Heterogeneous semiconductor photocatalysis is considered a promising technology due to its great potential to solve energy and environmental problems [1, 2]. Traditional photocatalyst TiO₂ is low cost, highly photocatalytically active, and shows excellent chemical and photochemical stabilities [3, 4]. However, the wide band gap of TiO₂ renders it responsive to only UV light, which accounts for only 4% of solar light. Meanwhile, the visible light region comprises 43% of the entire solar spectrum. Accordingly, much research is currently focused on the manufacture of efficient visible-light-response photocatalysts for the utilization of solar energy [5-8].

Among well-known visible-light-driven photocatalysts, silver orthophosphate (Ag₃PO₄), has attracted increasing attention since first being reported by Ye et al. [9] in 2010. Ag₃PO₄ has a body-centered cubic structure with space group P4-3n [10]. As shown in Fig. 1, all atoms are in four-coordinate environments, with the Ag atom coordinated by four O atoms, the P atoms coordinated by four O atoms, and the O atoms coordinated by three Ag atoms and one P atom [11]. Ag₃PO₄ has an indirect band gap of 2.36 eV and direct transition of 2.43 eV, and is considered a promising visible-light-driven photocatalyst [10]. Under visible light illumination, Ag₃PO₄ can achieve a quantum efficiency of ~90% for water oxidation using AgNO₃ as the electron acceptor [9], which is much higher than that of the other reported visible-light-driven photocatalysts [12]. In our previous work, under simulated solar light or visible LED light, Ag₃PO₄ drastically degraded dye pollutants in a short time [13, 14]. Ag₃PO₄ has excellent photocatalytic activity due to its superior electronic structure characteristics [10]. Using density-functional-theory-based calculations, Umezawa et al. [15] concluded that the excellent photocatalytic performance of Ag₃PO₄ was partly attributed to the highly dispersive band structure of the conduction-band minimum (CBM), which results from Ag s/Ag s hybridization without localized d states. Zhu et al. [11] employed first-principles density functional theory incorporating the LDA+U formalism to investigate the origin of Ag₃PO₄ photocatalytic activation. They demonstrated that Ag₃PO₄ has a highly disperse conduction band and the PO₄^3- inductive effect, which helps separate electron/hole pairs. Furthermore, the defect states of Ag vacancies act as capture traps for photoexcited holes, which promotes the separation of electron/hole pairs excited by visible-light irradiation.

Although Ag₃PO₄ possesses outstanding visible-light-driven photocatalytic performance, the photocorrosion of Ag₃PO₄ is the main obstacle to its practical application [10, 12]. The conduction band (CB) of Ag₃PO₄is located at +0.45 V vs. NHE (normal hydrogen electrode, pH = 0), indicating that H₂O cannot be reduced to H₂ by Ag₃PO₄. Therefore, the electrons in the Ag₃PO₄ CB can reduce Ag⁺ ions released from the crystal lattice of Ag₃PO₄ to Ag. This phenomenon destroys the Ag₃PO₄ structure and decreases its photocatalytic performance [10]. To meet industrial requirements, much effort has been devoted to further enhance the photocatalytic performance and stability of Ag₃PO₄, including metal deposition [16, 17], assembly with carbon materials [18-20], immobilization on support materials [21, 22], and combination with other semiconductors to form type-Ⅱ heterojunctions or Z-scheme systems [10, 12, 23]. In particular, mimicking natural photosynthesis in Ag₃PO₄-based all-solid-state Z-scheme systems could produce a wide absorption range, high charge-separation efficiency, strong redox ability, and long-term stability [23]. In this brief review, we provide an overview of the main achievements in Ag₃PO₄-based Z-scheme photocatalytic systems, focusing on their construction, application, and reaction mechanism, followed by a brief discussion of the future development of this class of photocatalytic system.

Photocatalytic reaction processes mostly involve three steps. (ⅰ) Generation of electrons and holes through the absorption of light with higher energy than the band gap of the semiconductor photocatalyst; (ⅱ) charge separation and migration onto the photocatalyst surface; and (ⅲ) reduction/oxidation reactions on the photocatalyst surface [24, 25]. To utilize solar light, the band gap of the photocatalyst must be narrow. However, a narrow band gap allows photogenerated electrons and holes to easily recombine. Therefore, it is difficult for a single photocatalyst to simultaneously utilize solar light and possess high photocatalytic activity. This problem can be overcome by constructing heterogeneous photocatalytic systems (usually type-Ⅱ heterojunctions) [26]. In a type-Ⅱ heterostructured photocatalyst system (Fig. 2(a)) under visible light irradiation, photoinduced electrons (e^-) in the CB of photocatalyst Ⅰ (PC Ⅰ) can transfer to the CB of photocatalyst Ⅱ (PC Ⅱ), while holes (h⁺) in the valence band (VB) of PC Ⅱ can migrate to the VB of PC Ⅰ, which promotes the effective separation of electrons and holes, resulting in enhanced photocatalytic performance [26]. Although the recombination of electrons and holes is inhibited in type-Ⅱ heterostructured photocatalytic systems, the redox ability of photogenerated electrons and holeson reaction sites is decreased after migration (Fig. 2(a)) [23, 27]. Therefore, it is necessary to design and explore new photocatalytic systems to solve these problems. By mimicking natural photosynthesis, a newly-developed Z-scheme photocatalytic system can satisfy the aforementioned requirements [23, 27].

In nature, H₂O and CO₂ are converted into O₂ and carbohydrate through photosynthesis in green plants (Fig. 2(b)). This photosynthetic process, which looks like the letter "Z", is known as the Z-scheme system. As shown in Fig. 2(b), photosystems Ⅰ (PS Ⅰ) and Ⅱ (PS Ⅱ) are excited by solar light irradiation, and the photogenerated electrons jump to a higher electronic state (excitation), which is connected to an electron transfer chain (electron mediator). The electrons in PS Ⅰ flow from the electron transport chain, leading to the reduction of coenzyme NADP⁺ into NADP, which reduces CO₂ to carbohydrate, while water oxidation occurs in PS Ⅱ [28, 29]. Mimicking the natural photosynthesis of green plants, the artificial Z-scheme photocatalytic system has been explored, and was first used for water splitting, with an electron acceptor/donor (A/D) usually serving as the electron mediator [30-32]. As shown in Fig. 2(c), two different narrow-bandgap semiconductor photocatalysts are selected as PS Ⅰ and PS Ⅱ. Under visible light irradiation, photoinduced electron transfer from the CB of PS Ⅱ to the VB of PS Ⅰ is dependent on the redox reactions of the A/D pair. Therefore, the electron in the CB of PS Ⅰ and the hole in the VB of PS Ⅱ participate in water splitting. Commonly used A/D pairs are IO^3-/I^-, Fe³⁺/Fe²⁺, and NO₃^-/NO₂^-[23]. These redox mediators can strongly absorb visible light, decreasing the light absorption of the photocatalysts (PS Ⅰ and PS Ⅱ), making it difficult to maintain long-term stability and active states of these A/D pairs. Furthermore, this Z-scheme system (Fig. 2(c)) cannot be used to degrade pollutants in solution because the pollutants can hinder the redox reaction of the A/D pair. Accordingly, all-solid-state Z-scheme photocatalytic systems have been developed.

As shown in Fig. 3(a), two different narrow band gap photocatalysts were used as PS Ⅰ and PS Ⅱ. A solid conductor between the two photocatalysts forming the ohmic contact was employed as a solid-state electron mediator [23]. Under visible light, the electrons can be simultaneously excited from the VB of both PS Ⅰ and PS Ⅱ to the CB, leaving holes in the VB. The electrons from the CB of PS Ⅱ can directly recombine with the holes from the VB of PS Ⅰ through the ohmic contact [23, 32]. Therefore, the recombination of electrons and holes is inhibited within PS Ⅰ or PS Ⅱ, such that electrons in the CB of PS Ⅰ and holes in the VB of PS Ⅱ can be mostly reserved for forward reduction and oxidation reactions (Fig. 3(a)), respectively. Moreover, in this system, the shielding effect of irradiated incident light caused by the A/D pair can be eliminated. Without this limitation, all-solid-state Z-scheme systems can be used in both the gas and liquid phase.

As many defects can aggregate at the solid/solid contact interface, the energy levels of the solid/solid contact interface are quasicontinuous, which is similar to in a conductor. This indicates that the solid/solid contact interface has some properties similar to conductors. Therefore, the solid/solid contact interface can also form the ohmic contact [23]. Without a conductor as the electron mediator, a direct Z-scheme photocatalytic system can be formed when PS Ⅰ directly contacts PS Ⅱ (Fig. 3(b)). Similarly, the direct Z-scheme photocatalytic system can also retain oxidative holes and reductive electrons in different counterparts (Fig. 3(b)), resulting in enhanced photocatalytic activity. Evidently, all-solid-state Z-scheme photocatalytic systems can solve the problems of single photocatalyst and type-Ⅱ heterostructured photocatalytic systems (Fig. 3).

Ag₃PO₄ has a very deep valence band, located at approximately +2.9 V vs. NHE (pH = 0), making it well aligned for oxidation reactions [12]. Ag₃PO₄ is used as PS Ⅱ and coupled with a suitable visible-light-driven photocatalyst (PS Ⅰ) to form the Z-scheme system. Therefore, the photogenerated electrons in the CB of Ag₃PO₄ can recombine with the holes from PS Ⅰ, resulting in the improved photostability of Ag₃PO₄. Furthermore, the leaving holes in the VB of Ag₃PO₄ can retain the strong oxidizability, allowing the photocatalytic oxidation of water and organic pollutants [10, 12]. Therefore, Ag₃PO₄-based Z-scheme photocatalytic systems show great potential for solving energy and environmental problems in the future.

In Ag₃PO₄-based Z-scheme photocatalytic systems, metal Ag conductor has always been used as the electron mediator because Ag nanoparticles can be generated in situ during preparation or produced by the photoreduction of Ag₃PO₄ in the photocatalytic process. Recent progress in Ag₃PO₄-based Z-scheme photocatalytic systems with solid state electron mediators are summarized in Table 1.

In recent years, graphitic C₃N₄ (g-C₃N₄), a novel metal-free inorganic semiconductor, has attracted much attention in the photocatalysis of water splitting, CO₂ reduction, and environmental purification [33]. The band gap of g-C₃N₄ is about 2.7 eV, meaning that it can absorb visible light up to 460 nm. Furthermore, the CB minimum of g-C₃N₄ is extremely negative, so photogenerated electrons in the CB of g-C₃N₄ should have a high reduction ability [33, 34]. As described above, holes in the VB of Ag₃PO₄ have excellent oxidation abilities; therefore, combining Ag₃PO₄ with g-C₃N₄ to construct a Z-scheme photocatalytic system was expected to achieve enhanced photocatalytic performance. Katsumata et al. [35] synthesized a g-C₃N₄/ Ag₃PO₄ hybrid photocatalyst using a facile in situ precipitation route, confirming that a small amount of Ag was formed on the g-C₃N₄/Ag₃PO₄ surface from the electron-rich structure of g-C₃N₄ donating electrons to Ag₃PO₄. The as-prepared g-C₃N₄/Ag₃PO₄ containing 25 wt% g-C₃N₄ exhibited the highest photocatalytic performance for methyl orange (MO) degradation, and its activity and stability were better than that of pure Ag₃PO₄. The higher photoluminescence (PL) intensity of g-C₃N₄/Ag₃PO₄ compared to pure Ag₃PO₄ was attributed to the higher recombination rate between electrons in the CB of Ag₃PO₄ and holes in the VB of g-C₃N₄. Furthermore, superoxide radicals (O₂^*-) and holes were found to be the main reactive species for g-C₃N₄/Ag₃PO₄ in MO degradation. If photogenerated charge carrier transfer in g-C₃N₄ and Ag₃PO₄ occurs via a type-Ⅱ heterojunction mechanism, the photogenerated electrons in the CB of g-C₃N₄ would migrate to the CB of Ag₃PO₄. Due to the positive E_CB (+0.45 V vs. NHE) of Ag₃PO₄, the electrons in the CB of Ag₃PO₄ could not reduce O₂ to O₂^*- with a redox potential of -0.046 V, which was not in agreement with the radical capture experiments. The CB potential of g-C₃N₄ is -1.15 V, and electrons in the CB of g-C₃N₄ can reduce O₂ to O₂^*-. Therefore, g-C₃N₄/Ag₃PO₄ was confirmed as a typical Z-scheme photocatalyst with Ag as the electron mediator, and the photocatalytic mechanism for MO degradation by g-C₃N₄/Ag₃PO₄ is shown in Fig. 4(a). Under visible light irradiation, both Ag₃PO₄ and g-C₃N₄ are excited, leading to the formation of photogenerated electrons in the CB and holes in the VB. Electrons in the CB of Ag₃PO₄ shifted into Ag nanoparticles due to the more positive Fermi energy of Ag compared to the CB level of Ag₃PO₄. Simultaneously, the holes in the VB of g-C₃N₄ can move to Ag and combine with the electrons from Ag₃PO₄. However, the highly reducing electrons in the CB of g-C₃N₄ could react with O₂ to form O₂^*-, which would lead to MO degradation. The holes with strong oxidative abilities in the VB of Ag₃PO₄ would degrade MO directly (Fig. 4(a)). As previously reported, Ag₃PO₄ with (111) facets exhibited the best photocatalytic activity due to the (111) facet possessing the highest surface energy [36]. Fan et al. [37] developed a facile solvent evaporation method for the synthesis of g-C₃N₄/Ag₃PO₄(111) hybrid photocatalyst, during which metallic Ag was also formed. Under visible light irradiation (λ > 400 nm), the as-obtained g-C₃N₄/Ag₃PO₄(111) containing 10 wt% of g-C₃N₄ exhibited the highest photocatalytic activity in the degradation of methylene blue (MB) and NO gas, and possessed stable photocatalytic ability compared with pure Ag₃PO₄. The detection of reactive species confirmed that the enhanced photocatalytic activity of g-C₃N₄/Ag₃PO₄(111) composite was due to the efficient separation of electron/hole pairs through a Z-scheme system composed of Ag₃PO₄, Ag, and g-C₃N₄, in which Ag acted as the electron mediator. In another study, a facile precipitation method was employed to synthesize the Ag₃PO₄/g-C₃N₄ Z-scheme photocatalyst, which was used to degrade sulfamethoxazole in water. Under visible light irradiation, the sulfamethoxazole removal efficiency using Ag₃PO₄/g-C₃N₄ (m/m = 98/2) was higher than that of pure Ag₃PO₄ and g-C₃N_4.In the photodegradation process, in situ generated metallic Ag acted as an electron mediator between Ag₃PO₄ and g-C₃N₄ [38].

In addition to its application in photodegradation, the Ag₃PO₄/g-C₃N₄ Z-scheme system has also been used to convert CO₂ to fuels and oxidize H₂O to O₂. Fan et al. [39] prepared Ag₃PO₄/g-C₃N₄ composite using a simple deposition method, and then formed metallic Ag in situ by irradiating the Ag₃PO₄/g-C₃N₄ composite with a 500 W Xe lamp. Under simulated sunlight irradiation, the optimal Ag₃PO₄/g-C₃N₄ showed a CO₂ conversion rate of 57.5 μmol/(h·g_cat), which was 6.1 times higher than that of g-C₃N₄, and found that the main reduction product was CO. The results of CO₂ photoreduction and reactive species scavenging confirmed that the transfer of electrons and holes in Ag₃PO₄/g-C₃N₄ composite obeyed the Z-scheme photocatalytic mechanism, as shown in Fig. 4(b). Under simulated sunlight, the photogenerated electrons in the CB of Ag₃PO₄ shift to metallic Ag and then combine with holes from g-C₃N₄. The leaving electrons in the CB of g-C₃N₄ can reduce CO₂ to fuels. Using an electrostatic assembly and solution-based precipitation method, Yang et al. [40] reported a Z-scheme system of Ag₃PO₄/g-C₃N₄ for water photooxidation using AgNO₃ as the electron acceptor. The authors demonstrated that the morphology of g-C₃N₄ in the composite material can influence photoactivity in the O₂ evolution reaction. After exposure to white LED light, the evolved O₂ efficiency over the optimal Ag₃PO₄/g-C₃N₄ composite was higher than that over pure Ag₃PO₄. Electronic spin resonance (ESR) results indicated that hydroxyl radicals (OH^*) were formed in the oxygen evolution process. If the photocatalytic mechanism for oxygen evolution reaction was a type-Ⅱ heterojunction, the photogenerated holes in the VB of Ag₃PO₄ could migrate to the VB of g-C₃N₄, meaning that OH^* could not be generated. This confirmed that an in situ Z-scheme system was obtained by the generation of metallic Ag nanoparticles, which acted as a recombination center for photogenerated electrons from Ag₃PO₄ and holes from g-C₃N₄, maintaining active holes in the VB of Ag₃PO₄ for water oxidation (Fig. 4(c)). Using a facile in situ precipitation method, g-C₃N₄/tetrahedral Ag₃PO₄ hybrid photocatalysts with different mass ratios of g-C₃N₄ were prepared by Ktsumata et al. [41], in which the g-C₃N₄ content had a large influence on the photocatalytic performance. The 20 wt% g-C₃N₄/tetrahedral Ag₃PO₄ photocatalyst exhibited the highest photocatalytic activity for O₂ evolution from water because a Z-scheme system composed of Ag₃PO₄, Ag, and g-C₃N₄ was formed.

In addition to g-C₃N₄, researchers have coupled other photocatalysts with Ag₃PO₄ to construct Z-scheme photocatalytic systems with enhanced activity and stability. The solubility of Ag_nX (X =Br, I, S) is lower than that of Ag₃PO₄, allowing Ag₃PO₄/Ag_nX (X =Br, I, S) hybrids to be obtained via an in situ ion-exchange route [42]. Ag₃PO₄/AgI composite was fabricated using an in situ anion-exchange method by Fang et al. [43] and showed enhanced visible-light-driven photocatalytic performance for MO and phenol degradation compared to pure Ag₃PO₄ and AgI. XPS and TEM analysis confirmed that Ag nanoparticles formed in the early stage of the photocatalytic process, which could act as the charge transmission bridge between Ag₃PO₄ and AgI to construct the Z-scheme system, resulting in enhanced activity and stability. Similarly, other Z-scheme photocatalysts, such as AgBr/Ag₃PO₄ [44] and Ag₂S/Ag₃PO₄[45], have been synthesized via anion-exchange methods with in situ generated metallic Ag acting as the electron mediator in the photocatalytic process. Furthermore, other Ag₃PO₄-based Z-scheme systems using Ag as the electron mediator, including Ag₃PO₄/Ag/SiC [46], In₂O₃/Ag/Ag₃PO₄ [47], Ag₃PO₄/Ag/CuBi₂O₄ [48], SrTiO₃/Ag/Ag₃PO₄ [49], Ag₃PO₄/Ag/WO_2.72 [50], and Ag₃PO₄/Ag/MoS₂[51], have been prepared via liquid phase synthesis (see Table 1). In the Ag₃PO₄/Ag/WO_2.72 Z-scheme system, Ag₃PO₄ and WO_2.72 were used as PS Ⅰ and PS Ⅱ, respectively, due to the VB of WO_2.72 being located at a more positive position than that of Ag₃PO₄ and the CB of Ag₃PO₄ being located at a more negative position than that of WO_2.72.

Recently, a novel plasmonic Z-scheme system was developed due to the response of noble-metal nanoparticles (such as Ag) to visible light through the surface plasmon resonance (SPR) effect [52]. Tang et al. [53] prepared binary Ag₃PO₄/Ag₂MoO₄ hybrid materials using a facile solution phase reaction. XRD analysis showed that Ag nanoparticles were formed on the surface of the Ag₃PO₄/Ag₂MoO₄ composites in the initial stage of photocatalytic process, leading to the generation of the Ag₃PO₄/Ag/Ag₂MoO₄ photocatalyst, which exhibited enhanced photocatalytic activity and photostability toward the remediation of organic dye compared with pure Ag₃PO₄. This result was due to the plasmonic Z-scheme photocatalytic mechanism, which was confirmed by reactive species trapping experiments and band structure analysis of Ag₃PO₄ and Ag₂MoO₄. As shown in Fig. 5, under visible light irradiation, Ag₃PO₄ can absorb photons to produce photogenerated electrons and holes. The surface plasmonic resonance effect and dipolar character of metallic Ag allows Ag to also absorb visible light, with each absorbed photon inducing the efficient separation of an electron and hole. Plasmon-induced electrons in Ag nanoparticles are transported to the CB of Ag₂MoO₄ to react with O₂, forming O₂^*-active species, which can degrade organic pollutants. Simultaneously, photogenerated electrons in the CB of Ag₃PO₄ transfer to the Ag nanoparticles to recombine with holes produced by plasmonic absorption in the Ag nanoparticles, while the photogenerated holes in the VB of Ag₃PO₄ can directly oxidize the dye molecules.

In addition to Ag, some nonmetal materials with excellent conductivity can be employed as solid state electron mediators. Recent studies have confirmed that reduced graphene oxide (RGO) can be used as the electron mediator to promote charge transfer between PS Ⅰ and PS Ⅱ in the Z-scheme system [54, 55]. Liu et al. [56] reported a Ag₃PO₄@RGO@La, Cr:SrTiO₃ composite photocatalyst that showed superior anti-photocorrosion and photocatalytic activities in the degradation of both RhB and 2, 4-DNP. Reactive species trapping experiments and ESR results indicated that O₂^*- and holes played significant roles in the photocatalytic degradation processes. If charge carrier transfer between Ag₃PO₄ and La, Cr:SrTiO₃ occurred via a type-Ⅱ heterojunction mechanism, namely, the photogenerated electrons in the CB of La, Cr:SrTiO₃ migrate to the CB of Ag₃PO₄ with holes in the VB of Ag₃PO₄ simultaneously transferring to the VB of La, Cr:SrTiO₃, then the photogenerated electrons would accumulate in the CB of Ag₃PO₄ and be unable to reduce O₂ into O₂^*-, which would contradict the above experimental results. Therefore, charge transfer in Ag₃PO₄@RGO@La, Cr:SrTiO₃ composite was shown to obey the Z-scheme photocatalytic mechanism. Under visible light irradiation, photogenerated electrons in the CB of Ag₃PO₄ could transfer to the donor levels of La, Cr:SrTiO₃ via RGO and recombine with photogenerated holes. Meanwhile, the excited electrons in the CB of La, Cr:SrTiO₃ with a high reduction ability could react with O₂ to produce O₂^*-, and holes in Ag₃PO₄with excellent oxidation power could react with H₂O to produce OH^*, thereby achieving the highly efficient degradation of RhB and 2, 4-DNP. Furthermore, RGO could be used as a sheltering layer to protect Ag₃PO₄from photocorrosion.

A schematic of a direct Z-scheme photocatalytic system is shown in Fig. 3(b). As discussed above, without an electron mediator, the direct Z-scheme photocatalytic system could also retain a high redox ability. Recent studies on direct Ag₃PO₄-based Z-scheme photocatalytic systems are summarized in Table 2.

As described above, combining Ag₃PO₄ with g-C₃N₄ can form Z-scheme photocatalytic systems with Ag as the electron mediator [35-41]. However, direct Ag₃PO₄/g-C₃N₄ Z-scheme systems have also been reported. Yi et al. [57] synthesized g-C₃N₄/Ag₃PO₄ composite by mixing g-C₃N₄ and Ag₃PO₄ in water with stirring, finding that both the C₂H₄ photo-oxidative activity and stability of Ag₃PO₄ were enhanced by fabrication of the composite. According to the C₂H₄ degradation results and band structure analysis of g-C₃N₄ and Ag₃PO₄, it was concluded that charge transfer in the g-C₃N₄/Ag₃PO₄ composite did not follow a conventional type-Ⅱ heterojunction, and a direct Z-scheme photocatalytic mechanism was suggested. Chen et al. [58] used a facile in situ precipitation method to prepare Ag₃PO₄/g-C₃N₄ nanocomposites with different molar ratios of Ag₃PO₄ to g-C₃N₄that exhibited enhanced photocatalytic performance for methylene blue degradation compared to pure Ag₃PO₄ and g-C₃N₄ under visible light. Furthermore, XPS results showed no metallic Ag on the surface of the fresh and used Ag₃PO₄/g-C₃N₄ nanocomposites. The results of electron spin resonance (ESR), the photoluminescence (PL), and reactive species determination confirmed that the migration of photogenerated electrons/holes exhibited a direct Z-scheme mechanism in the Ag₃PO₄/g-C₃N₄ nanocomposite.

Other direct Z-scheme composites, such as Ag₃PO₄/ZnO [59], Ag₃PO₄/SnS₂ [60], Ag₃PO₄/SnSe₂ [61], Ag₃PO₄/Bi₂MoO₆ [62], Ag₃PO₄/MoS₂ [63], and Ag₃PO₄/WO₃ [64], have also been synthesized using liquid-phase methods, including precipitation/deposition, hydrothermal treatment, and an organic phase in situ growth strategy.

In a Z-scheme system of Ag₃PO₄/MoS₂ reported previously [51], Ag₃PO₄ nanoparticles were loaded on the surface of MoS₂ nanoslices. Inevitably, metallic Ag could be in situ produced in the initial stage of the photocatalytic process and act as the electron mediator. However, Fan et al. [63] reported the direct Z-scheme system of a Ag₃PO₄@MoS₂ core-shell heterostructure, which was constructed because the MoS₂ shell can protect Ag₃PO₄ from dissolution and photocorrosion during the photocatalytic process. The Ag₃PO₄@MoS₂ photocatalyst showed excellent photocatalytic activity and stability in the photodegradation of RhB and the photocatalytic selective oxidation of benzyl alcohols (BA) to benzaldehyde compared with Ag₃PO₄. The energy band structure and quenching effects of different scavengers showed that the superior photocatalytic activity of the as-synthesized Ag₃PO₄@MoS₂ composite originated from the direct Z-scheme charge carrier migration mechanism, as shown in Fig. 6(a). Under visible light irradiation, both Ag₃PO₄ and MoS₂ were excited. Subsequently, electrons in the CB of Ag₃PO₄ could recombine with holes in the VB of MoS₂ through the solid/solid contact interface. As a result, abundant electrons on the CB of MoS₂ and holes on the VB of Ag₃PO₄participated in the reduction reaction of dissolved O₂and the oxidation of RhB and BA, respectively.

Notably, in the direct Z-scheme system of Ag₃PO₄/WO₃ composite, WO₃ operates as PS Ⅱ rather than Ag₃PO₄ due to the VB of WO₃ lying at a more positive position than that of VB of Ag₃PO₄. The enhanced photocatalytic performance was mainly attributed to the improved separation of photogenerated charge carriers between Ag₃PO₄ and WO₃ via the direct Z-scheme mechanism (Fig. 6(b)). Under visible light irradiation, Ag₃PO₄ and WO₃ can be excited to produce electrons and holes, simultaneously, electrons in the CB of WO₃ transfer to the VB of Ag₃PO₄ to combine with holes. The holes retained in the VB of WO₃ can directly degrade the adsorbed organic dyes or oxidize OH^-to OH^* radicals, triggering a series of degradation reactions. The electrons left in the CB of Ag₃PO₄ can be consumed through a multi-electron reaction with oxygen and finally produce OH^* radicals, which accelerate the organic dye degradation [64].

Photocatalysis is a promising technology for solving the environmental and energy problems in the future. Traditional photocatalyst TiO₂ has a wide band gap that limits its practical application. Silver phosphate (Ag₃PO₄) is among the most active visible-light-driven photocatalysts reported, with the excellent photocatalytic performance ascribed to its superior electronic structure characteristics. However, photocatalyst stability remains a major problem of Ag₃PO₄ in both water oxidation and organic decomposition. To meet industrial requirements, much effort has been devoted to further improving the photocatalytic performance and stability of Ag₃PO₄, including metal deposition, assembly with carbon materials, immobilization on support materials, and combination with other semiconductors to construct type-Ⅱ heterojunctions or Z-scheme systems. In particular, Ag₃PO₄-based all-solid-state Z-scheme photocatalytic systems could effectively improve the separation/transportation of charge carriers and substantially enhance the photocatalytic activity and stability of Ag₃PO₄. Although great progress has been made, some remaining problems must be solved to further advance the use of Ag₃PO₄-based Z-scheme photocatalytic systems.

(1) As shown in Figs. 2(a) and 3(b), when the band gaps of the two semiconductors are staggered, there is much controversy regarding whether the type-Ⅱ heterojunction photocatalytic mechanism or the indirect Z-scheme photocatalytic mechanism occurs. Using Ag₃PO₄/g-C₃N₄ composite as an example, some reports have confirmed that the migration of photogenerated electrons and holes in Ag₃PO₄/g-C₃N₄ exhibits the direct Z-scheme mechanism [57, 58], while other studies have shown that Ag₃PO₄/g-C₃N₄ composite followed a type-Ⅱ heterojunction mechanism [65-67]. Therefore, an in situ detection technique should be developed to clearly distinguish these photocatalytic mechanisms.

(2) Recently, graphene, which has excellent electrical conductivity, has attracted great interest in photocatalysis [68]. Therefore, the graphene could be used as the electron mediator in the Ag₃PO₄-based Z-scheme photocatalysts to improve photocatalytic activity. Previous reports also showed that the size and morphology of Ag₃PO₄ can greatly influence its photocatalytic performance [12]. Therefore, the size and morphology of Ag₃PO₄ should be adjusted to design more efficient Ag₃PO₄-based Z-scheme photocatalytic systems.

(3) Previous reports of Ag₃PO₄-based Z-scheme photocatalysts have mainly focused on the photodegradation of organic pollutants in water (Tables 1 and 2), while their application to photocatalytic H₂ generation and CO₂ reduction require future development. Ag₃PO₄-based Z-scheme systems have been applied to treat harmful gases (ethylene and nitric oxide), and could, therefore, be extended to degrade other harmful gases, such as formaldehyde and toluene.

(4) Separation and reuse of the photocatalysts in a solution dispersion system must be considered. Combining Ag₃PO₄ and magnetic semiconductor materials to form Z-scheme photocatalytic systems could address this issue