The chemical composition and structure are fundamental factors that influence the performance of photocatalysts, fundamentally determining key performance indicators such as active sites, electron transport properties, and the adsorption capacity for reactants.

TiO₂is a hotspot in photocatalysis research due to its low cost, chemical stability, excellent light selectivity, strong catalytic activity, and non-toxicity. Zhu Jiangfeng^[39]used tetrabutyl titanate as the main raw material and prepared anatase titanium dioxide nanosheets (TiO₂-NS) with a high proportion of exposed (001) active crystal facets via a hydrothermal method. Under irradiation with 500 W UV light at a dominant wavelength of 365 nm, this photocatalyst can reduce and remove 80% of NO₃ ^-in 50 mL of solution at a concentration of 50 ppm, with a N₂selectivity of 71%. This represents an improvement of 33% (from 47% to 80%) and 10% (from 61% to 70%) compared to the commercial TiO₂P25, respectively. The enhanced catalytic activity of TiO₂-NS is attributed to the large number of exposed (001) active crystal facets, which fully expose 100% of the unsaturated Ti_5csites, thereby enhancing the adsorption of nitrogen-containing species. However, due to its relatively wide band gap (typically above 3.0 eV), TiO₂photocatalysts generally respond only to ultraviolet light.

Metal sulfides typically have a narrower bandgap energy E_g than metal oxides (such as TiO₂), enabling them to absorb more sunlight and achieve higher light utilization, thereby generating higher concentrations of photogenerated electrons under the same illumination conditions. In addition, the conduction band position of metal sulfides is usually close to the reduction potential of water or NO₃^-, which facilitates the participation of photogenerated electrons in the reduction of NO₃^-[46]. Therefore, metal sulfides are also regarded as one of the ideal catalyst materials for photocatalytic reduction of NO₃^-. Wang et al.^[47] used CuCl₂·2H₂O, In₂O₃, and thiourea as the main raw materials and prepared CuInS₂ with a chalcopyrite structure via a hydrothermal method, and investigated its photocatalytic NO₃^- reduction capability. The experimental results show that, in the presence of oxalic acid as a hole scavenger, 0.1 g of CuInS₂ can reduce 74.8% of 200 mL of NO₃^- solution at a concentration of 25 ppm within 2 hours, demonstrating a photocatalytic NO₃^- reduction capability similar to that of P25 TiO₂. The reason why the photocatalytic NO₃^- reduction performance of CuInS₂ is suboptimal lies in its insufficient conduction band potential, high carrier recombination rate, and mismatch between its band structure and the requirements of the reaction. Yang et al.^[48] used CuCl₂·2H₂O, FeCl₃·6H₂O, CrCl₃·6H₂O, and thiourea as the main raw materials and prepared CuFe_1-xCrxS₂ with a chalcopyrite structure via a hydrothermal method (x ≤ 0.4). The NO₃^- removal rate of this material reaches up to 55%, exhibiting a stronger ultraviolet photocatalytic reduction capability for NO₃^- than P25 TiO₂ (55% vs. 19%). This is attributed to its E_g of approximately 0.61–0.80 eV, which is significantly lower than that of TiO₂ (~3.2 eV), making it easier for photogenerated electrons to be excited.

Polyoxometalates (POMs) can integrate the properties of different metal centers, exhibiting outstanding redox capabilities, precisely tunable structural and compositional characteristics, and excellent photocatalytic activity. These properties give POMs a significant advantage in photocatalytic NO₃ ^-reduction reactions. Kang et al.^[49]used Bi(NO₃)₃·5H₂O and Na₂MoO₄·2H₂O as the main precursors. They first prepared Bi₂MoO₆nanosheets via a hydrothermal method and then created oxygen vacancies (Ov-Bi₂MoO₆) on the nanosheets using plasma etching technology. They investigated the impact of common active center oxygen vacancies on the photocatalytic NO₃ ^-reduction performance of Bi₂MoO₆nanosheets. The experimental results show that although Bi₂MoO₆and Ov-Bi₂MoO₆have nearly identical E _gvalues (2.4 eV vs. 2.42 eV), under the same reaction conditions, their NO₃ ^-reduction rates are 30% and 38%, respectively. For Ov-Bi₂MoO₆, on the one hand, the d-valence electron orbitals of Mo can hybridize with the p-valence electron orbitals of N, weakening the N—O bond and effectively activating NO₃ ^-; on the other hand, the newly formed oxygen vacancies can effectively capture oxygen atoms from NO₃ ^-, further weakening the N—O bond and activating NO₃ ^-, thereby promoting the formation of adsorbed NO₂ (NO₂*). As an intermediate product, NO₂* is subsequently reduced to N₂in subsequent reaction steps, eventually desorbing from the catalyst surface and restoring the oxygen vacancies (Figure 2).

In specific application environments, especially under harsh conditions such as acidic ones, metal-containing catalysts carry the risk of metal ion leaching, which not only affects the durability of the catalyst but may also cause secondary pollution. In contrast, non-metal catalysts exhibit superior stability and can effectively address the issue of metal ion leaching. h-BN boasts strong high-temperature and corrosion resistance, and its relatively negative conduction band potential (−2.5 V vs. NHE) endows its valence band electrons (e_cb ^-) with sufficient direct reduction capability for NO₃ ^-, making it suitable for use in the field of photocatalytic NO₃ ^-reduction. Jiang et al.^[50]conducted separate studies on the photocatalytic NO₃ ^-reduction performance of commercial h-BN, P25 TiO₂, and self-prepared g-C₃N₄. The experimental results indicate that under identical experimental conditions, h-BN exhibits a stronger catalytic ability for NO₃ ^-reduction, with a maximum NO₃ ^-reduction rate of 97.16% and a N₂selectivity in the product reaching 95.42%. The NO₃ ^-reduction rates of P25 TiO₂and g-C₃N₄are both approximately 80%, but the N₂selectivities in their catalytic products are 77.54% and 57.37%, respectively. The strong catalytic NO₃ ^-reduction capability of h-BN stems from three key factors: (1) The reducing power of electrons generated in the conduction band of h-BN is the strongest (the potentials of electrons generated in the conduction bands of h-BN, P25 TiO₂, and g-C₃N₄are −2.5, −0.4, and −0.83 V, respectively), thereby promoting electron-mediated reduction processes; (2) The specific surface area of h-BN is approximately 38.9 m²·g^-1, significantly larger than that of g-C₃N₄(3.8 m²·g^-1) and TiO₂(13.6 m²·g^-1), providing more active sites for the adsorption of nitrate and formate ions and facilitating the NO₃ ^-reduction reaction; (3) The redox potential of catalytic systems containing h-BN is lower, accelerating the reduction process of NO₃ ^-. Moreover, h-BN demonstrates excellent stability: after three cycles of reuse, its NO₃ ^-reduction rate and N₂selectivity still remain at 93% and 90.67%, respectively.

The initial pH of the solution is another key parameter influencing the photocatalytic NO₃ ^-reduction efficiency, primarily through the following mechanisms: (1) altering the surface charge state of the catalyst; (2) regulating the adsorption behavior of reactants; (3) affecting the selectivity of the reaction pathway. Studies have shown that the performance of the catalyst varies significantly under different pH conditions.

Jung et al.^[51]prepared atomically thin TiO₂ nanosheets containing Ti defects via protonation and exfoliation, and investigated their photocatalytic NO₃^- reduction performance under different pH conditions. The experimental results show that under neutral conditions (pH = 7), the quantum confinement effect raises the conduction band edge of the material to −0.8 V vs. NHE, thereby satisfying the thermodynamic requirements for NO₃^- reduction; under acidic conditions (pH < 5), excess H⁺ competes with NO₃^- for adsorption sites, inhibiting the catalytic reaction; and under alkaline conditions (pH > 9), OH^- covers the active sites on the catalyst surface, leading to a decrease in catalytic activity. Schmuki et al.^[52]prepared p-type and n-type TiO₂ nanotubes (p-TiO₂ NTs and n-TiO₂ NTs) using a hydrothermal method and investigated their photocatalytic NO₃^- reduction to ammonia under pH = 7 conditions. The results indicate that under 8 hours of light irradiation without the addition of a hole scavenger, the ammonia yield of p-TiO₂ NTs is approximately 90 μmol·g^-1, whereas the ammonia yield of n-TiO₂ NTs is only about 20 μmol·g^-1. Further studies reveal that this difference stems from the fact that p-TiO₂ NTs exhibit a superior photocurrent response and higher carrier density under neutral conditions, and their high hole mobility effectively promotes the water oxidation reaction, thereby significantly enhancing the photocatalytic NO₃^- reduction performance. Silveira et al.^[53]used FeTiO₃ as a catalyst to investigate the effect of solution pH on the photocatalytic NO₃^- reduction reaction. The experimental results show that under acidic conditions (pH = 2.5–5), the reduction efficiency of NO₃^- is significantly higher than in neutral or alkaline environments. The reasons for this phenomenon are as follows: (1) The photocatalytic reduction of NO₃^- requires the consumption of large amounts of H⁺, and the acidic environment can continuously supply the protons required for the reaction; (2) When the solution pH approaches the isoelectric point of the catalyst at 4.5, the surface charge effect enhances the adsorption capacity of NO₃^- on the catalyst surface, thereby increasing its conversion efficiency. In contrast, under neutral and alkaline conditions (pH > 6), OH^- in the solution competes with NO₃^- and its reduction intermediates (such as NO₂^-) for the active sites on the catalyst surface, and this competitive adsorption effect significantly inhibits the progress of the entire NO₃^- reduction process.

In the photocatalytic reduction of NO₃ ^-, the recombination of photogenerated charge carriers reduces catalytic efficiency. Hole scavengers can react with photogenerated holes, thereby inhibiting charge carrier recombination and enhancing catalytic efficiency. Studies have shown that the choice of hole scavenger and its photolytic properties significantly affect the efficiency of the entire catalytic process and the distribution of products. Yi Zhigang et al.^[54]systematically investigated the effect of formic acid (HCOOH) as a hole scavenger under different light sources. Using TiO₂as the catalyst, they conducted irradiation experiments using a medium-pressure mercury lamp (main wavelength 365 nm) and a UV-LED lamp (combined wavelengths 285 + 300 + 365 nm). The results showed that under medium-pressure mercury lamp irradiation, HCOOH could not be fully photolyzed, resulting in a NO₃ ^-reduction rate and N₂selectivity in the products of 68% and 85%, respectively; whereas under UV-LED lamp irradiation, these values increased to 83% and 93%, respectively. The reason for this phenomenon is as follows: on the one hand, the wavelengths in the spectrum of the medium-pressure mercury lamp lead to photon saturation, and the excess photons inhibit the reduction of NO₃ ^-; on the other hand, the specific wavelengths emitted by the UV-LED lamp promote the photolytic reaction of the hole scavenger HCOOH in the reaction system, generating the oxidized product CO₂ ^•-. This reaction helps enhance the subsequent conversion of the intermediate product HONO into N₂, thereby increasing the selectivity of N₂in the reaction system.

The study by Mariana et al.^[55]further elucidates the regulatory role of hole scavengers on reaction pathways. They prepared ZnO nanoparticles with a particle size of 100–150 nm using a biotemplate method and investigated the effect of HCOOH on product selectivity. The experimental results show that without the addition of HCOOH, irradiation with a medium-pressure mercury lamp at 125 W and a main wavelength of 365 nm for 4 hours enables 0.03 g of ZnO to reduce approximately 91.79% of 70 mL of NO₃ ^-solution with a concentration of 100 mg·L^-1, with NO₂ ^- as the primary product. After adding HCOOH, the reduction rate of NO₃ ^- increases to 98.48%, and the primary product shifts from NO₂ ^- to N₂. This change occurs because the H⁺generated from the decomposition of HCOOH promotes further reactions of NO₂ ^-, leading to the formation of N₂ and NH₄ ⁺ (2NO₂ ^- + 6e^- + 8H⁺ → N₂ + 4H₂O, NO₂ ^- + 6e^- + 8H⁺ → NH₄ ⁺ + 2H₂O).

The structure of supported catalysts is a crucial factor influencing their performance. In the field of photocatalysis, materials with two-dimensional structures are widely used^[70-73]. Zheng et al.^[56](Table 1) prepared a three-dimensional/two-dimensional structured TiO₂/Ti₃C₂/g-C₃N₄composite catalyst using urea and Ti₃C₂MXene powder as raw materials via an intercalation method. Experimental results show that after 40 minutes of irradiation, the conversion rate of NO₃ ^-and the N₂selectivity of this supported catalyst reach 93.03% and 96.62%, respectively, and it still maintains an NO₃ ^-removal rate of approximately 92% after five cycles. Further photocatalytic results indicate that in the TiO₂/Ti₃C₂/g-C₃N₄composite material, Ti₃C₂not only serves as a template for the preparation of two-dimensional g-C₃N₄but also acts as a channel for electron transfer between TiO₂and g-C₃N₄. Photogenerated electrons in the conduction band of TiO₂migrate through Ti₃C₂to the valence band of g-C₃N₄, while the corresponding holes remain in the valence band of TiO₂and are captured by formic acid in the aqueous solution, generating reductive COO^-. Subsequently, COO^-in the solution and photogenerated electrons in the conduction band of g-C₃N₄highly selectively reduce NO₃ ^-to non-toxic and harmless N₂.

In addition to three-dimensional/two-dimensional structures, core–shell structures are also a special type of structure that can enhance photocatalytic performance. Hou et al.^[57](Table 1)encapsulated Ag/SiO₂within crystalline TiO₂, forming an Ag/SiO₂@cTiO₂core–shell structure for photocatalytic reduction of NO₃ ^-to N₂. This structure achieved a high NO₃ ^-removal rate of 95.8% and a N₂selectivity of 93.6%, while still maintaining a NO₃ ^-removal rate of 92.2% after five cycles. Due to the difference in refractive indices between the SiO₂core and the TiO₂shell, light scattering at the core–shell interface is enhanced, significantly improving the utilization of ultraviolet light in the photocatalytic process^[74-75].

Z-type heterojunction photocatalysts combine strong redox capabilities with spatially separated reduction and oxidation active sites, and Z-type photocatalysts constructed from narrow-bandgap semiconductor photocatalysts can further broaden the range of light capture^[36,76-79]. To investigate the photocatalytic NO₃ ^-reduction performance of Z-type heterojunctions, Wang et al.^[58](Table 1) synthesized polyaniline (PANI) nanowires modified with phosphomolybdic acid (PMoA) via a simple ultrasonic method, thereby constructing a Z-type heterojunction and studying its photocatalytic NO₃ ^-reduction behavior. The experimental results indicate that when the photocatalyst dosage is 0.4 g·L^-1, the removal rate of NO₃ ^-and the selectivity for N₂are 95.2% and 96.7%, respectively. The construction of the heterojunction facilitates the smooth transfer of photogenerated electrons from the conduction band of PMoA to the valence band of PANI, where they recombine with light-induced holes within PANI, effectively quenching these holes and thereby significantly extending the lifetime of photogenerated electrons in the PANI conduction band.

The built-in electric field in Z-scheme heterostructures has a significant impact on their photocatalytic performance. Shi et al.^[62](Table 1)first used FeCl₃·6H₂O, Bi(NO₃)₃·5H₂O, carboxymethyl cellulose (CMC), and NH₄VO₃as raw materials to prepare the NH₂-MIL-101(Fe)/BiVO₄composite photocatalyst via a hydrothermal method. Experimental results show that under UV light irradiation, this photocatalyst can achieve a NO₃ ^-removal rate of up to 94.8% within 50 minutes, with a selectivity for N₂of 93.4%. During the reaction, the valence band electrons of BiVO₄and NH₂-MIL-101(Fe) are first activated by UV light and migrate to their respective conduction bands, leaving behind holes. As shown in Figure 3, the non-uniform distribution of photogenerated electrons and holes gives rise to a built-in electric field that extends from the valence band of NH₂-MIL-101(Fe) to the conduction band of BiVO₄. Driven by this built-in electric field, electrons in the conduction band of BiVO₄rapidly transfer to the valence band of NH₂-MIL-101(Fe), where they recombine with holes. Ultimately, electrons and holes are left in the conduction band of NH₂-MIL-101(Fe) and the valence band of BiVO₄, respectively, participating in oxidation and reduction reactions.

Metal-organic frameworks (MOFs), as porous structural materials, offer advantages such as high specific surface area and good thermal stability, making them an ideal catalyst support material^[80-83]. Xin et al.^[59](Table 1) used a solvothermal–sol–gel method based on AgMIL-101(Cr) to prepare a Z-scheme Ag/TiO₂/AgMIL-101(Cr) photocatalyst through co-doping with Ag and TiO₂(Figure 4)), which was then applied to the reduction of NO₃ ^-. Experimental results show that after 10 minutes of UV irradiation, both the conversion rate of NO₃ ^-and the N₂selectivity can reach 100%, and after five cycles of use, the conversion rate of NO₃ ^-still remains at 90%. This is mainly attributed to two factors: (1) The Ag/TiO₂/AgMIL-101(Cr) catalyst has a large specific surface area, enhancing its adsorption of NO₃ ^-and formic acid; (2) The construction of a Z-scheme heterojunction enables photogenerated electrons in the conduction band of TiO₂to directly recombine with photogenerated holes in the valence band of MIL-101(Cr), while preserving photogenerated holes in the valence band of TiO₂and electrons in the conduction band of MIL-101(Cr) for oxidation and reduction reactions, respectively, thereby improving the overall catalytic efficiency of the reaction.

Although Ag/TiO₂/AgMIL-101(Cr) exhibits excellent photocatalytic NO₃ ^-reduction capability, most current MOF materials can only display effective photocatalytic activity under ultraviolet irradiation, limiting their application scope. To address this issue, Zhao et al.^[60](Table 1)used a solvothermal method to synthesize, in situ, a Cu-doped NH₂-MIL-125 composite catalyst, which was then used for photocatalytic reduction of NO₃ ^-under visible light irradiation. Experimental results show that the catalyst achieves an ammonia production rate of 32.8 mg·g {}_{cat}^{-1}⋅h^-1with a selectivity of 94.8%. The study found that Cu doping, on the one hand, alters the coordination environment of the MOF and reconfigures the charge distribution; on the other hand, it can serve as an active site, enhancing the catalyst's absorption of visible light. In addition, DFT computational simulations indicate that Cu doping induces orbital hybridization, forming a stable bidentate adsorption model for NO₃ ^-and reducing the energy barriers for intermediate activation and NO₃ ^-adsorption.

The initial pH of the solution also has a significant impact on the photocatalytic performance^[53,84]. Hu et al.^[85](Table 1) first prepared an Ag/TiO₂photocatalyst via chemical reduction, then used formic acid as a hole scavenger and conducted photocatalytic NO₃ ^-reduction experiments under initial solution pH conditions of 1.07, 3.11, 5.06, 7.09, 9.04, and 11.07. The results showed that the conversion rate of NO₃ ^-was highest at a pH of 3.11; as the solution pH continued to increase, the conversion rate of NO₃ ^-gradually decreased. This phenomenon can be attributed to changes in the surface charge state of the catalyst. Under acidic conditions, the TiO₂surface carries a positive charge, which facilitates the adsorption of NO₃ ^-and formate ions (HCOO^-) from water, thereby promoting the oxidation of HCOO^-and the reduction of NO₃ ^-. During the oxidation of formic acid by holes, intermediate products such as the highly reductive carbon dioxide single-electron radical anion (CO₂ ^•-, E ⁰ = -1.8 V) or the hydrogen radical (H^•, E ⁰ = -2.3 V) are generated. These intermediates can reduce NO₃ ^-to NO₂ ^-or NH₄ ⁺, resulting in superior NO₃ ^-reduction performance under acidic conditions. In particular, when the pH is around 3.11, the oxidation of HCOO^-and the reduction of NO₃ ^-proceed more readily, making it easier to form CO₂ ^•-or H^•, and thus leading to a higher conversion rate of NO₃ ^-.

Tong et al.^[86]prepared a PdSn/NiO/NaTaO₃:La composite catalyst using solid-phase reaction and precipitation-reduction methods, and investigated the effect of initial solution pH (1.5, 4.0, 7.0, and 10.0) on its photocatalytic reduction of NO₃^-. The experimental results indicate that as the solution pH increases, the conversion efficiency of NO₃^- and the selectivity for NH₃ exhibit a marked decreasing trend, while the selectivity for N₂ gradually increases; no NH₃ was detected at pH = 10. A high-pH environment exerts a dual inhibitory effect: (1) it reduces the adsorption capacity of the NO₂^- intermediate on the catalyst surface, directly affecting the synthesis rate and yield of NH₃; (2) the equimolar OH^- produced by NO₃^- reduction cannot be effectively neutralized in an alkaline environment, leading to local high-concentration accumulation that hinders the deep conversion of NO₃^-. In contrast, free H⁺ in acidic media (such as formic acid systems) can rapidly neutralize OH^-, maintaining proton supply and promoting the NO₃^- → NH₃ conversion. This is consistent with the findings of Chen et al.^[87] (poly pyrrole-modified copper nanoparticles achieve a Faraday efficiency of 96% in 0.5 M H₂SO₄), jointly validating the promoting effect of acidic environments on NO₃^- reduction.

In the photocatalytic reduction of NO₃ ^-, the recombination of photogenerated electron-hole pairs can be suppressed by adding hole scavengers, thereby enhancing photocatalytic efficiency. To investigate the specific effects of hole scavengers, Soares et al.^[88](Table 1)used PdCl₂, Cu(NO₃)₂, and TiO₂as precursors to prepare a Pd-Cu/TiO₂composite photocatalyst via the equal-volume impregnation method. They then conducted experiments on the photocatalytic reduction of NO₃ ^-in water, using formic acid, oxalic acid, ethanol, methanol, and humic acid as hole scavengers. The experimental results showed that when formic acid was used as the hole scavenger, an 84% removal rate of NO₃ ^-was achieved within 4 hours, with a N₂selectivity of 83%. However, when oxalic acid or humic acid was used as the hole scavenger, no conversion of NO₃ ^-occurred. The study found that formic acid exhibits the best catalytic performance because the noble metal can decompose it into H₂and CO₂, while simultaneously serving as both a reducing agent and a pH buffer. Danielle et al.^[89]prepared a Zn/TiO₂composite catalyst using the photo-deposition method and further investigated the effect of hole scavenger concentration on photocatalytic performance, using formic acid as the hole scavenger. Under conditions where the NO₃ ^-concentration in the solution was the same, when the CHOOH:NO₃ ^-ratio in the solution was less than 8:1 (the theoretical amount of formic acid required to reduce NO₃ ^-), both the NO₃ ^-conversion rate and N₂selectivity were relatively low. As the amount of formic acid increased, both the NO₃ ^-conversion rate and N₂selectivity rose to nearly 100% (CHOOH:NO₃ ^- ≈ 16:1). However, when the amount of CHOOH continued to increase, N₂selectivity in the product began to decline. The reason for this is that when the CHOOH:NO₃ ^-ratio exceeds 15:1, the active sites on the surface of the photocatalyst become saturated, and the ratio of surface nitrogen sources to reducing agents decreases—meaning that an excess of hole scavenger (HCOOH) occupies more active sites.

In the photocatalytic reduction of NO₃ ^-, the active centers of supported catalysts also play a crucial role. The type and distribution of active centers directly influence the catalyst’s light absorption performance, charge separation efficiency, and interaction with NO₃ ^-. Appropriate active centers can facilitate the efficient transfer of photogenerated electrons, thereby enhancing the selectivity and rate of the reduction reaction. Moreover, modification and optimization of the active centers can regulate the surface properties of the catalyst, enhancing its ability to adsorb and activate NO₃ ^-and further improving catalytic performance. Hou et al.^[63](Table 1Table 1) synthesized Pd/GdCrO₃photocatalysts with Pd loadings of 0.5 wt%, 1 wt%, and 2 wt% using solid-state combustion combined with photo-deposition, with PdCl₂and GdCrO₃as precursors. Experimental results show that under irradiation by a 500 W high-pressure mercury lamp, the 1 wt% Pd/GdCrO₃catalyst exhibits superior performance in terms of NO₃ ^-removal rate and N₂selectivity, reaching 98.7% and 100%, respectively. GdCrO₃is excited by ultraviolet light, generating electrons and holes; the photogenerated electrons migrate to the conduction band and are then captured by the supported Pd (Figure 5). Subsequently, the photogenerated electrons and holes are consumed by NO₃ ^-and formic acid, respectively. Due to the negative conduction band edge of GdCrO₃and the co-catalytic effect of Pd, the composite catalyst exhibits high photocatalytic activity. The 1 wt% Pd/GdCrO₃composite material shows the highest photocatalytic activity; however, when the Pd loading is further increased, the photocatalytic activity decreases, likely because the higher concentration of Pd nanoparticles acts as a recombination center for charge carriers, increasing the probability of electron–hole recombination and thereby reducing photocatalytic activity.

The micro-electrolysis effect of the bimetal also accelerates the photocatalytic reaction process. Gong et al.^[64](Table 1)used formic acid as a hole scavenger to investigate the application of Zn/Ag bimetal in the photocatalytic reduction of NO₃^-. The experimental results show that the micro-electrolysis effect of the Zn/Ag bimetal can rapidly chemically reduce NO₃^- to NO₂^-, while the highly active reducing species CO₂^•-, generated by the decomposition of formic acid under UV light, further promotes the conversion of NO₂^- to N₂, achieving a NO₃^- conversion rate of 99.58%.

Load-type photocatalysts not only retain the advantages of single-phase catalysts but also form structures such as core-shell architectures and Z-scheme heterojunctions, which are conducive to enhancing photocatalytic efficiency. These structures improve the utilization of light sources, promote the efficient separation and transport of photogenerated charge carriers, and enhance the catalytic effect on the reduction of NO₃ ^-. Compared with single-phase photocatalysts, load-type photocatalysts can broaden the range of light absorption, enhance stability, and improve product selectivity. However, several challenges remain, including insufficient absorption and utilization of sunlight, as well as the complex structures of some catalysts and the stringent requirements of their preparation processes. Therefore, future research should focus on developing novel supports, expanding the range of light absorption, optimizing the interface between catalysts and supports, and improving catalytic reaction conditions to advance the application and commercialization of load-type photocatalysts.