Recent advancements of g-C₃N₄-based magnetic photocatalysts towards the degradation of organic pollutants: a review

A good photocatalytic material must possess low bandgap energy, the appropriate positioning of the conduction band (CB)/valance band (VB), suitable redox potentials, and enhanced visible-light activity [36]. In recent times, magnetic NPs based materials have been more effective towards the photocatalysis application. The benefits of using magnetic NPs over non-magnetic NPs are as follows: high photocatalytic efficiency [37], reproducibility [38], sustainability [39], and simple preparation procedure [40]. Other benefits of using magnetic NPs include separation of the catalysts from the post-reaction liquid using an external magnetic field [40]. The steps followed in the entire process have led to low cost, low time consumption, and better reusability property of the catalyst. Among the magnetic NPs, iron oxide nanoparticles (IONPs) have received considerable attention in the domain of photocatalysis. Few notable features of such NPs are highlighted as follows: paramagnetic/ferromagnetic in nature [41], possesses high chemical and structural stabilities, high saturation magnetization (M_s), narrow bandgap energy [42], high visible light activity, and excellent transport properties [43]. Different iron oxide phases are synthesized and employed in catalytic applications based on their characteristics and attributes. The percentage distribution of published papers based on the type of iron-oxide magnetic materials is shown in figure 1.

Zoom In Zoom Out Reset image size

Among them, Fe₃O₄, α-Fe₂O₃, β-Fe₂O₃, γ-Fe₂O₃, and MFe₂O₄ (M = Mn, Fe, Zn, Ni, Co) are very popular magnetic NPs used as photocatalysts, shown in figure 2. The α-Fe₂O₃ (hematite) and γ-Fe₂O₃ (Maghemite) are stable iron (III) oxides [44, 45], out of which α-Fe₂O₃ absorbs yellow light in the visible region, possesses high saturation magnetization value [46], and low bandgap energy [47]. Therefore, hematite-based NPs show considerable photocatalytic activity [48]. The Fe₃O₄ and γ-Fe₂O₃ have similar crystal structures; however, the Fe₃O₄ unit cell constitutes both Fe²⁺ and Fe³⁺ions [49]. Fe₃O₄ and γ-Fe₂O₃ both have high M_s values at room temperature and offer low bandgap energy [50, 51]. The value of permanent magnetization and coercivity depends on the magnetic nanomaterial's shape and size [52, 53]. The Fe₃O₄ has low resistivity because of the speedy exchange of electrons between Fe²⁺ and Fe³⁺ in the octahedral sites [54]. The generation of intermediate bandgap due to shifting electrons from VB to the CB results in enhanced visible light activity. In addition, they show a high surface-to-volume ratio, which enhances the degradation due to its particle size. The MFe₂O₄ are spinal ferrite NPs [55], where M is a divalent metal ion such as Fe²⁺, Zn²⁺, Mn²⁺, Co²⁺, Ni²⁺, and Mg²⁺ [56]. They display superparamagnetic behavior and possess a low EHP recombination rate [57]. The presence of metallic ions enhances their characteristics and attributes. It shows high colloidal stability, and the magnetic core enhances the reusability properties [58]. Even after a significant number of cycles, the photodegradation rates for the magnetic materials are quite high.

Zoom In Zoom Out Reset image size

However, the IONPs display photocatalytic activity under UV light [59], and the metal-based sulfides/ or nitrides are also not stable under an aqueous environment [60]. The fast recombination rate of EHP is the primary concern. Hence, the combination of IONPs with semiconductors judiciously paves the way towards developing photocatalysts to overcome the limitations mentioned above. The iron oxide-based semiconductor NPs must possess the following attributes.

• I.  
  The NPs must have enhanced visible-light photocatalytic activity, high stability, and high efficiency.
• II.  
  The preparation of the material should be straightforward and cost-effective.
• III.  
  The catalysts must have a low recombination rate of electrons and holes.
• IV.  
  The catalysts should be effortlessly recycled, employing a simple external magnet.
• V.  
  The efficiency must not decrease drastically even after a repeated number of cycles.

Various IONPs have been developed and used in the photocatalytic degradation of wastewater [61]. The choice of a semiconductor depends upon the positioning of VB and CB. The positive redox potential of the VB enhances the production of ·OH radicals [62], and the CB is negative enough to enhance the superoxide radical production (·O₂ ⁻) [63]. The semiconductor material absorbs the incident photons when the energy is greater than the bandgap. Consequently, the electron is excited from the VB to CB, leaving a hole in the VB that causes the formation of EHP. For effective separation of EHP, heterojunction-based semiconducting materials are realized that facilitate the transfer of electrons and holes based on the relative energy band positions [64].

The graphitic carbon nitride (g-C₃N₄) has become very promising over the years due to commercial availability, high visible-light quantum efficiency, biocompatibility, non-toxicity, robustness, and cost-effectiveness [65]. The g-C₃N₄ consists of earth-profuse elements Carbon (C) and Nitrogen (N) in the company of some content of Hydrogen (H) [66]. It is a metal-free polymeric n-type semiconductor [67] and is considered a 'golden' photocatalyst due to its variety of attractive features. The unique feature determining the photocatalytic performance of g-C₃N₄ is the tunability of the energy band-gap (lowest energy gap ∼2.7 eV), with the CB and VB potentials located at −1.09 and +1.56 eV, respectively [68–70]. Therefore, the researchers have explored g-C₃N₄ extensively as potential photocatalysts to degrade waterborne contaminants and pathogen inactivation. The photocatalytic effectiveness of bulk g-C₃N₄ is limited on account of high photo-generated EHP recombination [71], smaller surface area [72], and low electrical conductivity [73]. Several strategies have been explored, such as noble metal deposition [74], proto-nation [75], construction of heterojunction with another semiconductor [76], energy band engineering, designing of well-defined g-C₃N₄ nanostructure [28, 77], coupling with other semiconductors [78], forming hybrid materials [76], etc. All these strategies significantly improve the performance as well as unfold its properties and reliable applications. The integration of non-metal dopants such as phosphorus (P), Boron (B), and Sulphur (S) onto the surface of g-C₃N₄ induce more reactive sites with an increased SSA, reduce the energy bandgap, and enhance the visible light absorption [76, 79]. However, it has never been unlocked the door of commercialization due to poor recovery and recyclability of the non-metal doped photocatalysts [80]. Numerous separation strategies are followed, such as immobilizing the photocatalysts on different support systems, i.e. sand, zeolites, ceramics, etc. However, the SSA of the composites is reduced in most instances. The advantages and limitations of the g-C₃N₄ photocatalysts, including the charge transfer mechanism, are presented in figure 3. Thus, the development of highly active, easily recoverable and recyclable g-C₃N₄-based photocatalysts is still a major challenge. The magnetically separable photocatalysts are optimized by embedding magnetic material on the g-C₃N₄ framework. Different strategies to prepare magnetically separable g-C₃N₄-based photocatalysts are discussed in detail in the following sections.

Zoom In Zoom Out Reset image size

Various design strategies have been followed to date to synthesize magnetic photocatalysts to degrade pollutants, reduce CO₂, water splitting, selective organic transformation, etc. Transition-metal/or non-metal doping, including constructing heterojunction with g-C₃N₄ using transition metal oxides (figure 4), has become very popular nowadays. Pristine g-C₃N₄ does not possess any intrinsic spin polarization, which prompted researchers to induce magnetic moment through suitable elements doping. The spin-polarized transition metals such as V, Cr, Mn, Fe, Co, Ni, Cu, and Zn -doped g-C₃N₄ are incorporated to harness the induced spin moments [81–83] of the layered composites. It is to mention here that the transition metal atoms are more stable in the cavity of the g-C₃N₄ system than that of their bulk phase. These atoms tend to segregate and form metal clusters in the cavity. Thus, single transition metal atoms doping on the g-C₃N₄ sheets are reasonably stable and prevent the segregation on the cavity, which is in full agreement with the theoretical observations [84]. Depending on the ionic radius, the atoms with a larger radius (V, Cr, Mn, and Fe) are mostly symmetrically placed in the g-C₃N₄ cavity. Whereas, the Co, Ni, Cu and Zn (relatively smaller radius) are asymmetrically coordinated with g-C₃N₄ [84]. Interestingly, the transition metal atoms (V, Cr, and Fe-doped g-C₃N₄) interact ferromagnetically, resulting in a ferromagnetic ground state whereas Mn couples antiferromagnetically. On the other hand, the Cu and Zn doped g-C₃N₄ induce structural distortion in the sheets, and the composites become nonmagnetic [84]. But, Co and Ni-doped g-C₃N₄ does not possess well-ordered magnetic orientation [84]. However, the modified g-C₃N₄ sheets display prominent absorption at low energy, confirming their enhanced visible light absorption capacity due to reduced bandgap energy [85]. For example, Liu et al [86] have synthesized Fe-doped g-C₃N₄ via a facile and cost-effective approach and showed that the Fe atom doping significantly influences the electronic structure and catalytic properties of g-C₃N₄. As shown in figure 5, the doped Fe³⁺ acts as electron capture sites to facilitate the separation of photo-induced charge carriers [87, 88]. The captured electrons then react to reduce O₂ to ·O₂ ⁻ species. Meanwhile, Fe³⁺ can act as a hole mediator to oxidize OH⁻ or H₂O molecules to ·OH species. Moreover, Fe-doped g-C₃N₄ offers superior photo degradation efficiency for RhB under visible light irradiation. Van et al [87] have designed a new-style heat stirring method to prepare Fe-doped g-C₃N₄ nanosheets using Fe(NO₃)₃ · 9H₂O and Melamine as the starting raw materials. The introduction of single Fe atom doping on the exfoliated g-C₃N₄ nanosheets enhances the SSA (up to 132 m² g⁻¹). The large SSA not only absorbs light but also excites more EHPs. As a result, it shows complete degradation of RhB within 30 min of visible light irradiation. Many other transition metals (Co, Ni, Zn, Sc, Ti, V, Cr) have also been doped with g-C₃N₄ to improve the photocatalytic activity. Therefore, the doping of transition metal ions on the g-C₃N₄ framework can increase the SSA and the recombination of photo-generated carriers is suppressed by forming intermediate energy levels. Subsequently, the non-metal doping (S, P, B, N, O, F, etc) also shows interesting results due to the bandgap tunability of the g-C₃N₄ framework. It is evident from the theoretical study that only Boron-doped g-C₃N₄ nanosheets show ferromagnetism and half-metallicity [89]. However, the M_S value of transition metal or non-metal doped g-C₃N₄ composites is not high enough to separate it simply using an external magnet. Therefore, the recovery and recycling of doped g-C₃N₄ still remain a challenge. The relative position of the valence and conduction bands (with respect to eV versus NHE) of g-C₃N₄ and various magnetic materials [90–101] is displayed in figure 6. Considering the above-mentioned challenges, the magnetic materials, e.g. Fe₃O₄, Fe₂O₃, MFe₂O₄, and so forth, are grafted onto the surface of g-C₃N₄ to prepare magnetically separable photocatalyst. The catalysts can be easily separated from the treated suspension using an external magnetic field, which eases the recovery and reusability of the same. It is the most rational and feasible solution for separating magnetic photocatalysts from the suspension [102]. As per the web of science database (August 2021), the number of publications during 2013–2021 on the g-C₃N₄-based magnetic photocatalysts is presented in figure 7. In the following section, the magnetically separable g-C₃N₄ –based NCs are thoroughly discussed.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Hematite (Fe₂O₃), an n-type semiconductor [103], is considered as one of the most promising visible-light-driven photocatalyst owing to narrow bandgap ∼2.2 eV [104], excellent stability [105], corrosion-resistant [106], non-toxicity [107], and wider pH range [108]. Being earth-abundant elements and environmentally benign, iron-oxide's strong photo-oxidation abilities [109] have been exploited for contaminant degradation. However, the weak photo-reduction ability, shorter lifetime, and small diffusion length (in the range from 2 to 4 nm) of the photo-generated charge carriers limit their extensive application. Therefore, the Fe₂O₃ in their pure form cannot separate the photo-generated EHPs effectively [105, 106] and, hence couples with suitable semiconducting materials to form efficient photocatalytic system. In an effort, the Fe₂O₃ (strong oxidation ability) is combined with g-C₃N₄ (strong reduction ability) to construct a heterojunction system [109], where both g-C₃N₄ and Fe₂O₃ produce EHPs under visible-light irradiation. It is found that the different morphological (figure 8) features of Fe₂O₃ such as spherical nanoparticles [110–113], nanorods [114, 115], nanocubes [116], quantum dots [117, 118], porous nanotubes [119], hollow microspheres [120], rhombohedral shape [121], hexagonal plate-like [122, 123], spindle-like [124, 125], monolith-like [126], cocoon-like [127], yeast-like [128] etc. affects the charge transfer mechanism and photocatalytic activity. Therefore, various methods are followed to synthesize Fe₂O₃/g-C₃N₄ NCs such as calcinations [103], wet impregnation [129], hydrothermal [106, 107, 130], thermal treatment [131], microwave-assisted [132], NaCl assisted templating [133], etc. Using the Fe₂O₃/g-C₃N₄ composites, photocatalytic degradation of various organic pollutants and antibiotics such as Rhodamine B (RhB), Direct Red 81, Methyl Orange (MO), Cr (VI), dinitrogen, Methylene Blue (MB), Phenol, Tetracycline (TC), Ciprofloxacin, etc. have been attempted. The salient feature of the composites responsible towards superior photocatalytic performances is the formation of heterojunction interface (Type II and Z scheme) between Fe₂O₃ and g-C₃N₄ [106, 134].

Zoom In Zoom Out Reset image size

In the case of Type II heterojunction, as shown in figure 9, the photo-generated electrons can migrate from the CB of g-C₃N₄ into the CB of Fe₂O₃ during the reduction reaction [135, 136]. On the other hand, the holes migrate from VB of Fe₂O₃ into the VB of g-C₃N₄ for oxidation reaction as the CB and VB edge potentials of g-C₃N₄ are more negative than Fe₂O₃. But as the CB edge of Fe₂O₃ is more positive than the redox potential of O₂/O₂ ⁻(−0.33e V versus NHE), the electrons on CB of Fe₂O₃ cannot reduce O₂ to ·O₂ ⁻. Hence, the poor reduction ability of electrons and higher recombination rate of photogenerated EHPs in Fe₂O₃ reduce the photocatalytic activity of such a system. Whereas in the case of Z-scheme heterojunction [137, 138], as shown in figure 10, the photo-generated electrons from the CB of Fe₂O₃ migrates into the VB of g-C₃N₄ and recombines with holes. This process leaves behind photo-generated charges in the VB of Fe₂O₃ and CB of g-C₃N₄ for redox reactions [106, 134, 139]. Hence a direct Z-scheme structure, instead of Type II heterojunction, is recommended to form the g-C₃N₄/Fe₂O₃ hybrid. As a result of such systems, the rate of EHP recombination decreases, and the charge separation efficiency increases, leading to enhanced photoactivity [129]. Recently, Rajiv et al have reported that the hydrothermally synthesized Fe₂O₃ NPs decorated g-C₃N₄ display a high SSA (416.3 m² g⁻¹), and more active sites are expected to improve photocatalytic activity [110]. In contrast, Hou et al [140] have synthesized similar composite via calcination method and have reported relatively lower SSA with higher photocatalytic activity. This superior performance in activity is attributed to the highly suppressed charge carrier recombination via the Z-scheme charge transfer mechanism compared to the Type II system. Similarly, numerous Fe₂O₃ decorated g-C₃N₄ NCs have been synthesized, and their charge transfer mechanisms are thoroughly studied by varying morphologies. It can be concluded that Fe₂O₃ NPs decorated g-C₃N₄ nanocomposites offer relatively higher SSA/active sites compared to other morphologies. Therefore, the photocatalytic activity of the composites depends not only on the number of active sites but also on the types of the charge transfer mechanism. The charge transfer mechanism of the Z-scheme system exhibits superior photocatalytic performance than a typical Type II heterojunction structure [141]. Moreover, many metallic and semi-metallic elements such as Ag, Cu, Au, and carbon (Graphene derivatives) are deposited at the g-C₃N₄/Fe₂O₃ composite interface, which delivers routes for the fast charge transfer process. Therefore, many strong charge carriers are left to participate in the indirect Z scheme mechanism [142].

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Moreover, the content of Fe₂O₃ significantly affects the photocatalytic activity of Fe₂O₃/g-C₃N₄ composites. The significant enhancement in the photocatalytic activity is achieved close to the optimal value by increasing the content of Fe₂O₃. Recently, Huang et al have synthesized α-Fe₂O₃/g-C₃N₄ NCs by calcinations [103]. Under the visible-light irradiation, the NCs shows photocatalytic degradation efficiency ∼87% for RhB at about 3 h. The possibilities of agglomeration and non-uniform dispersion property of α-Fe₂O₃ in the α-Fe₂O₃/g-C₃N₄ NCs reduce the contact surface area between g-C₃N₄ and α-Fe₂O₃. As a result, the photocatalytic activity is retarded considerably. However, Xu et al have reported relatively higher degradation efficiency by increasing SSA for α-Fe₂O₃/g-C₃N₄ NCs prepared through solvothermal calcination technique (a two-step synthesis method) [105]. In another study, α-Fe₂O₃/g-C₃N₄ composite (∼0.5 wt%) is prepared by a facile hydrothermal method which displays higher photocatalytic activity for the reduction of Cr(VI) than pure g-C₃N₄ and α-Fe₂O₃[130]. The α-Fe₂O₃ NPs are dispersed on the surface of g-C₃N₄, and no aggregation is observed by Xiao et al Table 1 summarizes [143–159] several preparation methods of α-Fe₂O₃/g-C₃N₄ composites and their uses as photocatalysts.

Recently, the magnetite (Fe₃O₄) NPs have shown considerable interest on account of their superparamagnetic and high saturation magnetization [160]. The multifunctional properties observe interdisciplinary applications, such as waste-water treatment [161], drug delivery, ultrahigh density magnetic recording [162], magnetic fluid [162], sensors [160], solid-phase extraction [163], lithium storage capacity, bio-nanotechnology and protein separation [164], etc. Alongside, Fe₃O₄ NPs are considered as one of the widely utilized iron oxide compounds for several advantages such as low cost [20], large SSA, low toxicity [164], etc. Moreover, the easy separation techniques, efficient recovery, and reasonable recyclability [16, 132] make them an ideal candidate for photocatalytic application. However, without any surface modification, the Fe₃O₄ NPs have a greater tendency to deform and aggregate during catalyst recovery, which is a vital issue from the application perspective [14, 161, 164]. Therefore, the Fe₃O₄ NPs are incorporated with photoactive materials (g-C₃N₄ nanosheets) to retrieve desirable photocatalytic performance, as demonstrated in numerous studies.

The advancement in the photocatalytic performance is attributed to the energy matching band structure observed in the g-C₃N₄/Fe₃O₄ NCs, as shown in figure 11. Upon visible light illumination, the photo-excited electrons move from the CB of g-C₃N₄ to the CB of Fe₃O₄ since the CB level of Fe₃O₄ is slightly lower than that of g-C₃N₄. Furthermore, the negligible bandgap energy (∼0.1 eV) and high conductivity (1.9 × 10⁶ Sm⁻¹) [163, 164] of the Fe₃O₄ NPs decline the chance of direct recombination of photo-generated excitons and facilitates the fast carrier transportation in the Fe₃O₄/g-C₃N₄ NCs. Thus, the available electrons in the CB of Fe₃O₄ are great reductants that competently reduce the absorbed O₂ on the catalyst surface into various ROS to produce water, CO₂, etc, through the generation of highly ROS ·OH (as shown in figure 11). The production of radicals is expressed by reactions as follows: [162]

$$\begin{eqnarray*}&&{{\rm{H}}}_{{\rm{2}}}{{\rm{O}}}_{2}\to {{\rm{O}}}_{2}+{{\rm{H}}}_{{\rm{2}}}{\rm{O}},\end{eqnarray*}$$

$$\begin{eqnarray*}&&{\rm{g}}-{{\rm{C}}}_{{\rm{3}}}{{\rm{N}}}_{{\rm{4}}}/{{\rm{Fe}}}_{{\rm{3}}}{{\rm{O}}}_{{\rm{4}}}+{\rm{h}}{\rm{\nu }}\to {\rm{g}}-{{\rm{C}}}_{{\rm{3}}}{{\rm{N}}}_{4}\left({{\rm{h}}}^{+}{/{\rm{e}}}^{-}\right)/{{\rm{Fe}}}_{{\rm{3}}}{{\rm{O}}}_{4}\left({{\rm{h}}}^{+}{/{\rm{e}}}^{-}\right),\end{eqnarray*}$$

$$\begin{eqnarray*}&&{{\rm{O}}}_{2}+{{\rm{e}}}^{-}\to \cdot {{{\rm{O}}}_{2}}^{-},\end{eqnarray*}$$

$$\begin{eqnarray*}&&{{\rm{H}}}_{{\rm{2}}}{{\rm{O}}}_{2}+{{\rm{e}}}^{-}\to \cdot {\rm{OH}}+{{\rm{OH}}}^{-},\end{eqnarray*}$$

$$\begin{eqnarray*}&&{{\rm{H}}}_{{\rm{2}}}{{\rm{O}}}_{2}+{{\rm{O}}}_{2}^{-}\to \cdot {\rm{OH}}+{{\rm{OH}}}^{-}+{{\rm{O}}}_{2},\end{eqnarray*}$$

$$\begin{eqnarray*}&&{{\rm{Fe}}}_{}^{2+}+{{\rm{H}}}_{2}{{\rm{O}}}_{2}\to {{\rm{Fe}}}^{3+}+\cdot {\rm{OH}}+{\rm{OH}},\end{eqnarray*}$$

$$\begin{eqnarray*}&&{{\rm{Fe}}}^{3+}+{{\rm{H}}}_{{\rm{2}}}{{\rm{O}}}_{2}\to {{\rm{Fe}}}^{2+}+\cdot {\rm{OH}}+{{\rm{H}}}^{+},\end{eqnarray*}$$

$$\begin{eqnarray*}&&{\rm{OH}},\cdot {{\rm{O}}}_{2}^{-},{{\rm{h}}}^{+}+{\rm{organic}}\,{\rm{pollutant}}\to {{\rm{CO}}}_{2}+{{\rm{H}}}_{{\rm{2}}}{\rm{O}}.\end{eqnarray*}$$

Zoom In Zoom Out Reset image size

In a report, Kumar et al [161] have successfully fabricated g-C₃N₄/Fe₃O₄ NCs via a facile and efficient in situ growth mechanism. In contrast, Jia et al [164] have fabricated the same nanocomposites via a simple hydrothermal route. In both cases, g-C₃N₄ nanosheets are decorated with Fe₃O₄ nanoparticles with a diameter size in the range of  ∼6–8 nm to prevent the possibility of nanoparticle agglomeration. Interestingly, compared to in situ growth, a large SSA of about 200.83 m² g⁻¹ and 115.1 m² g⁻¹ of g-C₃N₄/Fe₃O₄ are observed by Jia et al and Yang et al employing the hydrothermal route and calcination-coprecipitation method, respectively [163]. The materials preparation method significantly contributes to improving the photocatalytic effectiveness for degrading RhB under visible-light irradiation [163, 164]. Liu et al have reported photocatalytic performance of hydrothermally prepared Fe₃O₄/g-C₃N₄ NCs under photo Fenton eradication of RhB using H₂O₂ oxidizing agent [162]. Further, the NCs are recovered successfully and reused even after 5–6 successive cycles without any loss of photocatalytic activity. As per the literature, different preparation techniques are followed to synthesize NCs, and various parameters are tabulated below in table 2.

The coupling of Fe₃O₄ with g-C₃N₄ can resolve several problems like the separation and recycling of magnetic catalysts. However, the commercialization aspect of g-C₃N₄ is restricted because of the increasing recombination rate of the photo-generated EHP and lower visible-light absorption ability. Accordingly, many semiconductors with suitable band energies are combined to enhance the visible-light absorption capacity as well as to prolong the lifetime of photo-induced EHPs. Consequently, the g-C₃N₄-based ternary and quaternary NCs have attracted significant attention from researchers to get the collective advantages of the combined systems. These NCs are further sub-divided into four categories based on the band energies, as shown in figure 12.

Zoom In Zoom Out Reset image size

2.3.1. Fe₃O₄/g-C₃N₄-based ternary NCs

Various narrow and wide bandgap semiconductors such as Ag₃VO₄, AgBr, AgI, Ag₃PO₄, Ag2CrO₄, BiOI, NiWO₄, MnWO_4, CoWO_4, CuWO_4, CoMoO_4, MoO_3, CdS, Bi₂S₃, TiO_2, ZnO, etc, are combined with g-C₃N₄ to form heterojunction structure and boost the photoactivity. Combining a narrow bandgap semiconductor with pure g-C₃N₄ enhances visible-light absorption. At the same time, a combination of wide bandgap semiconductors with an appropriate band edge decreases the rate of photo-generated EHP recombination. Hence, the trade-off can be addressed by selecting semiconducting materials with suitable energy band edges. The photocatalytic activity is well explained by producing and characterizing several magnetically separable ternary NCs, as displayed in figure 13.

Zoom In Zoom Out Reset image size

The coupling of noble metal NPs with semiconductor surfaces exhibits encouraging results caused by the surface plasmon resonance (SPR) effect [165]. The plasmonic effects of the noble metal NPs boost both the visible light absorption ability and broaden the spectral response range. Consequently, the Ag NPs, which are in close proximity with g-C₃N₄, favor the separation of charge carriers as the Fermi level of Ag is lower than the CB of g-C₃N₄ [166]. For example, Zhu et al have prepared a well-dispersed Ag/Fe₃O₄/g-C₃N₄ NCs photocatalyst to boost the photocatalytic activity by retaining the magnetic property. This improvement is related to the trapping of electrons by Fe₃O₄ nanoclusters and transfers to Ag, leading to superior separation efficiency of EHPs and optimal light harvesting [165]. Further, a series of Fe₃O₄/g-C₃N₄/Ag₃VO₄ and Fe₃O₄/g-C₃N₄/Ag₂CrO₄ NCs are prepared following a facile refluxing method [22, 167]. The residual magnetizations of the composites are high enough to make them separate with the help of a magnet from the treated solution in successive runs. The NCs of Fe₃O₄/g-C₃N₄/Ag₃VO₄ (60%) and Fe₃O₄/g-C₃N₄/Ag₂CrO₄ (20%) exhibit broad absorption covering the whole visible region as compared to the absorption edge (470 nm) of pure g-C₃N₄. Superior photocatalytic activities for the degradation of RhB are found, which is many folds higher than pure g-C₃N₄ and g-C₃N₄/Fe₃O₄. The enhancement of photocatalytic activity is related to the effective separation of charge carriers through a transfer of photo-generated electrons and holes, schematically shown in figure 14(a) [90, 162–172]. The energy bandgap of g-C₃N₄, Ag₃VO₄, and Ag₂CrO₄ are 2.7, 2.2, and 1.8 eV, respectively, which boosts the visible-light absorption and generates EHP. The CB and VB of g-C₃N₄ are more negative than Ag₃VO₄/Ag₂CrO₄. As a result, the photo-generated electrons quickly transfer from g-C₃N₄ to Ag₃VO₄/Ag₂CrO₄, and holes migrate in the reverse direction. Therefore, the effective separation of photo-generated charge carriers occurs, leading to a decrease in the recombination rate. Among various narrow bandgap semiconductor NCs mentioned in the below table, it is observed that g-C₃N₄/Fe₃O₄/Bi₂S₃ shows relatively higher photocatalytic performance than others [16]. This enhancement in performance can be understood by the effective charge transfer mechanism within the system. The main ROS (·O₂ ⁻ and ·OH) suggests that the composite has substantial reduction and oxidation ability which leads to complete degradation of organic pollutants within 40 min of visible light irradiation [16]. On the other hand, wideband gap semiconductor NCs, such as g-C₃N₄/Fe₃O₄/AgCl (40%), prepared by a facile method, show 7.7 times higher photocatalytic activity than pure g-C₃N₄. Moreover, the NCs hold a maximum magnetization of 9.5 emu per gram, which is indeed an excellent feature to exploit potentially advantageous magnetic separation techniques. The energy band-gap of AgCl (3.25 eV) is higher than Ag₃VO₄/Ag₂CrO₄. Upon visible-light illumination, the EHPs are generated merely on the g-C₃N₄ counterpart of the nanocomposite. The VB and CB energies of AgCl are more positive than those of g-C₃N₄. Hence, the photo-induced electrons effortlessly migrate from the CB of g-C₃N₄ to that of CB of AgCl [168]. Therefore, the recombination rate decreases due to the effective separation of electrons and holes from one to another (figure 14(b)).

Zoom In Zoom Out Reset image size

In other studies, a series of ternary composites (g-C₃N₄/Fe₃O₄/MWO₄, where M = Cu, Co, Mn, Ni) are synthesized using a facile refluxing-calcination method [90, 169–171]. The ternary photocatalyst (M = Cu, Co, Mn, Ni) shows improved photocatalytic performance about 10.5, 12.5, 7, and 12 times higher for RhB; 17, 27.5, 10, and 30 times higher for MB; 12.5, 45.7, 25, and 52 times higher for MO; and 42.5, 102, 31 and 100 times higher for Fuchsine relative to pristine g-C₃N₄. These ternary photocatalysts also possess a large BET SSA and excellent absorption ability of visible light. It is revealed that the ·O₂ ⁻ is the leading active species responsible for enhancing photodegradation reaction for ternary composites. Whereas, both ·O₂ ⁻ and ·OH are the main ROS in the case of g-C₃N₄/Fe₃O₄/ZnO NCs as reported by Raha et al [172] and Yang et al [173]. Compared to different wideband gap semiconductor NCs (table 3), g-C₃N₄/Fe₃O₄/ZnO nanocomposite shows relatively higher degradation efficiency. This enhancement is attributed to the efficient separation of charge carriers by the composite counterparts, which consequently suppress the rate of EHPs recombination. Moreover, incorporating narrow and wide bandgap semiconductors with g-C₃N₄/Fe₃O₄ reduces overall bandgap energy and increases the visible light absorbance capacity. Therefore, for ternary NCs, the main ROS ·O₂ ⁻ is responsible for the efficient separation of photogenerated charge carriers and the strong reduction ability of the CB electrons. However, the photocatalytic performances of ternary NCs are moderate due to the absence of sufficient active sites (i.e. low SSA). Additionally, the M_s value of the composites depends on the content of Fe₃O₄ and reduces significantly with increasing the counterparts. Hence, the content of Fe₃O₄ has to be optimized while preparing NCs in order to achieve better photocatalytic performances.

Table 3. Photocatalytic performances of Fe₃O₄/g-C₃N₄-based ternary and quaternary magnetic photocatalyst.

Photocatalyst	Method of preparation	Saturation magnetization (emu g−1)	Reactive species	Specific surface area (m2 g−1)	Pore diameter (nm) and pore volume (cm3 g−1)	Targeted pollutant	Light source	Degradation rate/ time(min)	Runs	References
Fe3O4/g-C3N4/AgBr	Chemical Co-precipitation	19.5	·O2 −	—	—	RhB	50 W LED	98.3%/420 min	5	[20]
Fe3O4/g-C3N4/Ag3VO4 (60%)	Refluxing	5.1	·O2 −	—	—	RhB	50 W LED	98%/240 min	5	[167]
Fe3O4/g-C3N4/AgCl (40%)	In situ growth	9.5	·O2 −	—	—	RhB	50 W LED	98.3%/360 min	5	[168]
Fe3O4/g-C3N4/Ag2CrO4 (20%)	Refluxing	12.9	·O2 −	—	—	RhB	50 W LED	95%/300 min	5	[22]
Ag(3wt%) /Fe3O4/g-C3N4	Photo-deposition	12.68	·O2 −, h+	—	—	Tetracycline (TC)	300 W Xenon lamp	88%/90 min	—	[165]
Fe3O4/g-C3N4/AgI (20%)	Refluxing	16.9	·O2 −	—	—	RhB	50 W LED	98%/270 min	5	[174]
Fe3O4/g-C3N4/BiOI (20%)	Refluxing	8.7	·O2 −, h+	27.1	—	RhB	50 W LED	∼100%/180 min	5	[175]
Fe3O4/g-C3N4/WO3	Hydrothermal	20.41	·O2 −, ·OH	87.6	—	TC	300 W Xenon lamp	89%/120 min	—	[176]
Fe3O4/g-C3N4/Bi2S3	Refluxing	8.12	·O2 −, ·OH	23.1	19.59/0.1539	RhB	50 W LED	100%/40 min	4	[16]
Fe3O4/g-C3N4/CuWO4 (30%)	Refluxing-Calcinations	21.2	·O2 −	38.5	15.63/0.152	RhB	50 W LED	100%/270 min	4	[169]
Fe3O4/g-C3N4/CoWO4 (10%)	Refluxing-Calcinations	7.04	·O2 −	—	—	RhB	50 W LED	100%/150 min	4	[170]
Fe3O4/g-C3N4/MnWO4 (10%)	Refluxing-Calcinations		·O2 −	45.04	—	RhB	50 W LED	100%/360 min		[171]
Fe3O4/g-C3N4/NiWO4 (10%)	Refluxing-Calcinations	6	·O2 −	38.50	15.44/0.149	RhB	50 W LED	100%/240 min	4	[90]
Fe3O4/g-C3N4/TiO2	Sol-gel	12.1	—	—	—	Dinitro butyl phenol (DNBP) and MB	500 W Xenon lamp	—	6	[177]
Fe3O4(10%)/g-C3N4/CdS	Heat treatment	20.15	·OH	84.9		TC, Ciprofloxacin, Gatifloxacin, Tetracycline hydrochloride, Enrofloxacin hydrochloride	300 W Xenon lamp	100%/15 min 92% 91% 90% 89%	5	[178]
petal-like Fe3O4-ZnO@g-C3N4	Hydrothermal–in situ-calcination	2.7	·O2 −, h+	18.4	—	SMX	10 W UVC lamp	95%/90 min	5	[179]
Fe3O4-C3N4-Ag2MoO4	Refluxing-Calcinations	1.44	·OH	—	—	4-Chlorophenol (4-CP)	Visible light	80%/240 min	4	[180]
Fe3O4/g-C3N4/rGO	Hydrothermal	4.7	·O2 −, h+	33.215	—	TC	Visible light	90%/120 min	5	[181]
Zn doped Fe3O4/g-C3N4	Polythermal	—	—	—	—	Cephalexin	350 W Xenon lamp	91%/300 min	6	[182]
Fe3O4/g-C3N4/TiO2	Solid combustion, hydrothermal and wetness impregnation	0.1	—	66.2	—	MB	30 W (UV-C) Mercury lamp	96%/210 min	—	[183]
Fe3O4/g-C3N4(50%)/ZnO	—	∼5	·OH, ·O2 −	—	—	MO AYR OG	500 W Xenon lamp	97.8%/150 min 98.05%/150 min 83.35%/150 min	5	[184]
Fe3O4/Pd/mpg-C3N4	Three step solution	—	·O2 −, ·OH	—	—	E. coli S. Aureus	Solar irradiation	99.9%/120 min 99.8%/120 min	4	[185]
Fe3O4/g-C3N4/CoMoO4 (30%)	Refluxing-Calcinations	20.1	·O2 −	41.69	—	—	50 W LED	—	4	[186]
Zn doped Fe3O4/g-C3N4(20%)	Polyol thermal	—	·OH	—	—	Cephalexin (CEX)	350 W Xenon lamp	91%/300 min	6	[182]
Fe3O4/GO/g-C3N4 + H2 O2	Sonicated-hydrothermal	—	·O2 −, h+	23.1	37.1/0.084	Phenol	—	∼100%/120 min	3	[187]
Ag/Fe3O4/g-C3N4	Hydrothermal	—	·OH	—	—	Diazinon (DZN)	UV light Vissible light	100%/60 min 58.4%/60 min	—	[188]
Fe3O4/g-C3N4/TiO2	Hydrothermal	—	·O2 −	65.74	19.8/0.377	RhB MO	500 W Xenon lamp	96.4%/80 min 90%/120 min	4	[189]
Fe3O4-QDs/Bi2O4/g-C3N4	Calcination and hydrothermal	0.77	·OH	78.1	15.7/—	RhB	250 W Xenon lamp	78.4/160 min	4	[190]
Fe3O4/ZnO@ g-C3N4	Facile chemical route	16.16	·O2 −, ·OH	23.117	—	Pantoprazole	23 W white LED	97.09%/90 min	4	[172]
g-C3N4/ZnO@Fe3O4	—		·O2 −, ·OH	26.53	16.32/	RhB	Sunlight	100%/180 min		[173]
Fe3O4@Bi2O3/g-C3N4	Solvothermal-ultrasonicated-stirring	4.53	·O2 −, ·OH, h+	—	393/—	Indigo carmine (IC)	400 W Hg lamp	94.9%/120 min	4	[191]
Fe3O4/g-C3N4/ZnO	Disperse-magnetic stirring	∼8	·O2 −, ·OH	—	—	RhB	250 W Xenon lamp	100%/150 min	3	[192]
Fe3O4/g-C3N4/CdS	Chemical liquid phase	>54.98	—	54.7	>20/—	Ciprofloxacin (CPFX) RhB	250 W Xenon lamp	81%/330 min 77%/330 min	3	[193]
Fe3O4/porous Titanosilicate/g-C3N4	In-situ	9	·O2 −, ·OH, h+	480	3–3.5	RhB	400 W Hg lamp and 65 W Phillips (42 W m−2) and sunlight	∼100%/90 min ∼100%/50 min ∼100%/30 min	5	[194]
Fe3O4/g-C3N4/MoO3 (30%)	Calcination	26.3	·O2 −	72.7	13.85/0.25	Tetracycline (TC)	1000 W Xenon lamp	94%/120 min	—	[195]
Fe3O4/C/g-C3N4	Sonication and in-situ precipitation	—	·O2 −, h+	125.1	2.5/0.078	TC	500 W Xenon lamp	∼100%/120 min	5	[196]
Fe3O4/g-C3N4/diatomite + H2O2	Electrostatic self-assembly	4	·O2 −, ·OH	—	203/—	RhB	500 W Xenon lamp	98%/40 min	5	[197]
Fe3O4/Graphene/S- g-C3N4	Thermal polymerization, Impregnation & co precipitation	5–10	·O2 −, ·OH, h+	57.11	—	Ranitidine by product of N-nitrosodimethylamine	300 W Xenon lamp	100%/60 min	4	[198]
Fe3O4/g-C3N4/AgI/Bi2S3 (30%)	Refluxing	7.13	·O2 −	—	—	RhB	50 W LED	99%/60 min	4	[24]
Fe3O4/g-C3N4/Ag3VO4/Ag2CrO4 (20%) +H2 O2	Refluxing	6.12	·O2 −	—	—	RhB	50 W LED	∼100%/180 min	—	[199]
Fe3O4/g-C3N4/Ag3PO4/AgCl (30%)	Ultrasonic-irradiation	8.78	h+	—	—	RhB	50 W LED	100%/150 min	4	[200]
Fe3O4/g-C3N4/AgI/Ag2CrO4 (20%)	Chemical Co-precipitation	8.5	·O2 −	—	—	RhB	50 W LED	99.4%/150 min	4	[201]
Fe3O4/g-C3N4/Ag3PO/Co3O4 (20%)	Impregnation	13.5	·O2 −	—	—	RhB	50 W LED	∼100%/150 min		[202]
Fe3O4/ g-C3N4/Ag2WO4/AgBr (30%)	Refluxing	9.61	·O2 −	——	——	RhB	50 W LED	100%/150 min	4	[203]
Fe3O4@SiO2@Bi2WO6@g-C3N4	—	—	·OH	36.56	/0.19	RhB	350 W high stress Xenon lamp	∼85%/70 min	5	[207]
Fe3O4/BiOCl/g-C3N4/Cu2O	Co-precipitation	26	·O2 −, ·OH	65	2–11/0.032	Sulfamethoxazole (SME)	800 W Xenon lamp and Sun light	99.5%/60 min and 92.1%/120 min	5	[204]
Fe3O4@SiO2@MoS2/g-C3N4	—	5.31	·OH	46.71	21.67/0.196	RhB	350 W Xenon lamp	97%–98%/60 min	6	[205]
Ultrathin g-C3N4/TiO2//Fe3O4@SiO2	Sol-gel	8	·OH	—	—	Ibuprofen real sewage	330 W Xenon lamp	97%/15 min	3	[206]
rGo/Fe3O4/g-C3N4/TiO2 +H2 O2	Co-precipitation and hydrolysis	—	—	—	—	MB	UV light	97%/20 min	10	[207]
Fe3O4@SiO2@g-C3N4/TiO2 core–shell	Hydrothermal	—	·O2 −, ·OH	—	—	RhB MO	Sunlight	∼100%/180 min ∼100%/240 min	5	[208]
g-C3N4/NiO/ZnO/Fe3O4	Hydrothermal	36.76	·O2 −, ·OH	38.78	3.101/	Esomeprazole	23 W white LED	95.05%/70 min	5	[209]
Fe3O4/g-C3N4/C/MnO2	In-situ and low temperature assembling	—	—	120.25	6.35/0.31	MO	400 W metal halide lamp	94.11%/140 min	4	[210]
Fe3O4@ns-TiO2/Ag/g-C3N4 yolk-shell	Hydrothermal and chemical reduction	15.14	·O2 −, ·OH	130	6/	Mo	500 W Xenon lamp	∼70%–80%/360 min	4	[211]

2.3.2. g-C₃N₄/Fe₃O₄-based quaternary NCs

As discussed, the ternary NCs are composed of narrow bandgap semiconductors that exhibit broad absorption in the whole visible region. In contrast, the NCs composed of wide bandgap semiconductors effectively reduce the direct recombination of photo-generated EHP. Meanwhile, to achieve both the goals, several quaternary NCs (as shown in figure 15) are synthesized and extensively explored in the degradation of dyes. For example, Choi et al [199] have prepared quaternary NCs of g-C₃N₄/Fe₃O₄/Ag₃VO₄/Ag₂CrO₄ using two narrow bandgap semiconductors. In contrast to ternary composites, the quaternary NCs promise substantial improvement in the photodegradation process due to the generation of EHPs in their individual counterparts upon visible-light illumination. Accordingly, the complementing interfaces amongst g-C₃N₄, Ag₃VO₄, and Ag₂CrO₄ in the NCs significantly improve the separation of photo-induced EHPs. Further, the combination of narrow and wide bandgap semiconductors, for example, g-C₃N₄/Fe₃O₄/AgBr/Ag₂WO₄ [203], the electrons and holes are generated only in g-C₃N₄ and AgBr, not in Ag₂WO₄ due to the large bandgap. The CB and VB energies of g-C₃N₄ are more negative than AgBr and Ag₂WO₄, which assists the electrons in moving from CB of g-C₃N₄ to CB of AgBr/Ag₂WO₄ effortlessly. Therefore, the effective separation of photo-generated EHPs leads to enhanced photocatalytic activity. The literature demonstrates that the g-C₃N₄/Fe₃O₄/AgBr/Ag₂WO₄ (30%) NCs exhibits 94.7, 17.4, and 26.7-fold enhanced activity relative to g-C₃N₄, g-C₃N₄/Fe₃O/Ag₂WO₄ (30%), and g-C₃N₄/Fe₃O/AgBr (30%), respectively for the degradation of MO [203]. The possible mechanism of EHPs separation and improvement in photoactivity of the quaternary NCs is presented in figure 16. In another study, Kumar et al have prepared Fe₃O₄/BiOCl/g-C₃N₄/Cu₂O heterojunction via the facile co-precipitation method [204]. The composites show superior photocatalytic activity for Sulfamethoxazole (SME) degradation with 99.5% (60 min) under visible light. In general, the CB edge of g-C₃N₄ (−1.09 eV) lies above Cu₂O (−0.295 eV) and BiOCl (0.62 eV). It leads to effortless movement of electrons from g-C₃N₄ to Cu₂O and BiOCl, and the holes flow in the reverse direction. During the formation of p–n–p (BiOCl-g-C₃N₄-Cu₂O) heterojunction, the energy bands of p-type and n-type semiconductors move upward and downward direction, respectively. The electrons transport is favorable from BiOCl and Cu₂O to g-C₃N₄, which strengthens an inbuilt electric field. Upon light illumination, this inbuilt electric field propels the charge carriers to flow and minimizes carrier recombination. The photocatalytic performance of various ternary and quaternary NCs is summarized in table 3 [174–198, 200, 201, 202, 205–213]. So, for quaternary NCs, the leading active species are ·O₂ ⁻ and ·OH, which promotes simultaneous photo-reduction and photo-oxidation at the reduction and oxidation sites, respectively. Therefore, the quaternary NCs show relatively higher photocatalytic performance compared to ternary NCs. This superior performance is also attributed to the efficient separation of EHPs which consequently suppresses the recombination rate.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The most popular magnetic photocatalysts including transition metal oxides are termed as spinel ferrites. These spinel ferrites are stable both thermally and chemically and formulated by way of MFe₂O₄, where M is a transition metal with a +2-oxidation number. In the context of anchoring sites of M(II) and Fe(III), three types of spinel arrangements are possible, labeled as normal ferrite, inverse ferrite, and mixed ferrite structures [214]. Recently, various spinel ferrites are thoroughly studied in different domains due to their unique internal magnetic assets, availability of numerous photo-active sites [215], narrow bandgap, and upright response to visible light. Among all ferrites, spinel Zinc ferrite (ZnFe₂O₄) nanoparticles (with reasonably 1.9 eV narrow bandgap) put forth an ever-increasing consideration for its potential application in solar photocatalysis, photo-chemical H₂ production, and CO₂ reduction [29, 216]. Besides, these materials are very convenient to synthesize following cost-effective techniques and possess substantial magnetic property and sensitivity. For example, ZnFe₂O₄ has a face-centered cubic (FCC) crystal structure with divalent Zn²⁺ ions in the tetrahedral site and Fe³⁺ ions in the octahedral site [217]. Below Neel temperature (∼10.5 K), ZnFe₂O₄ shows Anti-ferromagnetic behavior and becomes paramagnetic at room temperature [216]. The magnetic properties of ZnFe₂O₄ nanoparticles allow effective separation and reusability of the catalyst after the treatment or purification process, providing a favorable and cost-effective route for practical application [92, 218]. However, the shorter lifetime of the photoexcited carriers is the major limitation, responsible for the poor photocatalytic activity of spinel ZnFe₂O₄ [92, 215, 217]. Therefore, a heterojunction structure comprising ZnFe₂O₄ and g-C₃N₄ is materialized to boost the photocatalytic performance [215, 217, 218].

The heterojunction structures create a built-in electric field at the interface, diminishing the recombination rate by driving the photo-generated carriers immediately in the opposite direction to improve the redox process [219]. Moreover, the photoreaction mechanisms are studied through possible charge transfer route analysis as shown in figure 17.The traditional type II double charge transfer route structure is expressed in figure 17(a), where the diffusion of electrons takes place from the CB of ZnFe₂O₄ to CB of g-C₃N₄ and the diffusion of holes takes place from the VB of g-C₃N₄ to the VB of ZnFe₂O₄. In such case, the photooxidation reaction (·OH generation) is not possible as the oxidation potential of ZnFe₂O₄ is more negative than the standard redox potential of H₂O/·OH. The figure 17(b) describes the Z-Scheme charge transfer route [220–223], where the CB electrons of ZnFe₂O₄ is moved to combine with VB holes of g-C₃N₄ on the contact interface. Therefore, the reduction potential of ZnFe₂O₄ is more negative than that of O₂/·O₂ ⁻ redox potential and the oxidation potential of g-C₃N₄ is more positive than ZnFe₂O₄. Hence, the enhanced photocatalytic activity via efficient charge separation is expected via Z-scheme charge transfer mechanism [220]. Zhang et al have prepared ZnFe₂O₄/g-C₃N₄ nanocomposite through a one-step solvothermal route [223]. The structural characterization of the NCs reveals uniform dispersion of magnetic ZnFe₂O₄ lying on the g-C₃N₄ sheets. The synergistic interfacial effect, high solubility in the aqueous medium, and large specific area of the g-C₃N₄/ZnFe₂O₄ composites augment the catalytic degradation of MO. More interestingly, the catalytic performances of the ZnFe₂O₄/g-C₃N₄ composite strongly depend on the content of g-C₃N₄. Moreover, the high saturation magnetization (63 emu g⁻¹) of the g-C₃N₄/ZnFe₂O₄ (80%) composite facilitates an efficient and fast separation of catalyst from the treated suspension through an external magnet. In another study, Borthakur et al [216] have fabricated g-C₃N₄/ZnFe₂O₄ using a template-free chemical co-precipitation method. The g-C₃N₄@ZnFe₂O₄/H₂O₂ composites exhibit notable photocatalytic activity under the visible light towards discoloration of rhodamine B. The as-prepared (10%) ZnFe₂O₄@g-C₃N₄ NCs degrades 93.7% of RhB within 30 min, nine folds higher than that of ZnFe₂O₄. Interestingly, the ZnFe₂O₄/g-C₃N₄ (10%) NCs do not show any significant loss in photocatalytic activity even after five successive runs. Several studies show that the performance of heterojunction-based NCs still lacks many aspects and needs further amendments. Das et al have developed n–n heterojunction-based on polypyrrole sensitized ZnFe₂O₄/g-C₃N₄ following an in-situ polymerization method. Polypyrrole (PPY), a conducting organic polymer, possesses a large no. of π- conjugated backbones, facilitating better charge transfer in a photocatalytic process. It exhibits high electrical conductivity, exciting redox properties, fantastic stability, and is environmentally compatible [215]. The polypyrrole sensitized n–n heterojunctions display fantastic photocatalytic activity and are used as a potential candidate in removing the non-biodegradable waterborne contaminants.

Zoom In Zoom Out Reset image size

Inverse spinel ferrite such as CoFe₂O₄ shows significant recognition in the process of photocatalytic discoloration of pollutants due to their fascinating magnetic properties, low bandgap energy, high coercivity, stable crystalline structure, and low solubility. Subsequently, the high saturation magnetization eases the isolation of NPs from the solution effortlessly [224, 225]. The heterojunction interface of CoFe₂O₄ NPs anchored g-C₃N₄ enhances the photocatalytic efficiency due to favorable energy band positions and complementing material properties [226–230], as shown in figure 18. In an investigation, the CoFe₂O₄/g-C₃N₄ hybrid has been synthesized by Huang et al [231], employing a calcination method. The synthesized hybrid composites display superior photocatalytic activity in the presence of H₂O₂ oxidant upon visible-light illumination. The traits of CoFe₂O₄/g-C₃N₄ at the heterojunction interfaces are harnessed to boost EHPs generation and separation. In that framework, Yao et al [93] have synthesized CoFe₂O₄/g-C₃N₄ NCs via a facile self-assembly method with mass ratio 2:1, which completely degrades RhB within 210 min under artificial light irradiation. The enhanced photocatalytic performance is attributed to the Z-scheme carrier transportation amongst the interfaces of g-C₃N₄ and CoFe₂O₄ [226]. The strong magnetic properties of the NCs help isolate the catalysts from the reaction medium using an external magnet. A plausible photocatalytic mechanism is proposed to figure out the generation process of EHPs in the NCs under visible light. The CB/VB energies of g-C₃N₄ and CoFe₂O₄ are −1.09/1.56 eV and +0.42/1.75 eV, respectively, with respect to NHE. The photo-induced electrons/holes in the CB/VB of g-C₃N₄ are transported towards the respective bands of CoFe₂O₄, facilitating the charge separation and diminishing the recombination rate. In the same framework, honey mediated sol-gel auto combustion synthesis method is reported by Inbaraj et al [232] The as-prepared NCs reveal broad absorption in the visible region and display excellent degradation efficacy towards MB of about 77% within 30 min Besides, the NCs also remove a significant amount of heavy metal such as lead from water in a short time. In another attempt to narrow down the bandgap energy (1.85 eV), Kamal et al [233] have incorporated CoFe₂O₄ onto sulfur-doped g-C₃N₄ via the ultrasonication method. The aggregated CoFe₂O₄ NPs on a sulfur-doped g-C₃N₄ sheet effectively improve the catalytic activity. In the presence of H₂O₂, the NCs demonstrate an excellent degradation efficiency of 96% within a reaction time of 20 min [233].

Zoom In Zoom Out Reset image size

Very recently, Cu₂V₂O₇/CoFe₂O₄/g-C₃N₄ ternary NCs have been synthesized via a cost-effective hydrothermal approach. The nanocomposite displays better catalytic activity towards the eradication of 4-NP to 4-AP using NaBH₄ upon visible-light illumination, and the process completes in just 60 s at room temperature. Moreover, the catalyst shows outstanding stability and reusability up to ten successive runs without any substantial loss in the activity [225]. Another magnetic semiconductor is the inverse spinel Nickel ferrite (NiFe₂O₄) [234, 235], Calcium ferrite (CaFe₂O₄), and Cadmium ferrite (CdFe₂O₄) [101] with an optical bandgap of ∼2 eV and appropriate positioning of CB and VB energies corresponding to redox potential. Few notable features of these ferrites (M = Ni, Ca, Cd, etc) include broad visible-light absorption, high electrical resistivity, and chemical stability, making them perfect choices. For inverse spinel structure, the M²⁺ and Fe³⁺ ions reside equally in numbers at the octahedral sites and balance through Fe³⁺ ions located at the tetrahedral sites, bearing antiparallel spins lead to ferrimagnetism. The Fe²⁺ ions are responsible for the system's n-type behavior, and the conductivity arises due to the jumping of electrons from Fe²⁺ to Fe³⁺. The presence of Fe³⁺ ions and transferring of holes from the M³⁺ site to the Fe³⁺ site is responsible for the p-type behavior of the system [236, 237]. However, the quick recombination of photo-generated EHPs makes MFe₂O₄ a poor photocatalyst. Therefore, the researchers have followed different compelling methodologies, including combining with inorganic metal oxide semiconductors. The MFe₂O₄/g-C₃N₄ composites have well-matched energy levels, which helps to abet the EHP recombination. Gebreslassie et al [238, 239] have fabricated NiFe₂O₄/g-C₃N₄ nanocomposite with a modification of direct Z-scheme charge transport mechanism via a facile one‐pot hydrothermal process. The same has been evaluated for the eradication of MO upon visible-light illumination and found to be 4.4 and 3 times higher than individual NiFe₂O₄ and g-C₃N₄, respectively. The proposed degradation mechanism [237, 240–243] of MFe₂O₄ (M = Ni, Ca, Cd)/g-C₃N₄ NCs is shown in figure 19. Researchers have reported excellent photocatalytic activity of NiFe₂O₄/Boron [244] and NiFe₂O₄/Phosphorus [76] doped g-C₃N₄ prepared via the sol-gel, and Fe₃O₄@NiFe₂O₄/Phosphorus doped g-C₃N₄ via co-precipitation method. The results reveal that all these are attributable to the increased SSA, improved charge transfer rate, and bandgap narrowing via non-metal doping.

Zoom In Zoom Out Reset image size

Copper ferrite (CuFe₂O₄), a p-type magnetic semiconductor, has a moderate bandgap of 1.4 eV and offers good chemical or thermal stability. Considering this, CuFe₂O₄ is employed in various catalytic applications. The matching electronic band structures of CuFe₂O₄ and g-C₃N₄ conveniently enhance the process of photo-generated charge separation, and the magnetic property helps the segregation of photocatalysts from the suspension. For example, Yao et al have reported CuFe₂O₄/g-C₃N₄ core–shell photocatalyst, which exhibits an enhanced photocatalytic activity compared to pristine CuFe₂O₄ or g-C₃N₄. The combination of g-C₃N₄ and CuFe₂O₄ under visible-light illumination displays an excellent response towards the degradation of Orange II dye in the presence of an H₂O₂ oxidizing agent [19, 245].

Compared to other ferrites (NiFe₂O₄, CoFe₂O₄, and CuFe₂O₄) [246], the manganese ferrite (MnFe₂O₄) displays moderately high adsorption, high mechanical hardness, biocompatibility, high magnetic susceptibility, and photochemical stability. But, pure MnFe₂O₄ is photo catalytically inactive because of the lower VB edge potentials and unsatisfactory performance towards photoelectric conversion [247]. Hence, the magnetic NCs of MnFe₂O₄/g-C₃N₄/TiO₂ is successfully synthesized using the chemical impregnation approach by Vignesh et al and effectively utilizes it as a solar light-driven photocatalyst for the degradation of MO. In another attempt, the magnetic g-C₃N₄@MnFe₂O₄-Graphene photocatalyst is synthesized through a facile combination of solvothermal and impregnation synthesis approach [248]. The composite undergoes a photo-Fenton-like reaction and exhibits superlative catalytic performance to eradicate many antibiotics upon visible-light illumination. The synergistic interfacial effects, high visible-light response, and large SSA appear to be responsible for the fast and effortless separation and transportation of photo-excited EHPs. From this perspective, we can conclude that the degree of magnetization plays a vital role in the isolation of catalysts without any significant loss even after successive runs. Table 4 provides information about various combinations of ferrites, and g-C₃N₄ NCs, which were designed and executed for the purification of aqueous solutions consisting of numerous organic–inorganic contaminants [96, 249–266]. These magnetic heterojunctions display exciting properties, such as high SSA, enough magnetism, convenient particle size, effortless recovery and outstanding photocatalytic performance in various AOPs. Therefore, from the energy band diagram and above discussion, it can be inferred that most of the g-C₃N₄/ferrites (MFe₂O₄) composites (type-II or Z-scheme) such as (M = Zn, Cu, Mn, and Mg) shows stronger reduction ability rather than oxidation. The type II and Z-scheme mechanism of the g-C₃N₄/ferrites are schematically shown in figures 20 and 21, respectively [96, 267]. On the other hand, ferrites with M = Ni, Co, Ca, and Cd offers strong oxidation ability to the g-C₃N₄/ferrite composite for simultaneous photoreduction and oxidation reactions on the active sites. Besides, the self-redox properties of M and Fe atoms in MFe₂O₄ induced H₂O₂ to be facile for the generation of ·OH. The redox reactions are as shown in figure 22. The features mentioned above inspire the designing of g-C₃N₄-based ferrite NCs. Based on the literature review, a comparison of photocatalytic performances among the different magnetic NCs is made towards the degradation of RhB dye under 300 W Xenon lamp irradiation (figure 23). Among the binary composites, NiFe₂O₄/P-doped g-C₃N₄ and α-Fe₂O₃/few-layer g-C₃N₄ (Z-Scheme) show excellent photocatalytic performances. Hence, it is suggested that the combining magnetic materials with exfoliated g-C₃N₄ or doped g-C₃N₄ is preferable to constructing new materials with superior performances.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The First‐principles calculation using density functional theory (DFT) is a fundamental and powerful technique to determine the material's intrinsic properties by optimizing the geometric structure and calculating the system energy [268]. Further, the geometric optimization is performed using the ultra-soft projector augmented wave (PAW) pseudo-potential method, and the GGA (generalized gradient approximation) depicts the functional exchange-correlation interaction [269, 270]. The computational study demonstrates that heteroatoms [271] doping modifies the chemical and electronic properties of the g-C₃N₄ sheets and improves photocatalytic activity significantly. In the theoretical models [272], the nitrogen linker is primarily replaced by various heteroatoms or aromatic groups to obtain the 2D or 3D multilayered structure [273]. Many transition metal ions, including noble metals, are incorporated into the caves of the g-C₃N₄ [274] to improve carrier mobility by narrowing the energy band gap and increasing the visible light absorption ability. Similar types of cave doping using the alkali-metal ions [272, 275] cause nonuniform charge distribution on the sheets favorable for free carrier generation, transportation and separation. Hence, understanding the effects of element doping and interaction between different components or layers, including the photocatalytic reaction process on catalysts' surfaces, are crucial for designing new materials. To date, the major theoretical studies are concentrated on the electronic and magnetic behavior of the g-C₃N₄ catalysts.

The precise architecture of g-C₃N₄ has two accepted crystal structures: tri‐s‐triazine (C₆N₇) and s‐triazine (C₃N₃) rings [276], as shown in figure 24. Tri-s-triazine- based g-C₃N₄ is found to be a more stable photocatalyst at ambient conditions [277, 278]. For non-metal doped tri‐s‐triazine‐based g-C₃N₄ systems, Stolbov et al and Ma et al have revealed that B and S atoms preferred to substitute the C and two‐coordinated N atoms, respectively (shown in figure 25) [279, 280]. In another study by Yousefi et al, a similar verdict is observed for the s-triazine-based system [281]. The monolayer of g-C₃N₄ exhibits a larger bandgap than bulk g-C₃N₄ because of the quantum confinement effect. The bandgap of bulk g-C₃N₄ is indirect, whereas monolayer g-C₃N₄ is direct. The work function (Φ) of g-C₃N₄ is around 4.50 eV, which weakly correlates with the number of layers [276]. The energy bandgap (E_g) of g-C₃N₄ is experimentally estimated to be 2.7 eV; however, the DFT calculation estimates higher values depending on the constructed model. A model for g-C₃N₄ is proposed with a supercell consisting of 27 Carbon atoms and 36 nitrogen atoms. A vacuum spacing of 15 Å is considered to avoid the interaction of the complex system with periodic counterparts [282]. The DFT study demonstrates that the oxidation and reduction reactions of different species take place at the sites, overlapping of atomic orbitals (P_z) of nitrogen and Carbon [283], creating the VB (+1.6 eV) and CB (−1.1 eV) of g-C₃N₄. Thus, the C and N atoms act as sites where reduction and oxidation of O₂ and H₂O occur, respectively [278, 284]. Till now, comprehensive knowledge of DFT calculation methods and results of g-C₃N₄ atomic arrangements has expounded, such as pristine g-C₃N₄, nonmetal doped g-C₃N₄, metal decorated g-C₃N₄, and g-C₃N₄ based composites [276]. Recently, many alkali metals decorated g-C₃N₄ and transition metals decorated g-C₃N₄ systems have been investigated by DFT calculation. For example, Zhang et al have studied Ti or Sc decorated monolayer g-C₃N₄ system and introduced it as a promising candidate for hydrogen storage due to its excellent hydrogen adsorption capacity [285, 286]. Zhang et al have found that the decoration of transition metal atoms (Sc, Ti, V, Cr, Mn, Fe, Co, and Ni) [285, 286] resulted in various band edge potentials of bulk g-C₃N₄. It can be inferred that the photogenerated electrons of Sc‐, Ti‐, and V‐decorated g-C₃N₄ systems had strong reduction ability. The band edge potentials of Cr‐, Mn‐, and Fe‐decorated g-C₃N₄ systems met the requirement of overall water splitting [275, 287]. Figure 26 shows different symmetric and asymmetric cave doping of transition metals. Recently, computational investigation of g-C3N4-based metal oxides gained enormous attention towards understanding the interaction at the interface by calculating the charge density difference in the composite or comparing the work functions of different components.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size