5.1 Introduction

Owning to the limited global fossil fuel storage, the demand of increasing energy poses serious challenges to energy security and environmental protection [1]. A lot of attention has been paid on the development of sustainable energy resources [2]. Among various types of sustainable energy resources, solar energy is considered to be one of the cleanest and most practical energy resources because of its inexhaustibility, universality, high capacity, and environment‐friendly nature [3, 4]. At present, there are three types of typical methods to directly use solar energy: photothermal, photovoltaic, and photocatalytic approaches [3]. The efficient conversion of solar energy into chemical energy via photocatalytic approaches has been considered one of the most promising solutions to the energy crisis and environment pollution [5–7]. During the photocatalytic processes, photocatalysts play a crucial role in achieving high conversion efficiency of solar energy. First, photocatalysis simply begins with the photoabsorption of photocatalysts [8]. When the bandgap energy (E_g) of the catalyst is lower than the energy of a photon, the photon can be absorbed to excite an electron to the conduction band (CB) and leave behind the equal number of hole in the valence band (VB) [9–12]. Then, the separation and migration process of charge carriers can occur, and finally, these charge carriers react with targeting reagents in the surface of the photocatalysts. If the generated electrons and holes have enough activity, the active species and free radicals could be produced [9, 13]. These active species and free radicals are of vital importance for the solar energy conversion processes, such as environmental remediation [14], selective transformations for chemicals synthesis [15, 16], reduction of CO₂ [17–20], water splitting to hydrogen [21–23], and other photoelectrochemical processes [24].

In 1972, TiO₂ photoelectrode was first reported on photoelectrochemical water splitting [25]. Subsequently, Carey et al. reported TiO₂ photocatalytic decomposition of organic pollutants in 1976 [26]. Since then, a lot of researchers have set out to the study of metal‐semiconductor‐based photocatalysts, such as TiO₂ [27], BiVO₄ [28], ZnO [29], Cu₂O [30], Fe₂O₃ [31], CdS [32], and SrTiO₃ [33]. However, the recombination of these photoinduced electron–hole pairs in the semiconductor photocatalysts brought about the decrease in photocatalytic performance. To date, the related research findings cannot satisfy with request for application because of the lack of highly efficient and stable photocatalytic system. These semiconductors also need to overcome their own defects. For example, the sunlight absorption efficiency of TiO₂, ZnO, and SrTiO₃ are very low because they are only active in the UV region, while ZnO, Cu₂O, and CdS suffer from photocorrosion [8]. These methods of elemental doping [34], facet engineering [35], creation of mesoporous structure [36], construction of semiconductor heterostructures [9, 37], and semiconductor–metal hybrids [38] have been used to overcome the above‐mentioned shortcomings.

To realize an ideal photocatalytic performance, a perfect photocatalyst system should have the following characteristics: (i) appropriate bandgap and band alignment to satisfy the kinetic requirement of the targeted reaction, (ii) high photoabsorption efficiency and charge carriers migration, (iii) extended charge separation time, and (iv) outstanding chemical and photic stability [39, 40]. However, it is extremely difficult to satisfy the above‐mentioned requirements for the conventional metal‐oxide semiconductors materials [41, 42]. Most of the metal‐semiconductor‐based photocatalysts are often based on metal cations possessing d⁰ and d¹⁰ electronic configurations while have been confined by the relatively limited properties for many years [22]. Therefore, it is very essential for the research and development of novel photocatalysts. Recently, some new types of metal‐free materials, such as carbon [43], phosphorus [44, 45], graphitic carbon nitride [46, 47], hexagonal boron nitride [48, 49], and boron carbide [50], have emerged and grown on the basis of lightweight, rich resources, and low cost, which provide new opportunities for photocatalysis. With abundant carbon resources, carbon‐based, metal‐free catalysts will enable a significant reduction in costs while maintaining high efficiency with economic viability for photocatalysis applications.

Until now, there have been some review literatures focusing on carbon‐based metal materials, which have been intensively investigated as a new class of semiconductor photocatalysts. However, there was little or no emphasis on the systematic summarization of photocatalysts from the perspective of carbon‐based, metal‐free materials. Hence, the aim of this chapter is to provide a comprehensive review of the topic by summarizing the important developments in this emerging field of active research on photocatalysis from the viewpoint of carbon‐based, metal‐free materials, including graphene, carbon quantum dots (CQDs), and carbon nitride.

5.2 Graphene‐Based, Metal‐Free Photocatalysis

5.2.1 Graphene and Graphene Oxide

Graphene, a single‐atom‐thick sheet composed of hexagonal structured sp²‐hybridized carbon atoms, has received considerable attention in various energy‐related applications since the first fabrication of the star materials using adhesive tape to repeatedly split graphite crystals in 2004 [51]. The unique two‐dimensional structure of graphene makes it to be the thinnest and strongest material in the universe [52]. Moreover, pristine graphene endows many apparent merits and excellent properties, including unconventional quantum Hall effect [53–55], relativistic Dirac fermions [56], high thermal conductivity (5000 W m⁻¹ K⁻¹) [57], high transparency (97.7%) [58], superior carrier mobility (200 000 cm² V⁻¹ s⁻¹) [59], good electrical conductivity (2000 S m⁻¹) [60], extremely large specific surface area (2630 m² g⁻¹) [61], excellent environmental compatibility, high mechanical stiffness (1060 GPa) [62], and high adsorption capacity [63]. However, the zero‐gap semiconductor nature of graphene significantly limits the applicability of graphene in photoenergy conversion [64]. As a functionalized graphene, graphene oxide (GO) is made of various oxygen functional groups on the basal plane and sheet edge of graphene. Because of the presence of oxygen bonding, the structure and electronic properties of GO change and the sp³ hybridization appears in graphene [65, 66]. As the bandgap increases with the oxidation level in GO, fully oxidized GO, partially oxidized GO, and graphene are insulator, semiconductor, and conductor, respectively [67]. As shown in Figure 5.1, in the condition of hydrazine vapors, the optical bandgap of reduced graphene oxide (rGO) shows a gradual decrease from 3.5 to 1 eV with the increase in the C:O ratio [68].

5.2.2 Graphene‐Based, Metal‐Free Catalysts for Photocatalysis

In the field of photocatalysis, some researchers have demonstrated that GO alone can be a potential metal‐free photocatalyst [69, 70]. As early as 2010, Yeh et al. found that, under UV or visible light irradiation, partially oxidized GO can serve as a metal‐free photocatalyst for stable H₂ generation from water in a 20 vol.% aqueous solution or pure water [69]. Figure 5.2 shows a schematic energy diagram to overview H₂ generation using GO as a photocatalyst. Because of the highly hydrophilic nature, GO can effectively catalyze water splitting under irradiation without cocatalyst. In spite of the bandgap reduction and increased conductivity of GO during photocatalytic reaction, the H₂ evolution still remained constant on account of the stable photogenerated charges created by the oxygenated sites [69].

Thereafter, Hsu et al. reported that the conversion rate of photocatalytic CO₂ to methanol is up to 0.172 mmol g cat⁻¹ h⁻¹ using GO as a metal‐free photocatalyst and the catalytic performance is six times higher than that of the pure TiO₂ under visible light [70]. As shown in Figure 5.3, the possible photocatalytic mechanism is that the adsorbed CO₂ and H₂O can react with the photoinduced electrons and holes on the irradiated GO to generate CH₃OH.

However, the problem that GO is easily reduced by light irradiation should be noted. Once GO is reduced, the proportion of sp²‐hybridized carbon atoms to sp³‐hybridized ones will increase. In addition, VB position also has an upward shift. However, there is no change for CB position.

Recently, graphene has attracted many researchers in photocatalytic fields, and it has been proved as an efficient electron acceptor to boost charge transfer and lessen recombination of the electron–hole pairs [2]. As a graphene derivative, GO also has a number of advantages, such as high flexibility, large specific surface area, and excellent solution resolvability. These merits allow graphene nanosheets as a matrix to easily combine with other semiconductors to generate electronic bridges [71]. As noted, compared with inorganic metal compounds, metal‐free semiconductors, such as C₃N₄, possess the superiorities of low cost, ease of acquisition, and environment‐friendly feature, thus leading to a lot of applications in environmental and energy fields [72]. Therefore, the combination of graphene or graphene derivatives with metal‐free semiconductors will improve the photocatalytic activity of composite photocatalysts.

In addition, Long and coworkers first introduced a novel metal‐free catalyst (a layered C₃N₃S₃ polymer (CNS/graphene hybrids)) for selective photocatalytic oxidation of benzylic alcohols under visible light [73]. Actually, the CNS has already been reported by their group, which is a novel organic semiconductor photocatalyst for generating hydrogen from water without the aid of a sacrificial electron donor under visible light irradiation [74]. Because the addition of rGO can improve the separation and transport of the photoinduced electrons and holes, a series of the layered rGO‐CNS hybrids with different rGO ratios were prepared by the in situ stepwise polymerization approach to elect an optimal photocatalyst [73]. The fabrication route of the “sandwich” rGO‐CNS polymer hybrids is shown in Figure 5.4a. This “sandwich” structure can effectively improve the photocatalytic efficiency by holding back the restacking of graphene sheets during the reduction process. Figure 5.4b shows the photocatalytic activity of the pure CNS and the xrGO‐CNS hybrids, where the “x” is the rGO content by weight ratio. The results appeared that the photocatalytic conversion enhanced with the increasing rGO content and 0.3rGO‐CNS hybrids reached the maximum conversion (51.5%) with 100% selectivity. The contrastive photocatalytic activities of 0.3rGO‐CNS hybrids prepared by different ways are shown in Figure 5.4c. Considering that graphene/g‐C₃N₄ composites play a very important role in the field of photocatalysis, the relevant content will be discussed in Section 5.4 of this chapter.

5.3 Carbon‐quantum‐dot‐Based, Metal‐Free Photocatalysis

CQDs composed of sp²/sp³‐hybridized carbon atoms are a novel class of carbon nanomaterials with ultrafine sizes below 10 nm, prominent fluorescence, and various surface functional groups [75]. Spectacularly, CQDs have both the unique fluorescence of quantum dots and the outstanding electronic properties of carbon materials. In 2004, Xu et al. first discovered CQDs by a gel eletrophoretic method for the refinement of single‐wall carbon nanotubes [76]. Two years later, a simple surface passivation method was designed to the synthesis of CQDs with strong fluorescence [77]. Since then, CQDs have drawn plenty of attention in drug delivery [78–80], bioimaging [81], sensors [82, 83], and photovoltaic devices [84] because of their unique properties including intense photoluminescence, outstanding photostability, and excellent biocompatibility. Considering the merits of low cost, ease of synthesis, excellent performance, and nontoxicity, it is not surprising that CQDs have been applied in photocatalytic solar‐to‐energy conversion [85–88]. Here, we aim to summarize the work reported to date utilizing metal‐free CQDs or CQD composites as catalysts for photocatalysis and we also will place emphasis on their applications in water splitting into hydrogen, photocatalytic reduction of CO₂, and organic synthesis.

5.3.1 Synthesis of Carbon Quantum Dots

5.3.1.1 Top‐down Approaches

In general, synthetic approaches of CQDs can be subdivided into “top‐down” and “bottom‐up.” For one, top‐down approaches refer to exfoliate macroscopic carbon structures (carbon fibers, carbon nanotubes, graphene oxide, soot, coal, carbon black, activated carbon, etc.) into quantum‐sized CQDs by laser [89], electrochemical approach [90–94], hydrothermal method [95, 96], ultraphonic [97], ball milling [98], and arc discharging [76]. In 2006, the laser ablation method was first adopted to prepare CQDs with graphite as a raw material under argon atmosphere in the presence of water vapor [77]. Then, Li et al. reported a simple method to prepare CQDs with visible, tunable, and stable photoluminescence by laser passivation of nanocarbon particles [99]. In 2011, our group adopted the electrochemical approach to prepare the functional graphene quantum dots (GQDs) with green luminescence as potential electron acceptors for photovoltaics [90]. Subsequently, we reported the well‐confined two‐dimensional (2D) and 3D CQDs honeycomb frameworks by the simple electrodeposition of the functional CQDs within preformed SiO₂ templates and the preparation methods for the unique structural samples were shown in Figure 5.5 [91]. Then, the functional GQDs prepared by our previously reported electrochemical approach were also assembled into one‐dimensional (1D) nanotube arrays as metal‐free platform for the Raman enhancement applications [92]. Further, we modified the electrochemical approach to prepare N‐doped carbon nanotubes (N‐CNTs), which can emit blue luminescence and exhibit an excellent electrocatalytic activity comparable with that of a commercially available Pt/C catalyst for the oxygen reduction reaction in an alkaline medium, using N‐containing tetrabutylammonium perchlorate in acetonitrile as the electrolyte to introduce N atoms into the resultant GQDs in situ [93].

Hydrothermal method was typically used to fabricate GQDs using rGO as a raw material by chemical cutting and deoxidation in an alkaline medium (Figure 5.6) [96]. Nowadays, ball milling has been widely applied in material synthesis [100]. Wang et al. prepared CQDs with multiple surface states by high‐energy ball milling of a mixture of activated carbon and KOH [98]. These CQDs displayed dual‐wavelength photoluminescence. The first emission peak is independent of excitation wavelength and the second one is not, as well as dual‐wavelength electrochemiluminescence.

5.3.1.2 Bottom‐up Approaches

Bottom‐up approaches include microwave synthesis [101], thermal decomposition [102], template‐based routes [103], and plasma treatment [104]. The raw materials of synthesizing CQDs have molecular precursors such as citrate, carbohydrates, and polymer–silica nanocomposites [105]. Microwave synthesis has many advantages, including effectively cutting down the reaction time and providing homogeneous heating to uniform size distribution of CQDs [101, 106]. In 2009, Zhu et al. first adopted the microwave approach to synthesize CQDs only for several minutes and these CQDs owned the excellent features of bright, stable luminescence and excellent water dispersion [106]. Recently, thermal decomposition has been applied for dehydration and carbonization of organics to synthesize CQDs because of the advantages of simplicity to operate, solvent‐free, short reaction time, low cost, and scalable production [107, 108]. Ma et al. used thermal decomposition for the synthesis of GQDs doped by various elements from a wide range of precursors [107]. The growth process of N‐ and O‐doped GQDs and graphene‐like structures were shown in Figure 5.7. Thereafter, Martindale et al. employed straightforward thermal decomposition of citric acid at 180 °C to prepare CQDs (6.8–2.3 nm) with a relatively broad size distribution [102]. In addition, Jiang et al. described and achieved that a novel one‐step process combined the synthesis of GQDs and the grafting of primary amine functionalities onto the surface of these GQDs using an all‐in‐one small submerged‐arc plasma reactor with low power input [104].

Then, Liu et al. reported a template‐based route to fabricate multicolor photoluminescent CQDs (1.5–2.5 nm) with the polymer–silica nanocomposites as precursors (see Figure 5.8) [103].

5.3.2 Carbon‐quantum‐dot‐Based, Metal‐Free Catalysts for Photocatalysis

In 2010, the photocatalytic properties of CQDs were first found by Li et al. [109]. They fabricated the CQDs with the size of 1.2–3.8 nm by the facile one‐step, alkali‐assisted electrochemical method. These high‐quality CQDs have size‐dependent photoluminescence and excellent up‐conversion luminescence properties (quantum yield about 12%). Then, CQDs/TiO₂ and CQDs/SiO₂ nanocomposites were synthesized by a conventional sol–gel method and their photocatalytic property was evaluated by the degradation reaction of methyl blue (MB) under a 300 W halogen lamp. The results confirmed that MB can be thoroughly reduced after the different illumination time (CQDs/TiO₂, 25 min; and CQDs/SiO₂, 15 min). Since then, CQD‐based photocatalysis has attracted enormous research attention to the fields of photocatalytic degradation, photocatalytic hydrogen production, and photocatalytic conversion of CO₂ [75, 110].

The photocatalytic hydrogen production has long been considered one of the most promising approaches for transforming solar energy to hydrogen energy and meet the energy demands of a growing population [111]. CQDs have a lot of merits, including adjustable fluorescence properties, outstanding electron transfer capacities, and low cost and are thus suitable as photocatalysts. In 2011, Cao et al. found that poly(ethylene glycol) (PEG)‐functionalized CQDs themselves act as photocatalysts and efficiently reduce CO₂, but had the performance of photocatalytic hydrogen production under certain reaction conditions [112]. When light was exposed to an aqueous CuSO₄ solution in the presence of CQDs, they found a color change and precipitate, which was ascribed to the reduction of cupric ion by the H₂ produced. Therefore, metal‐free, CQD‐based photocatalysis began to attract the attention of researchers [113, 114].

Later, Yang et al. found that pure CQDs without any modification and cocatalyst, which was prepared via ultrasonic hydrothermal process, can serve as a photocatalyst for photocatalytic hydrogen generation and the photocatalytic performance of CQDs with methanol as the sacrificial donor was 34.8‐folds higher than that of commercial Degussa P25 photocatalyst under the same test conditions [115]. Moreover, the hydrogen generation rate of pure CQDs was 423.7 mmol g⁻¹ h⁻¹, and its photocatalytic performance almost showed no change after four cycles of testing. Based on the structural characteristics required for photocatalytic hydrogen generation from water, the nitrogen‐doped graphene oxide quantum dots (NGO‐QDs), which were prepared by heat treatment of GO in NH₃ and then oxidization of NH₃‐treated GO using modified Hummers' method, were capable of catalyzing overall water splitting to generate H₂ and O₂ under visible light [116]. The bandgap of the prepared NGO‐QDs were approximately 2.2 eV and thus can absorb visible light to generate excitons. In addition, the oxygen functionalities and doped nitrogen atoms of NGO‐QDs brought about the n‐ and p‐type conductivity, respectively. Therefore, as a metal‐free photocatalyst, NGO‐QDs can effectively generate H₂ and O₂ from water under visible light illumination and the possible mechanism for photocatalytic water splitting is shown in Figure 5.9.

The large amounts of CO₂ emissions cause more and more serious environmental issues because of excessive utilization of fossil fuels [17, 117, 118]. The photocatalytic conversion of CO₂ into chemicals and fuels has been considered a promising approach to solve the problems of global warming and energy crisis [18]. However, the chemical conversion of CO₂ is extremely challenging because CO₂ is an extremely stable molecule and its CO bond has a higher dissociation energy of ∼750 kJ mol⁻¹ [119]. The metal‐free, CQD‐based photocatalysts have been investigated to convert carbon dioxide under visible light illumination [110]. Sahu et al. reported that bare CQDs in an aqueous solution without any surface passivation owned the capability of transforming the CO₂ into formic acid [120]. And then, they still built a baseline in the use of CQDs for capturing visible photons to the photoreduction of CO₂. Then, Yang et al. demonstrated a p‐type silicon nanowire with nitrogen‐doped graphene quantum sheets (N‐GQSs) as a photocathode for the selective photoconversion of CO₂ into CO [121].

Recently, some groups have focused on the photocatalytic oxidation of alkanes and alcohols using the CQDs to settle issues of low efficiency and poor selectivity. In addition, CQDs were also prepared for the activation of oxidant H₂O₂ because of the electron transfer of CQDs under light illumination. A kind of near infrared ray (NIR) light‐driven photocatalyst was fabricated using CQDs with the size of 1–4 nm by Li et al. The catalytic mechanism is shown in Figure 5.10 [123]. Through the alkali‐assisted electrochemical method, a high conversion of 92% and a selectivity of 100% can be achieved in the presence of H₂O₂ for the oxidation of benzyl alcohol to benzaldehyde. Then, they also found that 5–10 nm CQDs prepared by electrochemical ablation of graphite had the excellent catalytic performance for a series of organic reactions (esterification, Beckmann rearrangement, and aldol condensation) in water solution under visible light irradiation [122].

In addition, Li et al. reported that the various sizes of functional CQDs decorated with hydrogen sulfate groups (S‐CQDs) possessed high photocatalytic performance for the ring‐opening reactions of epoxides with methanol and other primary alcohols owing to the photoinduced proton‐generating capacity in the solution and the preparation process, transmission electron microscopy (TEM) image, particle size distribution, and Raman spectrum of S‐CQDs, as shown in Figure 5.11 [124].

Furthermore, CQDs with size of 5 nm synthesized by an electrochemical etching method were directly used as excellent heterogeneous photocatalysts for hydrogen bond catalysis in aldol condensations because of highly efficient electron‐accepting capabilities under visible light irradiation [125]. In this paper, Han et al. found that CQDs have the excellent photocatalytic abilities with 89% yield for catalyzing 4‐cyanobenzaldehyde, which was attributed to their photoinduced electron‐accepting properties. The superficial OH bond will be strengthened, whereas the CO bond will be activated by the efficient electron‐accepting properties of CQDs. The photoenhanced mechanism of hydrogen bond of CQDs under visible light irradiation is shown in Figure 5.12.

5.4 Graphitic Carbon‐Nitride‐Based, Metal‐Free Photocatalysis

5.4.1 Graphitic Carbon Nitride

Unlike TiO₂, which is only active in the UV region, graphitic carbon nitride (g‐C₃N₄) possesses a medium bandgap of c. 2.7 eV, with the CB and VB positions at c. −1.1 eV and c. +1.6 eV versus normal hydrogen electrode, respectively [126, 127]. Thus, g‐C₃N₄ can be used as the metal‐free photocatalyst for visible light photocatalytic water splitting as reported by Wang et al. in 2009 [126]. It is known as the next‐generation photocatalyst because of its facile synthesis, appealing electronic band structure, high physicochemical stability, and “earth‐abundant” nature. g‐C₃N₄ is in the form of two‐dimensional sheets consisting of tri‐s‐triazines interconnected via tertiary amines (Figure 5.13) [126]. As the metal‐free photocatalyst, g‐C₃N₄ can be easily fabricated by thermal condensation of some low‐cost abundant nitrogen‐rich precursors such as melamine [128], cyanamide [129], dicyandiamide [130], urea [131, 132], and thiourea [133, 134] (Figure 5.14).

X‐ray powder diffraction (XRD) patterns are usually used to determine the phase of carbon nitrides. The XRD patterns of g‐C₃N₄ feature two pronounced diffraction peaks at c. 27.4° and c. 13.0° (Figure 5.15a) [126]. For graphitic materials, the former can be indexed as the 002 peak characteristic for interlayer stacking of aromatic systems and the latter can be indexed as the 100 peak that corresponds to the interplanar separation. UV–Vis diffuse reflectance spectra (Figure 5.15b) [126] are commonly used to evaluate the bandgaps (E_g) of g‐C₃N₄ samples. Roughly, E_g can be estimated using the following simple equation: E_g = 1240/λ_g (nm), in which λ_g is the absorption band edge of a given sample. X‐ray photoelectron spectroscopy (XPS) measurements are used to investigate the status of carbon (Figure 5.15c) and nitrogen elements (Figure 5.15d) in g‐C₃N₄, including sp²‐bonded carbon in CC (c. 284.6 eV) and NCN (c. 288.1 eV), the sp²‐bonded nitrogen in CNC (c. 398.7 eV), the nitrogen in tertiary N(C)₃ groups (c. 400.3 eV), and the presence of amino groups (CNH, c. 401.4 eV) caused by imperfect polymerization [134]. Consequently, elemental analysis is employed to determine the elemental content of g‐C₃N₄ materials such as the C and N percentages and the C:N ratio. Importantly, g‐C₃N₄ is recognized as the most stable allotrope among various carbon nitrides under ambient conditions. Thermogravimetric analysis (TGA) reveals that g‐C₃N₄ is thermally stable even in air up to 600 °C, which can be attributed to its aromatic CN heterocycles. Owing to the strong van der Waals interactions between the layers, g‐C₃N₄ is chemically stable in most solvents such as water, alcohols, diethyl ether, toluene, N,N‐dimethylformamide (DMF), tetrahydrofuran (THF), glacial acetic acid, and 0.1 M NaOH aqueous solution [47, 135]. The theoretical specific surface area of ideal monolayer g‐C₃N₄ can be up to 2500 m² g⁻¹ similar to the layered structure in graphite [136]. More importantly, many simple strategies can be used to adjust the properties of g‐C₃N₄ without a significant change of the overall composition by including only two earth‐abundant elements: carbon and nitrogen [47, 137]. Moreover, its polymeric nature allows control over the surface chemistry via molecular‐level modification and surface engineering. Also, g‐C₃N₄ can serve as a host matrix of outstanding compatibility to various inorganic nanoparticles on account of its polymeric nature. This ensures the structure of g‐C₃N₄ with sufficient flexibility. Apparently, these features are very beneficial for the fabrication of g‐C₃N₄‐based composite materials.

The unique aforementioned characteristics of g‐C₃N₄ make this material a very promising photocatalyst for various applications [133]. In recent years, great and fruitful efforts have been made in the field of g‐C₃N₄‐based photocatalysis. To date, most of them demand the utilization of cocatalysts to form surface junctions, thus optimizing light harvesting and improving charge separation and surface catalytic kinetics to achieve efficient photocatalytic activity [138]. However, the widely used cocatalysts usually contain rare and noble metals, e.g. Pt, that are not suitable for large‐scale practical applications. Therefore, the development of high‐performance visible‐light‐driven photocatalysts that are both stable and cost‐efficient nature is required for photocatalysis. In this section, we present an overview on the recent advances of g‐C₃N₄‐based, metal‐free photocatalysts.

5.4.2 Synthesis of Pristine g‐C₃N₄ and its Functionalization

5.4.2.1 Effect of Nitrogen‐Rich Precursors and Reaction Parameters

It is widely reported that these crucial factors include precursor types, pretreatment and modification of precursors, and reaction temperature, duration, and atmosphere and significantly influence the physicochemical properties of g‐C₃N₄, such as C:N ratio, specific surface area, porosity, absorption edge, and nanostructures [139]. g‐C₃N₄ was prepared with various proportions of C:N by calcining melamine at different temperatures in a semiclosed system [128]. When the reaction temperatures increased from 500 to 580 °C, the C:N ratio increased from 0.721 to 0.742 and the bandgaps decreased from 2.8 to 2.75 eV. Because of the structural defects and incomplete condensation, amine functional groups existed with about 2% of hydrogen content and the molar ratio of C to N was smaller than that of the ideal g‐C₃N₄ (0.75) [132]. It is difficult to further lower the hydrogen content by a facile condensation approach; thus, fabrication of an ideal, fully condensed g‐C₃N₄ with a stoichiometric C:N ratio of 0.75 is still a challenge. Actually, for better interactions with reactant molecules, the g‐C₃N₄ with a trace amount of amine groups is advantageous by increasing the g‐C₃N₄ surface activeness [140, 141]. However, more defects could undesirably hinder the charge transportation and separation, resulting in a low photocatalytic activity.

Urea is a superior precursor for preparing g‐C₃N₄ with large specific surface area because it produces sheet‐like g‐C₃N₄ with much smaller thickness [142, 143]. For example, Dong et al. treated urea at 550 °C for different time and found that the thickness of the obtained g‐C₃N₄ was reduced from 36 to 16 nm, and the specific surface area was improved from 31 to 288 m² g⁻¹ as the time increased from 0 to 240 min [142].

To get a relatively larger specific surface area, the g‐C₃N₄ could be prepared by modifying precursors before thermal treatment. Zhang et al., for instance, heated a mixture of melamine and sulfur (S₈) at 650 °C in a N₂ flow for 2 h to obtain g‐C₃N₄ materials with larger surface areas and narrower bandgaps than those of the g‐C₃N₄ prepared from melamine only [144]. Interestingly, such sulfur‐mediated synthesis did not induce the sulfur doping but just affected the polymerization process.

In addition, the unique properties and chemical structures of g‐C₃N₄ are also strongly influenced by the reaction atmospheres such as air, N₂, H₂, Ar, He, and NH₃ through causing disordered structures, defects, and C and N vacancies. Compared with the O vacancies reported by many works [145–147], the fundamental role of C and N vacancies in altering the optical and electronic properties is more promising for the nitride‐based photocatalysts [148, 149]. Holey g‐C₃N₄ (HGCN) nanosheets with abundant in‐plane holes and a large number of C vacancies were prepared by thermal treat of bulk g‐C₃N₄ under NH₃ atmosphere (Figure 5.16a) [150]. The in‐plane holes not only endowed HGCN with more exposed active edges and cross‐plane diffusion channels that strikingly speed up the mass transfer and photogenerated charge diffusion but also provide numerous boundaries and hence decrease the van der Waals interactions to mitigate severe aggregation. The HGCN had a larger specific surface area of 196 m² g⁻¹ compared with bulk g‐C₃N₄ (6 m² g⁻¹) because of the rich in‐plane holes. Also, the abundant in‐plane holes and crumpled structure caused an extremely high pore volume of 1.38 cm³ g⁻¹ for HGCN, accelerating the kinetics of photoactivity by enhancing mass transfer. This could be further supported by the scanning electron microscopy (SEM) (Figure 5.16b,c) and TEM (Figure 5.16d,e) analyses. From the SEM images, the HGCN nanosheets were corrugated with a thickness of c. 20 nm and comprised a number of in‐plane holes with diameters ranging from tens to hundreds of nanometers. This was in agreement with the observation from the TEM images.

Considering the specific surface areas, electronic properties, and bandgaps of g‐C₃N₄ photocatalysts developed from various precursors and different reaction parameters, we can find that these features of g‐C₃N₄ could be altered by the appropriate control over the reaction parameters and the starting materials in order to improve photocatalytic efficiency.

5.4.2.2 Nanostructure Design of g‐C₃N₄

Nanotemplating method is another promising approach to tailor the structural properties and interlayer interactions of g‐C₃N₄ through the introduction of different morphologies and ordered porosity in the bulk g‐C₃N₄. Indeed, the controllable nanostructure design of g‐C₃N₄, such as porous g‐C₃N₄, 1D g‐C₃N₄ nanofiber, hollow g‐C₃N₄ nanospheres, and so forth, has been exhaustively explored by researchers through hard‐templating or soft‐templating methods in liquid precursors [151–155]. The porosity, structure, morphology, surface area, and size can be easily tuned by appropriate templates. The structure ordered and high porous g‐C₃N₄ obtained larger surface areas and more active sites, beneficial for various applications as photocatalysts.

The hard template approach is considered to be a flexible technique for the development of nanostructured g‐C₃N₄ materials. In this approach, hard templates can endow g‐C₃N₄ with various structures and geometries as well as hierarchical pore architectures from nanometers to micrometers in length scales [156–158]. Different nanostructured silica such as silica nanoparticles or nanospheres and mesoporous silica are utilized as hard templates to construct mesoporous g‐C₃N₄ materials, which exhibited accessible open pores, large surface area, and improved light‐harvesting features because of multiple scattering effects and enhanced mass diffusion of reactant molecules [129]. For instance, uniform‐sized silica nanospheres (SNSs) were used as a hard template to prepare mesoporous g‐C₃N₄ (Figure 5.17a) [159]. First, the infiltration and polymerization of cyanamide were completed within the narrow gaps between the SNSs. Then, the SNSs were removed by the HF treatment to get the porous g‐C₃N₄ (Figure 5.17b–e). As such, the resultant g‐C₃N₄ possessed inverse opal structure with spherical pores manifesting the size of used SNSs.

The relatively “greener” soft template process simplifies the synthesis route of g‐C₃N₄ compared with the hard template strategy [160, 161]. The frequently used soft‐structure‐directing agents include ionic liquids, amphiphilic block polymers, and surfactants, determining the highly porous nanoarchitectures of g‐C₃N₄ [154]. Mesoporous g‐C₃N₄ can be fabricated by thermal polymerization reaction of dicyandiamide with numerous soft templates such as nonionic surfactants, amphiphilic block polymers, and also ionic surfactants [154]. Some of the resulting g‐C₃N₄ samples possessed an enlarged specific surface area, a large pore volume, and a high conductivity. For example, the soft templates (Triton X‐100 and ionic liquids) led to the mesoporous g‐C₃N₄ structures with large surface areas (Figure 5.18a,b) [154]. Furthermore, Yan developed a worm‐like porous g‐C₃N₄ by replacing dicyandiamide with a less‐reactive melamine precursor and employing pluronic P123 surfactant as a soft template (Figure 5.18c) [153]. Also, the ionic liquid 1‐butyl‐3‐methylimidazolium tetrafluoroborate (BMIMBF₄) behaves as a soft modifier excellently tuning the surface morphology and chemistry, texture, and semiconductor properties of g‐C₃N₄ in a facile one‐step process since the BF₄⁻ anions can enter the CN condensation scheme [162]. The resulting materials demonstrated a well‐developed “morel‐like” and sponge‐like mesopore architecture with a narrow pore size distribution (Figure 5.18d) [162].

Without doubt, the template‐free method possesses several advantages over the hard template and soft template syntheses. By this approach, the versatile morphologies and desired size, such as nanorods, quantum dots, “seaweed” architecture, microspheres, nanofibers, and so forth, can be simply achieved for g‐C₃N₄ materials [130, 163–168]. For instance, a seaweed‐like g‐C₃N₄ architecture was presented by Qu and coworkers via direct calcination of the freeze‐drying‐assembled, hydrothermally treated dicyandiamide fiber network (Figure 5.19a) [163]. The hydrothermal treatment and freeze‐drying technique is vital for the development of 1D fiber‐like architectures. The calcination of the treated precursors led to network morphologies of interlocking fibers, which has a width of hundreds of nanometers and a length of tens of microns similar to “seaweed” (Figure 5.19b–d). The g‐C₃N₄ nanofibers were rich in mesopores with the pore size smaller than 20 nm (Figure 5.19e). This work presents a very simple method for designing the g‐C₃N₄ and developing high‐performance photocatalysts for hydrogen evolution.

5.4.2.3 Exfoliation of Bulk g‐C₃N₄

Single‐layer g‐C₃N₄ exhibits a unique 2D feature similar to the graphene and its fabrication from bulk g‐C₃N₄ is also similar to the development of graphene from the bulk graphite. Because of the stacking of g‐C₃N₄ layers, the bulk g‐C₃N₄ displays a very small specific surface area (<10 m² g⁻¹) far below its theoretical value of monolayer (2500 m² g⁻¹) [169]. Exfoliating g‐C₃N₄ into a few layers is an efficient strategy to promote the photocatalytic performance of g‐C₃N₄ by causing attractive surface, optical, and electronic properties [170–173]. Many exfoliation methods have been reported so far, such as the ultrasonication‐assisted liquid exfoliation approach, the post‐thermal oxidation etching route, and the combined thermal delamination with a sonication process [171, 174–178]. Between the g‐C₃N₄ layers, the van der Waals forces present weak [179]. Therefore, the acoustic energy of the ultrasonic wave can be transmitted to overcome the van der Waals forces between the layers to obtain the ultrathin g‐C₃N₄ nanosheets in an appropriate solvent. This liquid exfoliation has become a widely known approach by most researchers for exfoliating bulk g‐C₃N₄ because of its facile and convenient process. Recently, utilizing the freeze‐drying‐assembled dicyandiamide as precursors, Qu and coworkers developed a solvothermal exfoliation strategy to generate the single atomic layer of g‐C₃N₄ nanosheet with in‐situ‐formed in‐plane mesopores [178]. Remarkably, the resultant atomically thin mesoporous g‐C₃N₄ nanomesh exhibits dramatically enhanced photocatalytic efficiency for hydrogen evolution under visible light.

5.4.2.4 Elemental Doping of g‐C₃N₄

Doping of g‐C₃N₄ can introduce additional elements and impurities into the g‐C₃N₄ framework to distinctly tune the optical, electronic, and other physical properties. For metal‐free photocatalysts, the elemental doping is considered as the nonmetal doping. Hitherto, numerous studies on the nonmetal doping such as O [180–183], C [184, 185], P [186–188], S [189, 190], B [191–193], I [194, 195], F [196], and some combination thereof [197–199] into g‐C₃N₄ have been extensively reported. Bandgap engineering of g‐C₃N₄ via element doping plays a predominant role to modulate the light absorption and redox band potentials for the targeted photocatalytic applications.

5.4.2.5 Copolymerization of g‐C₃N₄

The photocatalytic activity of g‐C₃N₄ is originated from the π‐conjugated structure. However, because of the defective polymerization and the aromatic π system of g‐C₃N₄ with drawbacks, rapid recombination of charge carriers, inadequate utilization of sunlight, and small surface areas considerably restrict the photochemical functions [200, 201]. Thus, during the copolymerization process, the delocalization of the π electrons can be further extended by modifying and anchoring the existing molecular structure of g‐C₃N₄ with another structure‐matching aromatic groups or organic additives [202–206]. Copolymerization approach as a typical molecular doping can adjust the conventional π systems, electronic properties, optical absorption, band structure, and importantly photocatalytic performance [207–210]. Wang and coworkers have developed a new strategy of copolymerization inspired by the Schiff base chemistry, which is frequently used to extend the π‐conjugated electronic structures of polymers [211]. Until now, they have successfully advanced this novel strategy and demonstrated a family of modified g‐C₃N₄ with grafted organic functional aromatic groups as comonomers such as barbituric acid, 2‐aminobenzonitrile (ABN), phenylurea, diaminomaleonitrile (DAMN), 2‐aminothiophene‐3‐carbonitrile (ATCN), quinolone, pyromellitic dianhydride, 2,6‐diaminopyridine, and 2,4,6‐triaminopyrimidine [203, 211–213].

5.4.3 g‐C₃N₄‐Based, Metal‐Free Catalysts for Photocatalysis

5.4.3.1 Photocatalytic Water Splitting

The photocatalytic reaction of water splitting for H₂ generation involves the following steps: charge carriers are generated upon light absorption, then an electron–hole pair separates and transfers to the surface active sites, and finally a reduction reaction occurs with the surface‐adsorbed proton to generate H₂ [214–216]. To extend the research on sustainable metal‐free g‐C₃N₄ photocatalyst, Suryawanshi et al. investigated on the nanocomposite with nanotubes to understand the electronic and morphological changes in g‐C₃N₄ [217]. Various concentrations of multiwall carbon nanotubes (MWCNT) were mixed with g‐C₃N₄ for the synthesis of an all‐carbon photocatalyst. A twofold increase in H₂ evolution activity was observed under visible light irradiation for the optimized metal‐free photocatalyst system, 0.5% MWCNT/g‐C₃N₄. Lately, further progress toward the development of a metal‐free photocatalyst was made by functionalization of g‐C₃N₄ with an electron acceptor carbon nanoparticles derived from zeolitic imidazolate framework (ZIF). The metal‐free bifunctional catalyst system of ZIF‐derived carbon (C‐ZIF)/g‐C₃N₄ composite was prepared by a facile thermal condensation process (Figure 5.20a) [218]. Precursors for both the carbon nanoparticles and the g‐C₃N₄ were mixed together and treated in one step at 650 °C under N₂ flow to obtain the composite photocatalyst. Carbon nanoparticle with a size of 60 nm was in situ derived from a ZIF during the thermal conversion of melamine into g‐C₃N₄. The composite with encircled carbon nanoparticles grown on sheets of g‐C₃N₄ is provided in Figure 5.20b. Before photocatalytic test, the composite was washed with HCl solution to remove the residual Zn and to make sure it is an all‐carbon system. In comparison with the pristine and Pt‐loaded g‐C₃N₄, all the composites with various concentrations of carbon nanoparticles displayed better activity toward H₂ evolution (Figure 5.20d). The rate of H₂ evolution over 1% carbon‐nanoparticle‐functionalized g‐C₃N₄ (32.6 mol h⁻¹) was 36 times higher than that of the pure g‐C₃N₄ (0.9 mol h⁻¹), under visible light irradiation. Interestingly, under experimental conditions, this metal‐free carbon composite performed 2.8 times better than 3% Pt‐loaded g‐C₃N₄ (11.6 mol h⁻¹). The photoluminescence (PL) spectra (Figure 5.20c) exhibited a small quenching upon loading of g‐C₃N₄ with Pt metal. However, a significant decrease in the PL intensity was observed for the optimized composite with carbon nanoparticles, suggesting an improved efficiency for separation of charge carriers. This trend of the charge carrier dynamics was consistent with the photocatalytic activity.

5.4.3.2 Photocatalytic Reduction of CO₂

It is well known that carbon at the highest valence state in CO₂ molecule is a thermodynamically stable molecule and its transformation to other chemical states requires to overcome high activation energy barriers. The photoreduction of CO₂ can occur in multiple pathways, generating various kinds of gaseous and liquid‐phase products of hydrocarbon and oxygenate. Among the metal‐free photocatalysts, the GO/g‐C₃N₄ nanocomposite has been most frequently studied for the photoreduction of CO₂. In another study, mesoporous phosphorylated g‐C₃N₄ (MPCN) was prepared from bulk g‐C₃N₄ (BCN) by concentrated phosphoric acid posttreatment. Material characterizations proved mesopores formation, exfoliation, and joint PO₄³⁻ groups on MPCN (Figure 5.21a–d) [219]. Transient photocurrent responses and electrochemical impedance spectra (EIS) indicated that the photocurrent for MPCN was higher than that of BCN when light was on (Figure 5.21e). Furthermore, the smaller diameter of semicircle arc of MPCN revealed that it had a lower electron transfer resistance value (Figure 5.21f) than BCN. Therefore, the mobility and separation efficiency of the photoinduced carriers were efficaciously enhanced via phosphorylation of the catalyst. The light control experiments indicated that light and CO₂ gas were required for CO₂ photocatalytic reduction. Therefore, the carbon source of CO and CH₄ should be CO₂ rather than g‐C₃N₄. After 1 h of simulated solar light irradiation, the fuel gas production using MPCN (7.91 μmol for CO, 15.6 μmol for CH₄, 4.19 μmol for O₂, and 1.18 μmol for H₂) was much higher than that using MCN (mesoporous g‐C₃N₄) and PCN (PO₄³⁻ modified g‐C₃N₄). More importantly, it was found that the total main products of PCN and MCN (6.27 μmol for CO; 7.85 μmol for CH₄) were also lower than those of MPCN. It indicated the synergistic cooperation of phosphorylation‐induced mesoporous structure and phosphate group over g‐C₃N₄ for photocatalytic reduction of CO₂ (Figure 5.21g). The corresponding bandgap values of MPCN (2.83 eV) and BCN (2.69 eV) were compared (Figure 5.21h). It indicated that the CB of MPCN and BCN should be −0.78 and −0.64 V, respectively. Therefore, MPCN displayed stronger photocatalytic reduction ability than BCN. The photocatalytic mechanism was proposed and shown in Figure 5.21i. The synergistic cooperation of mesoporous structure and phosphate groups resulted in the enhancement of photocatalytic solar‐to‐fuel conversions.

5.4.3.3 Photocatalytic Removal of NO_x

Nitric oxide (NO_x) emission into the atmosphere has caused an increasing concern about the impact to environmental pollution, which can result in the formation of haze, acid rain, and photochemical smog [220]. Up to now, a variety of solutions have been developed to remove NO_x including biofiltration, wet scrubbing, adsorption and chemical reaction, selective catalytic reduction, photocatalysis, and so forth. Among them, the photocatalytic removal of NO_x is regarded as an alternative to conventional methods because of the high efficiency and low‐cost nature of the process [221, 222]. Remarkably, g‐C₃N₄ can also be the most widely adopted metal‐free photocatalyst [223, 224].

From the morphology‐controlled perspective, a honeycomb‐like g‐C₃N₄ was synthesized via thermal condensation of urea with the addition of water, as shown in Figure 5.22a [225]. The water bubble formed at high temperature serves as a soft template for the formation of a honeycomb‐like structure. In addition, prolonging the condensation time could change the porous honeycomb structure to a velvet‐like nanostructure. Although the as‐prepared honeycomb‐like g‐C₃N₄ has a comparatively small surface area of around 39 m² g⁻¹, it still displays much enhanced photocatalytic activity with 48% NO_x removal at 30 min under visible light (Figure 5.22b).

5.4.3.4 Photocatalytic Degradation of Organic Pollutants

Environmental pollution caused by noxious organic substances released to water streams has emerged as a seriousproblem over the past few decades. Various wastewater treatment technologies have been developed, and some of them have been successfully industrialized, such as bioremediation using microorganisms, membrane filtration, and chemical oxidation. The safe, robust, low‐cost technologies capable of tackling persistent organic pollutants of low concentrations are urgently needed. Photocatalytic degradation of organic pollutants has been considered a green technique with great potential to decompose organic pollutants of high toxicity and low concentration from wastewater under UV or visible light irradiation [226–228]. During such a photocatalytic process, photoinduced electrons can be captured by the dissolved oxygen in water to generate superoxide radicals (^•O²⁻), while the hole can not only directly oxidize organic pollutants but also react with water to produce a hydroxyl radical (^•OH). All these reactive species contribute to transform the organic pollutants in water to nontoxic small molecules or CO₂ [229, 230].

For instance, the porous g‐C₃N₄ can be prepared by pyrolysis of dicyandiamide in air with urea as the bubble template [231]. The surface area of the porous g‐C₃N₄ can be regulated by controlling the dicyandiamide:urea mass ratio and pyrolysis temperature. Promising activity was obtained using the porous g‐C₃N₄ for photodegradation of methylene blue dye and phenol in an aqueous solution under visible light. In addition, they also realized the conversion of g‐C₃N₄ morphology from nanoplates to nanorods by a facile reflux process (Figure 5.23a) [164]. The photocatalytic activity of g‐C₃N₄ nanorods for dye degradation was found to be higher than that of nanoplates under visible light irradiation. Such a phenomenon occurred because of the increase in active facets and elimination of surface defects (Figure 5.23b). Moreover, a 3D ordered macroporous g‐C₃N₄ photocatalyst was constructed by a facile thermal condensation assisted by a colloidal crystal as the template. The as‐prepared ordered macroporous g‐C₃N₄ photocatalyst exhibits enhanced photodegradation activity for rhodamine B (RhB) dye and the removal efficiency can reach up to 100% within 40 min under visible light, which is about 5.3 times higher than that of pristine g‐C₃N₄. Such a great improvement is attributed to the unique 3D ordered macroporous structure facilitating light absorption and charge separation and transfer. Other morphologies of g‐C₃N₄ such as ultrathin nanosheets and nanofibers have also been reported by different synthesis strategies [133, 232]. Most of them have shown promising photocatalytic activities for degradation of organic contaminants including dyes, phenols, and antibiotics. Forming a composite with other functional components is another way to improve the activity of g‐C₃N₄ for photocatalytic degradation of organic pollutants. Ge et al. reported that the polyaniline (PANi)/g‐C₃N₄ composite photocatalyst can be prepared by in situ deposition tactics in an ice bath [233]. The composite structure displays higher photocatalytic activity for degradation of MB under visible light irradiation than the pure g‐C₃N₄ and TiO₂ (P25). PL spectra and photocurrent response results suggest a synergistic effect between PANi and g‐C₃N₄ in promoting charge carrier separation and transfer, thus improving photocatalytic activity.

5.4.3.5 Photocatalytic Organic Synthesis

Organic chemicals are essential for the manufacture of a great number of products and other chemicals including pharmaceuticals, pesticides, and food additives [234, 235]. Conventional industrial routes for many important organic chemicals typically require harsh operating conditions, such as high temperature and pressure, and some dangerous and toxic redox agents including lithium aluminum hydride and permanganate. In this regard, photocatalytic technology could open up a facile route for organic synthesis as it could utilize light as an energy source to drive chemical reactions under milder conditions and avoid the use of hazardous redox agents. As discussed above, semiconductor photocatalysis could generate both oxidizing holes and reducing electrons on the photocatalyst at the same time upon photoexcitation. It is, therefore, well suited for the synthesis of organic chemicals either through oxidative and/or reductive pathways. As an emerging field for photocatalytic applications, there are many knowledge gaps and technological difficulties in photocatalytic organic synthesis using semiconductor materials, especially metal‐free material systems. Apart from the conversion and yield, another key issue with the photocatalytic process for targeted organic synthesis is to improve the selectivity of the products. It is worth noting that each photocatalyst needs to be optimized for a specific organic synthesis reaction in a case‐by‐case manner, as the selectivity control should depend on the molecular structure and properties of the specific organic substrate as well as the photocatalyst properties [235]. A variety of strategies have been developed to improve the yields and selectivity in photocatalytic organic synthesis. Examples of such strategies include (i) tuning the VB and CB potentials of the given semiconductor photocatalyst by band engineering to enhance the reactivity; (ii) modifying the surface properties of the given photocatalysts to control the adsorption of the reactant on the photocatalyst surface, the desorption of products from the photocatalyst surface, and the kinds of interfacial charge transfer; and (iii) optimizing the reaction conditions (e.g. temperature, solvent, and pH value).

The mesoporous g‐C₃N₄ photocatalyst has also been used to activate O₂ for the highly selective oxidation of benzyl alcohols to the corresponding aldehydes under visible light [236]. In another work, mesoporous g‐C₃N₄ can act as an effective photocatalyst for oxidation of the primary CH bonds in toluene to benzaldehyde in the presence of O₂ [237].

5.4.3.6 Photocatalytic Bacteria Disinfection

Recent advances have demonstrated the potential capabilities of some metal‐free photocatalysts, such as g‐C₃N₄‐based photocatalysts, graphene‐based photocatalysts, and elemental photocatalysts, for water disinfection under visible light irradiation. For instance, Quan and coworkers found that the atomic single‐layer g‐C₃N₄ has excellent photocatalytic activity for inactivation of Escherichia coli under visible light [238].

5.5 Graphene/g‐C₃N₄ Metal‐Free Catalysts for Photocatalysis

As a metal‐free organic semiconductor, g‐C₃N₄ is a promising analog of graphene, which is rapidly developing because of its excellent chemical and electronic properties [150, 175]. Therefore, the combination of graphene or graphene derivatives with C₃N₄ will improve the photocatalytic activity of composite photocatalysts. Thus, graphene/g‐C₃N₄ composite photocatalysts have drawn great attention because of their outstanding physicochemical and electrical properties. In the aspect of theoretical calculation, the interfacial effect, stacking patterns, and the correlations between electronic structures and related properties of the hybrid graphene/g‐C₃N₄ nanocomposite were systematically investigated and calculated by the first‐principles method [239–241].

The metal‐free graphene/g‐C₃N₄ catalysts were fabricated through different approaches and applied in photocatalysis because of an appropriate graphene ratio resulting in high light utilization, improved oxidation capability, and excellent electron transport property. As early as 2011, Xiang et al. prepared a series of graphene/g‐C₃N₄ composite photocatalysts for improving visible light photocatalytic H₂ generation efficiency by an impregnation–chemical reduction strategy as follows: first, melamine and the different specified volume of GO were dispersed in distilled water, then the mixture was reduced by adding hydrazine hydrate, and finally graphene/g‐C₃N₄ composite was obtained by thermal treatment at 550 °C under flowing nitrogen. Their work demonstrated that the optimal graphene content is 1.0 wt% in graphene/g‐C₃N₄ composite and the corresponding H₂ generation rate was 451 μmol h⁻¹ g⁻¹, which was more than three times better than that of pure g‐C₃N₄ [242]. However, an extremely simple method was adopted to synthesize g‐C₃N₄/rGO photocatalyst for enhancing photocatalytic capability by direct mixing of GO and g‐C₃N₄ under the condition of ultrasonication [243]. Thereafter, a series of g‐C₃N₄/rGO nanocomposites with different rGO contents were synthesized by a facile thermal treatment of cyanamide and GO [244]. They found that g‐C₃N₄/rGO nanocomposites were formed by cross‐linked COC bonds rather than non‐covalent bonds during thermal conversion (Figure 5.24a). In addition, g‐C₃N₄/rGO nanocomposites exhibited the excellent photocatalytic efficiency for photocatalytic degradation of RhB and 4‐nitrophenol under visible light (Figure 5.24b–e).

For increasing visible light photoreduction rate of CO₂ to methane, Ong et al. adopted a facile one‐pot impregnation–thermal reduction method to synthesize sandwich‐like metal‐free graphene/g‐C₃N₄ photocatalyst by direct polymerization of urea in the presence of GO as a structure‐guiding agent at 520 °C (Figure 5.25a). A charge transfer mechanism of graphene/g‐C₃N₄ photocatalyst was illustrated in Figure 5.25b, showing the formation of CH₄ by reduction of CO₂ with H₂O. When the graphene content is 1.5 wt%, graphene/g‐C₃N₄ possessed a 2.3 times enhancement in comparison with pure g‐C₃N₄ [131].

Similarly, graphene/g‐C₃N₄ composites were also constructed using dicyandiamide as the monomers of g‐C₃N₄ via this facile impregnation–thermal reduction approach [245, 246]. Moreover, a novel g‐C₃N₄/rGO hybrid material was used for the photocatalytic decomposition of organic pollutants under sun solar light by the reaction of an aqueous suspension of GO with cyanamide at a low temperature of 100 °C [247]. In generally, GO and g‐C₃N₄ will not be successfully coupled together because of their negative polarity [248, 249]. To solve this problem, Ong et al. achieved the surface protonation modification of g‐C₃N₄ (pC₃N₄) by vigorously stirring in the acid suspension and successfully obtained 2D/2D rGO/pC₃N₄ nanostructures for improving photocatalytic reduction of carbon dioxide to methane by the π–π stacking and electrostatic attraction of oppositely charged materials (Figure 5.26) [250]. Owing to the effective charge transfer across the rGO/pC₃N₄‐layered heterojunction to prevent the recombination of electron–hole pairs, rGO/pC₃N₄ possesses significantly 5.4 and 1.7 times enhancement of CH₄ evolution over pure pC₃N₄ and rGO/C₃N₄ nanocomposites, respectively. Furthermore, a similar strategy was adopted to synthesize rGO/g‐C₃N₄ composite with well‐contacted 2D/2D heterostructure for the photocatalytic degradation of methylene blue by coupled pC₃N₄ with GO during the photo‐assisted electrostatic assembly [251]. To further improve the photocatalytic efficiency of g‐C₃N₄, Zhao et al. designed a novel g‐C₃N₄/rGO/cellulose acetate (g‐C₃N₄/rGO/CA) composite photocatalytic membrane with photoresponse, excellent photocatalytic activity and low cost for the removal of organic contaminants, complete bacterial inactivation, and surface water treatment in the integrated process of filtration with visible light photocatalysis [252]. Furthermore, the novel graphene/g‐C₃N₄ photocatalysts were produced by wrapping polymeric rGO and g‐C₃N₄ sheets in α‐sulfur crystals for the application of bacterial disinfection [253]. With the different wrapping sequences, the prepared samples have different photocatalytic inactivation activity toward E. coli K‐12 CElls and the different inactivation mechanism.

5.6 CQDs/g‐C₃N₄ Metal‐Free Catalysts for Photocatalysis

As a promising metal‐free catalyst, CQDs/g‐C₃N₄ also has attracted interest in the field of photocatalysis. In 2015, Liu and coworkers prepared CQDs/g‐C₃N₄ nanocomposites composed of low cost and environmentally benign materials by heating a mixture of ammonia‐treated CQDs and urea powder at 550 °C and then applied as metal‐free photocatalysts for effectively generating H₂ from water under visible light illumination [254]. The CQDs/g‐C₃N₄ nanocomposites without any sacrificial agents can split water into H₂ and O₂ with unprecedented quantum efficiency (QE) under different wavelengths of light (QEs of 16% at 420 nm, 6.29% at 580 nm, and 4.42% at 600 nm). After 200 cyclic tests over 200 days, the composition, structure, and photocatalytic activity of the nanocomposites had almost no change. Subsequently, CQDs/g‐C₃N₄ nanocomposite was constructed via a simple one‐step hydrothermal method as metal‐free photocatalysts for hydrogen production from water in the NIR region (Figure 5.27) [255].

5.7 Summary and Outlook

Until now, the direct conversion of solar energy to energy fuels and chemical energy has been regarded as one of the green sustainable ways to address the energy and environmental crisis in the future. Therefore, the development of active, earth‐abundant, and stable catalyst materials that are capable of photocatalysis is vitally important in clean and sustainable energy conversion technology. Because of abundant carbon sources in earth, the development of carbon‐based, metal‐free catalysts will certainly bring about significant economic, environmental, and social benefits. In recent years, carbon‐based, metal‐free materials, including graphene, CQDs, and g‐C₃N₄, have been developed as low‐cost and highly efficient photocatalysts to reduce or replace metal‐based catalysts in the application fields of photocatalytic degradation, photocatalytic hydrogen production, and photocatalytic conversion of CO₂. This chapter highlights on the important developments and the versatile advantages of carbon‐based, metal‐free photocatalysis in the different photocatalytic regions.

To date, although some considerable achievements have been realized, there are still many issues and challenges to rationally construct highly efficient carbon‐based, metal‐free photocatalysts toward various applications and deeply understand the underlying enhancement mechanism of the photocatalysts. In comparison with metal‐based catalysts, the reaction mechanism and relationship between the active sites and their catalytic performance of metal‐free carbonaceous catalysts are immature; many aspects need further research and exploration. Accordingly, more studies are also needed to make full use of the optimal catalytic structure and mechanism of carbon‐based, metal‐free photocatalysts. Although there are many problems and challenges, we firmly believe that under the efforts of many researchers, substantial breakthroughs in the practical applications of carbon‐based, metal‐free photocatalysts are expected to occur in the near future.

References

1. 1 Dai, L., Xue, Y., Qu, L. et al. (2015). Chem. Rev. 115: 4823–4892.
2. 2 Xiang, Q., Cheng, B., and Yu, J. (2015). Angew. Chem. Int. Ed. 54: 11350–11366.
3. 3 Zhang, P., Wang, T., Chang, X., and Gong, J. (2016). Acc. Chem. Res. 49: 911–921.
4. 4 Walter, M.G., Warren, E.L., McKone, J.R. et al. (2010). Chem. Rev. 110: 6446–6473.
5. 5 Karkas, M.D., Verho, O., Johnston, E.V., and Akermark, B. (2014). Chem. Rev. 114: 11863–12001.
6. 6 Wang, H., Zhang, L., Chen, Z. et al. (2014). Chem. Soc. Rev. 43: 5234–5244.
7. 7 Yuan, Y.‐P., Ruan, L.‐W., Barber, J. et al. (2014). Energy Environ. Sci. 7: 3934–3951.
8. 8 Cao, S. and Yu, J. (2016). J. Photochem. Photobiol., C 27: 72–99.
9. 9 Kou, J., Lu, C., Wang, J. et al. (2017). Chem. Rev. 117: 1445–1514.
10. 10 Yang, M.‐Q., Zhang, N., Pagliaro, M., and Xu, Y.‐J. (2014). Chem. Soc. Rev. 43: 8240–8254.
11. 11 Hisatomi, T., Kubota, J., and Domen, K. (2014). Chem. Soc. Rev. 43: 7520–7535.
12. 12 Ma, Y., Wang, X., Jia, Y. et al. (2014). Chem. Rev. 114: 9987–10043.
13. 13 Kou, J. and Varma, R.S. (2012). ChemSusChem 5: 2435–2441.
14. 14 Khin, M.M., Nair, A.S., Babu, V.J. et al. (2012). Energy Environ. Sci. 5: 8075–8109.
15. 15 Lang, X., Chen, X., and Zhao, J. (2014). Chem. Soc. Rev. 43: 473–486.
16. 16 Xiao, F.‐X., Miao, J., and Liu, B. (2014). J. Am. Chem. Soc. 136: 1559–1569.
17. 17 Chang, X., Wang, T., and Gong, J. (2016). Energy Environ. Sci. 9: 2177–2196.
18. 18 Habisreutinger, S.N., Schmidt‐Mende, L., and Stolarczyk, J.K. (2013). Angew. Chem. Int. Ed. 52: 7372–7408.
19. 19 Tu, W., Zhou, Y., Liu, Q. et al. (2013). Adv. Funct. Mater. 23: 1743–1749.
20. 20 Tu, W., Zhou, Y., and Zou, Z. (2014). Adv. Mater. 26: 4607–4626.
21. 21 Chen, J., Wang, D., Qi, J. et al. (2015). Small 11: 420–425.
22. 22 Zhang, G., Lan, Z.‐A., and Wang, X. (2016). Angew. Chem. Int. Ed. 55: 15712–15727.
23. 23 Singh, W.M., Pegram, D., Duan, H. et al. (2012). Angew. Chem. Int. Ed. (7): 1653–1656.
24. 24 Tachibana, Y., Vayssieres, L., and Durrant, J.R. (2012). Nat. Photonics 6: 511–518.
25. 25 Fujishima, A. and Honda, K. (1972). Nature 238: 37–38.
26. 26 Carey, J.H., Lawrence, J., and Tosine, H.M. (1976). Bull. Environ. Contam. Toxicol. 16: 697–701.
27. 27 Yu, J., Low, J., Xiao, W. et al. (2014). J. Am. Chem. Soc. 136: 8839–8842.
28. 28 Van, C.N., Chang, W.S., Chen, J.‐W. et al. (2015). Nano Energy 15: 625–633.
29. 29 Wang, X., Liao, M., Zhong, Y. et al. (2012). Adv. Mater. 24: 3421–3425.
30. 30 Huang, W.C., Lyu, L.M., Yang, Y.C., and Huang, M.H. (2012). J. Am. Chem. Soc. 134: 1261–1267.
31. 31 Zhou, X., Xu, Q., Lei, W. et al. (2014). Small 10: 674–679.
32. 32 Hu, Y., Gao, X., Yu, L. et al. (2013). Angew. Chem. Int. Ed. 52: 5636–5639.
33. 33 Chen, X., Choing, S.N., Aschaffenburg, D.J. et al. (2017). J. Am. Chem. Soc. 139: 1830–1841.
34. 34 Asahi, R., Morikawa, T., Irie, H., and Ohwaki, T. (2014). Chem. Rev. 114: 9824–9852.
35. 35 Liu, G., Yang, H.G., Pan, J. et al. (2014). Chem. Rev. 114: 9559–9612.
36. 36 Linares, N., Silvestre‐Albero, A.M., Serrano, E. et al. (2014). Chem. Soc. Rev. 43: 7681–7717.
37. 37 Zhou, P., Yu, J., and Jaroniec, M. (2014). Adv. Mater. 26: 4920–4935.
38. 38 Jiang, R., Li, B., Fang, C., and Wang, J. (2014). Adv. Mater. 26: 5274–5309.
39. 39 Qu, Y. and Duan, X. (2013). Chem. Soc. Rev. 42: 2568–2580.
40. 40 Tu, W., Zhou, Y., and Zou, Z. (2013). Adv. Funct. Mater. 23: 4996–5008.
41. 41 Han, C., Zhang, N., and Xua, Y.‐J. (2016). Nano Today 11: 351–372.
42. 42 Liu, S., Tang, Z.‐R., Sun, Y. et al. (2015). Chem. Soc. Rev. 44: 5053–5075.
43. 43 Allen, M.J., Tung, V.C., and Kaner, R.B. (2010). Chem. Rev. 110: 132–145.
44. 44 Yasaei, P., Kumar, B., Foroozan, T. et al. (2015). Adv. Mater. 27: 1887–1892.
45. 45 Brent, J.R., Savjani, N., Lewis, E.A. et al. (2014). Chem. Commun. 50: 13338–13341.
46. 46 Cao, S., Low, J., Yu, J., and Jaroniec, M. (2015). Adv. Mater. 27: 2150–2176.
47. 47 Wang, X., Blechert, S., and Antonietti, M. (2012). ACS. Catal. 2: 1596–1606.
48. 48 Nag, A., Raidongia, K., Hembram, K.P.S.S. et al. (2010). ACS Nano 4: 1539–1544.
49. 49 Raidongia, K., Angshuman, N., Hembram, K.P.S.S. et al. (2010). Chem. Eur J. 16: 149–157.
50. 50 Goldstein, A., Geffen, Y., and Goldenberg, A. (2001). J. Am. Chem. Soc. 84: 642–644.
51. 51 Novoselov, K.S., Geim, A.K., Morozov, S.V. et al. (2004). Science 306: 666–669.
52. 52 Geim, A.K. (2009). Science 324: 1530.
53. 53 Zhang, Y.B., Tan, Y.W., Stormer, H.L., and Kim, P. (2005). Nature 438: 201–204.
54. 54 Novoselov, K.S., McCann, E., Morozov, S.V. et al. (2006). Nat. Phys. 2: 177–180.
55. 55 Novoselov, K.S., Jiang, Z., Zhang, Y. et al. (2007). Science 315: 1379–1379.
56. 56 Novoselov, K.S., Geim, A.K., Morozov, S.V. et al. (2005). Nature 438: 197–200.
57. 57 Balandin, A.A. (2011). Nat. Mater. 10: 569–581.
58. 58 Dreyer, D.R., Ruoff, R.S., and Bielawski, C.W. (2010). Angew. Chem. Int. Ed. 49: 9336–9344.
59. 59 Du, X., Skachko, I., Barker, A., and Andrei, E.Y. (2008). Nat. Nanotechnol. 3: 491–495.
60. 60 Wu, Z.‐S., Ren, W., Gao, L. et al. (2009). ACS Nano 3: 411–417.
61. 61 Stankovich, S., Dikin, D.A., Dommett, G.H. et al. (2006). Nature 442: 282–286.
62. 62 Lee, C., Wei, X., Kysar, J.W., and Hone, J. (2008). Science 321: 385–388.
63. 63 Lazar, P., Karlicky, F., Jurecka, P. et al. (2013). J. Am. Chem. Soc. 135: 6372–6377.
64. 64 Yeh, T.‐F., Teng, C.‐Y., Chen, L.‐C. et al. (2016). J. Mater. Chem. A 4: 2014–2048.
65. 65 Jung, I., Dikin, D.A., Piner, R.D., and Ruoff, R.S. (2008). Nano. Lett. 8: 4283–4287.
66. 66 Boukhvalov, D.W. and Katsnelson, M.I. (2008). J. Am. Chem. Soc. 130: 10697–10701.
67. 67 Lahaye, R.J.W.E., Jeong, H.K., Park, C.Y., and Lee, Y.H. (2009). Phys. Rev. B 79: 125435–125438.
68. 68 Mathkar, A., Tozier, D., Cox, P. et al. (2012). J. Phys. Chem. Lett. 3: 986–991.
69. 69 Yeh, T.‐F., Syu, J.‐M., Cheng, C. et al. (2010). Adv. Funct. Mater. 20: 2255–2262.
70. 70 Hsu, H.‐C., Shown, I., Wei, H.‐Y. et al. (2013). Nanoscale 5: 262–268.
71. 71 Dreyer, D.R., Park, S., Bielawski, C.W., and Ruoff, R.S. (2010). Chem. Soc. Rev. 39: 228–240.
72. 72 Zhao, Z., Sun, Y., and Dong, F. (2014). Nanoscale 7: 15–37.
73. 73 Xu, J., Luo, L., Xiao, G. et al. (2014). ACS Catal. 4: 3302–3306.
74. 74 Zhang, Z., Long, J., Yang, L. et al. (2011). Chem. Sci. 2: 1826–1830.
75. 75 Yu, H., Shi, R., Zhao, Y. et al. (2016). Adv. Mater. 28: 9454–9477.
76. 76 Xu, X.Y., Ray, R., Gu, Y.L. et al. (2004). J. Am. Chem. Soc. 126: 12736–12737.
77. 77 Sun, Y.‐P., Zhou, B., Lin, Y. et al. (2006). J. Am. Chem. Soc. 128: 7756–7757.
78. 78 Feng, T., Ai, X., An, G. et al. (2016). ACS Nano 10: 5587–5587.
79. 79 Tang, J., Kong, B., Wu, H. et al. (2013). Adv. Mater. 25: 6569–6574.
80. 80 Zheng, M., Liu, S., Li, J. et al. (2014). Adv. Mater. 26: 3554–3560.
81. 81 Antaris, A.L., Robinson, J.T., Yaghi, O.K. et al. (2013). ACS Nano 7: 3644–3652.
82. 82 Zhu, S., Meng, Q., Wang, L. et al. (2013). Angew. Chem. Int. Ed. 52: 3953–3957.
83. 83 Ding, C., Zhu, A., and Tian, Y. (2014). Acc. Chem. Res. 47: 20–30.
84. 84 Zhang, Y.‐Q., Ma, D.‐K., Zhang, Y.‐G. et al. (2013). Nano Energy 2: 545–552.
85. 85 Hou, J., Cheng, H., Yang, C. et al. (2015). Nano Energy 18: 143–153.
86. 86 Fang, S., Xia, Y., Lv, K. et al. (2016). Appl. Catal., B 185: 225–232.
87. 87 Zhao, Y., Zhao, B., Liu, J. et al. (2016). Angew. Chem. Int. Ed. 55: 4215–4219.
88. 88 Linic, S., Aslam, U., Boerigter, C., and Morabito, M. (2015). Nat. Mater. 14: 567–576.
89. 89 Cao, L., Wang, X., Meziani, M.J. et al. (2007). J. Am. Chem. Soc. 129: 11318–11319.
90. 90 Li, Y., Hu, Y., Zhao, Y. et al. (2011). Adv. Mater. 23: 776–780.
91. 91 Fan, Y., Cheng, H., Zhou, C. et al. (2012). Nanoscale 4: 1776–1781.
92. 92 Cheng, H., Zhao, Y., Fan, Y. et al. (2012). ACS Nano 6: 2237–2244.
93. 93 Li, Y., Zhao, Y., Cheng, H. et al. (2012). J. Am. Chem. Soc. 134: 15–18.
94. 94 Zhang, Z., Zhang, J., Chen, N., and Qu, L. (2012). Energy Environ. Sci. 5: 8869–8890.
95. 95 Gao, S., Chen, Y., Fan, H. et al. (2014). J. Mater. Chem. A 2: 6320–6325.
96. 96 Pan, D., Zhang, J., Li, Z., and Wu, M. (2010). Adv. Mater. 22: 734–738.
97. 97 Zhuo, S., Shao, M., and Lee, S.‐T. (2012). ACS Nano 6: 1059–1064.
98. 98 Wang, L., Chen, X., Lu, Y. et al. (2015). Carbon 94: 472–478.
99. 99 Li, X., Wang, H., Shimizu, Y. et al. (2011). Chem. Commun. 47: 932–934.
100. 100 Pan, H., Liu, L., Guo, Z. et al. (2003). Nano Lett. 3: 29–32.
101. 101 Tang, L., Ji, R., Cao, X. et al. (2012). ACS Nano 6: 5102–5110.
102. 102 Martindale, B.C.M., Hutton, G.A.M., Caputo, C.A., and Reisner, E. (2015). J. Am. Chem. Soc. 137: 6018–6025.
103. 103 Liu, R., Wu, D., Liu, S. et al. (2009). Angew. Chem. Int. Ed. 48: 4598–4601.
104. 104 Jiang, H., Chen, F., Lagally, M.G., and Denes, F.S. (2010). Langmuir 26: 1991–1995.
105. 105 Lim, S.Y., Shen, W., and Gao, Z. (2015). Chem. Soc. Rev. 44: 362–381.
106. 106 Zhu, H., Wang, X., Li, Y. et al. (2009). Chem. Commun. 1: 5118–5120.
107. 107 Ma, C.‐B., Zhu, Z.‐T., Wang, H.‐X. et al. (2015). Nanoscale 7: 10162–10169.
108. 108 Zhai, X., Zhang, P., Liu, C. et al. (2012). Chem. Commun. 48: 7955–7957.
109. 109 Li, H.T., He, X.D., Kang, Z.H. et al. (2010). Angew. Chem. Int. Ed. 49: 4430–4434.
110. 110 Shiral Fernando, K.A., Sahu, S., Liu, Y. et al. (2015). ACS Appl. Mater. Interfaces 7: 8363–8376.
111. 111 Moniz, S.J.A., Shevlin, S.A., Martin, D.J. et al. (2015). Energy Environ. Sci. 8: 731–759.
112. 112 Cao, L., Sahu, S., Anilkumar, P. et al. (2011). J. Am. Chem. Soc. 133: 4754–4757.
113. 113 Wei, J., Wang, H.Z., Zhang, Q.H., and Li, Y.G. (2015). Chem. Lett. 44: 241–243.
114. 114 Yang, P.J., Zhao, J.H., Wang, J. et al. (2015). ChemPhysChem 16: 3058–3063.
115. 115 Yang, P.J., Zhao, J.H., Wang, J. et al. (2015). RSC Adv. 5: 21332–21335.
116. 116 Yeh, T.F., Teng, C.Y., Chen, S.J., and Teng, H.S. (2014). Adv. Mater. 26: 3297–3303.
117. 117 Goeppert, A., Czaun, M., Jones, J.P. et al. (2014). Chem. Soc. Rev. 43: 7995–8048.
118. 118 Mikkelsen, M., Jørgensen, M., and Krebs, F.C. (2010). Energy Environ. Sci. 3: 43–81.
119. 119 Xie, S., Zhang, Q., Liu, G., and Wang, Y. (2016). Chem. Commun. 52: 35–59.
120. 120 Sahu, S., Cao, L., Meziani, M.J. et al. (2015). Chem. Phys. Lett. 634: 122.
121. 121 Yang, K.D., Ha, Y., Sim, U. et al. (2016). Adv. Funct. Mater. 26: 233–242.
122. 122 Li, H., Liu, R., Kong, W. et al. (2014). Nanoscale 6: 867–873.
123. 123 Li, H.T., Liu, R.H., Lian, S.Y. et al. (2013). Nanoscale 5: 3289–3297.
124. 124 Li, H., Sun, C., Ali, M. et al. (2015). Angew. Chem. Int. Ed. 54: 8420–8424.
125. 125 Han, Y., Huang, H., Zhang, H. et al. (2014). ACS Catal. 4: 781–787.
126. 126 Wang, X., Maeda, K., Thomas, A. et al. (2009). Nat. Mater. 8: 76–80.
127. 127 Yan, S.C., Lv, S.B., Li, Z.S., and Zou, Z.G. (2010). Dalton Trans. 39: 1488–1491.
128. 128 Yan, S.C., Li, Z.S., and Zou, Z.G. (2009). Langmuir 25: 10397–10401.
129. 129 Wang, X., Maeda, K., Chen, X. et al. (2009). J. Am. Chem. Soc. 131: 1680–1681.
130. 130 Bai, X., Yan, S., Wang, J. et al. (2014). J. Mater. Chem. A 2: 17521–17529.
131. 131 Ong, W.J., Tan, L.L., Chai, S.P., and Yong, S.T. (2015). Chem. Commun. 51: 858–861.
132. 132 Zhang, Y., Liu, J., Wu, G., and Chen, W. (2012). Nanoscale 4: 5300–5303.
133. 133 Ong, W.J., Tan, L.L., Ng, Y.H. et al. (2016). Chem. Rev. 116: 7159–7329.
134. 134 Zhang, G., Zhang, J., Zhang, M., and Wang, X. (2012). J. Mater. Chem. 22: 8083–8091.
135. 135 Dong, G.P., Zhang, Y.H., Pan, Q.W., and Qiu, J.R. (2014). J. Photochem. Photobiol. C 20: 33–50.
136. 136 Sano, T., Tsutsui, S., Koike, K. et al. (2013). J. Mater. Chem. A 1: 6489–6496.
137. 137 Wang, Y., Wang, X., and Antonietti, M. (2012). Angew. Chem. Int. Ed. 51: 68–89.
138. 138 (a) Kudo, A. and Miseki, Y. (2009). Chem. Soc. Rev. 38: 253–278.(b) Hensel, J., Wang, G., Li, Y., and Zhang, J.Z. (2010). Nano Lett. 10: 478–483.(c) Li, Q., Guo, B.D., Yu, J.G. et al. (2011). J. Am. Chem. Soc. 133: 10878–10884.(d) Johnson, T.C., Morris, D.J., and Wills, M. (2010). Chem. Soc. Rev. 39: 81–88.
139. 139 Xiao, H., Wang, W., Liu, G. et al. (2015). Appl. Surf. Sci. 358 (Part A): 313–318.
140. 140 Cui, Y., Zhang, G., Lin, Z., and Wang, X. (2016). Appl. Catal., B 181: 413–419.
141. 141 Tay, Q., Kanhere, P., Ng, C.F. et al. (2015). Chem. Mater. 27: 4930–4933.
142. 142 Dong, F., Wang, Z., Sun, Y. et al. (2013). J. Colloid Interface Sci. 401: 70–79.
143. 143 Yuan, Y.P., Xu, W.T., Yin, L.S. et al. (2013). Int. J. Hydrogen Energy 38: 13159–13163.
144. 144 Zhang, J., Zhang, M., Zhang, G., and Wang, X. (2012). ACS Catal. 2: 940–948.
145. 145 Chen, X., Liu, L., Yu, P.Y., and Mao, S.S. (2011). Science 331: 746–750.
146. 146 Chen, X., Liu, L., and Huang, F. (2015). Chem. Soc. Rev. 44: 1861–1885.
147. 147 Zuo, F., Bozhilov, K., Dillon, R.J. et al. (2012). Angew. Chem. Int. Ed. 51: 6223–6226.
148. 148 Niu, P., Liu, G., and Cheng, H.M. (2012). J. Phys. Chem. C 116: 11013–11018.
149. 149 Li, S., Dong, G., Hailili, R. et al. (2016). Appl. Catal., B 190: 26–35.
150. 150 Liang, Q., Li, Z., Huang, Z.H. et al. (2015). Adv. Funct. Mater. 25: 6885–6892.
151. 151 Liu, J., Wang, H., Chen, Z.P. et al. (2015). Adv. Mater. 27: 712–718.
152. 152 Yang, Z., Zhang, Y., and Schnepp, Z. (2015). J. Mater. Chem. A 3: 14081–14092.
153. 153 Yan, H. (2012). Chem. Commun. 48: 3430–3432.
154. 154 Wang, Y., Wang, X., Antonietti, M., and Zhang, Y. (2010). ChemSusChem 3: 435–439.
155. 155 Savateev, A., Chen, Z.P., and Dontsova, D. (2016). RSC Adv. 6: 2910–2913.
156. 156 Kailasam, K., Epping, J.D., Thomas, A. et al. (2011). Energy Environ. Sci. 4: 4668–4674.
157. 157 Huang, J., Ho, W., and Wang, X. (2014). Chem. Commun. 50: 4338–4340.
158. 158 Li, X., Masters, A.F., and Maschmeyer, T. (2015). ChemCatChem 7: 121–126.
159. 159 Fukasawa, Y., Takanabe, K., Shimojima, A. et al. (2011). Chem. Asian J. 6: 103–109.
160. 160 Zhang, Y., Schnepp, Z., Cao, J. et al. (2013). Sci. Rep. 3: 2163.
161. 161 Xu, J., Wang, Y., and Zhu, Y. (2013). Langmuir 29: 10566–10572.
162. 162 Wang, Y., Zhang, J., Wang, X. et al. (2010). Angew. Chem. Int. Ed. 49: 3356–3359.
163. 163 Han, Q., Wang, B., Zhao, Y. et al. (2015). Angew. Chem. Int. Ed. 54: 11433–11437.
164. 164 Bai, X., Wang, L., Zong, R., and Zhu, Y. (2013). J. Phys. Chem. C 117: 9952–9961.
165. 165 Han, Q., Zhao, F., Hu, C. et al. (2015). Nano Res. 8: 1718–1728.
166. 166 Zhou, J., Yang, Y., and Zhang, C.‐Y. (2013). Chem. Commun. 49: 8605–8607.
167. 167 Gu, Q., Liao, Y., Yin, L. et al. (2015). Appl. Catal., B 165: 503–510.
168. 168 Wang, Y., Li, Y., Ju, W. et al. (2016). Carbon 102: 477–486.
169. 169 Chen, D., Wang, K., Xiang, D. et al. (2014). Appl. Catal., B 147: 554–561.
170. 170 Liu, G., Wang, T., Zhang, H. et al. (2015). Angew. Chem. Int. Ed. 54: 13561–13565.
171. 171 Niu, P., Zhang, L., Liu, G., and Cheng, H.‐M. (2012). Adv. Funct. Mater. 22: 4763–4770.
172. 172 Cheng, F., Wang, H., and Dong, X. (2015). Chem. Commun. 51: 7176–7179.
173. 173 Schwinghammer, K., Mesch, M.B., Duppel, V. et al. (2014). J. Am. Chem. Soc. 136: 1730–1733.
174. 174 Wang, W., Yu, J.C., Shen, Z. et al. (2014). Chem. Commun. 50: 10148–10150.
175. 175 Yang, S., Gong, Y., Zhang, J. et al. (2013). Adv. Mater. 25: 2452–2456.
176. 176 Ma, L., Fan, H., Li, M. et al. (2015). J. Mater. Chem. A 3: 22404–22412.
177. 177 Xu, H., Yan, J., She, X. et al. (2014). Nanoscale 6: 1406–1415.
178. 178 Han, Q., Wang, B., Gao, J. et al. (2016). ACS Nano 10: 2745–2751.
179. 179 Zhang, X., Meng, Z., Rao, D. et al. (2016). Energy Environ. Sci. 9: 841–849.
180. 180 Liu, J., Li, W., Duan, L. et al. (2015). Nano Lett. 15: 5137–5142.
181. 181 Guo, S., Zhu, Y., Yan, Y. et al. (2016). Appl. Catal., B 185: 315–321.
182. 182 Bu, Y. and Chen, Z. (2014). Electrochim. Acta 144: 42–49.
183. 183 Dong, G., Ai, Z., and Zhang, L. (2014). RSC Adv. 4: 5553–5560.
184. 184 Dong, G., Zhao, K., and Zhang, L. (2012). Chem. Commun. 48: 6178–6180.
185. 185 Zhang, P., Li, X., Shao, C., and Liu, Y. (2015). J. Mater. Chem. A 3: 3281–3284.
186. 186 Ma, T.Y., Ran, J., Dai, S. et al. (2015). Angew. Chem. Int. Ed. 54: 4646–4650.
187. 187 Zhang, L., Chen, X., Guan, J. et al. (2013). Mater. Res. Bull. 48: 3485–3491.
188. 188 Lan, D.H., Wang, H.T., Chen, L. et al. (2016). Carbon 100: 81–89.
189. 189 Xu, C., Han, Q., Zhao, Y. et al. (2015). J. Mater. Chem. A 3: 1841–1846.
190. 190 Ma, X., Lv, Y., Xu, J. et al. (2012). J. Phys. Chem. C 116: 23485–23493.
191. 191 Lu, C., Chen, R., Wu, X. et al. (2016). Appl. Surf. Sci. 360 (Part B): 1016–1022.
192. 192 Raziq, F., Qu, Y., Zhang, X. et al. (2016). J. Phys. Chem. C 120: 98–107.
193. 193 Pan, H., Zhang, H., Liu, H., and Chen, L. (2015). Solid State Commun. 203: 35–40.
194. 194 Zhang, G., Zhang, M., Ye, X. et al. (2014). Adv. Mater. 26: 805–809.
195. 195 Han, Q., Hu, C., Zhao, F. et al. (2015). J. Mater. Chem. A 3: 4612–4619.
196. 196 Wang, Y., Di, Y., Antonietti, M. et al. (2010). Chem. Mater. 22: 5119–5121.
197. 197 Lin, Z. and Wang, X. (2014). ChemSusChem 7: 1547–1550.
198. 198 Lu, Y.C., Chen, J., Wang, A.J. et al. (2015). J. Mater. Chem. C 3: 73–78.
199. 199 Ma, H., Li, Y., Li, S., and Liu, N. (2015). Appl. Surf. Sci. 357 (Part A): 131–138.
200. 200 Zhang, J., Zhang, M., Sun, R.‐Q., and Wang, X. (2012). Angew. Chem. Int. Ed. 51: 10145–10149.
201. 201 Zhang, J., Guo, F., and Wang, X. (2013). Adv. Funct. Mater. 23: 3008–3014.
202. 202 Chu, S., Wang, Y., Guo, Y. et al. (2013). ACS Catal. 3: 912–919.
203. 203 Zhang, G. and Wang, X. (2013). J. Catal. 307: 246–253.
204. 204 Fan, X., Zhang, L., Cheng, R. et al. (2015). ACS Catal. 5: 5008–5015.
205. 205 Ye, X., Cui, Y., and Wang, X. (2014). ChemSusChem 7: 738–742.
206. 206 Zhang, J., An, X., Lin, N. et al. (2016). Carbon 100: 450–455.
207. 207 Ho, W., Zhang, Z., Lin, W. et al. (2015). ACS Appl. Mater. Interfaces 7: 5497–5505.
208. 208 Chen, S., Wang, C., Bunes, B.R. et al. (2015). Appl. Catal., A 498: 63–68.
209. 209 Shiraishi, Y., Kanazawa, S., Kofuji, Y. et al. (2014). Angew. Chem. Int. Ed. 53: 13454–13459.
210. 210 Rahman, M.Z., Ran, R.J., Tang, Y. et al. (2016). J. Mater. Chem. A 4: 2445–2452.
211. 211 Zhang, J., Chen, X., Takanabe, K. et al. (2010). Angew. Chem. Int. Ed. 49: 441–444.
212. 212 Qin, J., Wang, S., Ren, H. et al. (2015). Appl. Catal., B 179: 1–8.
213. 213 Zhang, J., Zhang, G., Chen, X. et al. (2012). Angew. Chem. Int. Ed. 51: 3183–3187.
214. 214 Pan, Y.X., Zhou, T.H., Han, J.Y. et al. (2016). Catal. Sci. Technol. 6: 2206–2213.
215. 215 Zhang, J., Yu, J., Zhang, Y. et al. (2011). Nano Lett. 11: 4774–4779.
216. 216 He, F., Chen, G., Zhou, Y. et al. (2015). Chem. Commun. 51: 16244–16246.
217. 217 Suryawanshi, A., Dhanasekaran, P., Mhamane, D. et al. (2012). Int. J. Hydrogen Energy. 37: 9584–9589.
218. 218 He, F., Chen, G., Zhou, Y. et al. (2016). J. Mater. Chem. A. 4: 3822–3827.
219. 219 Ye, L., Wu, D., Chu, K.H. et al. (2016). Chem. Eng. J. 304: 376–383.
220. 220 Cai, W., Zhong, Q., Zhao, W., and Bu, Y. (2014). Appl. Catal., B 158: 258–268.
221. 221 Ai, Z., Ho, W., Lee, S., and Zhang, L. (2009). Environ. Sci. Technol. 43: 4143–4150.
222. 222 Ding, X., Song, X., Li, P. et al. (2011). J. Hazard. Mater. 190: 604–612.
223. 223 Wu, H., Chen, D., Li, N. et al. (2016). Nanoscale 8: 12066–12072.
224. 224 Zhang, W., Zhang, J., Dong, F., and Zhang, Y. (2016). RSC Adv. 6: 88085–88089.
225. 225 Wang, Z., Guan, W., Sun, Y. et al. (2015). Nanoscale 7: 2471–2479.
226. 226 Turchi, C. (1990). J. Catal. 122: 178–192.
227. 227 Li, C., Chen, G., Sun, J. et al. (2015). New J. Chem. 39: 4384–4390.
228. 228 Dong, F., Li, Y., Wang, Z., and Ho, W.K. (2015). Appl. Surf. Sci. 358: 393–403.
229. 229 Huang, Z.F., Song, J., Pan, L. et al. (2014). Chem. Commun. 50: 10959–10962.
230. 230 Dong, H., Chen, G., Sun, J. et al. (2014). Dalton Trans. 43: 7282–7289.
231. 231 Zhang, M., Xu, J., Zong, R., and Zhu, Y. (2014). Appl. Catal., B 147: 229–235.
232. 232 Tahir, M., Cao, C., Mahmood, N. et al. (2014). ACS Appl. Mater. Interfaces 6: 1258–1265.
233. 233 Ge, L., Han, C., and Liu, J. (2012). J. Mater. Chem. 22: 11843–11850.
234. 234 Corrigan, N., Shanmugam, S., Xu, J., and Boyer, C. (2016). Chem. Soc. Rev. 45: 6165–6212.
235. 235 Friedmann, D., Hakki, A., Kim, H. et al. (2016). Green Chem. 18: 5391–5411.
236. 236 Su, F., Mathew, S.C., Lipner, G. et al. (2010). J. Am. Chem. Soc. 132: 16299–16301.
237. 237 Li, X.H., Wang, X., and Antonietti, M. (2012). ACS Catal. 2: 2082–2086.
238. 238 Zhao, H., Yu, H., Quan, X. et al. (2014). Appl. Catal., B 152: 46–50.
239. 239 Du, A., Sanvito, S., Li, Z. et al. (2012). J. Am. Chem. Soc. 134: 4393–4397.
240. 240 Li, X., Dai, Y., Ma, Y. et al. (2014). Phys. Chem. Chem. Phys. 16: 4230–4235.
241. 241 Cui, J., Liang, S., Wang, X., and Zhang, J.‐M. (2015). Phys. Chem. Chem. Phys. 17: 23613–23618.
242. 242 Xiang, Q., Yu, J., and Jaroniec, M. (2011). J. Phys. Chem. C 115: 7355–7363.
243. 243 Liao, G., Chen, S., Quan, X. et al. (2012). J. Mater. Chem. 22: 2721–2726.
244. 244 Li, Y., Zhang, H., Liu, P. et al. (2013). Small 9: 3336–3344.
245. 245 Fu, Y., Zhu, J., Hu, C. et al. (2014). Nanoscale 6: 12555–12564.
246. 246 Zhang, Y., Mori, T., Niu, L., and Ye, J. (2011). Energy Environ. Sci. 4: 4517–4521.
247. 247 Oh, J., Lee, S., Zhang, K. et al. (2014). Carbon 66: 119–125.
248. 248 Xu, Y., Xu, H., Wang, L. et al. (2013). Dalton Trans. 42: 7604–7613.
249. 249 Sun, L., Qi, Y., Jia, C.‐J. et al. (2014). Nanoscale 6: 2649–2659.
250. 250 Ong, W.‐J., Tan, L.‐L., Chai, S.‐P. et al. (2015). Nano Energy 13: 757–770.
251. 251 Chen, J., Xu, X., Li, T. et al. (2016). Sci. Rep. 6: 1–13.
252. 252 Zhao, H., Chen, S., Quan, X. et al. (2016). Appl. Catal., B 194: 134–140.
253. 253 Wang, W., Yu, J.C., Xia, D. et al. (2013). Environ. Sci. Technol. 47: 8724–8732.
254. 254 Liu, J., Liu, Y., Liu, N. et al. (2015). Science 347: 970–974.
255. 255 Xia, X., Deng, N., Cui, G. et al. (2015). Chem. Commun. 51: 10899–10902.