Photocatalytic Self-Fenton System of g-C₃N₄-Based for Degradation of Emerging Contaminants: A Review of Advances and Prospects

Abstract

The discharge of emerging pollutants in the industrial process poses a severe threat to the ecological environment and human health. Photocatalytic self-Fenton technology combines the advantages of photocatalysis and Fenton oxidation technology through the in situ generation of hydrogen peroxide (H₂O₂) and interaction with iron (Fe) ions to generate a large number of strong reactive oxygen species (ROS) to effectively degrade pollutants in the environment. Graphite carbon nitride (g-C₃N₄) is considered as the most potential photocatalytic oxygen reduction reaction (ORR) photocatalyst for H₂O₂ production due to its excellent chemical/thermal stability, unique electronic structure, easy manufacturing, and moderate band gap (2.70 eV). Hence, in this review, we briefly introduce the advantages of the photocatalytic self-Fenton and its degradation mechanisms. In addition, the modification strategy of the g-C₃N₄-based photocatalytic self-Fenton system and related applications in environmental remediation are fully discussed and summarized in detail. Finally, the prospects and challenges of the g-C₃N₄-based photocatalytic self-Fenton system are discussed. We believe that this review can promote the construction of novel and efficient photocatalytic self-Fenton systems as well as further application in environmental remediation and other research fields.

1. Introduction

Since the middle of the 20th century, the rapid development of industrialization has appeared together with increasingly serious environmental problems, and the continuous release of toxic substances has led to the discovery of more and more emerging pollutants (e.g., antibiotics, dyes, etc.) in surface water, resulting in serious environmental pollution problems [1,2,3,4,5]. An increasing number of pollutants such as industrial dyes, antibiotics and agricultural chemicals exist widely in very low concentrations in the environment and thus are difficult to biodegrade, so the effective removal of these toxic species is related to human and biological health as well as sustainable reproduction [6,7,8].

In response to these problems, scientists have developed a series of water pollution treatment methods such as adsorption, flocculation, photodegradation, water splitting, and advanced oxidation processes (AOPs) [9,10,11,12,13,14,15]. Among them, AOPs are regarded as the most potential green treatment process to solve the problem of water pollution due to its strong oxidation capacity and without selectivity in the oxidation process, which has attracted global attention [16,17]. However, the traditional photocatalytic degradation technology relies on the active species produced by the semiconductor catalyst under light irradiation, which leads to a slow reaction rate, limited reaction applicability, and high light dependence, thus affecting its degradation performance. It is worth noting that photo-Fenton technology is considered to be one of the most cost-effective AOPs due to its nature of using sufficient ferrous ions (Fe²⁺) to activate hydrogen peroxide (H₂O₂) reactions to generate hydroxyl radicals (•OH) with strong oxidation capacity and sufficiently degrade organic pollutants to eventually mineralize into harmless small molecules (carbon dioxide, water, etc.), thus having important application potential in fields such as environmental remediation and waste treatment [18,19,20,21,22]. However, the single Fenton technology greatly limits its application in the field of degraded organic pollutants due to the following defects: (i) a large amount of H₂O₂ is added, resulting in a high processing cost [23,24]; (ii) the slower reduction kinetics of Fe³⁺/Fe²⁺ hinders the reaction efficiency and fails to adequately mineralize organic pollutants [25,26,27,28,29]. From the perspective of sustainable development, the photocatalytic self-Fenton system can organically combine photocatalytic with Fenton oxidation technology, which can not only solve the risk problem of adding extra H₂O₂ in the traditional Fenton technology, but can also accelerate the Fe²⁺/Fe³⁺ cycle, thus improving the removal efficiency of emerging pollutants [30]. Photocatalytic self-Fenton technology is mainly through the use of sunlight, which makes the photocatalyst, through photocatalytic oxygen reduction reaction (ORR) in situ, generate hydrogen peroxide (H₂O₂), and promotes the occurrence of heterogeneous self-Fenton reactions through the chain reaction with Fe²⁺ to generate strong oxidizing radical active species (e.g., •OH, ${\mathrm{O}}_2^{-}$, ¹O₂), thus degrading and mineralizing organic pollutants in the water environment to form environmentally friendly species [31,32,33].

In recent years, researchers have made unremitting efforts to develop a series of photocatalytic materials with efficient H₂O₂ production such as inorganic semiconductors like TiO₂ [34], CdS [35], graphite carbon nitride (g-C₃N₄) [36,37,38], and triazine covalent organic framework (CTF) [39]. Among them, g-C₃N₄, as a new organic semiconductor material, is a potential ORR photocatalyst due to its unique melamine (melam) structure, which is conducive to the reduction of O₂ molecules to H₂O₂ through the two-electron path [40,41,42,43,44]. In addition, g-C₃N₄ has the advantages of high stability, easy preparation, low cost, and non-toxicity, and has attracted widespread attention in the field of photocatalysis since it was first applied by Wang et al. [45]. For example, Ma et al. used doped phosphorus (P) atomic modified g-C₃N₄ to promote the separation of photon-generated carriers, which increased the yield of in situ generated H₂O₂, and added FeSO₄·7H₂O as the source of Fe²⁺ in the system to promote the degradation effect of photocatalytic-self-Fenton technology on organic pollutants [46].

However, the existing reviews have not provided a complete and detailed overview of g-C₃N₄-based photocatalytic self-Fenton systems. Therefore, this review summarizes the research progress of a g-C₃N₄-based photocatalysis self-Fenton system. In the following sections, we will start with the photocatalysis self-Fenton reaction mechanism and comprehensively summarize the modification methods of g-C₃N₄-based photocatalysis self-Fenton technology for the efficient degradation of emerging pollutants (Scheme 1) including heterojunction, element doping, morphology regulation, defect engineering, and other modification methods. At length, the prospects and challenges of the photocatalytic self-Fenton system are discussed from the perspective of high efficiency, green, and safety.

2. Fundamental Mechanisms of Degradation of Pollutants by Photocatalysis Self-Fenton Technology

The process of the degradation of emerging pollutants by the photocatalytic-self-Fenton system (Scheme 2) is mainly divided into two steps: (i) the photocatalyst can selectively and efficiently synthesize in situ H₂O₂ by solar-driven (i.e., photocatalytic process); (ii) the chain reaction between H₂O₂ and Fe²⁺ generates highly oxidized •OH, and then realizes efficient, green, and mineralization of various emerging pollutants (i.e., photo-Fenton process) [47].

2.1. Mechanisms of the H₂O₂ Production by Photocatalyst

A prerequisite for the efficient photocatalytic-self-Fenton degradation of emerging pollutants is the ability of the photocatalyst to produce a sufficient amount of H₂O₂. Photo-driven H₂O₂ production based on photocatalysts involves a series of complex and continuous photocatalytic reactions, which mainly consist of three consecutive reaction steps: (i) after the photocatalyst absorbs light energy greater than its band gap, the electrons (e⁻) are excited to migrate from the valence band (VB) to the conduction band (CB), leaving holes (h⁺) in the VB (Equation (1)); (ii) photo-excited electrons and holes migrate to the surface of the photocatalyst and participate in the surface redox reactions to produce reactive oxygen species (ROS); (iii) during this photocatalytic reaction, the carriers inevitably tend to recombine to form electron–hole pairs, while the surviving carriers migrate to the catalyst surface to participate in H₂O oxidation and O₂ reduction reactions (Equations (2) and (3)) and generate H₂O₂.

$$\mathrm{photocatalyst}+\mathrm{hv}\to \mathrm{photocatalyst} \left({ \mathrm{h}}^{+}+{\mathrm{e}}^{-}\right)$$

(1)

$${\mathrm{e}}^{-}+{\mathrm{O}}_2\to {\mathrm{O}}_2^{-}$$

(2)

$${\mathrm{O}}_2+2{\mathrm{e}}^{-}+2{\mathrm{H}}^{+}\to {\mathrm{H}}_2{\mathrm{O}}_2$$

(3)

In addition, since the research process of H₂O₂ production mainly focuses on the reduction of O₂, it can also be achieved through the two-step single-e⁻ oxygen reduction reaction (Equations (4)–(6)) [48,49,50]. Specifically, during the ORR, the photo-induced electrons of the photocatalyst conduction band can react with O₂ adsorbed on the surface to produce ${\mathrm{O}}_2^{.-}$ (Equation (2)), while O₂ can also react with H⁺ to produce H${\mathrm{O}}_2^{.-}$ (Equation (4)) as well as through the single-electron reduction process to produce HO₂⁻ (Equation (5)), and finally, the reaction with H⁺ to produce H₂O₂ (Equation (6)). It is worth noting that although four-electron ORR is beneficial to O₂ generation (Equation (7)), the selectivity for the formation of H₂O₂ from oxygen is diminished [51,52,53]. In addition, as the generation of ${\mathrm{O}}_2^{-}$ needs more negative potential (−0.33 V) than the one-step double electron reaction (+0.68 V) and involves numerous unpredictable reactions, the two-step one-electron ORR reduces the H₂O₂ yield and selectivity [41,54]. In summary, the one-step double electron⁻ oxygen reduction pathway is more advantageous for the production of light-induced H₂O₂.

$${\mathrm{O}}_2+{\mathrm{H}}^{+}\to {\mathrm{HO}}_2^{-}$$

(4)

$${\mathrm{HO}}_2^{.-}+{\mathrm{e}}^{-}\to {\mathrm{HO}}_2^{-}$$

(5)

$${\mathrm{HO}}_2^{-}+{\mathrm{H}}^{+}\to {\mathrm{H}}_2{\mathrm{O}}_2$$

(6)

$${\mathrm{O}}_2+4{\mathrm{e}}^{-}+4{\mathrm{H}}^{+}\to 2{\mathrm{H}}_2\mathrm{O}$$

(7)

2.2. Mechanisms of the Fe²⁺/H₂O₂-Mediated Fenton Reaction to Remove Emerging Pollutants

In 1894, Fenton technology was first proposed by the famous scientist Fenton for water treatment, mainly because ferrous ion (Fe²⁺) reacts with H₂O₂ to produce •OH with high oxidation capacity, which oxidizes and decomposes many refractory pollutants [55]. In addition, Haber et al. first studied the traditional homogeneous Fenton reaction between Fe²⁺ and H₂O₂ and formed an extraordinary complicated chain Fenton reaction, which is mainly divided into three chain steps: the initiation, propagation, and termination reaction [56]. In detail, these steps include (i) the production of •OH through oxidation reactions, at the same time, Fe²⁺ is oxidized to Fe³⁺ (Equation (8)); (ii) •OH with strong oxidation capacity and high electronegativity can react with organic compounds to form organic radical intermediates and hydroxylation products through the redox reaction and dehydrogenation reaction (Equation (9)) [47,57,58]; (iii) emerging pollutants react with •OH to form H₂O andCO₂ (Equation (10)).

$${\mathrm{Fe}}^{3+}+{\mathrm{e}}^{-}\to {\mathrm{Fe}}^{2+}$$

(8)

$${\mathrm{Fe}}^{2+}+{\mathrm{H}}_2{\mathrm{O}}_2\to {\mathrm{Fe}}^{3+}+{\mathrm{OH}}^{-}+•\mathrm{OH}$$

(9)

$$•\mathrm{OH}+\mathrm{emerging} \mathrm{pollutants}\to {\mathrm{CO}}_2+{\mathrm{H}}_2\mathrm{O}$$

(10)

However, it is inevitable that excess H₂O₂ may further react with the generated •OH to produce ${\mathrm{HO}}_2^{-}$, while Fe³⁺ can also react with H₂O₂ and then be reduced to Fe²⁺; H₂O₂ is oxidized to ${\mathrm{HO}}_2^{-}$. Therefore, the use of the initiation and propagation responses in the Fenton reaction system to avoid these unwanted termination reactions remains a considerable challenge.

2.3. The Influence Factors of H₂O₂ Generation

Efficient photocatalytic H₂O₂ production mainly depends on the suitable reaction conditions. First of all, suitable solvents (ethanol and isopropanol) provide a substantial role for photocatalytic H₂O₂ production as they can not only serve as electron donors to consume holes, but also promote the separation and migration of photo-generated carriers, so that more electrons can be used to reduce O₂ [59,60]. In the next place, since the content of protons is closely related to the production of H₂O₂ based on the ORR pathway, in general, acidic conditions (pH < 7) are conducive to proton coupled electron transfer and promote H₂O₂ production, whereas alkaline conditions (pH > 7) may produce excess protons, leading to the further reaction of H₂O₂ with H₂O [61,62,63]. At length, sufficient O₂ is the key to ensure a high H₂O₂ yield, so the O₂ consumed can be replenished continuously through the continuous supply of oxygen and four-electron water oxidation reactions during the reaction.

2.4. The Influence Factors of Fenton Reaction

The activity of the Fenton reaction greatly determines the mineralization efficiency of the pollutants, and the conventional homogeneous Fenton reaction is mainly affected by the following aspects: (i) pH value, as the activity of the Fenton reaction is limited by the pH value of the reaction solution because the Fenton reaction is conducive to the reaction of ferrous ions with H₂O₂ to form ·OH under acidic conditions (pH = 3–4) [64]; (ii) H₂O₂ concentration, as the concentration of H₂O₂ is critical to the activity of Fenton reaction, so a low concentration will inhibit the reaction rate, while a high concentration will not only lead to low utilization and precipitation, but Fe²⁺ will also be oxidized to Fe³⁺ at the beginning of the reaction, thus inhibiting the production of •OH [65,66]; (iii) Fe²⁺ concentration, as the reaction rate is slow when the concentration is too low, at high concentration, H₂O₂ will be reduced, and Fe²⁺ will be oxidized to Fe³⁺ [67,68].

3. Graphite Carbon Nitride (g-C₃N₄) Material

3.1. Synthetic Methods and Morphology of g-C₃N₄

Due to the small specific surface area of g-C₃N₄, the low utilization efficiency of visible light and the easy recombination of the photogenerated electrons and holes, researchers have improved the photocatalytic activity of g-C₃N₄ through different synthesis methods and morphological regulation [69,70,71].

As one of the most commonly used synthesis methods of g-C₃N₄, thermal polymerization, which has the advantages of simple operation, safety, and easy control, produces the triazine ring structure g-C₃N₄ by the pyrolysis polymerization of nitrogen-containing organic compounds such as melamine, urea, and thiourea at high temperature [72,73,74,75]. For example, Li et al. used melamine as a precursor to obtain g-C₃N₄ with curled edges through two-step thermal polymerization [76]. However, it is worth noting that compared with traditional low-dimensional materials, g-C₃N₄ obtained by the thermal polymerization of precursors is mostly of a block or lamellar structure, which is unfavorable to the photogenerated carrier separation and transport, thus limiting its photocatalytic activity. However, the solvothermal method can, by the thermal polymerization of organic precursors, form g-C₃N₄ with a nanoparticle or nanosheet structure under high temperature solvent conditions to solve the limitations of thermal polymerization method [77,78]. For example, Zhou et al. successfully prepared g-C₃N₄ metal-free composite (CNQDs/CN) modified with g-C₃N₄ quantum dots (CNQDs) by a simple solvothermal method, which significantly improved the photogenerated charge separation and migration rate of g-C₃N₄, thus showing excellent photocatalytic degradation activity [79].

Furthermore, the porous g-C₃N₄ with different nanostructures and morphologies can be obtained by depositing organic precursors in the template (silica gel, alumina) channels and conducting pyrolysis or chemical transformation [80,81]. Goettmann et al. used cyanamide as the precursor and SiO₂ nanoparticles as the template; after heat treatment and condensation, g-C₃N₄ with high specific surface area and pore structure was obtained, which significantly increased its photocatalytic activity and light absorption capacity [82]. In addition to the above commonly used synthesis methods, there are some other methods such as microwave assisted synthesis, the sol–gel method, electrochemical synthesis, etc. that can achieve accurate control of the morphology and structure of g-C₃N₄ by regulating the reaction conditions, etc., thus providing diversified options for its application in the field of photocatalysis [83,84].

3.2. Band Structure of g-C₃N₄

In addition, graphite phase carbon nitride (g-C₃N₄), as a kind of carbide with a unique 2D layered structure, with easy to obtain raw materials, a medium band gap (2.70 eV), excellent chemical/thermal stability, and can absorb visible light at about 450–460 nm, is considered to be the most promising non-metallic visible light catalyst after TiO₂ [85,86,87,88]. In particular, CN has the highest nitrogen content among known 2D carbon materials and the unique sp²-hybridization of C and N has been used to construct π-conjugated structures [89,90,91,92]. In addition, from the perspective of band position, the CB position of g-C₃N₄ (−1.27 V vs. NHE) is more negative than the standard potential (${\mathrm{O}}_2^{-}$: −0.33eV vs, NHE) of O₂/${\mathrm{O}}_2^{-}$, so the electrons on the conduction band (CB) of g-C₃N₄ can effectively reduce O₂ to ${\mathrm{O}}_2^{-}$, which is more conducive to the generation of H₂O₂ in the photocatalytic self-Fenton reaction process through the double-electron oxygen reduction path. More importantly, the low valence band (VB) (1.4 eV vs. NHE) of g-C₃N₄ effectively prevents the oxidation of H₂O₂ during light irradiation [46,93,94,95]. Therefore, g-C₃N₄ is considered as a promising candidate for the in situ generation of H₂O₂. Unfortunately, some inherent scientific problems including the easy recombination of electron–hole pairs, small specific surface area, and low utilization of visible light, limit the practical application of original g-C₃N₄ in the actual photocatalytic self-Fenton system [96,97,98,99]. In order to solve the above limitations, as shown in

Table 1. Summary of the recent classification of pollutants degraded by the photocatalytic self-Fenton system based on g-C₃N₄. (N/A is not available).

Year	Photocatalysts	Light Source	Iron Ion Source	Sacrificial Agent	Pollutants	H2O2 Yield	Pollutant Removal (%)	TOC Removal (%)	Ref.
2020	P-g-C3N4	Vis	FeSO4·7H2O	-	2,4-DCP	23.5 μM h−1	91	42.7	[46]
	Ti3+-TiO2-C3N4por	Vis	-	-	MO	6.32 mM h−1	97.9	89	[105]
	M101-Ux	Vis	MIL101	-	2,6-DCP/2,4,5-TCP	23 μM h−1	98.7/97.3	16/14	[106]
	TE-g-C3N4	Vis	FeCl3	-	RhB	N/A	~100	-	[107]
	α-Fe2O3/g-C3N4	Vis	α-Fe2O3	IPA	RhB/TC	7.5 μM h−1	96/95	92/86	[93]
	Al0-CNTs/CNTs-Fe-Cu	Vis	FeSO4 7H2O	-	SMR	N/A	85	60	[108]
	Metal-free GCN	Vis	Fe(II)	Isopropyl alcohol	GA	N/A	~93	88.8	[109]
2021	ZnO/g-C3N4	Vis	FeSO4	IPA	Bactericidal	5312.45 μM L−1	97.4	-	[110]
	SCNO@CdS	Vis	FeCl2·4H2O	IPA	RhB	79,971.0 μM g−1 h−1	~83	-	[111]
	P-C3N4/O-C3N4	Vis	FeSO4 7H2O	IPA	MTZ	179 μM h−1	91.6	-	[112]
	g-C3N4/PDI	Vis	FeSO4 7H2O	-	Bactericidal	112.95 μM h−1	~100	-	[113]
2022	GCN	Vis	FeSO4 7H2O	IPA	2,4-DCP	21.59 mM h−1 g−1	88.8	-	[114]
	UPCN	Vis	FeCl3·6H2O	IPA	OTC	23.91 μM L−1	86.23	48..6	[115]
	Cg-C3N4	Vis	FeSO4 7H2O	IPA	4-CP	217.26 μM g−1 h−1	~98	59.64	[116]
	KBCN	Vis	-	-	RhB/CR	57.7 μM h−1	93.3/96.6	-	[117]
	OPCN	Vis	FeCl3·6H2O	IPA	2,4-DCP	25.06 μM h−1	93	42.22	[24]
	CUCN	Vis	FeSO4 7H2O	-	RhB	14.81 μM h−1	up to 100	63.77	[118]
	TFMS-CP	Vis	FeCl3	IPA	PH/HQ/MO	1.02 μM h−1	92.6/84.6/88.2	-	[119]
2023	MoS2-v/TCN	Vis	-	IPA	RhB	1879 μM g−1 h−1	98.5	-	[120]
	PC@PL	Vis	LaFeO3	-	MB	7.86 μM L−1	97	-	[121]
	Coral-B-CN	Vis	FeSO4 7H2O	-	4-CP	314.55 μM g−1 h−1	99.6	70.3	[122]
	Cv-PCNNS	Vis	FeSO4 7H2O	IPA	MTZ	984.8 μM L−1 h−1	90.7	62	[123]
	CPBA-CN	Vis	FeCl2·4H2O	-	4-CP	156.4 μM h−1	~97	74.6	[124]
	KCN	Vis	-	IPA	BPA	1.76 mM h−1	95.6	63.8	[125]

, scientists have carried out various physical and chemical modifications of g-C₃N₄ (Scheme 3) from the aspects of morphological control, defect engineering, element doping, heterojunction construction, etc., to enhance the visible light capture, carrier migration, and separation rate to further improve the photocatalytic self-Fenton degradation activity of g-C₃N₄ [100,101,102,103,104]. Furthermore, we integrated the full text abbreviations into

Table 2. Summary of the full text abbreviations.

Abbreviated Pre-Name	Abbreviated Name
Hydrogen peroxide	H2O2
Reactive oxygen species	ROS
Superoxide anion radical	${\mathrm{O}}_2^{-}$
Singlet oxygen	1O2
Graphite carbon nitride	g-C3N4
Oxygen reduction reaction	ORR
Advanced oxidation processes	AOPs
Hydroxyl radicals	∙OH
Valence band	VB
Conduction band	CB
2,4-dichlorophenol	2,4-DCP
Methyl Orange	MO
2,4,5-Trichlorophenol	2,4,5-TCP
Rhodamine B	RhB
Tetracycline	TC
2,6-dichlorophen	2,6-DCP
Sulfamerazine	SMR
Gallic acid	GA
Metronidazole	MTZ
4-Chlorophenol	4-CP
Phenol	PH
Hydroquinone	HQ
Oxytetracycline	OTC
Methylene Blue	MB
Metronidazole	MTZ

.

4. Various Modification Strategies for Photocatalytic Self-Fenton Based on g-C₃N₄

4.1. Morphology Control

The regulation of the morphology of the photocatalyst can improve the specific surface area and expose more reactive sites, thus affecting the photocatalytic activity to a certain extent [126,127,128]. Unfortunately, the practical application of the original g-C₃N₄ is limited due to the small specific surface area, the high photo-generated carrier recombination rate, and weak van der Waals force interaction between layers [129,130,131]. Currently, the morphological adjustment work of g-C₃N₄ is mainly focused on high-dimensional porous engineering and low-dimensional nanostructure control including templating, supramolecular preorganization, and the exfoliation method (e.g., liquid phase ultrasound [132], chemical oxidation [133], and gas template stripping [134], etc.), successfully realizing the transformation from the bulk phase structure to nanosheets. For instance, Cui’s group prepared mesoporous CN (mpg-CN) nanosheets with a specific surface area of 180 m²g⁻¹ using 12 nm silica particles as a template through the silicon hard template method [135]. Shortly after, Xu et al. used the thermal oxidation etch stripping approach to fabricate a two-dimensional (2D) graphene-like carbon nitride-like structure (GA-C₃N₄) with a specific surface area of 30.1 m²g⁻¹ [136]. It is worth noting that the construction of porous and hierarchical 2D g-C₃N₄ can not only enhance the light absorption, but also promote photo-induced charge separation and migration by shortening the transmission distance, which is conducive to the photocatalytic in situ production of H₂O₂ as an intermediate to enable the production of •OH for the degradation of visible light-driven pollutants [129].

The specific surface area (Figure 1a) and pore volume (Figure 1b) of TE-g-C₃N₄ nanosheets prepared by Li et al. by the thermal exfoliation method have been greatly improved compared with the bulk g-C₃N₄ (B-g-C₃N₄), providing more abundant reaction sites and light capturing ability [107]. In addition, due to the improvement in the TE-g-C₃N₄ photo-generated charge separation efficiency (Figure 1c), the in situ generation of H₂O₂ was promoted, and the efficient degradation of the organic pollutant rhodamine B (RhB) was quickly completed (Figure 1d) in the presence of Fe³⁺ (supplied by FeCl₃) as the catalyst. To our knowledge, this work is the first to incorporate a practical application of photo-Fenton degradation pollutants without the external addition of H₂O₂.

In addition, the construction of three-dimensional (3D) structures also has a significant impact on the photocatalytic activity. Compared with the conventional 2D structure, the 3D structure has the advantages of larger specific surface area, a large number of active sites, and better light refraction, promoting photo-generated carrier separation and inhibiting carrier recombination [137]. Chen et al. used a novel catalyst (CNT-Fe-Cu) combined with Al⁰-carbon nanotubes (CNTs) to construct a photocatalytic self-Fenton system under visible light irradiation (Figure 2a) [108]. Among them, the Al⁰-CNTs composite, due to its 3D tubular structure, has a large specific surface area (Figure 2b), which promotes photo-generated electron separation and migration; H₂O₂ can be generated in situ under oxygen aeration, while the CNT-Fe-Cu composite can convert H₂O₂ into •OH and O₂⁻ radicals. The deposition of Fe and Cu metals on the surface of CNTs improved the catalytic activity (Figure 2c), resulting in the efficient degradation of sulfamerazine (SMR) at neutral pH. Although morphology control is a very important means by which to improve the photocatalytic performance of materials, the degree of modification is still low, so other modification strategies are often combined to improve the photocatalytic reactivity. Dang et al. modified CN by combining morphology regulation and P doping using isopropyl alcohol as the hole scavenger, where the H₂O₂ generation rate under visible light was up to 1684 μmol h⁻¹ g⁻¹, which was 11.2 times the volume of pristine CN [138]. Due to the synergistic effect of the two modification methods, the change in morphology significantly enhances the specific surface area and light response, and the P element doping effectively regulates the band structure and promotes the separation of photoinduced carriers, thus enhancing the photocatalytic activity.

4.2. Defect Engineering

In recent years, introducing defects (such as oxygen vacancy [139], sulfur vacancy [140], carbon vacancy [141], and nitrogen vacancy [142]) with appropriate concentrations in photocatalysts has been proven to be an effective way to improve photocatalytic activity. On the one hand, the introduction of defects in g-C₃N₄ can adjust its band structure, activate oxygen molecule, narrow the band gap, and effectively broaden the visible light absorption range. On the other hand, the defect can act as the capture center of the photo-generated carrier, which can effectively reduce the recombination rate of photo-induced electron–hole pairs and improve the photocatalytic H₂O₂ generation [143]. In a typical case, Zhang et al. introduced double defect sites (-C≡N groups and N-vacancy) sequentially into g-C₃N₄ (Nv-C≡N-CN), forming an electron-rich structure that resulted in a more local charge density distribution to improve the visible light absorption and photo-generated carrier separation and migration [144]. In addition, the N-vacancy and -C≡N groups can activate O₂ and promote the adsorption of H⁺, respectively, thus significantly improving the H₂O₂ generation. As a result, Nv-C≡N-CN reached the H₂O₂ generation rate of 3093 mmol g⁻¹ h⁻¹, which was 8.7 times stronger than the original CN (PCN) under visible light. Notably, the introduction of N defects can narrow the band gap of the photocatalyst and form a defect state, thereby enhancing light capture and accelerating the photo-generated carrier separation and migration [145].

Wu et al. constructed a photocatalytic self-Fenton system (GCN-PSFs) for the degradation of 2,4-dichlorophenol (2,4-DCP) by introducing carbon defects into garland g-C₃N₄ (Figure 3a) [114]. Notably, in the absence of a sacrificial agent, the introduction of carbon defects not only accelerated charge separation, but also improved the double-electron ORR, resulting in a H₂O₂ yield of 507.82 μM g⁻¹ h⁻¹ for garland g-C₃N₄ (GCN) (Figure 3b). Subsequently, compared with pure BCN and GCN, the photocatalytic degradation 2,4-DCP activities of BCN-PSFs and GCN-PSFs formed by adding Fe²⁺ to the photocatalytic degradation system could reach 45.8 and 88.8%, respectively (Figure 3c). The improvement in the photocatalytic degradation activity is mainly attributed to the following reasons: (i) the introduction of carbon defects accelerates the separation of photo-generated electron and holes, improving the adsorption capacity of O₂ and the production of H₂O₂ by double-electron ORR; (ii) the stable production of H₂O₂ can effectively transform into •OH to degrade organic pollutants.

In addition, under suitable basicity, Yue et al. used the one-step thermal polymerization method of urea in KOH-KCl to construct a novel carbon nitride (CN) with abundant surface N defects and compressed π–π layer stacking for the photocatalytic self-Fenton removal of organic pollutants in water (Figure 4a) [125]. The high-resolution transmission electron microscopy (HR-TEM) images in Figure 4b display two different spacing lattice fringes, which are due to the stacking distances of different π–π layers at the surface and bulk phases, forming a double structure [146]. Due to the unique surface N defect that transfers photo-generated electrons to O₂, KCN-0.5 exhibited an efficient in situ H₂O₂ production of 3.69 mM h⁻¹ g⁻¹ under λ = 420 nm and an aqueous solution containing IPA. As expected, KCN-0.5 exhibited excellent photocatalytic self-Fenton degradation and mineralized organic pollutants (tetracycline hydrochloride, bisphenol A, ciprofloxacin, rhodamine B) efficiency (Figure 4c,d). Therefore, the regulation of defects can promote the generation of H₂O₂ and further activate the H₂O₂ into ·OH (H₂O₂ + e⁻ → ·OH + OH⁻), so as to improve the photocatalytic self-Fenton degradation activity of emerging pollutants.

4.3. Constructed Heterojunctions

Coupling g-C₃N₄ to other photocatalysts with a staggered band structure (including metals, metal oxides, inorganic semiconductors, conductive polymers, etc.) by synthetic means to form heterostructures is also an effective way to improve the separation and migration of photogenerated carriers and improve their photocatalytic activity [147,148,149]. Moreover, compared with other commonly used g-C₃N₄ structural modification methods, the establishment of heterogeneous structures has the advantage of being more controllable and facile [150]. Generally speaking, heterojunctions are divided into traditional heterojunctions (Type-I, II), Z-scheme heterojunctions, and emerging S-scheme heterojunctions (Scheme 4).

Due to the difference in band structure between the phosphorus-doped C₃N₄ (P-C₃N₄) and oxygen-doped C₃N₄ (O-C₃N₄), Li et al. prepared a P-C₃N₄/O-C₃N₄ Type-I heterojunction as an efficient photocatalytic self-Fenton reaction catalyst (Figure 5a) [112]. From the transmission electron microscopy (TEM), it is evident that P-C₃N₄/O-C₃N₄ had been successfully synthesized. Under visible light irradiation, the formation of a Type-I heterojunction effectively promoted the photo-induced carrier separation with the help of an internal-electric-field, and the in situ generation rate of H₂O₂ was as high as 179 μM g⁻¹ h⁻¹ (Figure 5b). In the presence of Fe²⁺, the photocatalytic self-Fenton degradation of metronidazole (MTZ) by the P-C₃N₄/O-C₃N₄ Type-I heterojunction in Fenton within 60 min was 91.6%, which is 10.3 times that of pure C₃N₄ (Figure 5c,d). However, in the conventional Type-I heterojunction, since the semiconductor band structure is easily included, electrons and holes tend to migrate to the other semiconductor for recombination, which is not conducive to carrier separation.

Fortunately, for the Type-II heterojunction, due to the interleaving semiconductor band structure, electrons and holes are easily separated in space, thus reducing carrier recombination to further promote photocatalytic self-Fenton activity. Geng et al. prepared a novel ZnO/g-C₃N₄ Type-II heterojunction photocatalytic self-Fenton system by the thermal polycondensation process (Figure 6a) for the in situ sterilization in natural water under simulated sunlight [110]. By constructing a Type-II heterojunction, the response ability of the ZnO/g-C₃N₄ catalyst to sunlight and photoinduced charge separation was improved (Figure 6b), and the in situ H₂O₂ production and sterilization performance of the 10 wt% ZnO/g-C₃N₄ heterojunction catalyst within 60 min were 5312.45 μmol⋅L⁻¹ and 97.4%, respectively (Figure 6c,d). It is worth mentioning that after the end of the disinfection process, the number of colonies and the concentration of hydrogen peroxide continued to decrease (Figure 6e), which may be due to the fact that there was still a certain concentration of hydrogen peroxide that could kill bacteria in the solution after the end of the photocatalysis. Unfortunately, due to the existence of electrostatic repulsion, Type-II heterojunctions still face the problem of poor redox capacity [151].

Based on the Z-scheme heterojunction simulated by natural plant photosynthesis, it could not only improve the photo-generated carrier separation efficiency, but also retained strong redox potential, thus overcoming the defects of Type-II heterojunction. Recently, Wang et al. prepared an efficient Z-scheme α-Fe₂O₃/g-C₃N₄ photocatalytic self-Fenton system (Figure 7a) by using α-Fe₂O₃ as the heterogeneous Fenton catalyst coupled with a g-C₃N₄ narrow band gap semiconductor [93]. The formation of a Z-scheme heterojunction not only promotes the separation of photogenerated electron–hole pairs, but also maintains the strong redox ability of electron–holes pairs in the g-C₃N₄ CB and α-Fe₂O₃ VB, which is conducive to the in situ formation of H₂O₂ (Figure 7b) and the efficient degradation of rhodamine B (RhB) and tetracycline hydrochloride (TC-H) (Figure 7c). Furthermore, Jiang et al. constructed the photocatalytic self-Fenton reaction system based on molybdenum disulfide with S-vacancy/carbon nitride nanotube (MoS_2-v/TCN) composites (Figure 7f) for in situ H₂O₂ generation and the decomposition of micropollutants/bacteria over a wide pH range [120]. The unique tubular hollow structure of TCN provides more active sites, reduces the migration distance of photogenerated carriers, and promotes the close and tight contact between TCN and MoS_2-v at the interface (Figure 7d). Combined with density functional theory (DFT) calculation, the construction of Z-scheme heterojunction promotes the rapid separation of photo-induced e⁻ and h⁺, greatly reduces the energy barrier production of •OOH and •HO₂⁻, and effectively improves the photocatalytic production of H₂O₂ (Figure 7e) and the degradation of organic pollutants by the MoS_2-v/TCN heterojunction.

In summary, the construction of a Z-scheme heterojunction is considered to be an effective modification method, which can improve the photocatalytic activity. Nevertheless, the Z-scheme heterojunction also has some drawbacks including unpredictable side reactions, sensitivity to pH, and controversial charge transfer routes [152]. In recent years, S-scheme heterojunctions composed of oxidized and reduced photocatalysts with staggered band structures have shown broad prospects in the field of photocatalysis, which promote the separation of photo-generated electrons and holes due to their internal electric field, band bending, and electrostatic interactions [153]. For instance, Pham et al. constructed S-scheme heterojunction α-Fe₂O₃/g-C₃N₄ nanocomposites that could efficiently degrade cephalexin and amoxicillin under visible light [154]. Meanwhile, Wei et al. reported for the first time an S-scheme heterojunction Co₃O₄/Bi₂MoO₆@g-C₃N₄ hollow microsphere that effectively promoted the separation of photogenerated electrons and holes due to the formation of internal electric field, and had a degradation rate of levofloxacin (LVFX) up to 95.21% under visible light irradiation [155]. In addition, it is worth noting that the construction of an S-scheme heterojunction photocatalyst based on g-C3N4 is also conducive to the production of H₂O₂. Xia et al. synthesized an S- scheme Zn-TCPP/CN composite by grafting g-C₃N₄ with supramolecular zinc porphyrin (Zn-TCPP) using ethanol as a sacrificial agent, and showed a H₂O₂ production rate with 532.7 μmol/L within 90 min, which is 3.1 times and 9.0 times of pure CN and Zn-TCPP, respectively [156]. Therefore, the development of new S-scheme photocatalysts based on g-C₃N₄ may be a future direction to improve the activity of photocatalytic self-Fenton degradation.

4.4. Element Doping

Element doping refers to the method of introducing metal/non-metal elements into the semiconductor in a certain way. In general, the non-metallic elements doped into g-C₃N₄ will replace the C and N elements of g-C₃N₄ itself, and the metal elements will be embedded in the internal structure of g-C₃N₄, and the energy band structure of g-C₃N₄ will change significantly after the element doping, improving the ability to capture sunlight and conductivity [157,158,159,160]. Furthermore, element doping is also an effective strategy to enhance photocatalytic activity by regulating the optical electronic properties of H₂O₂ generation [161]. Hu et al. synthesized hollow copper (Cu) doped graphitic carbon nitride (g-C₃N₄) microspheres, where Cu⁺ is inserted into the interstitial site in the form of a coordination bond (Cu(I)-N) and acts as a chemisorption site to activate the O₂ molecule, thus exhibiting excellent H₂O₂ production performance (4.8 mmol L⁻¹) [162].

Jing and co-workers fabricated a novel photocatalytic self-Fenton system based on a coral-like B-doped g-C₃N₄ (coral-B-CN) photocatalyst [122]. The transmission electron microscope (TEM) image in Figure 8a shows that coral-B-CN has a lamellar structure and an obvious pore distribution. Under simulated sunlight irradiation, as B heteroatom doping in coral-B-CN enhanced the molecular dipole, accelerated the charge separation rate, and enhanced the O₂ adsorption, the H₂O₂ production activity of coral-B-CN (Figure 8b) reached 104.9 μM. As expected, the efficient carrier separation and migration rate of coral-B-CN accelerated the Fe²⁺/Fe³⁺ cycle, and the resulting H₂O₂ reacted with Fe²⁺ to form the ·OH radical, resulting in the efficient degradation and mineralization of 4-CP in the photocatalytic self-Fenton system (Figure 8c). Significantly, the co-doping of multiple elements can often achieve the effect of “1 + 1 > 2” by utilizing the synergy of each component [163]. For example, in the photocatalytic self-Fenton system, Chen et al. constructed benzene and K⁺ co-doped g-C₃N₄ (KBCN) by the thermal polymerization approach to degrade rhodamine B (RhB) and Congo red (CR) via in situ generated H₂O₂ under visible light irradiation (Figure 8d) [117]. Since benzene and K⁺ co-doping increased the photo-generated charge separation rate and promoted the in situ formation of H₂O₂ in KBCN, the H₂O₂ yield of KBCN was 57.5 mM in pure water after 60 min of visible irradiation (Figure 8e). As shown in Figure 8f,g, KBCN for the degradation of RhB and CR rate was 93.3% and 96.6%, respectively, after 60 min with H₂O₂-derived •OH radicals under the condition of natural pH and no Fe²⁺ ions added.

It must also be mentioned that a single modification strategy may have some limitations in optimizing the photocatalytic activity, and it is an effective method to improve the self-Fenton performance of photocatalysis by integrating different modification strategies to maximize the synergistic effect. Theoretically, the combination of morphology regulation and element doping is a feasible method because the morphology regulation can solve the shortcomings of low specific surface area and slow carrier propagation efficiency, while element doping has great benefits in solving the light absorption capacity. For instance, Xu’s research group constructed a photocatalytic self-Fenton system based on oxygen-doped porous g-C₃N₄ nanosheets (OPCN) [24]. The well-developed porosity and high specific surface area of OPCN effectively promote the reduction of double electrons to O₂, while the doping of oxygen containing groups enhances the water oxidation capacity, leading to the efficient generation of H₂O₂ (Figure 9a). Due to the efficient in situ formation of H₂O₂ by OPCN and the introduction of the FeCl₃.6H₂O reaction reagent containing Fe³⁺, the degradation rate of 2,4-DCP (Figure 9b) is 11.5 and 9.9 times higher than that of the g-C₃N₄ photocatalytic and Fenton system, while the mineralization rate (Figure 9c) is 11.4 and 4.2 times higher than that of g-C₃N₄ photocatalytic system, respectively. To sum up, the morphology of the porous nanosheets and the introduction of oxygen-containing groups promote the carrier transfer efficiency, strengthen the oxidation capacity of water, and lead to the highly selective generation of H₂O₂ on the surface of OPCN under visible light. At the same time, some electrons reduce Fe³⁺ to Fe²⁺, react with H₂O₂ to generate ·OH radicals, and degrade organic pollutants into water and CO₂ (Figure 9d).

4.5. Others

4.5.1. Heterojunction of g-C₃N₄ with Other Materials

The heterostructures formed between g-C₃N₄ and other materials (such as MOFs, metal oxides, single atom catalysts, etc.) can further enhance the photocatalytic reaction activity by improving the light absorption, optimizing the interfacial transport path between photogenerated carriers and enhancing the interfacial reaction characteristics. MOFs (metal–organic frameworks), as new porous nanomaterials with high porosity and low density that are constructed by metal ions and organic ligands, are widely favored in the field of photocatalysis [164,165,166,167]. For example, V. Muelas-Ramos et al. synthesized a g-C₃N₄/NH₂-MIL-125 composite material where the presence of MOF (NH₂-MIL-125) not only inhibited the recombination rate of the g-C₃N₄ photogenerated electron–hole pairs, but also increased the g-C₃N₄ specific surface area and pore volume, demonstrating an excellent photocatalytic degradation diclofenac rate [164].

However, compared with the conventional MOF materials, since metal oxides (e.g., TiO₂, WO₃, ZnO, etc.) have the advantages of stable structure, good biocompatibility, and convenient preparation, they are also widely used to form heterojunctions with g-C₃N₄ to improve the optical absorption of composite materials, photocarrier separation, and the adsorption activity of reactants [168,169]. Wang et al. synthesized in situ a compact and uniform core-shell TiO₂@g-C₃N₄ quantum heterojunction using urea and surface titanium dioxide as precursors [170]. Due to the unique structural advantages of the g-C₃N₄ shell and the compact and uniform contact interface with TiO₂, it promoted photo-generated electron transfer and ·OH surface adsorption, thus significantly improving the degradation activity of photocatalytic tetracycline.

Moreover, it is worth noting that single-atom catalysts are widely used in the field of photocatalysis because of their advantages such as atomic-level active sites, high catalytic performance, low metal consumption, and stable structure [171]. Therefore, the heterojunctions formed by the hybridization of a single atomic catalyst and g-C₃N₄ creates new opportunities for improving the photocatalytic degradation performance. For instance, Wang et al. synthesized single-atom dispersed Ag modified mesoporous graphite carbon nitride (Ag/mpg-C₃N₄) by the co-condensation method [172]. The introduction of single-atom Ag enhanced the absorption and utilization of light and the Ag/mpg-C₃N₄ photocatalyst could degradation 100% bisphenol A (BPA) and mineralization of 80% TOC within 60 min.

4.5.2. g-C₃N₄-Based Magnetic Photocatalysts

In addition, although g-C₃N₄ has the advantages of high thermodynamic and chemical stability and condonable energy band structure, it still suffers from the disadvantages of limited light absorption, low charge separation efficiency, and difficulty in recycling, which limit the photocatalytic performance of the material in practical applications [173]. Based on the advantages of g-C₃N₄-based magnetic photocatalysts such as excellent optical properties, thermal stability, and easy recycling, they have good prospects for the photocatalytic self-Fenton system. In 2021, Xiong et al. synthesized g-C₃N₄/diatomaceous earth/Fe₃O₄ composites by electrostatic self-assembly and added H₂O₂ to the system for the efficient degradation of rhodamine by the photocatalytic self-Fenton [174]. The incorporation of diatomaceous earth and magnetic material (Fe₃O₄) narrowed the band gap of the material, increased the light utilization rate, and improved the degradation ability of the g-C₃N₄-based photocatalyst. In addition, the composites had good recyclability. In 2022, Wang et al. synthesized magnetic Fe₃O₄/CeO₂/g-C₃N₄ composites used for the efficient photocatalytic degradation of tetracycline hydrochloride (TCH) by the hydrothermal method [175]. The optimal material could remove 96.63% of TCH after 180 min of reaction. In addition, the composite had a lower leaching rate of Fe ions. Therefore, the g-C₃N₄-based magnetic photocatalyst can efficiently utilize photo-Fenton properties through the synergistic effect of semiconductors and magnetic materials as well as reduce environmental hazards by recycling materials.

5. Conclusions and Prospects

To conclude, this review summarizes the practical applications of g-C₃N₄-based photocatalysts in the photocatalytic self-Fenton system based on the existing literature and discusses in detail the mechanisms of photocatalytic self-Fenton in the degradation of emerging pollutants and the key factors affecting the degradation activity. In addition, according to the inherent defects of g-C₃N₄-based photocatalysts such as low light absorption utilization, few active sites, photogenerated electron–hole recombination, etc., different modification strategies to improve the performance of photocatalytic self-Fenton degradation were discussed in detail including morphological modulation, construction heterojunctions, defect engineering, and element doping strategies. Although the existing research on g-C₃N₄-based photocatalytic self-Fenton systems has made considerable progress from the perspective of adding additional Fe ions, the pH application range of the reaction system is narrow (pH ≤ 3), and Fe³⁺ is difficult to separate and recycle after the reaction, which cannot meet the actual needs of an efficient, green, and safe deep treatment technology that can be applied to the degradation and mineralization of emerging pollutants in the environment. In order to better improve the activity of a photocatalytic self-Fenton system, some insights and prospects are provided, as follows:

(1)

In the photocatalytic self-Fenton process, the catalyst generates H₂O₂ by external interaction with Fe²⁺ in the photocatalytic phase, generating active species such as •OH by in situ activation, and promotes Fe²⁺/Fe³⁺ cycling for the efficient degradation of pollutants. However, most self-Fenton systems have problems such as metal leaching or agglomeration and difficult separation and recovery of Fe³⁺ after the reaction, leading to low material utilization and environmental pollution, so it is important to design and synthesize environmentally stable catalysts. Secondary contamination is avoided by constructing metal atom dispersion to hinder the leaching of iron atoms as well as the rational use of iron materials such as adding spinel ferrite materials to the system to form heterojunctions with the catalyst. The preparation of catalysts that are easy to synthesize, stable, easy to separate, and prevent metal ion leaching plays an important role in improving the performance of photocatalytic self-Fenton systems.

(2)

As a photocatalytic system, the photocatalytic self-Fenton system is based on three basic processes of photocatalysis, whose photogenerated carrier separation and migration efficiency and photo-response range affect its activity. Currently, most g-C₃N₄-based photocatalysts are limited to the visible light region, and the near-infrared light, which accounts for 50% of the solar spectrum, is not effectively utilized, resulting in low overall light utilization of the catalyst. Although coupling with semiconductors and heteroatom doping can adjust the band gap to improve light absorption, the improvement in the light absorption range is limited based on the inherent nature of semiconductors. Therefore, it is a good strategy to achieve a broad-spectrum photo-response range by combining with near-infrared response materials containing metal elements such as Ag and Cu to generate a local plasmon resonance effect or up-conversion process.

(3)

In addition to the catalyst properties themselves, external factors such as the pH of the reaction system and reaction solvent have significant effects on the catalyst activity. The conventional Fenton and Fenton-like reaction systems mainly generate active species such as hydroxyl groups via H₂O₂ and Fe²⁺ under acidic conditions, which limit the scope of catalyst applications, thus, it is important to design and synthesize catalysts with a wide pH response. Strategies such as the formation of iron complexes and the use of Fenton-like systems with a wide pH range can be used to prepare photocatalytic self-Fenton systems for the efficient degradation of pollutants.

(4)

Sacrificial agents play an important role in the photocatalytic Fenton system, acting as electron acceptors or donors, trapping photogenerated electrons or holes and promoting photogenerated carrier separation efficiency. However, the widespread use of sacrificial agents lacks economic benefits, so the reaction performance can be improved by adding some functional small molecules that are not sacrificial agents. For example, when ethylenediaminetetraacetic acid (EDTA) is added to the reaction system, the electron-rich carboxyl groups in EDTA form hydrogen bonds with O₂, which increases the electron density and bond length of O₂ molecules, thus facilitating O₂ activation and significantly increasing the H₂O₂ yield of the reaction system. In addition, functional small molecules such as oxalic acid and furfuryl alcohol were applied in the photocatalytic H₂O₂ production system, which significantly improved the H₂O₂ yield. Therefore, it is a direction worth exploring to effectively utilize the synergy between functional molecules and catalysts to improve the H₂O₂ yield and thus promote the subsequent Fenton reaction for the efficient degradation of pollutants.