Metal‐Free Photocatalysts for Conversion of H2O into Hydrogen Peroxide

Abstract Hydrogen peroxide (H2O2) is an important green oxidizing agent for environmental protection and chemical production. In comparison to the traditional anthraquinone method, photosynthesis is a green and energy‐saving process for H2O2 production. To improve the stability and practical application value of the H2O2 synthesized by photocatalysis, the H2O2 photosynthesis should be conducted in pure water without involving any sacrificial reagents. In this regard, organic semiconducting catalysts pose as a suitable candidate for photocatalytic H2O2 synthesis owing to their metal‐free nature to prevent H2O2 decomposition by the metal ions. In this Perspective, the H2O2 photosynthesis history is firstly introduced, followed by a review of the organic semiconductor photocatalysts reported to date. Finally, the main problems to thwart the advances of current pure H2O‐to‐H2O2 photosynthesis are discussed, followed by proposed solutions to address these issues in order to pave new ways for the development of highly efficient metal‐free organic photocatalysts for sustainable pure H2O‐to‐H2O2 conversion.


Introduction
Hydrogen peroxide (H 2 O 2 ) is a strong oxidizing agent which possesses antibacterial, antiviral, and disinfectant properties. [1] In addition, it is widely used in various industries including chemical synthesis, textiles and pulp bleaching, wastewater treatments, and electronic applications. [2] Since the decomposition of H 2 O 2 will only produce water (H 2 O) without other hazardous products, H 2 O 2 is regarded as an environmentally friendly chemical. [3] Although anthraquinone oxidation is the most commonly employed method for the production of H 2 O 2 , [4] there are still some unsettling problems in this method that should not be neglected. For instance, the anthraquinone oxidation involves a sequential process including hydrogena-tion, oxidation, extraction, and purification, which require large inputs of energy, and it is potentially dangerous. [5] Moreover, the solvents used in the production process may lead to severe environmental pollution, which conflicts with the original idea of green chemistry. In this regard, photocatalysis, a process capable of converting free, inexhaustible solar energy into chemically valuable substances with the aid of a suitable photocatalyst, is a remarkable approach to replace the anthraquinone process. [6] Through photocatalytic route, H 2 O and O 2 are the only raw materials needed with the presence of sunlight to provide energy for the formation of H 2 O 2 . [7] For example, it has been shown that H 2 O 2 can be photosynthesized from seawater directly in air by the hydrolysis of lignin. [8] This is a simple process that can take place at mild conditions (room temperature and atmospheric pressure), and more importantly, it does not produce harmful substances to contaminate the environment. [9] Therefore, the photocatalytic process is in good agreement with the concept of green chemistry.
In order to prevent H 2 O 2 decomposition by metal and ultraviolet (UV) light, [10] visible-light-responsive metal-free organic semiconductors are the current main focus regarding photocatalysts for achieving H 2 O 2 photosynthesis in high concentration. However, organic semiconductors often exhibit poor separation of photogenerated charge carriers, and hence, sacrificial reagents such as ethanol and methanol are usually added to consume the photoinduced holes to thwart the electron-hole pairs recombination for enhanced photocatalytic efficiency. However, the residual sacrificial reagents will severely affect the stability and purity of the H 2 O 2 produced, thereby affecting the practical application value of the H 2 O 2. Therefore, many organic semiconducting polymers, [11] such as graphitic carbon nitride (g-C 3 N 4 ), resins, covalent triazine frameworks (CTFs), and covalent organic framework (COFs), were modified by constructing opportune active sites at the molecular level to photosynthesize high concentration of H 2 O 2 from pure water without any sacrificial reagents. In this Perspective, the state-ofthe-art organic semiconductor photocatalysts with various modification approaches are reviewed (Scheme 1). We also raise the main issues that currently hinder the practical applications of H 2 O 2 photosynthesis, and lastly, we propose some potential solutions to solve these issues in order to bring the research community a step closer towards the commercialization of photocatalytic H 2 O 2 production from pure water.

A Brief History of H 2 O 2 Photosynthesis
Solar light-driven production of H 2 O 2 from water was first proposed by Baur and Neuweiler in 1921 by employing ZnO as a catalyst to produce H 2 O 2 in the presence of glycerol and glucose. [12] Since then, various inorganic materials such as TiO 2 , WO 3 , CdS, and so on have been studied for photocatalytic production of H 2 O 2 . [13] However, it was found that the metal element composition often decomposes the generated H 2 O 2 , leading to a low solar-to-chemical conversion (SCC) efficiency. In 2014, Shiraishi et al. firstly reported metal-free g-C 3 N 4 photocatalysts for H 2 O 2 photosynthesis, [14] Following this, the scope of application of organic polymer semiconductors was extended quickly. After years of study, researchers found that g-C 3 N 4 suffers from intrinsic low photogenerated charge carrier separation efficiency, narrow light absorption range, low crystallinity, and exhibits single active sites, which greatly limits the photocatalytic performance of H 2 O 2 synthesis. Therefore, finding new metal-free photocatalysts and the relevant modification methods to further improve the photocatalytic performance is the topic of current study.

Organic Semiconductors for H 2 O 2 Photosynthesis g-C 3 N 4
g-C 3 N 4 is an organic layered polymer composed of C and N elements with high thermal and chemical stability. [15] Owing to its excellent chemical stability and unique electronic band structure, it is widely studied in photocatalytic degradation of pollutants, hydrogen and oxygen production from water splitting, and organic synthesis. [16,17] In 2014, g-C 3 N 4 was first studied for H 2 O 2 photosynthesis in ethanol solution. [14] During the photocatalytic reaction over g-C 3 N 4 , the formed internal peroxides prevented one-electron reduction of O 2 . However, the H 2 O 2 yield over g-C 3 N 4 was still very low, only 30 μmol after 12 h of light illumination. The low yield could be attributed to several reasons: (1) the intrinsic weak absorption of visible light; (2) low separation efficiency of the photogenerated charge carriers; (3) low crystallinity of g-C 3 N 4 inhibiting charge transfer. Therefore, incessant research efforts were devoted to improving the efficiency of g-C 3 N 4 for H 2 O 2 photosynthesis. In order to enhance the water oxidation activity for H 2 O 2 photosynthesis in pure water, g-C 3 N 4 was compounded with homophthalic diimide (PDI). [18] The valence band position of the compound was positively shifted to promote the oxidation of water for the generation of H 2 O 2 , while ensuring that the conduction band position can still effectively reduce oxygen, thereby forming a dual pathway to generate H 2 O 2 from pure water ( Figure 1). Theoretical study revealed that the electrons and holes are respectively generated at C1, N4 and N2, N6 positions under solar excitation (the specific structure is shown in Figure 2). Particularly, the holes oxidize H 2 O to generate O 2 , whereas two electrons reduce O 2 to generate internal peroxide, which then convert into H 2 O 2 by protonation. The whole process effectively suppresses the single-electron and four-electron reduction of O 2 . Subsequently, reduced graphene oxide (rGO) was further hybridized with C 3 N 4 /PDI, [19] and the incorporation of rGO greatly improved the separation of photogenerated charge carriers ( Figure 1). By comparing the photocatalytic performance, the apparent quantum efficiency (AQE) at 420 nm was determined to be 6.1 %, and the average SCC efficiency was enhanced to 0.20 %. By further adding boron nitride (BN) to the g-C 3 N 4 /PDI/rGO system, the g-C 3 N 4 /PDI/rGO/BN composite exhibited an average SCC efficiency of 0.27 %. [20] Interestingly, the researchers found that the electron-hole recombination efficiency of the material was greatly inhibited by the introduction of aryl amino groups to g-C 3 N 4 , while promoting the oxygen reduction reaction (ORR), which could produce 2.0 mm h À 1 H 2 O 2 in the presence of household LED. [21] Besides, single atoms, which serve as the catalytic reactive sites, are often used in g-C 3 N 4 for H 2 O 2 photosynthesis from pure water (Figure 2). It was reported that Co single atom and anthraquinone (AQ) can be used as the oxidation and reduction co-catalysts of g-C 3 N 4 , respectively. [22] The spatial separation of the redox centers can effectively inhibit the complexation of electron-hole pairs while increasing the reaction active sites, rendering a H 2 O 2 yield of 60 μm h À 1 over Co 1 /AQ/C 3 N 4 . In addition, when Sb single atom is loaded on g-C 3 N 4 , [23] the conduction band electrons accumulate on the Sb sites whereas the holes focus on the N atoms near the single Sb atom, thereby suppressing the recombination of electron-hole pairs. With this elegant design, even in the absence of electron donor, there are still Sb with terminal adsorption conformation * OOH formation, which significantly suppressed the occurrence of undesirable 4e À ORR side reaction, thereby allowing the highly selective occurrence of the 2e À ORR reaction with an average SCC efficiency of 0.61 % for the optimal Sb-SAPC15. It is challenging to find a suitable modification method for g-C 3 N 4 to efficiently improve the photocatalytic efficiency. Therefore, a continuous search for more suitable organic semiconductor materials for H 2 O 2 photosynthesis is necessary.

Resins
Compared to g-C 3 N 4 , resins usually possess a narrower bandgap that can absorb more visible light to facilitate oxygen reduction and water oxidation ( Figure 3). In 2019, resorcinol-formaldehyde (RF) resin with unique π-conjugated and π-stacked benzenequinoline donor-acceptor (D-A) structure was reported for the first time for H 2 O 2 photosynthesis in pure water. [24] The narrow bandgap of RF resin greatly promotes oxygen reduction and water oxidation, in which H 2 O 2 was stably generated with an average SCC yield of 0.5 % without involving any sacrificial reagents. However, a defective valence band structure is present in the RF resin, which limits the transportation of electrons, thus leading to a low photocatalytic efficiency. To overcome the defective valence band structure, polythiophene (P3HT) was doped in the RF resin. [25] It was found that P3HT with suitable highest occupied molecular orbital (HOMO) can be used as a charge transfer polymer for efficient electron migration from the quinone-type acceptor site, thereby enhancing the water oxidation reaction (WOR) and ORR reactions for generating high concentration of H 2 O 2 solution with an SCC

ChemSusChem
Perspective doi.org/10.1002/cssc.202201514 efficiency of 1.0 %. In addition, monodisperse RF resin microspheres (MRFs) was also studied, in which the mesoporous microspheres can effectively modulate the charge distribution in the nano space to result in an SCC efficiency of 1.1 %. [26] This indicates that the changes in the morphology will lead to the differences in photocatalytic performance. Although resin addresses the issue of narrow forbidden bandwidth of g-C 3

CTF
CTFs own a large number of oxygen-reducing active sites (C=N), which is similar to g-C 3 N 4 . Meanwhile, organic water oxidation sites can also be artificially introduced into CTF through structural modification. Hence, CTF may be an ideal material for H 2 O 2 photosynthesis via 2e À dual channels. It was reported that the introduction of acetylene and diacetylene can serve as the active sites for WOR ( Figure 4). The comparative experiments and theoretical calculations revealed that the conjugated structures of CTF-EDDBN (4,4'-(ethyne-1,2-diyl)dibenzonitrile) and CTF-BDDBN (4,4'-(buta-1,3-diyne-1,4-diyl)dibenzonitrile) can promote charge separation for the formation of suitable intermediates and create a new pathway for the generation of H 2 O 2 by 2e À WOR. [27] Although the ORR and WOR pathways can be realized simultaneously for H 2 O 2 generation, the photogenerated charge carriers tend to recombine at the redox-active site, which limits the photocatalytic efficiency. Recently, researchers have rationalized the design of novel covalent heptazine frameworks (CHF) with spatially separated redox centers by rationalizing the structure at the molecular level. [28] For instance, s-heptazine ring, which is connected by three s-triazine rings, is rich in single atoms and thus is a strong electron-deficient group that can be used as an active site for O 2 reduction. From photocatalytic experiments, it exhibited excellent photocatalytic efficiency with an average SCC efficiency of 0.76 %, and the H 2 O 2 yield over CHF-DPDA (diphenyldiacetylene) was measured to be 69 μmol h À 1 , which was much higher than that of C 3 N 4 and CTF-BDDBN (g-C 3 N 4 : 2.5 μmol h À 1 ; CTF-BDDBN: 2.9 μmol h À 1 ). However, most of the triazine organic semiconductors are in amorphous form with low crystallinity that largely limits the transportation of photogenerated charge carriers. Therefore, it is necessary to search for other organic semiconductor materials with high crystallinity along with the presence of multiple active sites to further improve the photocatalytic efficiency.

COF
Compared to other organic semiconductor materials, COF with high crystallinity has great advantages in the field of photocatalysis ( Figure 5). It was firstly reported for H 2 O 2 photosynthesis with ethanol as a sacrificial reagent in 2020. [29] Recently, our research group developed bipyridine-based COF photocatalysts that showed excellent activity for two-channel photocatalytic synthesis of H 2 O 2 with a superior SCC efficiency of 1.08 %. [30] By comparing the monomer structure and the crystallinity of COFs, it was found that bipyridine serves as the main reactive site. Further theoretical study showed that COF-TfpBpy (Tfp: 1,3,5-triformylphloroglucinol; Bpy: 2,2'-bipyridine-5,5'-dia-mine) drove a one-step two-electron redox reaction to produce H 2 O 2 whereas COF-TfpDaaq (Tfp: 1,3,5-triformylphloroglucinol; Daaq: 2,6-diaminoanthraquinone) altered the redox pathway to a two-step single-electron process by changing the active sites. In addition, based on in-situ Fourier-transform infrared (FTIR) spectroscopy, the high photoactivity of COF-TfpBpy is attributed to the protonation of bipyridine monomer, which promotes the rate-limiting reaction (2e À WOR) and  improves Yeager-type oxygen adsorption to accelerate 2e À onestep oxygen reduction. In order to enhance the activity of COFs for H 2 O 2 photosynthesis, highly polar ionic sites were introduced in COFs materials to promote the charge carrier separation. [31] After ionization, the catalytic performance of COFs was increased by more than 5 times under visible light illumination.
Presently, most of the photocatalytic synthesis of H 2 O 2 experiments were performed in a two-phase system. However, the solubility of O 2 in water is low (0.25 μmol cm À 3 for oxygensaturated water), which greatly limits the H 2 O 2 photocatalytic synthesis activity. In addressing this problem, three-phase reaction interface was established with COFs, in which the H 2 O 2 generated at the three-phase interface was 2.9 mmol g À 1 h À 1 , [32] demonstrating a remarkable 15 times enhancement over the two-phase system. That being said, the photocatalytic production of H 2 O 2 from organic semiconductor materials is still in the nascent stage, and there is still much room for development. The search for suitable materials and efficient synthetic approaches to achieve stable and highly efficient photocatalytic production of H 2 O 2 from pure water at high yields requires the joint efforts of researchers from different fields.

Future Prospects
Researchers have basically reached a consensus in the field of photocatalytic synthesis of H 2 O 2 . In comparison to the inorganic materials for photocatalytic production of H 2 O 2 , organic polymeric semiconductor materials that do not contain any metal are more suitable for photocatalytic H 2 O 2 production, and, ideally, the process should not involve any organic reagents as sacrificial reagents. This method has gradually developed into one of the most effective H 2 O 2 synthesis methods with an aim of replacing the traditional anthraquinone oxidation. However, there is still a big chasm difficult to cross from laboratory extension to daily application. In order to solve this, a series of in-depth studies from the aspects of molecular-level design of organic polymeric semiconductor materials for improving the photocatalytic performance is necessary ( Figure 6).

Stability
Till date, most of the studies focus on the search for suitable photocatalysts or center on the modifications of the catalysts to improve the overall efficiency. However, the stability of the generated H 2 O 2 and organic semiconductors are often neglected. It is worth mentioning that H 2 O 2 can be decomposed by metal ions, UV light, heat, and the catalyst itself. For the H 2 O 2 solution generated from the photocatalytic synthesis, there will be some residual organic catalysts inevitably present in the solution, which may lead to undesirable H 2 O 2 decomposition. On the other hand, oxygen free radicals produced by the one-electron pathway during photocatalysis will degrade organic polymers ( * O 2À for oxygen reduction; * OH for water oxidation). Therefore, to solve the problem ofstability, it is necessary to improve the stability of the material itself and the stability of H 2 O 2 production. And improving the crystallinity of organic semiconductors, and then improve the efficiency of carrier separation of materials, the formation of multi-electron double channel selectivity, to a greater extent to inhibit the generation of oxygen free radical, thus improving the stability of H 2 O 2 production. Also increasing the degree of crystallinity of organic semiconductors will improve the stability of the material itself.

Synthesis temperature
According to the empirical rule of Van't Hoff, the reaction rate increases by two-to-four times for every 10 K increase in the reaction temperature. The experimental studies demonstrated that the photocatalytic efficiencies increase gradually with the increase in temperature for both resins and COFs. [25] However, the underlying mechanism is still unclear. For example, how the reaction temperature affects the adsorption of substrate molecules, the separation efficiency of charge carriers, and the activity of active sites still needs to be systematically investigated. If the temperature of the reaction system increases to 60°C under the irradiation of sunlight, the photocatalytic performance is expected to increase by more than 50 %. [30] Based on this phenomenon, the possibility of solar-induced photothermal synergistic reaction for the synthesis of H 2 O 2 could not be excluded. Therefore, visible to near-infrared response activities of the organic semiconductors are equally deserving of much attention due to the potential photothermal effect.

Active site
As mentioned previously, H 2 O 2 can be obtained by oxygen reduction or water oxidation via photocatalytic reactions. There are three ways for the oxygen reduction or water oxidation: 1e À process to reactive oxygen species (

Theoretical study
The findings from the theoretical study can provide reliable technical supports for designing efficient and stable organic photocatalysts. In the process of photocatalytic synthesis of H 2 O 2 , in situ spectral characterization techniques and theoretical calculations should be the most effective methods. The in situ spectral characterization such as in situ IR and Raman spectroscopies were generally used to confirm the conformational changes of the catalysts and the intermediate species in the reaction. However, at present, gas-solid in situ technology is commonly used, which is far from the actual liquid-solid reaction environment. Therefore, with an aim of revealing the process of photocatalytic synthesis of H 2 O 2 more accurately, liquid-solid in situ spectral technology should be introduced to understand the interaction mechanism between H 2 O and O 2 . In addition, theoretical calculations including band structures, differential charge density mappings, adsorption energies, and possible reaction pathways can also be used as supplementary evidence to support the results from in situ spectral characterization. With these, the reliability of the photocatalytic synthesis of H 2 O 2 mechanisms will be greatly improved.
All in all, this perspective summarizes the current problems for photocatalytic synthesis of H 2 O 2 over organic polymer semiconductors in terms of their stability, synthesis temperature, photocatalytic mechanisms, and active sites. In addition to the above-mentioned aspects, there are still many factors waiting for the researchers to explore and study in order to develop an organic photocatalytic system with remarkable efficiency.