorption, crystalline structure, and large lateral sizes of the films.

Fig. 16 Synthesis and applications of CTF films for water splitting. (a) Scheme of the synthesis of CTF films. (b) The fabrication of CTF films on a dimethylsulfoxide surface assisted by an imine precursor. (c) Transparent and free-standing CTF films. (d) Solvent-immersed CTF films. (e) Light absorption spectrum of CTF films. (f) The photocurrent curve and (g) photocatalytic hydrogen evolution performance of CTF films on glass for four cycles. (h) An image of CTF film deposited on a glass plate for photocatalytic hydrogen evolution. Reproduced with permission from ref. 128 Copyright© 2021, Springer Nature.

4.1.3 Heptazine-based COFs. Like triazine moieties, heptazine-based monomers have also been used to synthesize porous frameworks due to their nitrogen-rich, rigid, tricyclic aromatic backbone and their potential photocatalytic applications.^129–133 In particular, in 2013, Thomas and coworkers demonstrated the use of heptazine-based microporous polymer networks (HMP-1) as photocatalysts for visible light-driven hydrogen evolution.¹³⁴ Through a polycondensation reaction of cyanuric chloride with phenylenediamine, the synthesis of heptazine-based polymer (HMP-1) was achieved. Interestingly, HMP-1 exhibited good absorption in the visible region (400–600 nm) due to its aromatic network, which confers photocatalytic properties. The photocatalytic hydrogen generation experiments performed using HMP-1, Pt (3 wt%) as cocatalyst and TEA as SED showed H₂ generation at a rate of 2 μmol h⁻¹ at λ > 420 nm. Later, in 2016, the same group reported a donor–acceptor-based heptazine-based polymer, named HMP-3, by replacing phenylenediamine with benzothiadiazole moieties in the network.¹³⁵ Indeed, the molar ratio of heptazine to benzothiadiazole was varied to obtain HMP-3_4:3 and HMP-3_2:3, which showed relatively higher hydrogen evolution rates of 32 and 31 μmol h⁻¹, respectively. The higher HER of HMP-3 was attributed to the stabilization of photogenerated electrons and holes by a donor–acceptor effect. Motivated by these studies, Tan and coworkers synthesized a series of 2D covalent heptazine-based frameworks (CHFs) by covalently linking heptazine with electron-rich alkynyl, phenyl or alkynyl–phenyl units.¹³⁶ Furthermore, the overall water splitting performance was investigated using rationally constructed 2D covalent heptazine-based frameworks, which revealed that the 2D CHFs possess tunable band gaps in the range of 2.27–3.26 eV. The CHFs prepared using p-phenyl, p-phenylenediynyl, m-phenylenediynyl and phenyltriynyl showed considerable performance for total water splitting to produce H₂ and O₂. These studies demonstrate the potential utility of heptazine-based frameworks for photocatalytic water splitting to H₂ evolution.^137,138

4.2 Pyrene-based COFs

Pyrene-based COFs have attracted significant interest as promising candidates for photocatalytic hydrogen generation from water owing to their extended π-conjugation and visible light absorption properties. In this context, Lotsch and coworkers designed pyrene-based COFs denoted TEBPY, TENPY, and TEPPY by a condensation reaction of hydrazine with 1,3,6,8-tetrakis(4-ethynylbenzaldehyde)-pyrene (TEBPY), 1,3,6,8-tetrakis(6-ethynylnicotinaldehyde)-pyrene (TENPY), and 1,3,6,8-tetrakis(2-ethynylpyrimidin-5-carb-aldehyde)pyrene (TEPPY), respectively (Fig. 17a).¹³⁹

Fig. 17 Synthesis and applications of pyrene-based COFs. (a) Synthesis scheme for pyrene-based azine-linked COFs. (b) Time course of photocatalytic hydrogen evolution from a 10 vol% aqueous TEOA suspension and Pt-modified A-TEXPY-COFs. Reproduced with permission from ref. 139 Copyright© 2018, Wiley-VCH.

The azine-linked COFs A-TEBPY, A-TENPY, and A-TEPPY showed water splitting activity with hydrogen generation rates of 98, 22, and 6 μmol h⁻¹ g⁻¹, respectively (Fig. 17b). The mechanism of photocatalytic water splitting was believed to follow a radical cation pathway, and the stability of the radical cation was found to increase with decreasing nitrogen content. As the nitrogen content increases in the COFs, the hydrogen generation rate was found to decrease, highlighting the effect of structural tuning at the molecular level on photocatalytic activity – a structure–property correlation.

In an attempt to create a pyrene-based COF donor–acceptor COF system, Jiang and coworkers developed a donor–acceptor-based sp² carbon conjugated COF, named sp²c-COF, by solvothermal reaction between tetra(4-formylphenyl)pyrene (TFPPy) and 1,4-phenyldiacteonitrile (PDAN) and further functionalized it with 3-ethylrhodanine (ERDN) monomers as an end-capping group (Fig. 18a and b).¹⁴⁰ These sp²c-COF samples showed excellent stability in water, acidic (HCl), and basic (NaOH) conditions. The COF showed good hydrogen evolution activity with a rate of 1360 μmol h⁻¹ g⁻¹ under visible light (≥420 nm) (Fig. 18c–e). Interestingly, COF functionalized with ERDN end groups showed enhanced hydrogen generation with a rate of 2120 μmol h⁻¹ g⁻¹ upon irradiation with a wavelength of 420 nm, which was found to be approximately 1.6 times higher than that of pristine COF. The enhanced hydrogen evolution activity of sp²c-COF_ERDN has been ascribed to its band structure, improved light-harvesting capability, and heterojunction formation. In addition, the electron-deficient nature of EDRN introduces a push–pull effect in the COF network, facilitating facile electron migration and thereby preventing charge recombination.

Fig. 18 (a) Synthesis scheme of sp²-c-COF and sp²-c-COF_EDRN. (b) Mechanism of hydrogen generation from water using a sp² carbon-conjugated framework and Pt as a cocatalyst. (c) Apparent quantum yields (AQYs) of sp²c-COF_ERDN under irradiation with monochromatic light at 420 nm, 490 nm, 520 nm and 578 nm. (d) Hydrogen evolution was monitored over 5 h with sp²c-COF (blue circles), sp²c-COF_ERDN (red circles), sp²c-CMP (black circles) and imine-linked pyrene COF (triangles) as photocatalysts under irradiation with wavelengths ≥420 nm. (e) Long-term stability of sp²c-COF_ERDN over four repeated photocatalytic hydrogen evolution cycles under light irradiation (≥420 nm). Reproduced with permission from ref. 140 Copyright© 2019, Elsevier.

The design of thiazole-containing pyrene-based COFs (Tz-COF) exhibiting high crystallinity, surface areas, and good physicochemical stability has been reported by Cooper and coworkers.¹⁴¹ The COFs were prepared via C–H functionalization and subsequent oxidative annulation reaction of 2,6-diaminonaphthalene and 2,6-diaminoanthracene moieties (Fig. 19a). The one-pot synthesis of thiazole-based COFs was complete with thiazole ring (Tz) formation by a cascade reaction between aldehyde, amine, and sulfur involving the following steps: (i) Schiff base condensation between aldehyde and amine, (ii) electrophilic attack from sulfur and elimination S_n−1, (iii) cyclization, and (iv) oxidative aromatization of the ring. Photocatalytic investigations of crystalline Tz-COF-4 constituted by pyrene and thiazole building blocks displayed a hydrogen generation rate of 4296 μmol h⁻¹ g⁻¹ under visible light (λ > 420 nm) using ascorbic acid as a sacrificial electron donor, which was higher than that of the amorphous counterpart (Fig. 19b and c). The high hydrogen generation performance of Tz-COF-4 has been attributed to its large surface area and broad absorption of visible light.

Fig. 19 (a) Scheme of the synthesis of thiazole-linked Tz-COFs. (b) Comparison of the absorption profile and (c) hydrogen evolution rate under visible light irradiation (λ > 420 nm) by Tz-COF-4, IL-COF-2 and amorphous Tz-COF-4. Reproduced with permission from ref. 141 Copyright© 2020, American Chemical Society.

In another precedent of the donor–acceptor strategy, Chen and coworkers developed pyrene-based isostructural Py-XTP-BT-COFs (X = H, F, Cl) by condensation of 4,4′,4′′,4′′′-(pyrene-1,3,6,8-tetrayl)tetrabenzaldehyde (Py-CHO) with terphenyl-based diamines, 4,7-diphenylbenzo[c][1,2,5]thiadiazole (HTP-BT-NH₂) or 4,7-diphenyl-5,6-difluorobenzo[c][1,2,5]thiadiazole (FTP-BT-NH₂) or 4,7-dibromo-5,6-dichlorobenzo[c][1,2,5]thiadiazole (ClTP-BT-NH₂) (Fig. 20a).¹⁴² The water splitting studies showed higher hydrogen evolution activity of chlorinated COF (Py-ClTP-BT) with a rate of 177 μmol h⁻¹ over fluorinated, Py-FTP-BT (57 μmol h⁻¹) and nonhalogenated, Py-HTP-BT (21 μmol h⁻¹) (Fig. 20b). The presence of electron-withdrawing halogen atoms on the benzothialozole linker facilitates effective charge separation by suppressing electron–hole recombination, lowering the activation barrier for hydrogen generation. This is also reflected in the average lifetimes of the photogenerated electrons of 3.8, 2.85, and 2.13 ns for Cl, F, and nonsubstituted COFs. This work represents another important demonstration of the effect of structural modulation on the photocatalytic property of COFs, providing a useful strategy for the rational design of COFs for enhanced hydrogen evolution from water.

Fig. 20 (a) Scheme of synthesis of Py-XTP-BT COF (X = Cl, F, H). (b) Hydrogen evolution rates with and without Pt-cocatalysts. Reproduced with permission from ref. 142 Copyright© 2020, Wiley-VCH.

In a similar strategy, Chen and coworkers utilized functionalized benzothiadiazole (BT) and pyrene tetraamine building blocks to construct donor–acceptor photocatalytic NK-COFs through Schiff base condensation under solvothermal conditions (Fig. 21a).¹⁴³ Notably, NK-COFs possess a high specific surface area and mesoporosity (∼35 Å) and exhibit good chemical stability in boiling water and acidic and basic solutions. The photocatalytic water splitting studies revealed higher hydrogen generation activity (120 μmol h⁻¹) for NKCOF-108 composed of a fluorinated benzothiadiazole linker over the nonfluorinated counterpart NKCOF-110 (Fig. 21b). As discussed in the previous sections, the fluorination of benzothiadiazole enhances the electron-accepting nature of the linker, lowering the LUMO and HOMO levels and facilitating effective charge separation and migration from the electron-donor pyrene to the benzothiadiazole moiety, leading to enhanced hydrogen evolution performance.

Fig. 21 (a) Scheme of synthesis of NK-COFs and (b) hydrogen evolution performance of various NK-COFs. Reproduced with permission from ref. 143 Copyright© 2021, American Chemical Society.

In a similar strategy, Wen and coworkers recently developed PyTz-COF by imine condensation between electron donor 4,4′,4′′,4′′′-(pyrene-1,3,6,8-tetrayl)tetraaniline (PyTA) and electron acceptor 4,4′-(thiazolo[5,4-d]thiazole-2,5-diyl) dibenzaldehyde (TzDA) moieties (Fig. 22a).¹⁴⁴ To compare the photophysical properties of PyTz-COF with those of a non-donor-acceptor system, PyBp-COF was prepared, wherein the Tz linker was replaced by a biphenyl linker. The photoluminescence emission intensity of PyTz-COF was lower than that of PyBp-COF, suggesting effective charge separation in the former compared to that of the latter COF (Fig. 22b). The photocatalytic study of PyTz-COF for water splitting revealed an excellent hydrogen generation rate of 2072.4 μmol g⁻¹ h⁻¹ under simulated sunlight using ascorbic acid as a sacrificial electron donor (Fig. 22c). The enhanced HER of PyTz-COF has been ascribed to good visible light absorption and effective separation and transportation of charge carriers aided by the push–pull effect of the donor–acceptor COF.

Fig. 22 (a) Preparation of donor–acceptor PyTz-COF. (b) Photoluminescence spectra of PyTz-COF and PyBp-COF excited at 320 nm. (c) Time-dependent photocatalytic hydrogen generation activity of PyTz-COF under AM 1.5 light irradiation. Reproduced with permission from ref. 144 Copyright© 2020 Wiley-VCH.

The controlled deposition of platinum (Pt) as a cocatalyst in the COF matrix is of paramount importance for photocatalytic hydrogen evolution, as the uncontrolled growth of large and non-uniform Pt nanoparticles limits the number of exposed active surfaces, even at higher loadings. To address this issue, Deng and co-workers synthesised a PY-DHBD-COF via a Schiff base condensation reaction involving 1,3,6,8-tetra(4-formylphenyl)pyrene (PY-CHO) and 1,4-dihydroxybenzidine (DHBD) (Fig. 23a).¹⁴⁵ The resulting COF structure contained adjacent hydroxyl groups and imine bonds in its constituent units. The diffuse reflectance spectroscopy analyses showed that the COF material exhibited an absorption edge at ∼510 nm and the corresponding band gap was calculated to be 2.28 eV (Fig. 23b). The photocatalytic performance of the COF material for hydrogen generation under visible light irradiation was investigated using ascorbic acid as the sacrificial electron donor and Pt (H₂PtCl₆) as the cocatalyst. Remarkably, the highest observed photocatalytic hydrogen evolution rate reached 71 160 μmol h⁻¹ g⁻¹ over 60 h when the Pt loading amount was 3.0 wt% with an AQY of 8.4% (Fig. 23c). During the photocatalytic hydrogen evolution, it was observed that the Pt cocatalyst exhibited selective adsorption near the hydroxyl groups and imine-N sites within the COF framework. Subsequent photoreduction processes resulted in the in situ photodeposition of highly dispersed Pt-clusters on the 2D layered surface of the COF. This study represents a significant advance in the field of photocatalysis by providing a new direction for precise control of co-catalyst deposition at the atomic level – contributing to the development of advanced strategies for improving the efficiency and effectiveness of photocatalytic processes.

Fig. 23 Synthesis and photocatalytic activity of pyrene-based COF. (a) Schematic representation of the synthesis of PY-DHBD-COF. (b) Absorption spectra of PY-DHBD-COF [inset: Tauc plot]. (c) Recyclability of PY-DHBD-COF for hydrogen generation using 3 wt% Pt as cocatalyst. Reproduced with permission from ref. 145 Copyright© 2022, Springer Nature.

4.3 Porphyrin-based COFs

Porphyrins offer diverse applications in many fields, and the incorporation of porphyrins as building blocks in the construction of COF can significantly boost the chemical and photophysical properties of COF materials.^146–148 Furthermore, the redox property of porphyrins can also play an essential role in hydrogen evolution.¹⁴⁹ Porphyrin molecules possess a good visible light-harvesting ability, and the HOMO–LUMO gap can be engineered with suitable functionalization to design COFs with optimal properties for efficient water splitting for hydrogen generation. Hence, the design and application of porphyrin-based COFs as visible light-responsive catalysts for hydrogen evolution from water has gained significant interest from researchers worldwide.

Exfoliation has become a popular method for preparing layered materials since the discovery of exfoliated graphene. This method has enabled many promising electronic and optoelectronic applications of two-dimensional materials. To extend this advantage, the exfoliation of COFs has been attempted. However, due to strong hydrogen bonding and π–π stacking interactions among the layers, the exfoliation of COFs has limitations. To counter this problem, Osakada and coworkers proposed an exfoliation method for porphyrin COFs through the incorporation of metal ions (Mg and Cu) and an axial ligand (pyridine) to break the π–π stacking of the porphyrin COF via steric hindrance.¹⁵⁰ For this study, a chemically stable porphyrin-based DhaTph COF was synthesized by reacting 5,10,15,20-tetrakis(4-aminophenyl)-21H,23H-porphyrin and 2,5-dihydroxyterephthalaldehyde in the presence of acetic acid. Using PXRD data, it was confirmed that the distance between two adjacent layers (001) was 4 Å, which approximately matches the simulated structure. Using simultaneous axial coordination of pyridines and metal ions, the synthesis of two-dimensional porphyrin nanodisks (e-CON) with an average thickness of 1 nm was achieved and further verified by scanning electronic microscopy. Considering the presence of extended π-conjugation and few-layered COF nanodisks, these materials were applied for photocatalytic hydrogen generation in the presence of Pt nanoparticles/reduced-graphene oxide. Enhanced hydrogen evolution was observed for e-CON (Cu,epy)/Pt/RGO compared with DhaTph/Pt/RGO, probably due to the increase in surface area between e-CON and Pt/RGO. This result demonstrated the role of COF exfoliation in enhancing the HER performance.

In a pioneering work, Wang and coworkers demonstrated the rational design of a series of porphyrin-based two-dimensional MPor-DETH-COFs (M = H₂, Co, Ni, Zn) and investigated their photocatalytic hydrogen evolution activity.¹⁵¹ The COFs were synthesized by the Schiff base condensation of tetrakis(4-formylphenyl)pophyrin (p-MPor-CHO) and 2,5-diethoxy-terephthalohydrazide (DETH) under solvothermal conditions (Fig. 24a). The COFs showed good crystallinity and high specific surface areas in the range of 826–1020 m² g⁻¹ with pore channels of approximately 2.4 nm. UV-Vis absorption measurements of the COFs showed broad absorption in the visible region with narrow band gaps in the range of 1.77–1.88 eV and the longest excited-state lifetimes, indicating good charge-separation. In addition, the COFs showed good water-splitting activity under visible light in the presence of a Pt cocatalyst and triethanolamine as a sacrificial electron donor. The hydrogen generation rate of ZnPor-COF was found to be highest (413 μmol g⁻¹ h⁻¹), followed by NiPor- and H₂Por-COF, with HERs of 211 and 80 μmol g⁻¹ h⁻¹, respectively. Notably, CoPor-COF showed the lowest hydrogen evolution rate of 25 μmol g⁻¹ h⁻¹ (Fig. 24b), which has been ascribed to the Co²⁺ (3d⁷) configuration with predominant LMCT that suppresses hole migration along the porphyrin channels. However, in the case of Ni²⁺ (3d⁸), the LMCT process is partially hampered, and as a result, hole transport is partly favored. On the other hand, for ZnPor-COF with Zn²⁺ (3d¹⁰), the LMCT process is forbidden, and hence, the photogenerated holes are free to migrate across the porphyrin rings and the electrons via the Zn⋯Zn chain, leading to the suppression of electron–hole recombination and resulting in higher hydrogen generation activity. Furthermore, the COFs showed good photostability and recyclability for multiple cycles of reuse, showcasing the utility of porphyrin-based COFs for photocatalytic water splitting for hydrogen generation.

Fig. 24 (a) Schematic representation of the synthesis of BDF-TAPT-COFs; (b) hydrogen evolution rates for MPor-DETH-COFs (M = H₂, Co, Ni, and Zn). Reproduced with permission from ref. 151 Copyright© 2021, Springer Nature.

4.4 Triphloroglucinol-containing COFs

After the first discovery of a COF by applying 1,3,5-triformylphloroglucinol (TP) as a linker,¹⁵² a large number of COFs have been prepared using the same linker and applied for various applications. The Schiff-base reaction of 1,3,5-triformylphloroglucinol with di-, tri- or tetra-amine linkers initiates through imine formation, followed by subsequent tautomerization to yield chemically stable β-ketoenamine-linked COFs. Because of their high chemical stability and conjugated structure, COFs constructed by using the covalent linking of 1,3,5-triformylphloroglucinol with other conjugated linkers have shown promising activity for water splitting. Note that the role of 1,3,5-triformylphloroglucinol in these COFs is to induce chemical stability and maintain conjugation, whereas the chemical functionalities and moieties (e.g., triazine, diacetylene, sulphone, etc.) induced through amine linkers initiate the water splitting reaction. The following section discusses the photocatalytic hydrogen generation activity of 1,3,5-triformylphloroglucinol-containing COFs.

In a major demonstration of the effect of conjugation on the water-splitting activity, we reported the effect of diacetylene moieties on the photocatalytic hydrogen generation activity of COFs.⁹⁶ To analyze the role of diacetylene moieties in water splitting, β-ketoenamine-linked TP-EDDA and TP-BDDA COFs containing acetylene and diacetylene moieties, respectively, were prepared by solvothermal reaction of 1,3,5-triformylphloroglucinol with 4,4′-(ethyne-1,2-diyl)dianiline (EDDA) and/or 4,4′-(buta-1,3-diyne-1,4-diyl)dianiline (BDDA) linkers (Fig. 25a). The COFs exhibited good crystallinity, chemical stability and moderate surface areas (Fig. 25b). To further study the effect of conjugation on the optical and catalytic properties, an isoreticular TP-DTP COF was prepared by using 1,3,5-triformylphloroglucinol and 4,4′′-diamino-p-terphenyl (DTP) linkers. The UV-Vis spectra of the COFs showed an absorption edge at 500 nm for TP-DTP COF, whereas the absorption edges for TP-EDDA and TP-BDDA COF were redshifted to 520 and 525 nm, respectively (Fig. 25c). Interestingly, TP-BDDA COF showed better hydrogen evolution performance with a rate of 324 μmol h⁻¹ g⁻¹ over 60 hours, which was found to be approximately ten times higher than that of TP-EDDA COF (HER: 30 μmol h⁻¹ g⁻¹) and TP-DTP COF (HER: 20 μmol h⁻¹ g⁻¹). The enhanced hydrogen generation activity of TP-BDDA has been attributed to the presence of diacetylene moieties with extended π-conjugation. This resulted in facile separation and migration of photoexcited electrons to the catalyst surface for effective hydrogen generation. Using a series of COFs, we have demonstrated photocatalytic hydrogen generation using a COF-based photocatalyst without the presence of any heteronuclear functionalities (i.e., triazine, heptazine, etc.) inside the structure, emphasizing the importance of the diacetylene diad in photocatalysis.

Fig. 25 Analyzing the role of diacetylene moieties in water splitting. (a) Scheme of the synthesis of TP-EDDA and TP-BDDA COFs. (b) Nitrogen sorption analyses of TP-EDDA and TP-BDDA COFs. (c) UV-Vis diffuse reflectance spectra of TP-BDDA, TP-EDDA, and TP-DTP COFs. (d) Photocatalytic hydrogen evolution performance of TP-EDDA, TP-BDDA and TP-DTP COFs. Reproduced with permission from ref. 96 Copyright© 2017, American Chemical Society.

In a significant development, Cooper and coworkers reported the design of sulfone-based S-COF (sulfone) and FS-COF (fused sulfone) and demonstrated the role of crystallinity, porosity and COF wettability on photocatalytic hydrogen generation activity. The synthesis of S-COF and FS-COF was achieved by a Schiff base condensation reaction between 1,3,5-triformylphloroglucinol and two planar thiophene-sulfone derivatives, 3,7-diaminodibenzo[b,d]thiophene sulfone (SA) and 3,9-diamino-benzo[1,2-b:4,5-b′]bis[1]benzothiophene (FSA), respectively (Fig. 26a).⁸⁰ Owing to the irreversible keto–enol tautomerization of the products, the COFs formed were found to exhibit good chemical stability. Additionally, the effective π–π stacking between fused and planar FSA linkers facilitated the formation of highly crystalline FS-COF over S-COF. The photocatalytic water splitting studies showed impressive hydrogen evolution rates of 4.44 and 16.3 mmol h⁻¹ g⁻¹ by S-COF and FS-COF, respectively, under visible light (λ > 420 nm) irradiation with the Pt cocatalyst and ascorbic acid as a sacrificial electron donor (Fig. 26b). The photocatalytic activity was found to be consistent over 50 hours, validating the long-term durability of the COFs (Fig. 26c). It is noteworthy that the presence of sulphone moieties provided potential hydrophilic sites for enhancing the interaction/uptake of water by the COFs, thereby resulting in high water splitting activity.

Fig. 26 Synthesis and applications of sulphone-based COFs. (a) Scheme of the synthesis of FS-COF. Inset: structures of 3,7-diaminodibenzo[b,d]thiophene 5,5-dioxide and [1,1′:4′,1′′-terphenyl]-4,4′′-diamine linkers. (b) Photocatalytic hydrogen evolution performance using FS-COF, S-COF, TP-COF and FS-P polymer. (c) Photocatalytic hydrogen generation under visible light irradiation using FS-COF over 50 h. Reproduced with permission from ref. 80 Copyright© 2018, Springer Nature.

In subsequent efforts, Lotsch and coworkers prepared a TpDTz COF utilizing 1,3,5-triformylphloroglucinol (TP) and 4,4′-(thiazolo[5,4-d]thiazole-2,5-diyl)dianiline (DTz) precursors and studied light-driven hydrogen evolution from water using a non-noble metal-based nickel-thiolate (NiME) cluster as a cocatalyst.¹⁵³ Herein, the thiazolo[5,4-d]thiazole (TzTz) linkage provided high oxidative stability, excellent light absorbability, and high electron and hole mobility, rendering efficient photocatalytic performance to the COFs. Consequently, TpDTz COF exhibited good HER performance with a sustained hydrogen generation rate of 941 μmol g⁻¹ h⁻¹. Notably, the presence of the NiME cocatalyst was beneficial in increasing the COF excited-state lifetime during photocatalysis, resulting in better hydrogen evolution activity with long-term stability.

The modulation of active centers in the COF skeletons is of great importance for tuning the light absorption as well as photocatalytic water splitting performance, which, however, encounters severe challenges. In this regard, Chen and coworkers synthesized β-ketoenamine-linked Tp-PDA and Tp-DBN COFs through a Schiff base condensation reaction between 1,3,5-triformylphloroglucinol and phenylenediamine (PDA) or 2,5-diaminobenzonitrile (DBN) linkers, respectively (Fig. 27).⁹⁸ Tp-PDA and Tp-DBN COFs exhibited hydrogen evolution with consistent rates of 0.6 mmol g⁻¹ h⁻¹ and 1.8 mmol g⁻¹ h⁻¹ H₂ for up to 25 hours. The electron-withdrawing cyano (–CN) groups in Tp-DBN COF effectively participate in electron redistribution in the COF framework and develop efficient charge separation for photogenerated electron–hole pairs, further increasing the electron transfer rate during photocatalysis and enhancing the hydrogen evolution performance.

Fig. 27 (a) Schematic representation of the synthesis method for Tp-PDA and Tp-DBN COFs. Hydrogen evolution activity (b) and incident light (wavelength)-dependent AQYs (c) for hydrogen generation using Tp-DBN COF carried out with 3 wt% Pt in sodium ascorbate solution. Reproduced with permission from ref. 98 Copyright© 2021, Wiley-VCH.

To establish the structure–property correlation, Le and coworkers synthesized several β-ketoenamine-based COFs, named TpPa-X [X = H, Cl₂, SO₃H and (CH₃)₂], by utilizing 1,3,5-triformylphloroglucinol and p-phenylenediamine (Pa-H), 2,5-dichloro-1,4-phenylenediamine (Pa-Cl₂), 2,5-diaminobenzenesulphonic acid (Pa-SO₃H) or 2,5-dimethyl-1,4-phenylenediamine (Pa-(CH₃)₂) (Fig. 28a).⁸¹ The photocatalytic investigations of the COFs revealed HER activity with rates of 11.13, 11.73, 4.44, and 3.62 μmol m⁻² h⁻¹ for TpPa-H, TpPa-Cl₂, TpPa-SO₃H, and TpPa-(CH₃)₂, respectively (Fig. 28b). The higher hydrogen evolution performance of TpPa-Cl₂ has been correlated with the combined effects of optimal band structure, large specific surface area, better separation, and migration of charge carriers.

Fig. 28 (a) Scheme of the synthesis of Tp-COFs using the solvothermal method. (b) The hydrogen evolution rate of different TpPa-H, TpPa-Cl₂, TpPa-SO₃H, and TpPa-(CH₃)₂, TpBD, and Tp-DTP COFs. Reproduced with permission from ref. 81 Copyright© 2021, Elsevier.

To modulate the photophysical properties of COFs via a bottom-up multicomponent design establishing the structure-to-activity relationships of multivariate COFs in terms of their photocatalytic hydrogen evolution, Guo and coworkers prepared a series of β-ketoenamine-based COFs by combining different molar ratios (x = 0, 0.05, 0.1, 0.25, 0.5, and 1) of 4,4′-(benzo-2,1,3-thiadiazole-4,7-diyl)dianiline (BT-NH₂) and 4,4′-diamino-p-terphenyl (TP) linkers with 1,3,5-trifluoroformylglucinol.¹⁵⁴ The photocatalytic studies of the COFs carried out under visible light (>420 nm) irradiation revealed the best water splitting performance by Tp(BT_0.05TP_0.95)-COF with a hydrogen evolution rate of 9839 μmol g⁻¹ h⁻¹. Herein, it was disclosed that the introduction of benzothiadiazole-derived units (chromophores) enhances the light-harvesting ability of COF, and due to its electron-withdrawing nature, effective separation and migration of photogenerated charge carriers was achieved, leading to improved photocatalytic hydrogen generation activity.

Recently, chirality has been shown to play a key role in determining the water splitting performance of materials through spin control. In further efforts, Guo and co-workers have investigated the potential of chiral β-ketoenamine-linked COF as artificial enzyme-like catalysts for visible photocatalytic hydrogen evolution. This work also presents a promising strategy for replacing platinum (Pt) as a cocatalyst in photocatalytic systems, thereby expanding the possibilities for developing efficient and sustainable artificial enzyme systems based on COF. In this contribution, achiral β-ketoenamine-linked COFs were synthesised by a solvothermal method involving aldimine condensation of 1,3,5-triformylphloroglucinol and diamine-substituted linkers (Fig. 29a).¹⁵⁵ Subsequently, copper ion complexation was achieved by introducing Cu(OAc)₂ into the aqueous dispersion of the COF (Fig. 29b). Photocatalytic hydrogen evolution in water was performed using the achiral TpPa-Cu(II)-COF photocatalyst under visible irradiation (λ > 420 nm) and L-cysteine as a sacrificial electron donor. Unfortunately, the TpPa-Cu(II)-COF with a copper loading of 10.76 wt% exhibited an inferior hydrogen evolution rate of 3640 μmol h⁻¹ g⁻¹ after 6 h of irradiation (Fig. 29c).

Fig. 29 Investigation of the role of COF chirality in determining water splitting performance. (a) Synthesis scheme of achiral TpPa-Cu(II)-COF. (b) Structural tautomerism and Cu-complexation with a subunit. (c) Hydrogen evolution performances of achiral TpPa-Cu(II)-COF. (d) Comparison of hydrogen evolution performances for chiral COF using L/D-cystine as a sacrificial electron donor. (e) Comparison of turnover frequency between chiral and achiral COFs. Reproduced with permission from ref. 155 Copyright© 2022, Springer Nature.

Further, to explore the role of chirality in COFs for hydrogen evolution, the chiral TpPa-COF was synthesised using a chiral regulator to yield different enantiomeric forms: TpPa(Δ)-COF and TpPa(Λ)-COF, which were characterised using polarimetry and electronic circular dichroism. Based on these results, photocatalytic tests were carried out on the chiral TpPa-Cu(II)-COF using L- and D-cysteine as sacrificial electron donors. Remarkably, after 6 h of visible irradiation, the enantiomeric mixtures of TpPa(Δ)-Cu(II)-COF/L-cysteine and TpPa(Λ)-Cu(II)-COF/D-cysteine exhibited impressive hydrogen evolution rates of 14 720 and 12 800 μmol h⁻¹ g⁻¹, respectively (Fig. 29d), which are among the highest reported values for non-precious metal–organic photocatalysts. In comparison, the HER of TpPa(Δ)-Cu(II)-COF/L-cysteine was about four times higher than that of achiral TpPa-Cu(II)-COF/L-cysteine (3640 μmol h⁻¹ g⁻¹). In particular, the TpPa(Δ)-Cu(II)-COF/D-cysteine system achieved an HER of 10 480 μmol h⁻¹ g⁻¹, whereas the TpPa(Λ)-Cu(II)-COF/L-cysteine system achieved an HER of 9480 μmol h⁻¹ g⁻¹ (Fig. 29e). The remarkable superiority of the chiral TpPa-COF can be attributed to its unique ability to allow parallel stacking of adjacent layers, with each atom in the upper layer aligned exactly with that in the lower layer. In contrast, the achiral TpPa-COF tends to stack its layers in an antiparallel fashion, as evidenced by the optimised layer structures. In the process of coupling the two adsorbed hydrogen (H) atoms to form H₂, the proximity between the H atoms plays a crucial role in determining the kinetic favourability. Consequently, the chiral TpPa-COF, with its parallel layer stacking facilitating closer proximity of the H atoms, exhibits enhanced kinetic favourability in the H₂ formation process, highlighting the significant influence of chirality on the photocatalytic performance.

4.5 Development of hybrid COFs for enhanced hydrogen generation

Hybrid materials are denoted as the most emergent materials at the edge of technological advancements, in which material properties are achieved through a synergetic combination of more than one component on the molecular scale. This makes hybrid materials interesting for several applications. It has been observed that the hydrogen generation activity of COFs can be enhanced significantly by hybridizing with a synergetic combination of different conductive or photoactive materials. In this context, the design of heterostructures composed of COFs and other materials, such as semiconductor metal oxides/sulfides, carbon-based materials, porous metal–organic frameworks (MOFs), etc., has been reported. In addition, the stable two-dimensional network of the COFs acts as a matrix to support the growth of nanoparticles. The heterojunctions formed favor the facile separation of the charge carriers by suppressing their recombination, resulting in higher photocatalytic efficiency. This section discusses various examples of COF-based hybrids reported for photocatalytic water splitting for hydrogen generation.

4.5.1 COF-metal oxide/sulfide-based hybrids. Cadmium sulfide (CdS) is a widely studied visible light active photocatalyst owing to its band gap (2.4 eV) and optical properties.¹⁵⁶ However, CdS is normally coupled with other materials to overcome the limitations of photocorrosion and to improve its photocatalytic properties. In this direction, Banerjee and coworkers developed a CdS-COF hybrid by utilizing a stable TpPa-2 COF as a matrix for CdS nanoparticle growth and investigated the hybrids for hydrogen generation activity.¹⁰⁷ TpPa-2 was obtained by a solvothermal reaction of 1,3,5-triformylphloroglucinol (Tp) and 3,6-dimethyl-1,4-diaminophenyl (Pa) as building blocks. The effect of varying the COF content on the photocatalytic efficiency of the hybrid was studied. Interestingly, the introduction of COF was found to increase the photocatalytic activity of the hybrid. Of the various compositions synthesized, the CdS-COF (90 : 10) hybrid with 10 wt% COF showed the best hydrogen evolution activity with a rate of 3678 μmol h⁻¹ g⁻¹, which is multiple times higher than that of the pure CdS catalyst (124 μmol h⁻¹ g⁻¹). The enhanced hydrogen generation activity of the hybrid was attributed to improved photostability and suppressed recombination of CdS photogenerated holes and electrons.

To enhance the photocatalytic hydrogen evolution efficiency, numerous efforts have been devoted. This includes the deposition of noble metals (e.g., Pt) and the addition of sacrificial reagents, which have been proven very effective. However, Pt is a rare and expensive noble metal; therefore, there is an interest in replacing Pt with a low-cost cocatalyst. In this context, the incorporation of a noble metal-free cocatalyst, Ni(OH)₂, in the TpPa-2 COF matrix was studied by Dong coworkers to boost the HER performance of the COF.¹⁵⁷ The various compositions of the hybrid, Ni(OH)₂-X%/TpPa-2 (X: molar fraction), were prepared by in situ synthesis of Ni(OH)₂ in the presence of a known amount of COF. Among the various composites prepared, Ni(OH)₂–2.5%/TpPa-2 with 2.5 mol% Ni(OH)₂ showed the best hydrogen generation activity with a rate of 1895 μmol h⁻¹ g⁻¹, which was found to be nearly 26 times higher than that of pure TpPa-2. The enhanced HER activity of the hybrid has been ascribed to the synergetic interaction between COF and Ni(OH)₂. Additionally, the metallic Ni formed by in situ reduction of Ni(OH)₂ facilitates the effective separation of photogenerated electron–hole pairs. Notably, the hydrogen evolution activity of the hybrid was nearly equivalent to that of 0.3 wt% Pt-loaded TpPa-2, highlighting the potential utility of Ni(OH)₂ as a promising noble metal-free cocatalyst.

In these attempts, Zhang and coworkers designed Z-scheme heterostructured hybrids by embedding hematite (α-Fe₂O₃) onto TpPa-2 COF for effective visible-light-driven photocatalytic water splitting (Fig. 30a).¹⁵⁸ The α-Fe₂O₃/TpPa-2 COF hybrid was prepared by adding α-Fe₂O₃ with the reaction of TpPa-2 COF precursors 1,3,5-triformylphloroglucinol and 3,6-dimethyl-1,4-diaminophenyl under solvothermal conditions. The photocatalytic studies of the hybrid showed the best water splitting activity for the α-Fe₂O₃/TpPa-2 COF (3 : 7) hybrid with a hydrogen evolution rate of 3.77 mmol h⁻¹ g⁻¹ in the absence of a noble metal cocatalyst (Fig. 30b and c). Interestingly, the activity was approximately 53 times higher than that of pristine TpPa-2-COF and higher than that of various reported Z-scheme photocatalysts. The higher HER performance of the hybrid has been assigned to its strong light-harvesting ability and formation of a Z-scheme heterojunction, which promotes efficient charge-carrier separation and migration for the reduction of protons to hydrogen. Later, Zhang and coworkers further extended this strategy for covalent anchoring of TiO₂ nanostructures into TpPa-1 COF by a Schiff base reaction between surface-functionalized –CHO groups on TiO₂ and –NH₂ groups in the TpPa-1 COF reactant.¹⁵⁹ Thus, various hybrids with varying TiO₂ : TpPa-1-COF ratios were prepared, and their HER activity was tested under visible light (420 nm) using Pt as a cocatalyst and sodium ascorbate as a sacrificial electron donor. Among the hybrids, TiO₂-TpPa-1 COF (1 : 3) showed the best hydrogen evolution rate of 11.19 mmol h⁻¹ g⁻¹. Mechanistic studies revealed that the covalent bonds in the hybrid serve as a bridge in promoting electron–hole separation, enhancing the HER performance.

Fig. 30 One-pot synthesis of the α-Fe₂O₃/TpPa-2-COF hybrid for water splitting. (a) Scheme of the synthesis of α-Fe₂O₃/TpPa-2-COFs. (b) Wavelength dependence of AQE for α-Fe₂O₃/TpPa-₂-COF (3 : 7). (c) Hydrogen evolution rates of hybrids with different compositions. Reproduced with permission from ref. 158 Copyright© 2020, Royal Society of Chemistry.

The same research group further studied the integration of MoS₂ with TpPa-1 COF to prepare a MoS₂/TpPa-1 COF composite by in situ growth of COF in an exfoliated MoS₂ dispersion (Fig. 31a).¹⁶⁰ Various compositions were prepared by varying the loading of MoS₂ from 1 to 5 wt% and further tested for photocatalytic water splitting. The optimized composite with 3 wt% loading of MoS₂ exhibited a hydrogen evolution rate of 55.85 μmol h⁻¹ without the use of a noble metal cocatalyst under visible light (Fig. 31d and e). Notably, the HER performance of the hybrid was 32 times higher than that of pristine TpPa-1 COF (1.72 μmol h⁻¹). Herein, the introduction of MoS₂ has been found to enhance the charge separation for photogenerated electron–hole pairs, resulting in enhanced HER performance (Fig. 31b). The mechanism involves the transfer of photoinduced electrons from the COF to MoS₂ and their utilization for water reduction for hydrogen generation (Fig. 31c).

Fig. 31 The preparation and applications of MoS₂/TpPa-1-COF hybrids. (a) Synthetic route for the preparation of the MoS₂/TpPa-1-COF hybrid. (b) EPR spectra of TpPa-1-COF and MoS2-3%/TpPa-1-COF with light on and off. (c) Schematic illustration of photocatalytic hydrogen evolution under visible light irradiation using the MoS₂/TpPa-1-COF composite. (d) Photocatalytic hydrogen generation rates for TpPa-1 and MoS₂/TpPa-1 COF composites with variable MoS₂ content. (e) Comparison of the photocatalytic activity of TpPa-1-COF, MoS₂-3%/TpPa-1-COF, Pt-3%/TpPa-1-COF and physically mixed (3%) material. Reproduced with permission from ref. 160 Copyright© 2019, Royal Society of Chemistry.

Similarly, CTF-based hybrid systems have been explored for photocatalytic water splitting. Since CTFs are generally prepared at high temperatures and using ZnCl₂, the preparation of hybrids using these materials has not been explored much. In these attempts, Luo and coworkers applied a strategy to enhance the photocatalytic efficiency of CTF-1 by appending black phosphorus (BP) as a bridge joint to generate a sandwich-type Pt-containing CTF system (CTF-BP-Pt).¹⁶¹ Interestingly, CTF-BP-Pt exhibited a higher hydrogen evolution rate of 614.6 μmol h⁻¹ g⁻¹ over the CTF/Pt (167.5 μmol h⁻¹ g⁻¹) sample with directly doped Pt and the physical mixture of CTF-BP + Pt (95.1 μmol h⁻¹ g⁻¹). The superior hydrogen generation activity of CTF-BP-Pt has been attributed to the important role of black phosphorus nanosheets in enhancing the stability of Pt nanoparticles by effectively suppressing their aggregation.

CdS-based photocatalysts such as nanosized CdS powder, solid solutions and quantum dots have been widely explored for photocatalytic hydrogen generation due to their strong visible light absorption, suitable band edge levels and electronic charge transfer properties. Wu and coworkers reported a facile in situ growth of CdS quantum dots (QDs) on CTF-1 via a facile photoreduction method to further improve their hydrogen evolution performance in visible light.¹⁶² The CdS@CTF-1 composite displayed higher HER performance with a rate of 27.8 μmol h⁻¹, which was approximately four times higher than that of pure CdS. The enhanced HER activity of the hybrid has been attributed to the synergistic interaction between CTF-1 and CdS, facilitating enhanced separation and transport of photoinduced charge carriers. In similar attempts, Wu and coworkers modified CTF-1 with reduced graphene oxide (rGO), and CTF-1/rGO-x hybrids with different loadings of rGO (x = 1, 2, 5, and 10 wt%) were obtained by using a facile UV reduction-based strategy.¹⁶³ Interestingly, the hydrogen evolution rate of the composites was found to vary with the graphene oxide content, and the best hydrogen evolution activity with a rate of 894 μmol g⁻¹ h⁻¹ was observed for the CTF-1/rGO-2 composite, which was approximately 4.3-fold higher than that of pristine CTF.

Nanostructured transition metal phosphides (Ni, Co, Mo, W) have attracted high scientific interest due to their unique physical and chemical properties. To take advantage of those properties, Fan and coworkers developed a Ni₂P/CTF composite and studied the effect of varying Ni₂P loading on the hydrogen evolution performance.¹⁶⁴ Notably, the rate of hydrogen evolution in the hybrid was found to increase gradually with an increase in the Ni₂P loading and then decrease at higher loadings. The hybrid with 2.8% Ni₂P loading showed the best hydrogen evolution activity with a rate of 5844 μmol g⁻¹ h⁻¹ and AQE of 3.8%. The higher hydrogen evolution performance of the hybrid has been attributed to the facile transfer of photogenerated electrons from CTF to Ni₂P, resulting in effective charge separation and utilization. Interestingly, the hydrogen evolution rate of 2.8% Ni₂P/CTF was equivalent to that of 3.0% Pt-loaded CTF, highlighting the importance of Ni₂P as a promising noble-metal-free cocatalyst for enhanced visible light-assisted hydrogen generation.

4.5.2 MOF–COF hybrids. Recently, the hybridization of MOFs and COFs has attracted significant interest owing to the unique advantages of long-range ordering, high surface area, and excellent visible light harvesting properties of both MOFs and COFs.¹⁶⁵ In this regard, Lan and coworkers developed a new class of MOF/COF hybrid materials by covalently anchoring NH₂-UiO-66 on the surface of TpPa-1-COF and tested their photocatalytic hydrogen evolution.¹⁶⁶ Herein, TpPa-1 COF was used to construct the composite, considering its rigid structure formed via Schiff-base linkage, and NH₂-UiO-66 MOF was chosen due to its photocatalytic properties. The hybrid was prepared by introducing NH₂-UiO-66 during the synthesis of TpPa-1 COF (Fig. 32a). Notably, the NH₂-UiO-66/TpPa-1 COF hybrids feature a crystalline porous framework structure with a high surface area. Owing to the matching band gaps between COF and MOF, the hybrids exhibit efficient visible-light-assisted photocatalytic hydrogen evolution activity with a rate of 23.41 mmol g⁻¹ h⁻¹, which was approximately 20 times higher than that of pristine TpPa-1 COF (Fig. 32b). The higher photocatalytic efficiency of the hybrid has been attributed to the formation of a COF/MOF heterojunction that facilitates the transfer of photoinduced electrons from the COF to the MOF, at which the reduction of water to hydrogen takes place (Fig. 32c). It is worth highlighting that the hydrogen evolution activity of these MOF/COF hybrids was found to be superior to that of many previously reported examples of MOFs and COFs (Table 3).

Fig. 32 Preparation of MOF-COF composites for water splitting. (a) Scheme of the synthesis of the NH₂-UiO-66/TpPa-1-COF hybrid. (b) Hydrogen evolution performance of the NH₂-UiO-66/TpPa-1-COF hybrid with different compositions. (c) Proposed mechanism for H₂ generation using the NH₂-UiO-66/TpPa-1-COF hybrid. Reproduced with permission from ref. 166 Copyright© 2018 Wiley-VCH.

In an interesting strategy, Yan and coworkers utilized postsynthetic modification of metal nodes from MOFs to prepare MOF/COF hybrids for photocatalytic water splitting. To prepare the hybrid, a Zr-MOF (MOF-808) without –NH₂-containing ligands was first prepared and further modified by functionalization with p-aminobenzoic acid onto Zr₆ clusters to anchor –NH₂ groups.¹⁶⁷ Then, the MOF-808@TpPa-1-COF core–shell structure was prepared by an in situ condensation reaction via covalent linkage between the COF precursors (Fig. 33a). Different MOF@COF hybrids were prepared by varying the weight ratios of the components. The hybrid possessing high crystallinity and hierarchical pore structure was applied for photocatalytic water splitting, where a hydrogen evolution activity as high as 11.88 mmol h⁻¹ g⁻¹ was revealed for the MOF-808@TpPa-1 COF (6/4) hybrid. This activity was approximately 5.6 times higher than that of pure TpPa-1 COF and the physical mixture (Fig. 33b and c). The higher photocatalytic efficiency of the hybrid has been correlated to facile separation and transport of photogenerated charge carriers at the heterojunction.

Fig. 33 (a) Synthesis scheme of the MOF-808@TpPa-1-COF hybrid. (b and c) Hydrogen evolution performance of the hybrids. Reproduced with permission from ref. 167 Copyright© 2021, Royal Society of Chemistry.

4.5.3 COF-carbon-based hybrids. The application of graphitic carbon nitride (g-C₃N₄) as a photocatalyst has gained significant interest owing to its visible-light harvesting property (band gap of 2.7 eV), nontoxicity, low cost, good stability and metal-free nature.¹⁶⁸ However, the photocatalytic activity of bare g-C₃N₄ is inhibited by a high charge recombination rate, which can be overcome by functionalization with suitable materials to form heterojunction photocatalysts with improved photocatalytic performance. In this regard, the first g-C₃N₄-modified CN-COF hybrid was prepared by in situ solvothermal reactions of g-C₃N₄ with 1,3,5-triformylphluoroglucinol (TP) and 4,4′,4′′-(1,3,5-triazine-2,4,6-triyl)trianiline (TTA) moieties.¹¹⁰ Different ratios of COF : CN were prepared, and their HER activity was investigated under visible light irradiation (λ > 420 nm). The optimized catalyst with 0.3 wt% COF showed the best hydrogen evolution rate of 10.1 mmol h⁻¹ g⁻¹, which was approximately 6.7 and 44.7 times higher than g-C₃N₄ and COF, respectively. The imine linkages between g-C₃N₄ and the COF material provide higher absorption in the visible region and reduce charge recombination with efficient charge transfer. Thus, the improved HER performance of the hybrid has been ascribed to enhanced broader absorption of visible light and effective separation and migration of photoexcited electron–hole pairs.

To construct heterojunction photocatalysts with multidimensional features, Li and coworkers designed hybrid photocatalysts composed of β-ketoenamine-based COF and g-C₃N₄ with varying mass ratios (1 : x; x = 2.5, 5, 10, 15, 20).¹⁶⁹ The composites were prepared via in situ reaction of 2,4,6-triformylphloroglucinol and benzidine building blocks in the presence of a stripped g-C₃N₄ suspension. The photocatalytic study of the composites showed the best HER activity by COF-CN (1 : 10) with a rate of 384 μmol h⁻¹, which was found to be much higher than that of pure TpBD COF (1.35 μmol h⁻¹) and g-C₃N₄ (6.24 μmol h⁻¹). It is worth noting that the apparent quantum efficiency (AQE) for hydrogen generation by COF-CN (1 : 10) was approximately 15% (500 nm), which is the highest value among previously reported COFs or g-C₃N₄-based systems. The superior HER performance of the COF-CN hybrid has been attributed to the formation of a type-II heterojunction facilitating facile separation of photogenerated electron–hole pairs, resulting in enhanced photocatalytic activity. Similarly, Jiang and coworkers developed a COF/g-C₃N₄ hybrid by utilizing a donor–acceptor type TBTA COF through in situ syntheses by the condensation of 2,4,6-triformylphloroglucinol and 4,4′-(benzo-1,2,5-thiadiazole-4,7-diyl)dianiline (BTDA) in the presence of g-C₃N₄.¹⁷⁰ The obtained hybrids possess optimal band edge positions suitable for effective water splitting and hence exhibit better photocatalytic activity than the individual components. Notably, the hybrid with the TBTA COF : g-C₃N₄ (2.5 : 1) composition showed the best activity with a hydrogen evolution rate of 11.73 mmol g⁻¹ h⁻¹ in the absence of a noble metal cocatalyst. Interestingly, the rate of hydrogen generation was further enhanced to 26.04 mmol g⁻¹ h⁻¹ with the addition of the Pt cocatalyst, highlighting the importance of hybrid systems for photocatalysis.

The construction of covalently bonded COF heterostructures has proven to be an effective way to improve photocatalytic HER performance. Toward the development of such heterostructures, the preparation of hybrids and Z-schemes using TiO₂, g-C₃N₄, metal hydroxides, etc., have been attempted. Additionally, many hybrid materials have explored reduced graphene oxide (rGO) with high electron conductivity and a large surface area as an electron acceptor and mediator. To take advantage of the two-dimensional structure, porosity and conductivity of reduced graphene oxide, Zhang and coworkers developed a covalently connected rGO-COF (TpPa-1) hybrid photocatalyst for enhanced hydrogen evolution (Fig. 34a).¹⁰⁸ The hybrids were prepared by following a one-pot synthetic route by taking a known amount of GO during the preparation of TpPa-1 COF using DMF solvent. During the synthesis, simultaneous reduction and covalent functionalization of GO with COF take place. Visible-light-assisted water splitting studies using these composites revealed that the rGO-TpPa-1 COF hybrid with a rGO composition of 5% showed the best hydrogen evolution rate of 11.98 mmol g⁻¹ h⁻¹, which was approximately 4.85 and 2.50 times higher than those of the pure TpPa-1-COF and 5%rGO/TpPa-1-COF physical mixture, respectively (Fig. 34b and c). The mechanistic investigation demonstrated that the covalent grafting of rGO with TpPa-1 COF in the rGO-TpPa-1 hybrid can act as an electron collector and transporter, promoting the separation of photogenerated charges and transferring the active electrons (Fig. 34d). In addition, the sheet morphology of the rGO-TpPa-1 COF hybrid with a uniform distribution of TpPa-1 on the rGO layers facilitates the migration of photogenerated electrons, demonstrating the importance of COF-based hybrids with strong bonds for efficient photocatalytic reactions.

Fig. 34 Synthesis of COF-reduced graphene oxide hybrids for photocatalysis. (a) Scheme of the synthesis of the rGO-TpPa-1 COF hybrid. (b) The photocatalytic hydrogen evolution performance over five hours of a series of rGO (x%)-TpPa-1-COF hybrids. (c) Recycling test for photocatalytic hydrogen evolution of rGO (5%)-TpPa-1-COF. (d) Transient photocurrent response measurements of TpPa-1-COF and rGO (5%)-TpPa-1-COF vs. SCE. Reproduced from ref. 108 with permission Copyright© 2020, Royal Society of Chemistry.

Two-dimensional COFs with periodic structures, extended π-conjugation and layered stacking have emerged as promising materials for photocatalytic hydrogen evolution. However, the layer-by-layer assembly in 2D COFs is unstable during photocatalytic water splitting, resulting in disordered stacking and decreased HER performance. To address this concern, in a major development in the design of COF-based hybrids, Guo and coworkers prepared polyethylene glycol (PEG)-incorporated COF hybrids by postsynthetic modification of benzothiadiazole-containing BT-COF (Fig. 35a).¹⁷¹ The filling of the open channels of BT-COF with PEG led to the retention of the ordered stacking during the visible light-assisted photocatalytic reaction (Fig. 35b and c). This strengthening of interlayer π-stacking resulted in enhanced performance of PEG-filled BT-COF for water splitting and retained its structural stability over photocatalytic cycling (Fig. 35d). Thus, the PEG@BT-COF composite showed efficient photocatalytic performance with a hydrogen generation rate of 11.14 mmol h⁻¹ g⁻¹, which is higher than that of pure BT-COF (7.70 mmol h⁻¹ g⁻¹) (Fig. 35e and f). Notably, without PEG stuffing, the crystallinity and stability of the COF were reduced during photocatalysis, resulting in lower hydrogen evolution performance (Fig. 35e). Herein, Pt nanoparticles grown on PEG chains promote efficient photocatalytic performance. Thus, incorporating PEG in the one-dimensional pore channels of the COF enhanced the hybrid's stability and photocatalytic performance (Fig. 35g). This work highlights the importance of the rational design of COF-based hybrid catalysts to achieve enhanced photocatalytic water splitting efficiency.

Fig. 35 Synthesis of PEG-stabilized 2D COFs. (a) Synthesis of BT-COF through a Schiff-base reaction. (b) HR-TEM image of BT-COF showing one-dimensional channels. (c) Scheme of transformation of COF and PEG@COF during Pt-cocatalyst deposition. (d) PXRD patterns for the as-prepared BT-COF, the Pt-deposited BT-COF before and after 48 h of cycling, and the recycled 30% PEG@BT-COF after 48 h of photocatalysis. (e) Photocatalytic hydrogen evolution under visible irradiation for BT-COF, BT-COF (HOAc), 30% PEG@BT-COF, poly(TpBT), and TP-COF. (f) The apparent quantum efficiency of BT-COF and 30%PEG@BT-COF. (g) Photocatalytic hydrogen generation using BT-COF and 30%PEG@BT-COF. Reproduced with permission from ref. 171 Copyright© 2021, Springer Nature.

5. Structure–property relationship among COF structures and photocatalytic performance

The use of photocatalysts for water splitting to hydrogen generation is the cornerstone of photocatalysis. Many efforts have been devoted to enhancing the light absorption, charge separation, and lifetime of charge carriers to overcome their recombination process to render improved photocatalytic efficiency. There are many important parameters, such as crystallinity, porosity, band gap, band positions, visible light absorption, surface wettability, charge transport, acidity, basicity, donor–acceptor structures, planarity, surface modification, chemical stability, functionalization, modular nature, presence of reactive sites and functional groups, structural designability, two-dimensional structure, π–π stacking, etc., govern the photocatalytic water splitting performance of the COFs (Fig. 36).

Fig. 36 Most important properties responsible for determining the photocatalytic water splitting performance of COFs (generated using Word Cloud Generator by Monkey Learn).

Among these properties, some of the important parameters that determine the water-splitting performance and apparent quantum efficiency of COFs are as follows:

❖ Crystallinity: crystallinity is defined as the degree of long-range structural ordering comprising a crystal lattice within a material. The photocatalytic properties are strongly governed by the crystallinity of photocatalysts.¹⁷⁸ An increase in the crystallite sizes has a strong impact on enhancing the photocatalytic rate, exhibiting low trap density and long diffusion length and thus demonstrating high catalytic performance. Highly crystalline cocatalysts show good photocatalytic performance due to their favorable charge transport properties.

❖ Porosity: porosity is a measure of the void spaces in a material, and a catalyst facilitates the exposure of active sites for effective photocatalytic hydrogen generation performance. It is well known that hierarchically porous structures are desirable for photocatalysis, as these materials provide more active sites, easy mass transport and shorten the carrier diffusion length between the carrier-generated centers and the active sites. Additionally, high porosity offers easy access to photoactive groups, solvents, sacrificial electron donors, cocatalysts, etc. – crucial components to achieve efficient water splitting.

❖ Band gap: generally, photocatalytic processes are initiated when a semiconductor photocatalyst absorbs photons with energy equal to or higher than its optical band gap. The photocatalytic water splitting performance of a photocatalyst strongly depends on its band-gap energy, which, in principle, should be smaller than 3 eV to extend the light absorption into the visible region, efficiently utilizing solar energy for hydrogen generation. In addition, the minimum band gap of semiconductor photocatalysts for water splitting should be 1.23 eV. Overall, the ideal band gap of COFs should be approximately 2.0 eV to effectively utilize solar energy.

❖ Band positions: in addition to the band gap, the optimum valence band and conduction band positions relative to the water redox potential are basic requirements of photocatalytic materials, principally determining the feasibility of hydrogen and oxygen evolution from water splitting under light illumination. Mainly, hydrogen and oxygen evolution reactions are known to occur when the conduction band edge of the photocatalyst is more negative than the reduction potential of H⁺/H₂ and the valence band edge is more positive than the oxidation potential of O₂/H₂O.

❖ Surface wettability: surface wettability is a measurement of surface energy that influences the degree of contact with the physiologic environment. It is considered one of the critical parameters affecting the photocatalytic performance of COFs, which dominates the adsorption and desorption processes of reactants and products. The better surface wettability of COFs with water helps to enhance the dispersion of materials (e.g., sulfonyl COFs) and interaction among the active sites (e.g., ether or hydroxyl groups) present in the COF matrix with water, increasing the photocatalytic water splitting performance for hydrogen generation many-fold.

❖ Donor–acceptor structures: donor–acceptor (D–A) materials are constructed using an alternating array of donor and acceptor moieties. Two-dimensional COFs with π–π stacking among the adjacent layers can promote exciton migration and charge transport through the aligned one-dimensional π-columnar arrays, exhibiting attractive promise in photocatalytic applications. However, the photocatalytic performances are generally obstructed by insufficient charge separation and fast charge carrier recombination. To overcome these challenges, the design of donor–acceptor COFs with an intrinsic heterostructure of spatially separated donor and acceptor columns is essential. These D–A systems support effective charge separation with high electron mobility by facile transfer of photoexcited electrons from the donor molecule to the acceptor. However, strong donor and acceptor moieties are essential to boost facile charge migration. Hence, rational construction of donor–acceptor structures offers an intriguing option to enhance the photocatalytic hydrogen generation efficiency of COFs.

❖ Presence of photoactive moieties: by integrating photoactive moieties, the pore structures and electronic properties of two-dimensional COFs can be tuned, which can be further employed to enhance the photocatalytic properties of the COFs. To achieve efficient light absorption, generation, and separation/migration of excitons to the photocatalyst surface, the integration of various functional groups, donor–acceptor units, and photoactive moieties has been attempted. In these regards, generally, the light absorption can be extended toward the visible or infrared region by incorporating photoactive moieties such as triazine, heptazine, pyrene, acetylene, diacetylene, sulfonyl, and benzothiadiazole.

❖ Functionalization: the modular synthetic routes used for COFs allow the incorporation of suitable cocatalysts such as metals, metal oxide/sulfide and metal phosphide to achieve enhanced HER performance. Furthermore, the construction of heterojunctions, such as type II or Z-scheme systems, is a promising approach to facilitate enhanced photocatalytic hydrogen generation, aided by the easy separation of photogenerated carriers at the interface.

6. Summary and perspective

In summary, the literature covered in this review highlighted the importance of COF-based photocatalysts for visible-light-driven photocatalytic hydrogen generation. The modular nature of these materials facilitates rational incorporation of desired functionalities by judicious choice of organic linkers to obtain COFs with high surface area, optimal absorption of visible light and charge transport properties suitable for enhanced water splitting activity (Table 1). Furthermore, band gap engineering of COFs is a useful strategy to tailor the photocatalytic properties of COFs. Hence, the nature of the organic linkers employed will guide the structure and function of the resulting COFs and their photocatalytic performance toward efficient water splitting for hydrogen generation (Table 2). Thus, understanding the structure–property correlation is essential in developing COFs with promising photocatalytic properties for hydrogen evolution from water. From the aforementioned discussion, it is clear that the poor hydrogen generation performance of COFs stems from the loss of photogenerated electron–hole pairs by recombination. In this regard, the application of cocatalysts was proven to be essential to suppress this recombination of charge carriers and to facilitate facile migration of photogenerated electrons to the cocatalyst for effective hydrogen generation. In this regard, COFs composed of extended π-conjugation and donor–acceptor type linkers are promising for facilitating facile transport of photogenerated electrons by a push–pull mechanism from the donor moiety to the acceptor molecule and then effective transfer to the cocatalyst for the reduction of protons for hydrogen evolution.

Another major bottleneck that hampers photocatalytic efficiency is the loss of photogenerated charge carriers due to the presence of surface defects and imperfections. To overcome these challenges, the COF materials employed should possess good crystallinity, yielding defect-free and fewer grain boundaries. In this regard, the synthetic method employed to prepare COFs plays a vital role in generating COFs with high crystallinity. As discussed in the previous section, the COFs synthesized using the solvothermal route are found to possess high crystallinity over those prepared by other synthetic methods.

In addition, optimal absorption of light in the broader region of the solar spectrum covering the UV-Vis to near-infrared region (NIR) is recommended to achieve enhanced photocatalytic efficiency. However, most of the COFs reported thus far are active mainly in the UV-Vis region; hence, the future design of COFs should consider incorporating suitable functionalities/chromophores capable of harvesting light in the near-infrared (NIR) region. The fabrication of heterojunction photocatalysts can achieve this by coupling COFs with other suitable materials to generate composites with improved light-harvesting properties in the broad region of sunlight (Table 3). Furthermore, the COFs employed for photocatalysis should exhibit high photostability and long-term durability for sustainable water splitting to hydrogen evolution. In this context, heterojunction materials constructed by covalent attachment of COFs with 2D carbon-based materials such as g-C₃N₄, rGO, and graphene are found to be promising candidate materials for accomplishing improved catalytic efficiency.^44,172 The resulting 2D/2D heterostructures possess promising features of improved light absorption, facile charge separation and migration aided by heterojunction formation. In addition, covalent functionalization provides extra stability to the COF network, rendering high photostability and long-term durability for multiple photocatalysis cycles. Recently, hybrids composed of COFs and MOFs have also gained significant interest, as they incorporate unique features of crystallinity, porosity, and light harvesting essential for an effective photocatalytic system. Hence, the future design of COFs should consider heterogenization with suitable materials for improved photocatalytic performance for the HER.^173–175 However, it is worth mentioning that the best hydrogen generation activity of COFs has been achieved thus far in the presence of noble metal platinum cocatalysts, which limits their large-scale applications. Thus, the development of COFs exhibiting high H₂ generation activity with non-Pt metal-based cocatalysts is highly desired. In this direction, the application of non-noble metal-based oxides, sulfides and phosphides with superior HER performance offers potential significance and should be considered in the future design of COF-based photocatalysts for hydrogen generation.¹⁷⁶

Despite considerable efforts in designing various COF-based photocatalysts for hydrogen generation, the highest hydrogen evolution rate achieved thus far is not the best compared to conventional semiconductor materials. Furthermore, for practical applications of hydrogen as a clean energy source, the solar-to-fuel conversion efficiency needs to be enhanced along with the long-term durability of the catalyst. In addition, for large-scale implementation of hydrogen generation, several factors, such as the cost of the COF material, suitable reactor design, optimal exposure of sunlight on the photocatalyst, and facile separation of the catalyst from the reaction mixture, need to be optimized. In this regard, recent literature reports have demonstrated the application of planar panels of photocatalytic materials by coating powder material on suitable supports for large-scale hydrogen evolution applications.¹⁷⁷ To achieve large-scale hydrogen generation, the large-scale synthesis of COFs with high crystallinity and porous structure needs to be optimized. Unfortunately, the high cost of large-scale preparation of COF semiconducting materials limits photocatalytic hydrogen production on an industrial scale. In this respect, the use of building blocks derived from natural sources or low-cost chemicals (e.g. urea, melamine, etc.) can reduce the overall cost of COF photocatalysts. In addition, to further reduce the cost, a photocatalytic water splitting system operating without a cocatalyst and sacrificial electron donor would be ideal. Moreover, precise knowledge of the thermodynamic aspects of the photocatalytic hydrogen generation mechanism can provide deep insight into the problem that can be helpful for the development of highly efficient COFs. Overall, the COFs and COF-based hybrid materials show great promise as photocatalysts for efficient visible light-mediated hydrogen evolution, and it is anticipated that future research in this area will benefit the advancement of COFs with optimal properties for large-scale photocatalytic water splitting to achieve sustainable generation of green hydrogen and toward the realization of a hydrogen economy.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

K. P. acknowledges IIT Ropar for the institute's postdoctoral fellowship. C.M.N. acknowledges IIT Ropar and DST-SERB (grant no. CRG/2022/006762). P. P. acknowledges the financial support from S. N. Bose National Centre for Basic Sciences and the Department of Science & Technology (No. SRG/2022/000217). P. P. and D. D. D. thank the Spanish Ministry of Science and Innovation for the project TED2021-132847B-100. D. D. D. thanks NANOtec, INTech, Cabildo de Tenerife and ULL for laboratory facilities. P. P. and B. M. would also like to thank Prof. Tanusri Saha-Dasgupta (Director) and Prof. Rajib Kumar Mitra (Head of Department) for their support, and Prof. Goutam De for the scientific discussion.

References

1. K. Calautit, D. S. N. M. Nasir and B. R. Hughes, Renewable Sustainable Energy Rev., 2021, 147, 111230.
2. R. Cassia, M. Nocioni, N. C. Aragunde and L. Lamattina, Front. Plant Sci., 2018, 9, 473.
3. O. Ellabban, H. A. Rub and F. Blaabjerg, Renewable Sustainable Energy Rev., 2014, 39, 748–764.
4. M. Kokkonen, P. Talebi, J. Zhou, S. Asgari, S. A. Soomro, F. Elsehrawy, J. Halme, S. Ahmad, A. Hagfeldt and S. G. Hashmi, J. Mater. Chem. A, 2021, 9, 10527–10545.
5. J. Y. Kim, J. W. Lee, H. S. Jung, H. Shin and N.-G. Park, Chem. Rev., 2020, 120, 7867–7918.
6. K. Rogdakis, N. Karakostas and E. Kymakis, Energy Environ. Sci., 2021, 14, 3352–3392.
7. Y. Ma, Y. Ma, Q. Wang, S. Schweidler, M. Botros, T. Fu, H. Hahn, T. Brezesinski and B. Breitung, Energy Environ. Sci., 2021, 14, 2883–2905.
8. E. B. Agyekum, C. Nutakor, A. M. Agwa and S. Kamel, Membranes, 2022, 12, 173.
9. M. Yue, H. Lambert, E. Pahon, R. Roche, S. Jemei and D. Hissal, Renewable Sustainable Energy Rev., 2021, 146, 111180.
10. J. Luo, S. Zhang, M. Sun, L. Yang, S. Luo and J. C. Crittenden, ACS Nano, 2019, 13, 9811–9840.
11. C. Acar, I. Dincer and G. F. Naterer, Int. J. Energy Res., 2016, 40, 1449–1473.
12. C. Acer and I. Dincer, J. Cleaner Prod., 2019, 218, 835–849.
13. J. Qi, W. Zhang and R. Cao, Adv. Energy Mater., 2018, 8, 1701620.
14. W. Liu, Y. Cui, X. Du, Z. Zhang, Z. Chao and Y. Deng, Energy Environ. Sci., 2016, 9, 467–472.
15. A. Alnouss, G. McKay and T. Al-Ansari, Appl. Energy, 2020, 266, 144885.
16. M. Watanabe, H. Inomata and K. Arai, Biomass Bioenergy, 2021, 22, 405–410.
17. J. V. Karaeva, Int. J. Hydrogen Energy, 2021, 46, 34089–34096.
18. A. Fujishima and K. Honda, Nature, 1972, 238, 37–38.
19. Y. P. Xie, G. S. Wang, E. L. Zhang and X. Zhang, Chin. J. Inorg. Chem., 2017, 33, 177–209.
20. M. Humayun, F. Raziq, A. Khan and W. Luo, Green Chem. Lett. Rev., 2018, 11, 86–102.
21. N. Cheng, S. Stambula, D. Wang, M. N. Banis, J. Liu, A. Riese, B. Xiao, R. Li, T.-K. Sham, L.-M. Liu, G. A. Botton and X. Sun, Nat. Commun., 2016, 7, 13638.
22. H. Nishiyama, T. Yamada, M. Nakabayashi, Y. Maehara, M. Yamaguchi, Y. Kuromiya, Y. Nagatsuma, H. Tokudome, S. Akiyama, T. Watanabe, R. Narushima, S. Okunaka, N. Shibata, T. Takata, T. Hisatomi and K. Domen, Nature, 2021, 598, 304–307.
23. Q. Wang, M. Nakabayashi, T. Hisatomi, S. Sun, S. Akiyama, Z. Wang, Z. Pan, X. Xiao, T. Watanabe, T. Yamada, N. Shibata, T. Takata and K. Domen, Nat. Mater., 2019, 18, 827–832.
24. A. Ishikawa, T. Takata, T. Matsumura, J. N. Kondo, M. Hara, H. Kobayashi and K. Domen, J. Phys. Chem. B, 2004, 108, 2637–2642.
25. A. Ishikawa, T. Takata, J. N. Kondo, M. Hara, H. Kobayashi and K. Domen, J. Am. Chem. Soc., 2002, 124, 13547–13553.
26. G. Liao, Y. Gong, L. Zhang, H. Gao, G.-J. Yang and B. Fang, Energy Environ. Sci., 2019, 12, 2080–2147.
27. V. W.-h. Lau, M. B. Mesch, V. Duppel, V. Blum, J. Senker and B. V. Lotsch, J. Am. Chem. Soc., 2015, 137, 1064–1072.
28. J. A. Nasir, Z. U. Rehman, S. N. A. Shah, A. Khan, I. S. Butler and C. R. A. Catlow, J. Mater. Chem. A, 2020, 8, 20752–20780.
29. B. Debnath, S. Dhingra and C. M. Nagaraja, Sol. RRL, 2021, 5, 2100226.
30. S. Dhingra, M. Sharma, V. Krishnan and C. M. Nagaraja, J. Colloid Interface Sci., 2022, 615, 346–356.
31. M. Kaur and C. M. Nagaraja, ACS Sustainable Chem. Eng., 2017, 5, 4293–4303.
32. P. Wu, J. Wang, J. Zhao, L. Guo and F. E. Osterloh, Chem. Commun., 2014, 50, 15521–15524.
33. X. Wang, K. Maeda, X. Chen, K. Takanabe, K. Domen, Y. Hou, X. Fu and M. Antonietti, J. Am. Chem. Soc., 2009, 131, 1680–1681.
34. L. Shen, Z. Xing, J. Zou, Z. Li, X. Wu, Y. Zhang, Q. Zhu, S. Yang and W. Zhou, Sci. Rep., 2017, 7, 41978.
35. T.-X. Wang, H.-P. Liang, D. A. Anito, X. Ding and B.-H. Han, J. Mater. Chem. A, 2020, 8, 7003–7034.
36. Y. Deng, Y. Wang, Y. Chen and Z. Zhang, Chem.–Asian J., 2021, 16, 1851–1863.
37. H. Kim, N. Kim and J. Ryu, Inorg. Chem. Front., 2021, 8, 4107–4148.
38. T. Zhang, G. Xing, W. Chen and L. Chen, Mater. Chem. Front., 2020, 4, 332–353.
39. L. Tan and B. Tan, Chem. Soc. Rev., 2017, 46, 3322–3356.
40. N. B. McKeown, S. Makhseed and P. M. Budd, Chem. Commun., 2002, 2780–2781.
41. Y. Yuan and G. Zhu, ACS Cent. Sci., 2019, 5, 409–418.
42. M. Liu, L. Guo, S. Jin and B. Tan, J. Mater. Chem. A, 2019, 7, 5153–5172.
43. A. P. Côté, A. I. Benin, N. W. Ockwig, M. O'Keeffe, A. J. Matzger and O. M. Yaghi, Science, 2005, 310, 1166–1170.
44. C. S. Diercks and O. M. Yaghi, Science, 2017, 355, 923–931.
45. S. Yanagida, A. Kabumoto, K. Mizumoto, C. Pac and K. Yoshino, J. Chem. Soc., Chem. Commun., 1985, 474–475.
46. D.-L. Jiang, C.-K. Choi, K. Honda, W.-S. Li, T. Yuzawa and T. Aida, J. Am. Chem. Soc., 2004, 126, 12084–12089.
47. M. G. Schwab, M. Hamburger, X. Feng, J. Shu, H. W. Spiess, X. Wang, M. Antoniettib and K. Müllen, Chem. Commun., 2010, 46, 8932–8934.
48. S. Haldar, M. Wang, P. Bhauriyal, A. Hazra, A. H. Khan, V. Bon, M. A. Isaacs, A. De, L. Shupletsov, T. Boenke, J. Grothe, T. Heine, E. Brunner, X. Feng, R. Dong, A. Schneemann and S. Kaskel, J. Am. Chem. Soc., 2022, 20, 9101–9112.
49. N. Haung, P. Wang and D. Jiang, Nat. Rev. Mater., 2016, 1, 16068.
50. Z. Wang, L. Liu, Z. Li, N. Goyal, T. Du, J. He and G. K. Li, Energy Fuels, 2022, 6, 2927–2944.
51. B. He, Q. Zhang, Z. Pan, L. Li, C. Li, Y. Ling, Z. Wang, M. Chen, Z. Wang, Y. Yao, Q. Li, L. Sun, J. Wang and L. Wei, Chem. Rev., 2022, 11, 10087–10125.
52. X. Guan, Q. Fang, Y. Yan and S. Qiu, Acc. Chem. Res., 2022, 55, 1912–1927.
53. K. T. Tan, S. Ghosh, Z. Wang, F. Wen, D. Rodríguez-San-Miguel, J. Feng, N. Huang, W. Wang, F. Zamora, X. Feng, A. Thomas and D. Jiang, Nat. Rev. Methods Primers, 2023, 3, 1.
54. R. Liu, K. T. Tan, Y. Gong, Y. Chen, Z. Li, S. Xie, T. He, Z. Lu, H. Yanga and D. Jiang, Chem. Soc. Rev., 2021, 50, 120–242.
55. T. Skorjanc, D. Shetty, F. Gándara, Li. Ali, J. Raya, G. Das, M. A. Olson and A. Trabolsi, Chem. Sci., 2020, 11, 845–850.
56. T. Zhou, Y. Zhao, J. W. Choi and A. Coskun, Angew. Chem., 2019, 131, 16951–16955.
57. B. Gui, G. Lin, H. Ding, C. Gao, A. Mal and C. Wang, Acc. Chem. Res., 2020, 53, 2225–2234.
58. V. Singh and H. R. Byon, Mater. Adv., 2021, 2, 3188–3212.
59. K. Geng, T. He, R. Liu, S. Dalapati, K. T. Tan, Z. Li, S. Tao, Y. Gong, Q. Jiang and D. Jiang, Chem. Rev., 2020, 120, 8814–8933.
60. J. Sprachmann, T. Wachsmuth, M. Bhosale, D. Burmeister, G. J. Smales, Z. Kochovski, N. Grabicki, B. Esser and O. Dumele, ChemRxiv. Cambridge, Cambridge Open Engage, 2022,  DOI:10.26434/chemrxiv-2022-p0pjr.
61. D. Jiang, Bull. Chem. Soc. Jpn., 2021, 94, 1215–1231.
62. I. Ahmed and S. H. Jhung, Coord. Chem. Rev., 2021, 441, 213989.
63. S. Ghosh, A. Nakada, M. A. Springer, T. Kawaguchi, K. Suzuki, H. Kaji, I. Baburin, A. Kuc, T. Heine, H. Suzuki, R. Abe and S. Seki, J. Am. Chem. Soc., 2020, 142, 9752–9762.
64. T. Banerjee, K. Gottschling, G. Savasci, C. Ochsenfeld and B. V. Lotsch, ACS Energy Lett., 2018, 3, 400–409.
65. M. Z. Rahman, Md. G. Kibria and C. B. Mulllins, Chem. Soc. Rev., 2020, 49, 1887–1931.
66. M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A. Santori and N. S. Lewis, Chem. Rev., 2010, 110, 6443–6445.
67. N. Romero, R. Bofill, L. Francàs, J. G. Antón and X. Sala, Catalysts, 2021, 11, 754.
68. Q. Wang and K. Domen, Chem. Rev., 2020, 120, 919–985.
69. J. Li, D. Zhao, J. Liu, A. Liu and D. Ma, Molecules, 2020, 25, 2425.
70. H. Shen, W. Zhou, F. He, Y. Gu, Y. Li and Y. Li, J. Mater. Chem. A, 2020, 8, 4850–4855.
71. A. M. Elewa, M. H. Elsayed, A. F. M. EL-Mahdy, C.-L. Chang, L.-Y. Ting, W.-C. Lin, C.-Y. Lu and H.-H. Chou, Appl. Catal., B, 2021, 285, 119802.
72. M. Natali, R. Argazzi, C. Chiorboli, E. Iengo and F. Scandola, Chem.–Eur. J., 2013, 19, 9261–9271.
73. H. Ozawa, M.-A. Haga and K. Sakai, J. Am. Chem. Soc., 2006, 128, 4926–4927.
74. W. Zhang, Y. Wang, Z. Wang, Z. Zhong and R. Xu, Chem. Commun., 2010, 46, 7631–7633.
75. N. Fajrina and M. Tahir, Int. J. Hydrogen Energy, 2019, 44, 540–577.
76. J. You, Y. Zhao, L. Wang and W. Bao, J. Cleaner Prod., 2021, 291, 125822.
77. L. Stegbauer, K. Schwinhghammer and B. V. Lotsch, Chem. Sci., 2014, 5, 2789–2793.
78. V. S. Vyas, F. Haase, L. Stegbauer, G. Savasci, F. Podjaski, C. Ochsenfeld and B. V. Lotsch, Nat. Commun., 2015, 6, 8508.
79. T. Banerjee, F. Haase, G. Savasci, K. Gottschling, C. Ochsenfeld and B. V. Lotsch, J. Am. Chem. Soc., 2017, 139, 16228–16234.
80. X. Wang, L. Chen, S. Y. Chong, M. A. Little, Y. Wu, W.-H. Zhu, R. Clowes, Y. Yan, M. A. Zwijnenburg, R. S. Sprick and A. I. Cooper, Nat. Chem., 2018, 10, 1180–1189.
81. L. Yin, Y. Zhao, Y. Xing, H. Tan, Z. Lang, W. Ho, Y. Wang and Y. Li, Chem. Eng. J., 2021, 419, 129984.
82. G.-B. Wang, S. Li, C. –X. Yan, Q. –Q. Lin, F. –C. Zhu, Y. Geng and Y. –B. Dong, Chem. Commun., 2020, 56, 12612–12615.
83. J. Yang, A. Acharjya, M.-Y. Ye, J. Rabeah, S. Li, Z. Kochovski, S. Youk, J. Roeser, J. Grgneberg, C. Penschke, M. Schwarze, T. Wang, Y. Lu, R. v. d. Krol, M. Oschatz, R. Schomacker, P. Saalfrank and A. Thomas, Angew. Chem., Int. Ed., 2021, 60, 19797–19803.
84. C. Li, J. Liu, H. Li, K. Wu, J. Wang and Q. Yang, Nat. Commun., 2022, 13, 2357.
85. M. Ai, J.-W. Zhang, Y.-W. Wu, L. Pan, C. Shi and J.-J. Zou, Chem.–Eur. J., 2021, 27, 13669–13698.
86. X. Li, C. Yang, B. Sun, S. Cai, Z. Chen, Y. Lv, J. Zhang and Y. Liu, J. Mater. Chem. A, 2020, 8, 16045–16060.
87. Y. Li, W. Chen, G. Xing, D. Jiang and L. Chen, Chem. Soc. Rev., 2020, 49, 2852–2868.
88. H. Wei, S. Chai, N. Hu, Z. Yang, L. Wei and L. Wang, Chem. Commun., 2015, 51, 12178–12181.
89. X. Guan, Y. Ma, H. Li, Y. Yusran, M. Xue, Q. Fang, Y. Yan, V. Valtchev and S. Qiu, J. Am. Chem. Soc., 2018, 140, 4494–4498.
90. A. d. l. P. Ruigómez, D. R.-S. Miguel, K. C. Stylianou, M. Cavallini, D. Gentili, F. Liscio, S. Milita, O. M. Roscioni, M. L. R. González, C. Carbonell, Da. Maspoch, R. Mas-Ballesté, J. L. Segura and F. Zamora, Chem.–Eur. J., 2015, 21, 10666–10670.
91. M. Matsumoto, R. R. Dasari, W. Ji, C. H. Feriante, T. C. Parker, S. R. Marder and W. R. Dichtel, J. Am. Chem. Soc., 2017, 139, 4999–5002.
92. B. P. Biswal, S. Chandra, S. Kandambeth, B. Lukose, T. Heine and R. Banerjee, J. Am. Chem. Soc., 2013, 135, 5328–5331.
93. S. Karak, S. Kandambeth, B. P. Biswal, H. S. Sasmal, S. Kumar, P. Pachfule and R. Banerjee, J. Am. Chem. Soc., 2017, 139, 1856–1862.
94. X. Guan, F. Chen, Q. Fang and S. Qiu, Chem. Soc. Rev., 2020, 49, 1357–1384.
95. M. Chen, H. Li, C. Liu, J. Liu, Y. Feng, A. D. H. Wee and B. Zhang, Coord. Chem. Rev., 2021, 435, 213778.
96. P. Pachfule, A. Acharjya, J. Roeser, T. Langenhahn, M. Schwarze, R. Schomacker, A. Thomas and J. Schmidt, J. Am. Chem. Soc., 2018, 140, 1423–1427.
97. H. Yu, J. Zhang, X. Yan, C. Wu, X. Zhu, B. Li, T. Li, Q. Guo, J. Gao, M. Hu and J. Yang, J. Mater. Chem. A, 2022, 10, 11010–11018.
98. L. Wang, L. Zhang, B. Lin, Y. Zheng, J. Chen, Y. Zheng, B. Gao, J. Long and Y. Chen, Small, 2021, 17, 2101017.
99. M. S. Lohse and T. Bein, Adv. Funct. Mater., 2018, 28, 1705553.
100. H. Wang, H. Wang, Z. Wang, L. Tang, G. Zeng, P. Xu, M. Chen, T. Xiong, C. Zhou, X. Li, D. Huang, Y. Zhu, Z. Wanga and J. Tang, Chem. Soc. Rev., 2020, 49, 4135–4165.
101. H. Li, H. Li, Q. Dai, H. Li and J.-L. Brédas, Adv. Theory Simul., 2018, 1, 1700015.
102. J. Xie, Z. Fang and H. Wang, Polymers, 2022, 14, 1363.
103. X. Zhao, P. Pachfule, S. Li, T. Langenhahn, M. Ye, C. Schlesiger, S. Praetz, J. Schmidt and A. Thomas, J. Am. Chem. Soc., 2019, 141, 6623–6630.
104. W. Zhong, R. Sa, L. Li, Y. He, L. Li, J. Bi, Z. Zhuang, Y. Yu and Z. Zou, J. Am. Chem. Soc., 2019, 141, 7615–7621.
105. J. Chen, X. Taoa, C. Lia, Y. Mab, L. Taoa, D. Zhenga, J. Zhud, H. Lia, R. Lia and Q. Yanga, Appl. Catal., B, 2020, 262, 118271.
106. Z. Mi, T. Zhou, W. Weng, J. Unruangsri, K. Hu, W. Yang, C. Wang, K. A. I. Zhang and J. Guo, Angew. Chem., Int. Ed., 2021, 60, 9642–9649.
107. J. Thote, H. B. Aiyappa, A. Deshpande, D. D. Diaz, S. Kurungot and R. Banerjee, Chem.–Eur. J., 2014, 20, 15961–15965.
108. Y.-H. Yao, J. Li, H. Zhang, H.-L. Tang, L. Fang, G.-D. Niu, X.-J. Sun and F.-M. Zhang, J. Mater. Chem. A, 2020, 8, 8949–8956.
109. Y. Hou, C.-X. Cui, E. Zhang, J.-C. Wang, Y. Li, Y. Zhang, Y. Zhang, Q. Wang and J. Jiang, Dalton Trans., 2019, 48, 14989–14995.
110. M. Luo, Q. Yang, K. Liu, H. Cao and H. Yan, Chem. Commun., 2019, 55, 5829–5832.
111. Y. Wang, Q. Yang, F. Yi, R. Lu, Y. Chen, C. Liu, X. Li, C. Wang and H. Yan, ACS Appl. Mater. Interfaces, 2021, 13, 29916–29925.
112. K. Gottschling, G. Savasci, H. V. Gonzalez, S. Schmidt, P. Mauker, T. Banerjee, P. Rovo, C. Ochsenfeld and B. V. Lotsch, J. Am. Chem. Soc., 2020, 142, 12146–12156.
113. G. B. Wang, F.-C. Zhu, Q.-Q. Lin, J.-L. Kan, K.-H. Xie, S. Li, Y. Geng and Y.-B. Dong, Chem. Commun., 2021, 57, 4464–4467.
114. C. Liu, Y. Xiao, Q. Yang, Y. Wang, R. Lu, Y. Chen, C. Wang and H. Yan, Appl. Surf. Sci., 2021, 537, 148082.
115. 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.
116. J. Bi, W. fang, L. Li, J. Wang, S. Liang, Y. He, M. Liu and L. Wu, Macromol. Rapid Commun., 2015, 36, 1799–1805.
117. S. Kuecken, A. Acharjya, L. Zhi, M. Schwarze, R. Schomacker and A. Thomas, Chem. Commun., 2017, 53, 5854–5857.
118. Z. Cheng, W. Fang, T. Zhao, S. Fang, J. Bi, S. Liang, L. Li, Y. Yu and L. Wu, ACS Appl. Mater. Interfaces, 2018, 10, 41415–41421.
119. W. Huang, Q. He, Y. Hu and Y. Li, Angew. Chem., Int. Ed., 2019, 58, 8676–8680.
120. L. Gao, Y. Niu, S. Razzaque, B. Tian and S. Jin, ACS Catal., 2019, 9, 9438–9445.
121. M. A. Fávaro, D. Ditz, J. Yang, S. Bergwinkl, A. C. Ghosh, M. Stammler, C. Lorentz, J. Roeser, E. A. Quadrelli, A. Thomas, R. Palkovits, J. Canivet and F. M. Wisser, ACS Appl. Mater. Interfaces, 2022, 14, 14182–14192.
122. C. B. Meier, R. Clowes, E. Berardo, K. E. Jelfs, M. A. Zwijnenburg, R. S. Sprick and A. I. Cooper, Chem. Mater., 2019, 31, 8830–8838.
123. K. Wang, L. –M. Yang, X. Wang, L. Guo, G. Cheng, C. Zhang, S. Jin, B. Tan and A. I. Cooper, Angew. Chem., Int. Ed., 2017, 56, 14149–14153.
124. M. Liu, Q. Huang, S. Wang, Z. Li, B. Li, S. Jin and B. Tan, Angew. Chem., Int. Ed., 2018, 57, 11968–11972.
125. S. Zhang, G. Cheng, L. Guo, N. Wang, B. Tan and S. Jin, Angew. Chem., Int. Ed., 2020, 59, 6007–6014.
126. M. Liu, X. Wang, J. Liu, K. Wang, S. Jin and B. Tan, ACS Appl. Mater. Interfaces, 2020, 12, 12774–12782.
127. M. Liu, K. Jiang, X. Ding, S. Wang, C. Zhang, J. Liu, Z. Zhan, G. Cheng, B. Li, H. Chen, S. Jin and B. Tan, Adv. Mater., 2019, 31, 1807865.
128. X. Hu, Z. Zhan, J. Zhang, I. Hussain and B. Tan, Nat. Commun., 2021, 12, 6596.
129. P. Audebert, E. Kroke, C. Posern and S.-H. Le, Chem. Rev., 2021, 121, 2515–2544.
130. S. Kumar, V. R. Battula, N. Sharma, S. Samanta, B. Rawata and K. Kailasam, J. Mater. Chem. A, 2022, 10, 14568–14575.
131. N. Sharma, S. Kumar, V. R. Battula, A. Kumari, A. Giri, A. Patra and K. Kailasam, Chem.–Eur. J., 2021, 27, 10649–10656.
132. D. Chen, W. Chen, G. Zhang, S. Li, W. Chen, G. Xing and L. Chen, ACS Catal., 2022, 12, 616–623.
133. H. Cheng, H. Lv, J. Cheng, L. Wang, X. Wu and H. Xu, Adv. Mater., 2022, 34, 2107480.
134. K. Kailasam, J. Schmidt, H. Bildirir, G. Zhang, S. Blechert, X. Wang and A. Thomas, Macromol. Rapid Commun., 2013, 34, 1008–1013.
135. K. Kailasam, M. B. Mesch, L. Mçhlmann, M. Baar, S. Blechert, M. Schwarze, M. Schrçder, R. Schomcker, J. Senker and A. Thomas, Energy Technol., 2016, 4, 744–750.
136. Y. Zhao, C. Wang, X. Han, Z. Lang, C. Zhao, L. Yin, H. Sun, L. Yan, H. Ren and H. Tan, Adv. Sci., 2022, 9, 2202417.
137. V. R. Battula, S. Kumar, D. K. Chauhan, S. Samanta and K. Kailasam, Appl. Catal., B, 2019, 244, 313–319.
138. S. Kumar, N. Sharma and K. Kailasam, J. Mater. Chem. A, 2018, 6, 21719–21728.
139. L. Stegbauer, S. Zech, G. Savasci, T. Banerjee, F. Podjaski, K. Schwinghammer, C. Ochsenfeld and B. V. Lotsch, Adv. Energy Mater., 2018, 8, 1703278.
140. E. Jin, Z. Lan, Q. Jiang, K. Geng, G. Li, X. Wang and D. Jiang, Chem, 2019, 5, 1632–1647.
141. K. Wang, Z. Jia, Y. Bai, X. Wang, S. E. Hodgkiss, L. Chen, S. Y. Chong, X. Wang, H. Yang, Y. Xu, F. Feng, J. W. Ward and A. I. Cooper, J. Am. Chem. Soc., 2020, 142, 11131–11138.
142. W. Chen, L. Wang, D. Mo, F. He, Z. Wen, X. Wu, H. Xu and L. Chen, Angew. Chem., Int. Ed., 2020, 59, 16902–16909.
143. Z. Zhao, Y. Zheng, C. Wang, S. Zhang, J. Song, Y. Li, S. Ma, P. Cheng, Z. Zhang and Y. Chen, ACS Catal., 2021, 11, 2098–2107.
144. W. Li, X. Huang, T. Zeng, Y. A. Liu, W. Hu, H. Yang, Y.-B. Zhang and K. Wen, Angew. Chem., Int. Ed., 2021, 60, 1869–1874.
145. Y. Li, L. Yang, H. He, L. Sun, H. Wang, X. Fang, Y. Zhao, D. Zheng, Y. Qi, Z. Li and W. Deng, Nat. Commun., 2022, 13, 1355.
146. H. Wang, H. Ding, X. Meng and C. Wang, Chin. Chem. Lett., 2016, 27, 1376–1382.
147. A. M. H. Flores, G. Bengasi, K. Baba and N. D. Boscher, ACS Appl. Energy Mater., 2020, 3, 9848–9855.
148. J. M. Gonclaves, B. A. Iglesias, P. R. Martins and L. Angnes, Analyst, 2021, 146, 365–381.
149. M. Joseph and S. Haridas, Int. J. Hydrogen Energy, 2020, 45, 11954–11975.
150. Z. Fan, K. Nomura, M. Zhu, X. Li, J. Xue, T. Majima and Y. Osakada, Commun. Chem., 2019, 2, 55.
151. R. Chen, Y. Wang, Y. Ma, A. Mal, X.-Y. Gao, L. Gao, L. Qiao, X.-B. Li, L.-Z. Wu and C. Wang, Nat. Commun., 2021, 12, 1354.
152. S. Kandambeth, A. Mallick, B. Lukose, M. V. Mane, T. Heine and R. Banerjee, J. Am. Chem. Soc., 2012, 134, 19524–19527.
153. P. Biswal, H. A. V. Gonzalez, T. Banerjee, L. Grunenberg, G. Savasci, K. Gottschling, J. Nuss, C. Ochsenfeld and B. V. Lotsch, J. Am. Chem. Soc., 2019, 141, 11082–11092.
154. T. Zhou, X. Huang, Z. Mi, Y. Zhu, R. Wang, C. Wang and J. Guo, Polym. Chem., 2021, 12, 3250–3256.
155. W. Weng and J. Gao, Nat. Commun., 2022, 13, 5768.
156. L. Cheng, Q. Xiang, Y. Liao and H. Zhang, Energy Environ. Sci., 2018, 11, 1362–1391.
157. H. Dong, X.-B. Meng, X. Zhang, H.-L. Tang, J. W. Liu, J.-H. Wang, J.-Z. Wei, F.-M. Zhang, L.-L. Bai and X.-J. Sun, Chem. Eng. J., 2020, 379, 122342.
158. Y.-P. Zhang, H.-L. Tang, H. Dong, M.-Y. Gao, C.-C. Li, X.-J. Sun, J.-Z. Wei, Y. Qu, Z.-J. Lib and F.-M. Zhang, J. Mater. Chem. A, 2020, 8, 4334–4340.
159. C.-C. Li, M.-Y. Gao, X.-J. Sun, H.-L. Tang, H. Dong and F.-M. Zhang, Appl. Catal., B, 2020, 266, 118586.
160. M.-Y. Gao, C.-C. Li, H.-L. Tang, X.-J. Sun, H. Dong and F.-M. Zhang, J. Mater. Chem. A, 2019, 7, 20193–20200.
161. L.-L. Zheng, D. Wang, S.-L. Wu, X.-H. Jiang, J. Zhang, Q.-J. Xing, J.-P. Zou and S.-L. Luo, J. Mater. Chem. A, 2020, 8, 25425–25430.
162. H. Huang, B. Xu, Z. Tan, Q. Jiang, S. Fang, L. Li, J. Bi and L. Wu, J. Alloys Compd., 2020, 833, 155057.
163. Z. Tan, P. Zhang, Q. Chen, S. Fang, G. Huang, J. Bi and L. Wu, Catal. Sci. Technol., 2021, 11, 1874–1880.
164. N. Xu, B. Cai, Q. Li, Y. Liu, J. Tang, K. Wang, B. Xu and Y. Fan, J. Alloys Compd., 2021, 871, 159565.
165. Y. Peng, M. Zhao, B. Chen, Z. Zhang, Y. Huang, F. Dai, Z. Lai, X. Cui, C. Tan and H. Zhang, Adv. Mater., 2018, 30, 1705454.
166. F.-M. Zhang, J.-L. Sheng, Z.-D. Yang, X.-J. Sun, H.-L. Tang, M. Lu, H. Dong, F.-C. Shen, J. Liu and Y.-Q. Lan, Angew. Chem., Int. Ed., 2018, 57, 12106–12110.
167. H.-Y. Zhang, Y. Yang, C.-C. Li, H.-L. Tang, F.-M. Zhang, G.-L. Zhang and H. Yan, J. Mater. Chem. A, 2021, 9, 16743–16750.
168. J. Low, S. Cao, J. Yu and S. Wageh, Chem. Commun., 2014, 50, 10768–10777.
169. Y. Xing, L. Yin, Y. Zhao, Z. Du, H.-Q. Tan, X. Qin, W. Ho, T. Qiu and Y.-G. Li, ACS Appl. Mater. Interfaces, 2020, 12, 51555–51562.
170. C. Lin, C. Han, L. Gong, X. Chen, J. Deng, D. Qi, Y. Bian, K. Wang and J. Jiang, Catal. Sci. Technol., 2021, 11, 2616–2621.
171. T. Zhou, L. Wang, X. Huang, J. Unruangsri, H. Zhang, R. Wang, Q. Song, Q. Yang, W. Li, C. Wang, K. Takahashi, H. Xu and J. Guo, Nat. Commun., 2021, 12, 3934.
172. W. Zheng, C.-S. Tsang, L. Y. S. Lee and K.-Y. Wong, Mater. Today Chem., 2019, 12, 34–60.
173. C. Altintas, I. Erucar and S. Keskin, CrystEngComm, 2022, 24, 7360–7371.
174. Y. Li, G. Wen, J. Li, Q. Li, H. Zhang, B. Tao and J. Zhang, Chem. Commun., 2022, 58, 11488–11506.
175. R. Kumar, S. N. Ansari, R. Deka, P. Kumar, M. Saraf and S. M. Mobin, Chem.–Eur. J., 2021, 27, 1–31.
176. D. Shang, D. Li, B. Chen, B. Luo, Y. Haung and W. Shi, ACS Sustainable Chem. Eng., 2021, 9, 14238–14248.
177. A. Xiong, G. Ma, K. Maeda, T. Takata, T. Hisatomi, T. Setoyama, J. Kubota and K. Domen, Catal. Sci. Technol., 2014, 4, 325–328.
178. B. Mishra, A. Alam, B. Kumbhakar, D. D. Díaz and P. Pachfule, Cryst. Growth Des., 2023, 23(6), 4701–4719.

---