TiO₂-Based Catalysts with Various Structures for Photocatalytic Application: A Review

Abstract

TiO₂-based catalysts with various surface heterostructures (0D, 1D, 2D, and 3D) have been widely researched owing to their cost-effectiveness, high stability, and environmentally friendly nature, and can be used for many applications in various fields, including hydrogen production and pollutant degradation. However, there are also many existing problems limiting their practical application, such as their large band gap and rapid electron–hole recombination rate. Owing to the abundance of recent achievements in materials science, we will summarize the recent structural engineering strategies which provide favorable photocatalytic activity enhancements, such as enhanced visible light absorption, stability, an increased charge–carrier separation rate and improved specific surface area. Among the various structural engineering methods in this review, we will introduce TiO₂-based materials with different dimensional structures. Meanwhile, we also discuss recent achievements in synthesis methods and application of TiO₂-based catalysts in various fields. We aim to display a comprehensive overview which can be a guide for the development of a new generation of TiO₂-based catalysts according to their structural design for enhanced solar energy conversion.

1. Introduction

Titanium dioxide (TiO₂), a common commercial catalyst, has been regarded as a promising material in various applications, e.g., self-cleaning glass, the paper industry, cosmetics, refractory medicine, the food industry, electronic devices, and catalysts. Over the last forty years, TiO₂-structured materials have attracted much attention, owing to their advanced properties in the degradation of pollutants, water splitting, and hydrogen generation [1,2,3,4,5,6,7]. Up to now, various nanostructured TiO₂ materials, such as nanoparticles (0D), nanowires (1D), nanosheets (2D), and foam-like materials (3D), have been synthesized [8,9,10,11,12]. Compared with TiO₂ nanoparticles (0D), 1D, 2D, and 3D TiO₂-based catalysts provide a large specific surface area and a narrowed band gap, resulting in more active sites and displaying enhanced photocatalytic activity. Although there have been various applications of TiO₂ in many fields, there are still some disadvantages: the long distance between interfaces, causing slow charge carrier movement; the wide band gap, leading to a slower visible light response; and the fast recombination of photo-reduced electron–hole pairs [13,14,15]. Meanwhile, used TiO₂ nanoparticles are difficult to recycle and reuse in water, causing secondary pollution. For example, for the commercial catalyst P25, a two-phase type of TiO₂ (80 wt% anatase and 20 wt% rutile), the cost of collecting it from suspension water after a reaction is much higher than the price of the P25. Many efforts have been devoted to overcoming these disadvantages of TiO₂ nanoparticles, including doping them with other atoms, involving oxygen vacancies, surface modification, and band gap modification. Nevertheless, the above strategies cannot totally overcome the problems.

Most of the above-mentioned methods involve the sacrifice of UV adsorption performance when the response of visible light is enhanced. In addition, the modified TiO₂ nanoparticles are usually not stable. Meanwhile, some novel methods have also been developed, such as those involving the Schottky junction and the introduction of the surface plasmon resonance (SPR) effect [16,17,18], which will extend the light absorption range by decorating the surface with metal particles. It is sure that these methods have many advantages, such as improved light adsorption, a high charge carrier separation rate, and the ability to inhibit charge carrier recombination, which will significantly enhance the photocatalytic activity of TiO₂. However, due to the complex synthetic process, their practical application is limited. As such, it is difficult to control the distribution status of metal particles on the surface of TiO₂, resulting in an unexpected structure that reduces light absorption and the diffusion of charge carriers on the TiO₂ surface [19,20,21,22,23,24,25].

As for the various polymorphs of TiO₂, anatase is often considered the most effective polymorph for photocatalytic applications due to its higher conduction band edge, which provides greater reduction potential for electron transfer reactions, and its lower rate of electron–hole recombination compared to rutile, enhancing the lifetime of charge carriers and improving photocatalytic efficiency. Additionally, anatase typically forms smaller particles with a larger surface area, increasing the number of active sites available for photocatalytic reactions. The highly reactive {001} facets of anatase, which can be synthesized to expose a high percentage of these facets, further boost its photocatalytic performance. Rutile, another important polymorph of TiO₂, is generally considered less effective due to its lower conduction band edge and higher rates of electron–hole recombination, though it is more thermodynamically stable at larger particle sizes. Brookite, which is less commonly used due to its complex synthesis and less-studied properties, has a similar band gap to anatase and unique surface properties that may offer specific advantages in certain reactions. Overall, anatase is typically preferred for photocatalytic applications due to its superior electronic structure, lower recombination rates, and higher surface area, but combining anatase with rutile can sometimes yield synergistic effects by facilitating charge separation and improving efficiency. The choice of polymorph should be guided by the specific requirements of the intended photocatalytic application, with potential benefits from the inclusion of rutile or brookite in certain contexts.

Several parameters of TiO₂ structures significantly influence their photocatalytic performance, including the particle size, crystallinity, the surface-to-volume ratio, and surface chemistry. Optimizing the size of TiO₂ nanoparticles is crucial for maximizing photocatalytic activity due to their specific size-related properties. When TiO₂ is scaled down to the nanometer range, its energy band structure becomes discrete, leading to distinct photophysical, photochemical, and surface characteristics compared to those of its bulk counterpart, primarily due to the quantum size effect. Additionally, smaller nanoparticles help reduce electron–hole recombination because of their shorter charge transfer distances, which is essential, as the diffusion length of charge carriers must exceed the particle size to ensure efficient charge–carrier dynamics. Other studies have shown that different phases of TiO₂ exhibit size-dependent stability; the rutile phase is stable in nanoparticles larger than 35 nm, while the anatase phase is stable in nanoparticles larger than 11 nm [26,27,28]. However, when nanoparticles are smaller than several nanometers, photocatalytic activity diminishes due to dominant electron–hole recombination on the surfaces. Improving the crystallinity of TiO₂ can further enhance the photocatalytic performance by facilitating the transfer of photogenerated charges, which is often achieved through calcination at suitable temperatures to reduce the defect density. A higher surface-to-volume ratio in nanostructured materials increases the number of active sites available for catalysis, thereby enhancing reactant adsorption and product desorption. Compared to the standard nanostructured TiO₂, such as commercial P25 nanoparticles, modified morphologies, like nanotubes, nanofibers, nanorods, foams, and mesoporous structures, offer larger surface areas. Notably, TiO₂ nanotubes have demonstrated nearly double the photoconversion efficiency of spherical nanoparticles. However, an increased surface area with many dangling bonds may introduce recombination centers for excited carriers, potentially reducing photocatalytic activity [29,30,31]. Two-dimensional nanoplates and nanosheet structures have been investigated for improved photocatalytic activity by maximizing the active facets of TiO₂. Both theoretical and experimental studies have shown that the {001} facet of anatase TiO₂ is significantly more reactive than the thermodynamically stable {101} facet. Researchers have developed methods to synthesize high-purity anatase TiO₂ single crystals with a high percentage of {001} facets using hydrofluoric acid as a morphology-controlling agent, achieving up to 90% exposure of these reactive facets. Surface area optimization is also crucial in maximizing the reaction sites available for photocatalysis. For example, ultrathin TiO₂ nanosheets with highly active {001} facets and a high specific surface area have shown a twofold increase in photodegradation efficiency compared to that of P25 [32,33,34,35]. Additionally, three-dimensional TiO₂ structures, such as hollow structures, interconnected structures, and hierarchical superstructures, have shown promise as efficient photocatalysts. These structures benefit from a high surface-area-to-volume ratio due to void formation among the nano building blocks, along with unique optical and carrier transfer properties. The core–shell and hierarchical structuring of TiO₂-based nanostructures represents a significant strategy for overcoming the limitations of pure TiO₂ in photocatalysis. The focus on structural design and functionality aims to develop next-generation, highly efficient, TiO₂-based photocatalysts.

2. Photocatalytic Mechanism of the Catalysts

The practical efficiency of a photocatalyst is mainly determined by the effective separation of charge carriers, followed by efficient charge transfer. When a semiconductor absorbs photon energy, excitons—pairs of electrons and holes—are generated. Specifically, electrons in the valence band (VB) become excited and move to the conduction band (CB), leaving holes in the VB. These photogenerated electrons and holes then migrate to the surface of the semiconductor (Figure 1a). The photocatalytic reaction begins when a CO₂ molecule accepts an electron from the semiconductor’s CB. However, the efficiency of the photocatalyst can be compromised by charge recombination, a process in which an electron and a hole recombine on the surface, reducing the number of charge carriers available for the reaction [36,37,38]. The reduction of CO₂ is an energetically uphill reaction, meaning it requires significant energy input. For this process to occur efficiently, the positions of the CB and VB of the photocatalyst must overlap with the reduction potential of CO₂ and the oxidation potential of H₂O. CO₂ has a symmetrical structure and strong bond energy, making its photoreduction thermodynamically unfavorable and requiring a high negative reduction potential (−1.90 V vs. NHE at pH 7.00). However, when CO₂ reduction occurs in the presence of water, it becomes proton-driven, which can create more favorable conditions for the reduction reaction. In such cases, the reduction potential required for protons is less negative than that required for CO₂, facilitating the reaction under less extreme conditions [39,40,41]. Nevertheless, the hydrogen evolution reaction (HER) and CO₂ reduction reaction can compete with each other, as they both involve similar reaction mechanisms. The electrochemical potentials for the reactions involved in the photoreduction of CO₂ and the associated H₂O redox reactions are summarized in Figure 1b. These potentials indicate the formal electrochemical requirements for efficient CO₂ reduction and help in understanding the competitive nature of HER and CO₂ reduction in photocatalytic systems [42,43,44].

As for the various polymorphs, anatase is often considered the most effective polymorph for photocatalytic applications due to its higher conduction band edge, which provides greater reduction potential for electron transfer reactions, and its lower rates of electron–hole recombination compared to rutile, enhancing the lifetime of charge carriers and improving photocatalytic efficiency. Additionally, anatase typically forms smaller particles with a larger surface area, increasing the number of active sites available for photocatalytic reactions. The highly reactive {001} facets of anatase, which can be synthesized to expose a high percentage of these facets, further boost its photocatalytic performance. Rutile, another important polymorph of TiO₂, is generally considered less effective due to its lower conduction band edge and higher rates of electron–hole recombination, though it is more thermodynamically stable at larger particle sizes. Brookite, less commonly used due to its complex synthesis and less-studied properties, has a similar band gap to anatase and unique surface properties that may offer specific advantages in certain reactions. Overall, anatase is typically preferred for photocatalytic applications due to its superior electronic structure, lower recombination rates, and higher surface area, but combining anatase with rutile can sometimes yield synergistic effects by facilitating charge separation and improving efficiency. The choice of polymorph should be guided by the specific requirements of the intended photocatalytic application, with potential benefits from the inclusion of rutile or brookite in certain contexts.

Traditional heterojunctions, including type–I, type–II, and type–III, are categorized according to energy band position relationships, as shown in Figure 2, in which the energy band structures of photocatalyst I (PC I) and photocatalyst II (PC II) are embedded. Under visible light irradiation, photogenerated electrons and holes in the CB and VB of PC I are transferred to the CB and VB of PC II, respectively. Type–I heterojunctions cannot effectively improve photocatalytic performance owing to the rapid recombination of carriers accumulated in the semiconductor. The energy band structure of a type–III heterojunction is completely staggered, and this heterojunction has rarely been reported. Therefore, only type–II heterojunctions are described in this review. Type–II comprises two semiconductors with relatively interleaved band positions. The CB of PC I is more negative than that of PC II, whereas the VB of PC II is more positive than that of PC I [46]. Two different carrier transfer pathways may occur depending on the difference in EF values. The first pathway occurs in the traditional type-II heterojunction when PC I has a high CB but low EF. When the two semiconductors are in contact, their electrons diffuse, thereby forming an IEF with a direction from PC II to PC I. Next, the energy band of PC I bends downward, whereas that of PC II bends in the opposite direction, and EF is in equilibrium. Upon illumination, the photo-generated electrons are transferred from the CB of PC I to that of PC II, and the holes are transferred from the VB of PC II to that of PC I. Thus, the photo-generated electron–hole pairs are effectively separated, thereby improving the photocatalytic activity. The second pathway is a direct Z–scheme when PC I has a high CB and EF. In this case, we obtain the opposite IEF to that of the previous case. The photogenerated electrons are transferred from the CB of PC II to the VB of PC I after illumination. The holes in the VB of PC I after illumination transfer to the CB of PC II [47]. Meanwhile, the electrons in the CB of PC I and the holes in the VB of PC II do not participate in the carrier complex; however, they participate in the redox reaction, thereby forming spatial separate oxidation and reduction active sites, respectively. The traditional type-II heterojunction has good charge separation efficiency; however, its redox capability is lower than that of the direct Z–scheme. For traditional type-II heterojunction, the photo-generated electrons accumulate in the CB of PC II with a weak reduction potential, whereas the photo-generated holes accumulate in the VB of PC I with a weak oxidation potential [48]. The direct Z–scheme benefits from the superior CB and VB for redox reactions and possesses high carrier separation efficiency and strong redox ability. Type–II heterojunctions are the most widely used because of their simplicity of construction.

The traditional Z–scheme is also known as the liquid phase heterojunction. As shown in Figure 3a, the electrons of the CB of PC II, with a low energy band position, are captured by A, which is subsequently reduced to D. Similarly, the holes of the VB of PC I are captured by D, which is oxidized back to A, thereby completing the cycle. However, because the electron potential difference of PC I is larger than that of PC II, the electrons in PC I are more likely to participate in the reduction of A. Simultaneously, the holes in PC II are more likely to be involved in the oxidation of D. The traditional Z–scheme has limited prospects and was phased out because of its complicated structure, high uncertainty, and narrow range of application. As depicted in Figure 3b, the all-solid-state Z–scheme achieves the complexation of electrons in the CB of PC II and holes in the VB of PC I using electron mediators. The strongly reducing electrons and strongly oxidizing holes remaining in the CB and VB of PC II participate in the reduction and oxidation reactions, respectively. As shown in Figure 3c, a semiconductor with a high energy band position also has a high EF, which is necessary for constructing a direct Z–scheme. This results in the retention of useful redox-competent photogenerated electrons and holes in the CB of PC I and VB of PC II, whereas other photogenerated carriers are involved in carrier complexation. Therefore, the direct Z–scheme solves the problem of carrier complexation while retaining a strong redox ability. However, the solid–solid interfacial resistance of the direct Z–scheme is usually higher than that of the second–generation all-solid Z–scheme, owing to the direct contact between the two semiconductors, which is unsatisfactory. This can be improved by optimizing the solid–solid interface contact to a tight contact for increasing the conductivity [50]. The S–scheme is a good alternative to the direct Z–scheme to avoid the failures caused by the ancestors of the Z–scheme family [51]. The carrier motion mechanism of the S-scheme, shown in Figure 3d, is the same as that of the direct Z–scheme. Therefore, the novelty of the S–scheme heterojunction is uncertain when considering the carrier motion mechanism. The RPs possess higher band positions than those of the OPs. The S–scheme heterojunctions consist of one RP and one OP, where the RP possesses a higher EF than that of the OP.

3. Structures of TiO₂

Many studies have been devoted to modifying the structure of TiO₂ in terms of dimensions for the enhancement of photocatalytic activity. Figure 4 displays structured TiO₂ possessing various dimensions and morphologies. These TiO₂–based materials with various structures will not only enhance their photocatalytic performance, but also form a platform to decorate with other materials, such as metal particles, metal oxides, and semiconductors, to improve the photocatalytic performance [54,55,56,57]. Some parameters of TiO₂ structures will affect the photocatalytic performance, including size, surface area, and surface chemical state [58,59,60]. TiO₂ particles with optimized size will improve the photocatalytic activities owing to their size-related properties. Meanwhile, the band structure of nanoscale TiO₂ is quite different from that of bulk TiO₂, owing to the quantum effect [61]. Furthermore, nanosized catalysts will also reduce the electron–hole recombination rate owing to the shorter carrier transfer distance. Up to now, the stability of variously structured TiO₂ has been researched. Rutile TiO₂ is quite stable with a size larger than 35 nm, while anatase TiO₂ is more stable around 11 nm. Moreover, high crystallinity is another crucial factor in facilitating the photoinduced charge carrier transfer. Thus, calcination at a suitable temperature is always necessary to improve crystallinity. Furthermore, a high specific surface area of TiO₂ materials will help to improve photocatalytic efficiency by providing more surface sites that facilitate reactions, thus improving reactant adsorption and desorption during the reaction [62,63,64]. Based on previous reports, it has been proved that the photocatalytic performance of TiO₂ nanotube is about twice that of pristine TiO₂. In addition, 2D nanosheet structures have been explored to enhance photocatalytic activity. Theoretical and experimental research has proven that the {001} facet is much more active than the {101} facet. Yang et al. synthesized an anatase TiO₂ single crystal material with 47% {001} facet using hydrofluoric acid as a controlling agent. Subsequently, sheet-like TiO₂ materials with 90% exposed {001} facet have been developed. Meanwhile, the specific surface area is another crucial factor in improving active sites for the photocatalysis reaction. Hu et al. reported a kind of {001} facet-rich TiO₂ nanosheet with a specific surface area of 162.4 m²·g⁻¹, exhibiting twice higher photocatalytic performance than P25 [65]. Other 3D TiO₂ structures, such as hollow fibers, connected structures, and hierarchical structures, have also been researched to achieve efficient catalysts.

4. Structural Design of Various Dimensions of TiO₂ Catalysts with Different Heterostructures

Until now, many strategies have been developed to prepare the 0D, 1D, 2D, and 3D nanostructures with various morphologies, including nanoparticles, nanofibers, nanotubes, and flower–like structures, all of which had been reported in previous articles. This part will summarize the main preparation methods, such as hydrothermal, sol–gel, vapor deposition, and electrospinning [67,68,69,70].

4.1. Hydrothermal Method

The hydrothermal method is typically operated in the autoclave system, known as a sealed Teflon–lined autoclave housed within the stainless steel vessel. When heated in an oven, the inner temperature of the Teflon–lined autoclave will increase beyond the boiling point of the solution, producing high pressure [71]. Various dimensional TiO₂ materials can be achieved under these conditions. Figure 5 displays the wire-like TiO₂ catalysts synthesized by the hydrothermal method. The synthesis of structured TiO₂ materials has recently been regarded as an effective technique for producing large quantities of TiO₂ materials with various dimensions in aqueous solutions [72]. Regarding the reagents applied during the hydrothermal synthesis, the method is divided into the acid hydrothermal and alkali hydrothermal methods. For the former method, the reagents often include titanium salts with acid, leading to the formation of 1D TiO₂ wire–like structures. The latter method is often used to synthesize TiO₂ nanoparticles. Various mechanisms have been researched for these two methods, yielding different morphologies.

4.2. Sol–Gel Method

Meanwhile, the sol–gel method is another effective technique for synthesizing TiO₂ particles, involving the hydrolyzation of materials using a titanium precursor. For example, in the synthesis of 1D TiO₂ nanowires, confinement is always needed to realize the growth of the 1D structure. Typically, TiO₂ nanowires are synthesized using the sol–gel method, with titanium(IV) isopropoxide serving as the precursor. The as–synthesized TiO₂ nanowires have a diameter of approximately 10 nm [74,75,76]. Through the sol–gel method, TiO₂ particles coalesce to form TiO₂ nanowires. Jin et al. synthesized TiO₂ nanowires with titanium tetraisopropoxide, isopropanol, and acetic acid as the precursors [77]. To obtain the xerogel, the as-prepared sol was then dried in an oven for one day. Finally, the samples were calcined in a tube furnace at 600 °C with a suitable gas to obtain the well-crystallized TiO₂ nanowires.

4.3. Vapor Deposition Method

The vapor deposition method, also known as chemical vapor deposition (CVD) and physical vapor deposition (PVD) method, has been used to prepare high-purity 2D TiO₂ materials. Compared with other methods, TiO₂ materials prepared by CVD and PVD show higher purity and crystallinity. Kamat et al. have synthesized a kind of well–crystallized TiO₂ by CVD (Figure 6). The as-prepared TiO₂ nanorods were deposited on an FTO glass with a suitable height [78]. Nevertheless, this method is also limited by the high costs associated with large-scale production.

4.4. Electrospinning Method

Compared to the methods mentioned earlier, the electrospinning method is regarded as a simpler approach for the large-scale synthesis of TiO₂ catalysts, especially suited for the preparation of TiO₂ nanofibers [80,81,82,83]. This method uses an electric field to transform the precursors into fine fibers. Compared to the previous heating method, the electrospinning process involves four main steps: (1) forming an aqueous solution with titanium precursors; (2) adding a polymer to form a sol suitable for electrospinning; (3) electrospinning the precursors to produce the nanofibers; (4) calcining the as-synthesized nanofibers to obtain highly crystalline TiO₂ fibers [84,85,86]. Kim et al. synthesized electrospun TiO₂ fibers with various TiO₂ percentages (Figure 7) [87]. The average diameter of the TiO₂ fibers can vary from 100 to 500 nm by adjusting the parameters.

5. Photocatalytic Application of Various Dimensions of TiO₂-Based Catalysts

The goal in developing catalysts is to synthesize various morphologies that increase the surface area and provide more active sites. To date, TiO₂-based catalysts of various dimensions (0D, 1D, 2D, and 3D) have been successfully synthesized. Meanwhile, the combination of metals, semiconductors, and elements doped with TiO₂ is considered a useful approach to designing novel TiO₂–based catalysts, which enhance light conversion efficiency for photocatalytic applications. Furthermore, the utilization of the structure of TiO₂-based catalysts is limited, as will be discussed below (

Table 1. Advantages and disadvantages of various TiO₂ preparation methods.

Method	Advantages	Disadvantages
Hydrothermal Method	Produces highly crystalline materials with controlled morphology. Suitable for large-scale production. Can produce various TiO2 phases (anatase, rutile, brookite).	Requires high pressure and temperature conditions. Longer reaction times. Equipment can be expensive and complex.
Sol–Gel Method	Low-temperature process. Excellent control over composition and purity. Versatile and can produce complex shapes and coatings.	Shrinkage and cracking can occur during drying and calcination. Solvent removal and purification steps can be cumbersome.
Vapor Deposition Method	Produces high-purity and uniform thin films. Excellent control over film thickness and composition. Can be used for large-area coatings.	High equipment and operational costs. Requires vacuum or controlled atmosphere conditions. Limited to thin film applications.
Electrospinning Method	Can produce continuous nanofibers with high surface area. Simple and cost-effective setup. Versatile in terms of material choice and fiber morphology.	Difficult to achieve uniform fiber diameter. Limited to materials that can be dissolved or melted. Post-processing may be required to remove solvents.

).

5.1. Zero-Dimensional TiO₂-Based Catalysts

As shown in Figure 8a, four different 0D TiO₂ catalysts were synthesized with Au particles to investigate the influence of these particles on photocatalytic performance. The synthesis process typically involves using hydrofluoric acid to control the morphology of TiO₂. Hydrofluoric acid promotes the exposure of the {001} facet, thereby enhancing photocatalytic activity due to the high percentage of reactive Ti centers. Figure 8b displays SEM images of TiO₂ spheres, each with a diameter of 200 nm, illustrating their uniform size and shape. The deposition of Au particles on TiO₂ spheres, visible in Figure 8c, further enhances photocatalytic efficiency by facilitating electron transfer. Figure 8d,e show the formation of a core–shell structure in the Au/TiO₂ catalysts prepared by the hydrothermal method, which is highlighted as a simple yet effective technique for tuning core–shell structures. This method not only improves the structural integrity of the catalyst but also optimizes its photocatalytic properties. Our analysis suggests that the combination of morphological control through hydrofluoric acid and the structural benefits of the hydrothermal method significantly enhances the performance of TiO₂-based photocatalysts.

In Figure 9a, core–shell-structured TiO₂ catalysts were synthesized with the addition of a sacrificial agent, a configuration often utilized in photocatalytic applications. Previous studies have demonstrated that this unique structure enhances photocatalytic performance [90,91,92,93]. Furthermore, the incorporation of gold (Au) onto TiO₂ catalysts, as depicted in Figure 9b, has shown a remarkable 40% increase in hydrogen (H₂) production, attributed to leaching effects. The controllable surface structure of these catalysts is crucial for maintaining optimal photocatalytic performance. Various studies have explored the use of TiO₂/metal catalysts, such as Au, Pd, and Ag, to improve photocatalytic activity by mitigating electron–hole recombination rates [94,95,96]. In these configurations, photo-induced electrons migrate to the metal due to their lower Fermi levels, while holes remain stable on TiO₂ surfaces, thereby enhancing carrier lifetimes and overall photocatalytic efficiency.

In recent advancements, the construction of quantum-scale catalytic regions on anatase TiO₂ nanoparticles has been achieved by loading TiO₂ quantum dots (QDs) onto their surfaces. Figure 10 from the study by Zhou et al. provides a detailed visual representation of this nanostructure. The transmission electron microscopy (TEM) images reveal that the TiO₂ nanoparticles are uniformly sized, with an average diameter of approximately 20 nm, and maintain their shape consistently across different samples (Figure 10A,B). The TiO₂-QD1 sample, in particular, stands out, as it displays a significant number of 1–3 nm TiO₂ QDs on its surface, as indicated by the red arrows in the images (Figure 10C).

These QDs are formed through the hydrolysis and condensation of titanium isopropoxide absorbed on the surfaces of the TiO₂ nanoparticles, a process that does not alter the primary size and morphology of the TiO₂ nanoparticles. High-resolution TEM (HR-TEM) images further confirm the successful integration of these quantum dots, showing lattice fringes corresponding to the {101} and {200} crystal facets of anatase TiO₂ (Figure 10D). The presence of these QDs significantly enhances the photocatalytic properties of the TiO₂ nanoparticles by providing additional active sites for reactions and facilitating more efficient charge separation. The incorporation of TiO₂ QDs onto the nanoparticle surface creates a synergistic effect, improving the photocatalytic degradation of volatile organic compounds (VOCs) such as toluene. This is attributed to the formation of quantum-scale catalytic regions, which increase the reaction probability of electron–hole pairs and their corresponding intermediates. Such advancements in nanostructure engineering highlight the potential of 0D TiO₂-based materials in environmental applications, particularly in the photocatalytic degradation of pollutants.

Figure 11 from the research provides an in-depth schematic representation of the band structure and the reaction mechanism within the TiO₂–QD composite. This scheme elucidates the intricate processes occurring at the interface between the TiO₂ quantum dots (QDs) and the TiO₂ nanoparticles. The electron redistribution at this interface is crucial, leading to the formation of an upward band bending from the inner TiO₂ particles to the surface TiO₂ QDs. This band bending plays a pivotal role in promoting the accumulation of photogenerated holes at the QD side while the electrons accumulate at the TiO₂ particle side.

The enhanced charge separation due to this band bending significantly improves the photocatalytic activity of the composite. The photogenerated holes, now concentrated on the QD side, are highly effective in oxidizing adsorbed pollutants such as water and toluene, leading to the generation of reactive oxygen species (ROS) like hydroxyl radicals. Concurrently, the electrons that migrate to the TiO₂ particle side react with oxygen molecules to form superoxide radicals, further contributing to the photocatalytic degradation of pollutants.

Moreover, the quantum-scale active regions created by the TiO₂ QDs optimize the photocatalytic performance in several ways. First, the close proximity of oxidation and reduction sites within the quantum scale ensures efficient transfer and reaction of intermediates, reducing the recombination rate of electron–hole pairs. Second, the presence of TiO₂ QDs increases the number of active sites available for photocatalytic reactions, thereby enhancing the overall reactivity of the composite. Third, the engineered band structure ensures that the photogenerated carriers are effectively separated and utilized, maximizing the quantum efficiency of the photocatalytic process.

5.2. One-Dimensional TiO₂–Based Catalysts

The hydrothermal method is widely utilized in the production of one-dimensional (1D) TiO₂–based catalysts, as exemplified in Figure 12, which exhibits a spectrum of structures comprising TiO₂ nanotubes, nanorods, nanobelts, and nanowires. This technique shows immense potential for generating a multitude of TiO₂ configurations with varied dimensions. Over recent years, hydrothermal synthesis of TiO₂ structured materials has emerged as a highly efficient technique for generating substantial quantities of TiO₂ with customized dimensions in aqueous solutions. Notably, the hydrothermal synthesis process can be classified into acid and alkali hydrothermal methods, each characterized by distinct reagents and resulting outcomes [98,99,100,101]. In acid hydrothermal synthesis, titanium salts are typically combined with acidic reagents, yielding the formation of 1D TiO₂ wire-like structures. Conversely, alkali hydrothermal methods are preferred for the synthesis of TiO₂ nanoparticles. Detailed mechanistic investigations into these methods have unveiled a plethora of pathways governing the formation of diverse morphologies. This comprehensive understanding contributes to optimizing TiO₂ catalyst design and performance, enhancing their efficacy in various applications.

As illustrated in Figure 13a–d, core–shell TiO₂ nanowires exhibit significantly enhanced photocatalytic activity compared to both TiO₂ nanowires and P25, primarily attributed to their high surface area and improved separation efficiency of charge carriers. Comprehensive measurements have confirmed the transfer of electron–hole pairs on the surface of TiO₂ nanowires. Specifically, Figure 13e,f demonstrate that the core–shell TiO₂ nanowire heterostructure displays superior photoactivity due to the enhanced rate of electron–hole separation, leading to a prolonged lifetime of charge carriers. This observation underscores the critical role of heterostructuring in optimizing the performance of TiO₂-based photocatalysts by facilitating efficient charge separation and minimizing recombination losses. The findings highlight the potential of core–shell nanostructures as a promising avenue for enhancing the photocatalytic efficiency of TiO₂ nanowires, offering insights for the design and development of advanced photocatalytic materials for various environmental and energy-related applications.

As displayed in Figure 14, silver (Ag) nanocrystals with an approximate diameter of 3.8 nm were uniformly dispersed across the surface of TiO₂ nanotubes (Figure 14a,b). The Ag/TiO₂ nanotube composite demonstrated significantly higher visible light photocatalytic activity for the degradation of rhodamine B (RhB) compared to P25 and pure TiO₂ nanotubes. Remarkably, this composite could almost completely degrade 100% of RhB within 2 h. This superior performance is attributed to the surface plasmon resonance (SPR) effect of Ag nanocrystals, which enhances visible light absorption, coupled with the high adsorption capacity of TiO₂ nanotubes due to their large specific surface area (Figure 14c,d). Additionally, TiO₂@carbon core/shell nanofibers (TiO₂@C NFs) with varying carbon layer thicknesses were synthesized through a combination of electrospinning and hydrothermal methods. By adjusting the parameters of the hydrothermal fabrication process, the thickness of the carbon layers was precisely controlled, ranging from 2 to 8 nm. These TiO₂@C NFs exhibited improved photocatalytic efficiency for RhB degradation under visible light irradiation compared to pure TiO₂ nanofibers. This enhancement is likely due to the synergistic effect between the carbon acting as a sensitizer and the 1D structure of TiO₂, which facilitates high separation efficiency of photogenerated electrons and holes. Notably, TiO₂@C NFs with a 2 nm thick carbon layer exhibited higher photocatalytic activity than those with an 8 nm thick carbon layer (Figure 14e–h). The enhanced photocatalytic activity of Ag/TiO₂ nanotube composites can be primarily attributed to the SPR effect of Ag nanocrystals. This effect leads to increased absorption of visible light, which is crucial for photocatalytic processes. Furthermore, the large specific surface area of TiO₂ nanotubes allows for greater adsorption of RhB molecules, making the degradation process more efficient. The uniform distribution of Ag nanocrystals ensures that the SPR effect is maximized across the entire surface of the nanotubes, leading to consistent and efficient photocatalysis. In the case of TiO₂@C NFs, the carbon layer plays a critical role in enhancing photocatalytic performance. As an excellent conductor, carbon helps in the rapid transfer of photogenerated electrons, reducing the recombination rate of electron–hole pairs. This improved charge separation is essential for higher photocatalytic efficiency. The ability to control the thickness of the carbon layer is particularly advantageous, as it allows for optimization of the photocatalytic properties. A thinner carbon layer (2 nm) provides a better balance between light absorption and electron transfer, leading to superior photocatalytic activity compared to a thicker layer (8 nm). Overall, these findings highlight the significant potential of Ag/TiO₂ nanotube composites and TiO₂@C NFs in environmental applications, particularly in the degradation of organic pollutants under visible light. The combination of SPR effects, high specific surface area, and efficient charge separation mechanisms provides a robust framework for developing advanced photocatalytic materials. Future research could focus on further optimizing these composites, exploring their applications in other types of pollutants, and scaling up the synthesis processes for industrial applications.

5.3. Two-Dimensional TiO₂–Based Catalysts

Figure 15 illustrates the synthesis procedure of graphene/TiO₂/Fe composite catalysts, a process that draws inspiration from numerous reports detailing the synthesis of composite catalysts via the hydrothermal method. Motivated by the success of this approach, we synthesized graphene/TiO₂/Fe composite catalysts by incorporating Fe³⁺ as an additive. Subsequently, we employed the hydrothermal method to synthesize these composite catalysts, during which graphene oxide underwent reduction to form graphene. This strategy capitalizes on the unique properties of graphene and the catalytic capabilities of TiO₂ and Fe, with the hydrothermal method serving as a versatile and effective technique for the integration of these components into a synergistic composite structure. By elucidating the synthesis procedure in detail, we aim to provide insights into the rational design and fabrication of graphene–based composite catalysts, offering potential avenues for enhancing their performance in various catalytic applications.

The SEM images presented in Figure 16 provide valuable insights into the morphology of the synthesized samples. A comparative analysis reveals distinct differences between the pristine Fe nanowires and P25 (Figure 16a,c) and the samples wherein TiO₂ is loaded onto graphene nanosheets via the hydrothermal method, showcasing a notably uniform dispersion (Figure 16b,d). This uniform dispersion suggests that TiO₂ readily forms Ti–O–C bonds to connect with graphene, facilitating a strong interaction between the two materials. In contrast, other nanoparticles exhibit less pronounced agglomeration on the graphene nanosheets, indicative of the potential for enhanced dispersion and interfacial interactions. These observations underscore the importance of morphology and interfacial characteristics in determining the performance of composite materials, highlighting the potential of graphene-based TiO₂ composites for various applications requiring enhanced catalytic activity and stability.

Figure 17 presents the detailed microstructure and elemental distribution of the synthesized photothermal catalysts, specifically the Au–loaded TiO₂ nanoflakes (AT–NFs). The high-resolution transmission electron microscopy (HRTEM) images in Figure 17a,b illustrate the two–dimensional size of the nanoflakes, approximately 80 × 50 nm, with a thickness of around 9 nm. These nanoflakes exhibit a well-defined anatase crystal structure, as evidenced by the lattice fringes observed in Figure 17c, which correspond to the {101} crystal plane of anatase TiO₂ with a lattice spacing of 3.52 Å. Additionally, the presence of Au nanoparticles on the TiO₂ surface is confirmed by the distinct lattice fringes of Au, identified as the {111} crystal planes with a lattice spacing of 2.36 Å. The scanning transmission electron microscopy (STEM) and energy-dispersive X-ray spectroscopy (EDS) mapping in Figure 17e,f further reveal the distribution of Au nanoparticles on the TiO₂ nanoflakes. The brighter areas in the STEM images, which align with the traces of Au in the EDS mapping, confirm the uniform distribution of Au on the TiO₂ surface. This uniform distribution is crucial for enhanced photocatalytic activity, as it ensures efficient interaction between the TiO₂ and Au nanoparticles.

Figure 18 explores the impact of the photothermal effect on the hydrogen production efficiency of the synthesized catalysts. The experiments compared the hydrogen evolution rates under photoreforming (PR) and photothermal reforming (PTR) conditions. The results in Figure 18a show that PTR significantly enhances hydrogen production compared to PR. Specifically, after 3 h of irradiation, the hydrogen yield for PTR reached 8.2 mmol/g, which is 58% higher than the 5.2 mmol/g yield observed for PR. Temperature monitoring of the catalyst suspension under UV–visible light illumination, shown in Figure 18b, revealed a significant rise in temperature due to the local surface plasmon resonance (LSPR) effect of the Au nanoparticles. The temperature of the suspension increased from an initial 25 °C to 48.1 °C under full–spectrum light, demonstrating the efficient photothermal conversion. This temperature increase was directly correlated with an enhanced hydrogen evolution rate, which rose with increasing temperature, peaking at 4.2 mmol/g/h at 65 °C, as shown in Figure 18c. The cyclic performance tests in Figure 18d confirmed the stability and reproducibility of the catalysts. Both PR and PTR exhibited excellent cyclic performance, with PTR consistently yielding higher hydrogen production across multiple cycles. The cyclic tests involved alternating periods of light irradiation and cooling to room temperature, demonstrating the catalysts’ robustness and sustained activity over repeated use.

5.4. Three-Dimensional TiO₂-Based Catalysts

Figure 19a delineates the synthesis process of the 3D TiO₂ film, which involves several sequential steps aimed at achieving a structured and functionalized nanoarchitecture. Initially, a 5 μm thick epoxy template is prepared, possessing a structural unit cell of elongated body-centered tetragonal (BCT) geometry with planar and perpendicular structural periodicities of 1 μm. Subsequently, a high-temperature atomic layer deposition (ALD) procedure is employed to deposit TiO₂ onto the template surface. Following this, the epoxy template is removed via thermal decomposition at 500 °C, resulting in the formation of a hollow 3D TiO₂ structure through a templating process. Next, a layer of TiN is deposited onto the 3D TiO₂ structure using ALD at a temperature of 350 °C. This deposition process yields a film with a distinctive sandwich structure comprising TiN/TiO₂/TiN layers. Finally, thermal annealing at 550 °C transforms the TiN layer into N-doped TiO₂, resulting in a material with a large specific surface area, as depicted in Figure 19b–d. This synthesis approach demonstrates a systematic strategy for fabricating 3D TiO₂ films with tailored nanostructures and functional properties. By incorporating ALD and thermal treatment steps, precise control over the composition, structure, and surface characteristics of the resulting films is achieved. The sequential deposition and annealing processes enable the generation of a hierarchical nanoarchitecture with enhanced surface area and photocatalytic activity, offering potential applications in various fields.

Figure 20 presents an evaluation of the photocatalytic activity and stability of various 3D TiO₂ films, assessed through the degradation rate of methylene blue (MB) as a model organic pollutant. Upon irradiation with solar light for 3 h, the degradation percentage of MB using the 3D N–doped TiO₂ film is approximately 50%, significantly outperforming other samples. This substantial enhancement in degradation rate underscores the superior photocatalytic performance, facilitated by the enhanced charge separation rate and specific surface area (Figure 20a–c). Furthermore, the degradation ability of the 3D N-doped TiO₂ film is 1.5 times higher than that of the pristine TiO₂ film and 30% higher than that of the undoped TiO₂ film (Figure 20d). Notably, the difference in degradation percentage between undoped and N-doped TiO₂ films, regardless of their nanostructured nature, is negligible under irradiation with a solar light simulator. This observation suggests a low N–TiO₂ film quantum efficiency due to the decreased photo-induced charge carriers and enhanced recombination rate of electron–hole pairs. Moreover, the enhancement of photocatalytic activity attributed to N doping primarily arises from improved visible light absorption, with the 3D structure further accelerating this enhancement. Additionally, the 3D N-doped TiO₂ film exhibits remarkable stability, allowing for repeated use as a highly efficient visible photocatalyst without significant performance degradation (Figure 20e). Even after five cycles of reuse, the 3D structure of the TiO₂ film remains intact (Figure 20f), highlighting its robustness and potential for practical applications in water treatment and environmental remediation.

6. Summary and Perspectives

In this review, we have comprehensively summarized recent advancements in synthesizing various TiO₂–based catalysts featuring novel nanostructures. TiO₂, owing to its versatile properties, serves as an ideal base material for creating TiO₂-based materials with diverse dimensions. The development of various synthesis methods, such as hydrothermal, sol–gel, vapor deposition, and electrospinning, has significantly contributed to this progress. Each of these methods offers unique advantages and challenges in controlling the morphology and size of the TiO₂ nanostructures, with the annealing temperature playing a crucial role in determining the final properties of the synthesized materials. Significant research efforts have been directed at combining TiO₂ with various metal oxides to enhance its photocatalytic performance. These composite materials exhibit improved visible light absorption and reduced electron–hole pair recombination rates, leading to higher photocatalytic efficiency. Integrating innovative synthesis techniques has enabled the design and fabrication of TiO₂–based catalysts with optimized structures that enhance their performance in photocatalytic applications. Among the various nanostructured TiO₂ materials, 3D TiO₂–based materials have garnered particular attention due to their high specific surface area and extended light response wavelength. These characteristics result in superior photocatalytic activity compared to their lower-dimensional counterparts. The 3D structures provide more active sites for photocatalytic reactions and facilitate better charge separation and transport, which are critical for improving the overall efficiency of the catalysts.

Future research in this field should focus on several key areas to further enhance the performance and applicability of TiO₂–based photocatalysts. Firstly, developing more efficient and scalable synthesis methods that allow precise control over the morphology and size of TiO₂ nanostructures is essential. Exploring new synthesis techniques and optimizing existing ones could produce materials with even higher photocatalytic activity. Secondly, combining TiO₂ with other materials, such as novel metal oxides, carbon-based materials, and other semiconductor nanostructures, should be investigated in greater detail. Understanding the interaction mechanisms between TiO₂ and these materials at the molecular level could provide insights into designing more effective composite catalysts. Thirdly, the stability and durability of TiO₂–based photocatalysts under real–world conditions must be thoroughly examined. Long-term studies and practical applications should be conducted to ensure that these materials can withstand various environmental conditions without significant degradation in performance. Additionally, expanding the range of applications for TiO₂–based photocatalysts beyond environmental remediation to include areas such as energy conversion, water splitting, and CO₂ reduction could open up new avenues for research. Tailoring the properties of TiO₂ nanostructures to meet the specific requirements of these applications will be crucial for their successful implementation. In conclusion, while substantial progress has been made in synthesizing and applying TiO₂–based catalysts with novel nanostructures, there remain numerous opportunities for further research and development. By addressing the challenges and focusing on the future directions outlined above, the potential of TiO₂–based photocatalysts can be fully realized, leading to significant advancements in various technological and environmental fields.