The semiconductor photocatalyst is a material that is capable of absorbing light and generating electron-hole pairs, which can then drive a variety of chemical reactions, such as splitting of water [
28], reduction of CO
2 [
29], or degradation of pollutant emissions [
30,
31]. In general, photocatalysts can be divided into two categories. The first category is semiconductor photocatalysts, which comprise metal-containing semiconductors (such as oxides, nitrides, sulfides, and oxynitrides) and other non-metal semiconductors (such as graphitic carbon nitride (g-C3N4), graphene oxide (GO) [
32], and carbon quantum dots (CQDs)) [
33]. The second category consists of plasmonic photocatalysts, which are composed of metal nanoparticles (NPs), such as copper, gold, and silver, that can exhibit strong localized surface plasmon resonance (LSPR) properties under visible light [
34]. These plasmonic photocatalysts can use the unique properties of noble metals to enhance light absorption and thereby increase photocatalytic activity.
The band structure of photocatalytic materials, including bandgap width, band position, and band bending, plays a pivotal role in determining the efficiency of photocatalysis. By effectively controlling these parameters, one can significantly improve the photogenerated charge carrier migration and thus the catalytic activity. The band structure can be tuned through various methods, while the most used methods in the degradation of microplastics are 1) doping, where intentional impurities are introduced into a semiconductor to modulate its electronic properties, and 2) the creation of heterostructures, which involves the integration of two or more different semiconductor materials to form a composite with superior properties.
2.1. Element Doping
Doped photocatalysis has emerged as a prominent technique to optimize the photocatalytic abilities of nanostructures. Doping introduces defects into the ideal crystal lattice of the native semiconductor and modifies the electronic structure of the photocatalysts in order to improve their activity [
35,
36]. This process not only helps in retarding the rapid charge recombination but also enables the absorption of visible light. This is because it enhances the interfacial charge transfer by trapping either the valence band (VB) holes or conduction band (CB) electrons within these defective sites [
37]. Besides, doping also increases the sub-bandgap irradiation, as shown in
Figure 1, which allows electronic transitions from the defect states to the CB or from the VB to the defect states [
38,
39].
The variety of dopants mainly falls into two categories: metal ions and non-metal ions. Typically, metals can facilitate electron transfer and reduce the bandgap energy level due to their inherent potential to transfer electrons, making them more favored than non-metal dopants in modifying photocatalysts [
38,
40]. When metal (cationic) dopants, either transition or noble, are introduced into the photocatalyst matrix, these dopants play a pivotal role in adjusting the semiconductor's bandgap. Doping with different metal ions possessing varied valence states can alter the material's light absorption range, vastly expanding the semiconductor's visible absorption range. For example, researchers like Tripathi et al. [
41] demonstrated that Sn and Mn doping in TiO
2 nanoparticles, prepared via the sol-gel method, led to a red shift with the bandgap energy decreasing from 3.24 eV to as low as 2.21 eV for Sn and 2.56 eV for Mn. While Yao et al. [
42] found that the synthesized Fe(III) porphyrin-conjugated TiO
2 had a strong response in the 500–800 nm range whereas bare TiO
2 have no absorption above 400 nm. Essentially, when metal ions are infused into the photocatalyst, it causes the genesis of a new energy level, which then hampers the recombination of electron/hole pairs. This phenomenon plays a pivotal role in enhancing photocatalytic activity under visible light.
In contrast to metal ions, non-metal ions, such as B, C, N, F, S, and P, generally do not act as recombination centers for charge carriers. Instead, they elevate the semiconductor material's band position, narrowing the bandgap, or introducing a new energy level to trap charge carriers, thereby improving the migration efficiency of photogenerated charge carriers. Except for boosting the visible light response, Viswanathan et al. [
43] reported that nitrogen (N) doping also modified several other properties of TiO
2, such as its hardness, refraction index, and electrical conductivity. A more specific exploration into the effects of N-doping can be seen in the work of Hwang et al. [
44]
, as shown in
Figure 2. They synthesized N-doped TiO
2 nanorods (NTR) and revealed a significant reduction in their bandgap from the typical 3.0 eV, as found in pristine rutile TiO
2, down to 1.94 eV. Detailed spectral analysis attributed this reduction to Ti
3+ and oxygen vacancy defect states. Furthermore, they discovered that long N-doped nanorods (NTR-150) samples possess a narrower bandgap than the short ones (NTR-60), which is possibly due to their richer defect structures and higher N-doping level. This observation underscores the significant influence of both N-doping and defect structures in determining the bandgap of these N-doped TiO
2 nanorods. Consequently, this results in a superior photo-oxidation activity for NTR-150 when compared to NTR-60.
This improvement in photocatalysis efficiency has also been observed during the degradation of MPs by N-doped nanomaterials. For example, Ariza-Tarazona et al. demonstrated that the C, N-TiO
2 they synthesized could degrade 72% of high-density polyethylene (HDPE) in 50 hours. However, due to the high persistence of MPs, hydrothermal pretreatment has also been used as a pretreatment method for photodegradation. For instances, Kang et al. [
45] demonstrated the possibility of using the magnetic N-doped nanocarbon springs that were capable of activating peroxymonosulfate to generate reactive radicals and decompose MPs under hydrothermal conditions. Besides, the nano-flower shaped N doped-TiO
2 catalyst (Pt@N-TiO
2-1.5%) synthesized by Zhou et al. [
46], after undergoing hydrothermal pretreatment, demonstrated an ability to degrade polyethylene terephthalate that was eight times more effective than the untreated sample.
2.2. Heterojunction Construction
A heterojunction refers to the interface between two or more distinct semiconductors having different band structures, leading to specific band alignments. Traditionally, there are three primary types of heterojunction photocatalysts: type-I (straddling gap), type-II (staggered gap), and type-III (broken gap), as illustrated in
Figure 3. In a type-I heterojunction, the conduction band (CB) and the valence band (VB) of one semiconductor are positioned above and below the respective bands of the second semiconductor. This alignment causes both electrons and holes to accumulate in the same semiconductor under light exposure, resulting in ineffective electron-hole separation and reduced redox ability.
In contrast, the type-II heterojunction exhibits an alignment where both the conduction band and valence band of one semiconductor are higher than those of the other. Upon light irradiation, this configuration promotes electron migration to one semiconductor and hole migration to the other, achieving efficient spatial separation of electron-hole pairs. However, the redox ability of this heterojunction is also reduced due to distinct redox reactions occurring on separate semiconductors. The type-III heterojunction resembles the type-II alignment but with an extreme staggered gap, causing the bandgaps not to overlap, and thus, making electron-hole migration and separation between the semiconductors impossible. Among these, the type-II heterojunction stands out as the most promising for enhancing photocatalytic activity. The appropriate overlap of energy bands ensures that the electrons and holes accumulate in different materials, making them more conducive to the spatial separation of the electron-hole pairs, thus allowing their participation in ensuing redox reactions. Over the years, there has been significant emphasis on developing type-II heterojunction photocatalysts, such as TiO2/g-C3N4, BiVO4/WO3, and g-C3N4–BiPO4, which have demonstrated commendable electron-hole separation efficiency, broad light-absorption capability, and rapid mass transfer.
The type-II heterojunction, while theoretically promising for the spatial separation of photogenerated electron-hole pairs upon illumination, carries certain intrinsic drawbacks that hamper its practical application. Thermodynamically, the efficiency of separating these photogenerated electron-hole pairs comes at the expense of reducing the redox potential of both semiconductor photocatalysts involved. For specific reactions that necessitate a certain redox potential, this reduction may hinder them. From a kinetics perspective, the presence of electrostatic interactions can inhibit interfacial charge transfer, as photogenerated charges in one catalyst suppress those in another due to mutual repulsion.
To further improve the photocatalytic efficiency, the p-n heterojunction photocatalyst emerges as one promising alternative. The internal electric field of a p-n heterojunction can serve as a driving force to promote the separation of electrons and holes. To be more specific, holes in the p-region would diffuse into the n-region until the Fermi energy levels are equated, leaving a net negative charge in the p-region. Conversely, the n-region accumulates a net positive charge, forming a built-in electric field from the n-region to the p-region. Under light exposure, photo-generated electrons and holes move in opposite directions due to this internal field, thus achieving their separation. Moreover, Qin et al. [
48] synthesized an enhanced Ag
2O/Fe-MOF p-n heterojunction photocatalyst, which excelled in converting microplastics such as PEG, PE, and PET, while also producing H
2. Notably, it facilitated the selective transformation of MPs into value-added products, presenting a promising approach to environmental remediation through microplastic upcycling and hydrogen generation.
The Z-scheme heterojunction, inspired by the natural photosynthetic charge transfer process in plants [
49], serves as another promising alternative to the limitations of type-II heterojunctions. First introduced by Bard et al. [
50] in 1979, the concept aimed to simulate natural photosynthesis and enhance the redox potential of photocatalysts. A typical Z-scheme photocatalytic system consists of two staggered semiconductor photocatalysts, PS I and PS II, connected by an acceptor/donor (A/D) pair, thereby allowing the spatial separation of electron-hole pairs. During the photocatalytic process, electrons migrate from the conduction band (CB) of PS II to the valence band (VB) of PS I through this A/D pair, ensuring optimal redox ability. This system achieves both spatial separation of the redox sites and ensures that photocatalysts maintain appropriate valence band positions, preserving a strong redox reaction ability. For example, Zhou et al. [
51] have successfully developed a Z-scheme Bi
2O
3@N-TiO
2 heterojunction utilizing a combination of solvothermal and wet-impregnation techniques, as shown in
Figure 4. This advanced heterojunction showcases its potential by degrading approximately 10.23 ± 1.91 wt% of PET fiber-based microplastic (FMP) - a predominant form of FMP pollution in the environment - under alkaline conditions (pH = 9).
Although these architectures seek to address some of the limitations of the type-II heterojunction and improve charge separation efficiency, the introduction of an A/D pair, essential for facilitating charge transfer between the two semiconductors, raises new challenges. The Z-scheme is confined to liquid-phase reactions due to the need for the A/D pair to achieve sufficient migration rates [
52]. Potential side reactions, influenced by a larger potential difference, can disrupt the charge transfer process [
53]. Furthermore, the redox pairs might exhibit color, as seen in ferrous/ferric pairs, which could interfere with light absorption by the photocatalyst [
54]. The system's functionality also hinges on maintaining specific pH conditions. While the Z-scheme offers a unique approach to enhancing electron-hole separation, it comes with its set of challenges that constrain its broader applicability in photocatalysis.
This has spurred innovations in S-scheme heterojunctions, introduced by Fu et al. [
55] in 2019 as the “step-scheme heterojunction”, which aim to combine efficient charge separation with potent photo-redox capabilities. Distinctly, the S-scheme heterojunction comprises a reducing photocatalyst (RP) with a smaller work function and a higher Fermi level, juxtaposed with an oxidizing photocatalyst (OP) possessing a larger work function and a lower Fermi level. When structured in this staggered fashion, they can effectively facilitate the separation of electron-hole pairs that boast strong oxidation-reduction capabilities.
Visually, the structure of the S-scheme heterojunction bears resemblance to a type-II heterojunction. However, there's a key distinction in their functional mechanisms. As shown in
Figure 5, in a typical type-II heterojunction, photo-induced electrons and holes tend to accumulate on the conductive and valence bands of the RP and OP, respectively, which can diminish their redox potential. In contrast, the S-scheme maintains effective electrons and holes while allowing non-essential photogenerated carriers to recombine. This step-like electron transfer mechanism is the defining characteristic of the S-scheme heterojunction, thus its name.
The charge transfer process in an S-scheme heterojunction is underlined by three pivotal factors: an intrinsic electric field, band bending, and electrostatic interactions [
56,
57]. Due to the RP's smaller work function and elevated Fermi level, when OP and RP come into contact, electrons spontaneously diffuse from RP to OP, establishing an electron depletion layer and an accumulation layer. This makes OP negatively charged and RP positively charged, creating an intrinsic electric field. This field facilitates the transfer of photogenerated electrons from OP to RP. Furthermore, when both semiconductors touch, their Fermi levels align at a common energy level, causing band bending between OP and RP, prompting the recombination of conductive electrons in OP with valence holes in RP. The electrostatic interactions at the interface of the two semiconductors also encourage different electrons and holes to recombine. This intricate interplay ensures spatial separation of the photogenerated electron-hole pairs while retaining robust oxidation-reduction capabilities.
2.4. Photocatalysts with Support Materials
Addressing the intricacies and economic challenges associated with large-scale nanoparticle recovery [
87] necessitates innovative approaches. One prevailing strategy involves the deployment of supporting materials to bolster photocatalyst efficiency without compromising recyclability. For example, Alle et al. [
88] employed open-cell β-silicon carbide alveolar foams as the support material for TiO
2 (P25) NPs, which were able to degrade 50% from the carbon of polymethylmethacrylate (PMMA) nanobeads in 7 hours under UV-A light irradiation. Properly selected supporting materials can serve the dual purpose of enhancing photocatalysis efficiency while ensuring reusability.
To further enhance the effectiveness of photocatalysts, Zhang et al. [
89] developed a novel solid-phase photodegradation of MPs and demonstrated complete mineralization of PS MPs using 30 nm TiO
2 nanoparticle photocatalysts fabricated with Triton X-100 as a nonionic surfactant. Comprehensive mechanistic studies, including mass spectrometry and
in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS), revealed that hydrocarbon ion fragments, generated during polystyrene (PS) degradation, were eventually transformed into CO
2. With a degradation efficiency of 98.4%, this method represents the highest efficiency among photocatalysts reported to date.
In a synergistic blend of environmental remediation and resource production, some photocatalysts have demonstrated dual functionalities. For instance, Meng et al. [
90] reported a biomass-derived 3D MoS
2/RGO/cotton photocatalyst capable of degrading PE and producing solar-driven freshwater. After 60 minutes of solar irradiation, a 12% weight loss of PE was observed, with large cavities and cracks on the PE surface indicating decomposition by reactive oxygen species.