1. Introduction

In the past few decades, the fast developments of economy, science and industry have resulted in the tremendous consumption of fossil fuels including coal, oil, and natural gas, bringing out great issues of environmental pollution and energy shortage [1-5]. Since the groundbreaking work in terms of sunlight-driven H₂O splitting was carried out by Fujishima and co-workers in 1972 [6], using photocatalysis technique to convert solar energy into chemical energy has been widely studied in various reactions, such as pollutants decomposition, H₂/O₂ evolution, N₂ fixation, inorganic conversion and H₂O₂ synthesis [7-12]. Meanwhile, a variety of semiconductor photocatalysts (such as metal oxides, metal sulfides, and nitride) with different micro/nano structure have been exploited as advanced photocatalysts for various photocatalytic reactions [13-16]. As to photocatalysis process, the active photocatalyst materials are highly essential to make sure the photoinduced charges into a higher energy level to trigger photoredox reactions [17-20]. Light absorption, photoinduced hole-electron pair separation, surface charge migration and product desorption are involved in the whole photocatalysis system [21-25]. So far, various strategies have been developed to improve the photocatalytic efficiency, nevertheless, how to effectively utilize sun light via the slow photons effect has been always ignored in photocatalyst material design.

As to some materials with periodic structures (such as inverse opal), the light speed would be reduced as the irradiation light passes the material, which is also called as slow photon effect. Slow photon effect can enhance the residence time and absorption capability of irradiation light energy, improving the interaction between light and materials (Fig. 1a) [26-28]. In fact, the variable colors of butterfly wings, peacock feathers, beetle shells, and so on in nature often originates from the selective reflection and absorption of specific wavelengths by the periodic nanostructures of photonic crystals (Fig. 1b) [24], which often involve slow photons effect. Photonic crystal is a material with a periodic dielectric structure, which can precisely control the propagation of light in the same way that semiconductors control electrons. The most important characteristic of photonic crystals is the photonic bandgap. Inspired by the photonic crystal structures in nature, a series of artificially synthesized photonic crystal materials (such as SiO₂) have been reported (Fig. 1c) [9], which can improve the absorption of light with specific wavelengths and exhibit variable color via tuning the size of nanosphere. Typically, the inverse opal TiO₂ sample with pore size around 330 nm exhibits green color (Fig. 1d) under daylight [11]. The spectrum formed by the reflection of incident light via the photonic bandgap suggests that there are two boundaries for the reflection peak corresponding to a specific photonic bandgap. The boundary of the reflection peak that corresponds to longer wavelengths and lower energy is referred to as the red-edge, whereas the boundary associated with shorter wavelengths and higher energy is called the blue-edge (Fig. 1e) [11]. Theoretical analyses have shown that during the red-edge slow photon effect, incoming photons travel through the photonic crystal structured materials, while in the case of the blue-edge slow photon effect, the photons are situated within the pore structure of the 3DOM configuration [29,30]. When the energy of slow photons matches well with the bandgap width of the semiconductor photocatalyst, slow photons will be utilized by the photocatalyst and the light absorption efficiency will be greatly improved, thus improving photocatalytic performance.

As a type of nanomaterial with unique pore structure, photonic crystals, also denoted as three-dimensionally ordered macroporous (3DOM) materials, have been widely studied in catalytic fields [31-35]. The periodic internal pore channels can bring out the multiple light scattering and lead to slow photons effect so as to boost the photocatalytic procedure (Fig. 2a) [36]. Moreover, their high specific surface area can offer sufficient reactive sites for reactant adsorption and activation in different photocatalytic reactions (Fig. 2b) [37-40]. Thus, the photonic crystal structured materials endow great prospects in sustainable energy production alleviate current energy crisis. In this review, we firstly introduced the fundamentals and mechanisms of solar light induced catalytic energy conversion via slow photon effects. Then, the design strategies for inverse opal structured materials aimed at enhancing photocatalysis were outlined, emphasizing the advantages of these nanostructured catalysts in terms of reactive sites, mass diffusion, light absorption, and local photothermal effects, among other factors. Subsequently, the discussion then delves into the recent advancements in photocatalytic water splitting for hydrogen production, as well as the generation of clean fuels and valuable chemicals through photo/photothermal catalytic CO₂ conversion using inverse opal structured catalysts. Finally, the perspectives and challenges of solar light-driven sustainable energy conversion over photonic crystal catalysts are outlooked and elaborated. We sincerely hope that this important review will inspire innovative ideas for developing effective nanostructured photocatalysts and enhance our understanding of the solar energy conversion process, ultimately helping to reach the objectives of carbon peak and carbon neutrality.

2. Basics and mechanisms of photocatalytic solar fuel generation

Photocatalysis is a technology that harnesses light energy to facilitate chemical reactions. It primarily involves several essential steps: absorbing light, generating photoinduced charge carriers, separating electron-hole pairs, moving the photogenerated electrons and holes, and conducting surface photoredox reactions [41-43]. Photogenerated electrons typically undergo reduction reactions with the adsorbed oxidants (such as oxygen molecules, protons, CO₂) [44,45], while holes undergo oxidation reactions with reducing agents (such as water molecules or organic substances) [46,47]. The series of oxidation–reduction reactions triggered by photogenerated electrons and holes are the core of the photocatalytic process [48-50]. Photocatalytic technology has been utilized in various areas for clean and sustainable energy conversion, including hydrogen production through water splitting and the generation of fuels and valuable chemicals from CO₂ conversion. This technology enables the easy production of green energy and offers innovative ideas and opportunities for future energy transformation systems.

2.1. Photocatalytic hydrogen/oxygen production

To conduct photocatalytic water decomposition to release hydrogen and oxygen, the CB potential of semiconductor photocatalyst materials should be slightly negative than the hydrogen reduction potential H⁺/H₂, while the VB potential should be positive than the oxygen oxidation potential O₂/H₂O (Fig. 3a) [51-54]. When exposed to external light, if the energy of the incoming photons is equal to or exceeds the bandgap of a semiconductor, electrons can be excited from the valence band (VB) to the conduction band (CB). Meanwhile, holes stay in the VB of the semiconductor. This process leads to the separation of electrons and holes, which can then facilitate the reduction of water to create hydrogen gas or the oxidation of water to generate oxygen gas at various sites within the semiconductor photocatalysts [55]. In comparison to the hydrogen evolution from water splitting, the oxygen formation half reaction in photocatalytic water splitting is a more difficult step as it involves a four-electron transfer process (Fig. 3b). The co-catalysts were frequently adopted for photocatalytic oxygen evolution via water splitting [56]. The most active oxygen evolution reaction co-catalysts are precious metal oxides (RuO₂ and IrO₂) [57]. But the interface formed between semiconductors and co-catalysts affects charge separation and transfer [58]. How to regulate the interface to accelerate hole transport and achieve efficient oxygen production efficiency is a great challenge. Up to now, a variety of semiconductor materials including CdS, TiO₂, C₃N₄ and Bi based photocatalysts have been explored for photocatalytic water decomposition. However, how to maximumly utilize solar/visible light by these semiconductor photocatalysts for efficient hydrogen/oxygen production is still highly difficult. Designing photocatalyst with inverse opal structure can improve light harvesting ability owing to slow photon effect, thus accelerating photocatalytic water splitting.

2.2. Photocatalytic/photothermal catalytic CO₂ conversion

As to the CO₂ conversion, it is necessary to provide enough energy to broke the two linear double bonds (C=O) due to the thermal stable and chemically inert properties of CO₂ molecule [59-61]. In early stage, thermocatalysis approach for CO₂ conversion has been extensively studied. In the process of thermocatalytic CO₂ reduction, the input of thermal energy and the presentation of catalysts are prerequisites for activating CO₂ molecules [62]. In particular, the C=O bonds are split and subsequent generation of new bonds upon high-energy environments. The total thermocatalytic CO₂ reduction is commonly composed of following steps: CO₂ molecule adsorption and activation, intermediates evolution, and the generation and desorption of products. Among the above procedures, intermediates production is widely considered as the rate-determining program of the total progress, being closely dependent on the surface chemistry of the thermocatalyst materials. Moreover, the fast desorption of the products from the catalyst surface is also essential, which would decide the cycle stability of the thermocatalytic CO₂ conversion system [63]. Different from the thermocatalytic reaction, the irradiated light as driven force is adopted to initiate photocatalytic CO₂ reduction (Fig. 4a) [64-66]. In this case, semiconductor materials as photocatalysts are frequently utilized because their appropriate electronic band gap structures can effectively excited by ultraviolet (UV) or visible light. Once the semiconductor photocatalysts are excited, the photoinduced electrons and holes can be produced on the CB and VB of the photocatalysts. The photogenerated charges can directly shift to the photocatalyst surface to activate or combine the CO₂ molecules, followed by the generation of intermediate species to achieve the conversion of CO₂ (Fig. 4b) [67]. The main gas products of photocatalytic CO₂ reduction reaction are CO and CH₄, may accompanied by the decomposition of water to produce hydrogen and oxygen, and may also produce alkane and alkene products [68]. Therefore, in addition to improve the photocatalytic CO₂ reduction rate, the selectivity modulation of the products from CO₂ reduction is also important for the practical application [69,70]. In addition to thermocatalysis and photocatalysis, photothermal catalysis has recently emerged as a very promising method for CO₂ reduction. This approach takes advantage of the wide absorption of solar energy to enhance both thermochemical and photochemical processes, which together increase the rate of catalytic reactions and enable effective CO₂ conversion under relatively mild conditions (Fig. 4c) [71]. As a type of photothermal catalysis, introducing thermal energy into photocatalysis can improve the utilization efficiency of sunlight, promote the excitation and separation of photoinduced charge carriers, accelerate the diffusion of reaction molecules, and enhance the reaction performance [72,73].

3. Preparation of photonic crystals with slow photons effect

The preparation of uniform and high-quality photonic crystals is of great importance for the investigation of slow photons effect and application of photocatalytic fields. Specifically, creating an ordered three-dimensional macroporous structure is crucial as it approaches the complete photonic bandgap. This bandgap signifies that photonic crystals can reflect light in all directions of propagation within a specific frequency range. Up to now, there are two major approaches to fabricate photonic crystal materials: bottom-up method and top-down method.

3.1. Bottom-up method

The bottom-up method has been extensively adopted for synthesizing nanostructured materials with different morphology, which includes hydrothermal/solvothermal method, molecule self-assembly, co-precipitation method, sol gel method [74,75]. For the fabrication of photonic crystals, self-assembly and manual assembly techniques have been frequently reported. In general, self-assembly is simple and easy to be conducted, and is widely used for preparing nanomaterial with various structure and morphology. Although nanoparticle self-assembly can produce photonic crystals with an opal structure, the refractive index ratio of photonic crystals is generally much lower to achieve a full bandgap [76]. Therefore, photonic crystals with an opal structure only exhibit a pseudo photonic bandgap. Differently, inverse opal structured materials are expected to achieve a complete photonic bandgap. Researchers applied the photonic crystal template method to prepare the nearly complete photonic band gap inverse opal structure. As shown in Fig. 5a, colloidal nanoparticles initially create ordered photonic crystals with an opal structure via a self-assembly process. Subsequently, either the penetration method or the original sub-layer deposition method is employed to fill the gaps in the photonic crystal template with solid materials. Finally, the original photonic crystal template is removed by the chemical dissolution or high-temperature calcination to generate photonic crystals with inverse opal structure. For instance, via changing the size of colloidal nanoparticles, the inverse opal TiO₂ materials with different sizes can be obtained (Figs. 5b–j), which exhibit different colors under daylight due to the scattering of light with different wavelengths. So far, various semiconductor materials with inverse opal structures, such as metal sulfides, metal oxides, and carbon materials, have been used for efficient photocatalytic reactions due to their unique slow photon effect [38,40].

3.2. Top-down method

The top-down strategy adopt physical or mechanical methods to reduce the size of bulk materials to the nanometer or micrometer scale, which mainly including mechanical drilling and photolithography methods [77]. Typically, Yablonovich et al. for the first time prepared a uniform photonic crystals with unique slow photons effect via direct mechanical drilling method [78]. This drilling technique for creating photonic crystals by mechanically drilling periodic holes in the substrate material, primarily applicable to macroscopic periodic structures on the millimeter to micrometer scale. Photolithography strategy is based on photochemical reactions and etching techniques, which enables high-precision control of photonic crystal structures and is widely used in the preparation of complex periodic structures [79,80]. Specifically, the photoresist occurs chemical change (such as crosslinking or decomposition) under light illumination, a preset periodic pattern is transferred onto the photoresist through a mask plate, and then the pattern is loaded onto the substrate material through etching techniques, ultimately forming a photonic crystal with a periodic dielectric structure. In comparison to bottom-up method, the top-down strategy is relatively less reported for preparing photonic crystals for photocatalytic applications.

4. Slow photons for catalytic green energy conversion applications

Solar energy as the inexhaustible and renewable energy sources have aroused great interests in photocatalysis and photovoltaics fields [81]. However, the unsatisfactory solar energy conversion efficiency caused by the poor light absorption capability and inferior quantum conversion yield of materials restricted the advances of solar energy technology [82]. Although a variety of strategies have been developed to improve the utilization ability of sun light [83-85], the slow photons effect has been always ignored in material design. In this section, we systematically introduced the advanced progresses of designing photonic crystal structured materials with tunable slow photons effect and revealed the inherent mechanism of slow photons in accelerating photocatalytic green energy conversion.

4.1. Photocatalytic water splitting

The increasing fossil energy consumption and global energy shortage issue forces scientists to develop new carbon neutral energy sources (Fig. 6a) [86], especially of hydrogen energy, which is widely deemed as a green, sustainable and environmentally friendly energy feedstock, and applied in different fields in the word (Fig. 6b) [87]. Currently, there are various key methods for producing hydrogen, such as extracting hydrogen from fossil fuels, using electrolysis to split water, and innovative techniques like solar-driven water splitting, thermochemical water decomposition, and biomass hydrogen production [88]. Notably, solar-driven water splitting is particularly promising as it transforms light energy into chemical energy, allowing for storage and positioning it as a potential solution to the global energy crisis and environmental pollution issues.

At present, the most efficient solar powered water splitting for hydrogen generation is reached via coupling solar cells with water electrolysis devices (Fig. 7a) [89]. The solar to hydrogen (STH) energy conversion efficiency has been reported to be as high as 30% (Fig. 7b). Differently, the energy conversion efficiency of photocatalytic water splitting for hydrogen production is much lower (only about 1%). Nevertheless, the overall system design of photocatalytic hydrogen production is much simpler, the cost is lower, and it is easier to industrial scale, thus endowing promising commercial application prospects. As the core aspect of photocatalytic hydrogen production technology, photocatalyst materials directly decide the efficiency of the entire photocatalytic reactions. Therefore, the development, design, and modification of advanced photocatalysts have become a research hotspot in photocatalysis hydrogen evolution field.

In past few years, Su’s group has conducted a series of works to confirm the essential role of slow photons effect for boosting photocatalytic hydrogen production in aqueous solution by constructing various kinds of photonic crystal structured composites. In 2016, their group reported a ternary TiO₂-Au-CdS photocatalyst with uniform photonic crystal structure based on 3DOM TiO₂ for efficient photocatalytic water splitting to hydrogen generation [90]. The schematic process of the TiO₂-Au-CdS inverse opal synthesis is depicted in Fig. 8a. Firstly, the 3DOM TiO₂ was successively immersed into HAuCl₄ and citrate aqueous solution to uniformly decorate Au nanoparticles on the surface of 3DOM TiO₂. Then the generated nanoparticles are encapsulated with citrate ions and would couple with Cd²⁺ ions because of the Coulomb force. Finally, S²⁻ ions replace citrate ions to generate CdS on the surface of Au nanoparticles and TiO₂. As shown in Fig. 8b, the prepared ternary 3DOM TiO₂-Au-CdS displays yellow color, indicating the considerable portion of visible-light harvesting. The SEM image indicates that the 3DOM TiO₂-Au-CdS exhibits a very uniform 3DOM structure, with gold nanoparticles approximately 30 nm in size distributed across the surface of the 3DOM TiO₂ (Fig. 8c). This arrangement can trigger a slow photo effect and surface plasmon resonance (SPR) phenomenon, enhancing the harvesting of visible light (Fig. 8d). As a result, the obtained 3DOM TiO₂-Au-CdS photocatalyst brings out efficient hydrogen generation rate of 1.81 mmol h^-1 g⁻¹ upon visible-light excitation, being 13 times greater than that of the binary TiO_2nullCdS photocatalyst (Figs. 8e and f). The mechanism investigation illustrates that the photoinduced electrons on TiO₂ are transferred to recombine with the holes of CdS and under UV–visible light excitation, photoinduced electrons on CdS will couple with protons to generate hydrogen. Thus, the excellent hydrogen evolution over 3DOM TiO₂-Au-CdS sample can be ascribed to synergistic effect of the enhanced light harvesting by photonic crystal effect, the photosensitizing effect of Au and the prohibited rate of charge carriers recombination (Fig. 8g). Although the inverse opal photonic crystal structure can enhance light absorption through multiple scattering of incident light for efficient hydrogen evolution, the randomness of this scattering mode is difficult to quantitatively describe.

As discussed above, the slow photon effect is beneficial for enhancing the residence and absorption of light energy in materials, thereby enhancing photocatalytic hydrogen evolution performance. For a long time, the red edge slow photon effect has been considered to have a better photocatalytic enhancement effect than the blue edge slow photon effect. As a result, studies focused on improving photocatalytic performance through the slow photon effect primarily concentrate on the red edge slow photon effect, whereas investigations into the enhancement of photocatalytic performance via the blue edge slow photon effect are comparatively scarce. Therefore, further experiments are needed to demonstrate which of the red edge and blue edge slow photon effects is better in promoting photocatalytic performance. At present, researches on slow photon effects mainly focus on thin film samples with 3DOM structures. This is because thin film samples are easier to adjust the angle of incident light, thus can unveil the influence of incident light angle on slow photon effects [91,92]. Due to limitations in the preparation conditions of thin film samples, the effective mass of the samples is extremely low and difficult to control the film layer, resulting in generally poor photocatalytic performance [93]. How to achieve slow photon effects in 3DOM powder sample has become a long-standing challenge for researchers. The reason is that the orientation change of powder samples caused by continuous stirring during the photocatalytic liquid-phase reaction process can result in the slow photon effect not being able to continue to occur. Therefore, it is necessary to design 3DOM powder samples with specific framework and solve the problem of orientation change caused by stirring during liquid-phase reaction process, ensuring that the slow photon effect can continue to function.

In this context, their team developed a staircase-shaped 3DOM composite photocatalyst [94]. The self-assembly of PS nanospheres and template method are involved in this synthesis being similar to the aforementioned work, except that the obtained PS nanosphere template has a staircase-style configuration. From Fig. 9a, it can be seen that the synthesized TiO₂ inverse opal sample has a large area of intact 3DOM structure, and the staircase-style configuration is also clearly visible. The TiO₂ inverse opal with a size of 160 nm presents a blue color upon day light environment (inset of Fig. 9a), which is related to the periodic macroporous structure of the sample and the reflectance of blue light. The 3DOM structure and staircase-style configuration of the 3DOM TiO₂-Au-CdS sample remain intact after loading CdS and Au (Fig. 9b). The ternary 3DOM TiO₂-Au-CdS appears yellow color (inset of Fig. 9b) due to the yellow color of CdS itself covering the structural color of TiO₂ inverse opal. The staircase-shaped 3DOM composite helps minimize the impact of changes in the angle of incoming light on the reflection spectrum and preserves the slow photon effect during ongoing stirring (Fig. 9c). The HAADF-STEM (Fig. 9d) and HR-TEM (Fig. 9e) images demonstrate that the Au nanoparticles are well dispersed on the macropore of the obtained 3DOM photocatalyst. In order to illustrate the relationship between the structural color and pore size of the inverse opal structure, a series of 3DOM TiO₂ and 3DOM TiO₂-Au-CdS with different pore sizes were synthesized by changing the size of PS nanosphere template. At the boundary of the reflection spectrum, the propagation speed of the incident photons in the photonic crystal structure sharply decreases, and this part of the photons is called slow photons. When the energy of slow photons matches the energy corresponding to the bandgap width of the semiconductor materials, slow photons are absorbed by the semiconductor component. From the reflection spectra of 3DOM TiO₂-Au-CdS samples with different macropore sizes, it can be seen that the boundary positions of the reflection spectra are mostly in the visible light region (Fig. 9f). Therefore, when the energy of red edge slow photons or blue edge slow photons overlaps with the energy corresponding to the CdS bandgap width (about 2.4 eV), the slow photon effect will occur.

Due to the stirring during the photocatalytic hydrogen evolution process, the sample is in a randomly oriented state. To illustrate the relationship between the slow photon effect and the incident light angle of the synthesized powder sample, the reflection spectra of two different 3DOM TiO₂-Au-CdS samples were measured at different incident light angles (Figs. 9g and h). The findings show that the reflection spectrum of the solid 3DOM TiO₂-Au-CdS sample does not vary significantly with changes in the angle of incident light. This suggests that if a slow photon effect exists in the sample, it will continue to operate effectively throughout the photocatalytic hydrogen production process. The main reason is that the synthesized sample has a staircase-style structure, which avoids the orientation problem caused by the angle change of incident light, making the sample have the same structure in all directions. Differently, 3DOM TAC-250 has a blue-edge slow photon effect due to the overlap between the blue edge of reflection spectrum and absorption of CdS, whereas 3DOM TAC-160 presents a red edge slow photon effect, which could cause different photocatalytic activity.

To illustrate the effect of slow photon effect on the photocatalytic performance, visible-light photocatalytic hydrogen production was studied for the TiO₂-Au-CdS samples with different pore sizes. Due to the same specific surface area and mesoporous pore size distribution of the samples with different pore sizes, the factor causing differences in hydrogen production performance among different samples is the change in large pore size, which is the influence of slow photon effect. The reflection peak of sample 3DOM TAC-200 covers the intrinsic absorption region of CdS, indicating that the incident light that can be absorbed and utilized by CdS is majorly reflected by the inverse opal structure, resulting in poor hydrogen evolution performance (1.81 mmol h^-1 g^-1). The hydrogen production rate over bulk TAC under visible light is 1.66 mmol h^-1 g^-1, being inferior than that of 3DOM TAC-200. The findings indicate that although the 3DOM structure reflects a significant amount of the incident light that CdS can absorb, its ability to scatter light multiple times and facilitate rapid mass transfer enhances its photocatalytic performance compared to the TAC sample lacking the 3DOM structure. This highlights the importance of structural advantages in hydrogen production. The 3DOM TAC-160 sample exhibits a hydrogen production rate of 2.55 mmol h^-1 g^-1, associated with the red-edge slow photon effect, while the 3DOM TAC-250 sample shows a hydrogen production rate of 3.50 mmol h^-1 g^-1, linked to the blue edge slow photon effect. It can be deduced that the blue edge slow photon effect also has the capability to enhance photocatalytic performance, which is more obvious than the red edge slow photon effect. This discovery firstly confirmed that the blue-edge slow photon effect has a better photocatalytic performance enhancement effect than that of the red-edge slow photon effect. The primary reason is that during the occurrence of the red-edge slow photon effect, the light reflected by the structure consists of photons that have shorter wavelengths and higher energy. When the blue edge slow photon effect occurs, the incident light reflected by the structure is composed of photons with longer wavelengths or lower energy, which cannot be absorbed by the material. Therefore, from the perspective of the incident photon energy that can be effectively absorbed and utilized, the blue edge slow photon effect does have a better enhancement effect on photocatalytic performance (Fig. 9i).

Considering the special SPR effect and fantastic charge transfer performance of gold nanoparticles in the ternary TiO₂@Au@CdS system, this team conducted further work to study the effect of gold nanoparticles size on photocatalytic performance [95]. They precisely modulated the size of gold nanoparticles and systematically investigated the size effect of gold nanoparticles in the ternary 3DOM TiO₂@Au@CdS composites (Figs. 10a–i), achieving further improvement in the photocatalytic hydrogen generation. In their work, sodium citrate solution with different concentrations is used to reduce chloroauricacid to prepare Au nanoparticles of tunable particle sizes, and loaded them into a 3DOM TiO₂ framework. CdS was introduced in the ternary system as the photosensitive material through chemical bath deposition route. The prepared ternary materials with tunable Au nanoparticles size not only have the capability to absorb UV-visible light, but also effectively facilitate the photogenerated carriers dissociation via the constructed type Ⅱ heterojunction. The UV–vis absorption and reflection spectra characterizations illustrated that TiO₂@Au@CdS composites with perfect photonic crystal structures also exhibit blue-edge slow photon effects (Fig. 10j), ensuring maximized utilization of incident photons by the photocatalyst. Photocatalytic tests demonstrated that there are significant differences in the photocatalytic H₂ evolution rate as changing the size of Au nanoparticles in the 3DOM TiO₂@Au@CdS system. The highest photocatalytic hydrogen production rate is achieved when the Au nanoparticles size is around 10 nm (Fig. 11a). A variety of photoelectrochemical characterizations were conducted to unveil the performance improvement mechanism. The photocurrent intensity gradually increases with decreasing the size of Au nanoparticles, indicating that the SPR effect of Au nanoparticles is gradually enhanced (Fig. 11b). The EIS tests also exhibit the same conclusion (Fig. 11c). However, electrochemical impedance spectroscopy (EIS) characterization gives rise to difference EIS results, and hydrogen production rate reveals a strong correlation between hydrogen production activity and charge transfer impedance. This suggests that altering the size of gold nanoparticles can influence hydrogen production efficiency by impacting both SPR and EIS characteristics, with EIS having a more significant effect. Additionally, the author utilized the finite time difference method to simulate the intensity and distribution of the electromagnetic field within the ternary 3DOM TiO₂@Au@CdS system, based on structural analysis and electrochemical characterization findings (Fig. 11d). The electromagnetic field intensity was significantly enhanced at the interfaces between TiO₂-Au and Au-CdS, verifying that the charge transfer pathway follows the type Ⅱ heterojunction mechanism under visible light conditions. It is important to note that when gold nanoparticles are very small (5 nm), the enhancement of electromagnetic field intensity at the interface is minimal, suggesting that charge transfer at the interface faces significant resistance in this scenario. Combined with the slow photon effect and optimized SPR effect of Au nanoparticles, a significant improvement in the photocatalytic hydrogen production performance of 3DOM TiO₂@Au@CdS catalyst system was achieved (Fig. 11e).

In a photocatalytic system, the heterojunction plays a crucial role in separating photoinduced electron-hole pairs. Therefore, it is important to thoroughly investigate the impact of the slow photon effect in three-dimensional ordered mesoporous (3DOM) heterostructure composites and to explore the advantages of combining slow photons with the heterojunction in photocatalytic hydrogen production systems [96-99]. Typically, Liu and co-workers prepared a series of CdS@ZnO core-shell inverse opal (CdS@ZnO-csIO) heterostructure photocatalyst for visible light driven hydrogen production [100]. In this composite photocatalyst, the utilization efficiency of photons is effectively enhanced and the separation of photogenerated charge carriers has been promoted, therefore improving the photocatalytic hydrogen evolution efficiency. As shown in Fig. 12a, they adopted the continuous ion layer adsorption reaction (SILAR) technique to uniformly decorate CdS nanoparticles onto the framework of ZnO inverse opal (ZnO-IO) and obtained CdS@ZnO-csIO composite photocatalyst. By adjusting the macropore size of ZnO-IO, the CdS@ZnO-csIO composites with different pore sizes were obtained (Figs. 12b–d), which can achieve the precise modulation of the photonic bandgap position in these inverse opal composites. Apparently, these CdS@ZnO-csIO inverse opal composites exhibit a uniform and stable inverse opal structure with macropore sizes and uniform pore walls. ZnO nanoparticles exhibit the same grain orientation at a certain scale, which facilitates the rapid migration of photogenerated electrons. Moreover, there is a tight contact interface between CdS and ZnO phases. The tight contact interface is also expected to result in efficient electron transfer and carrier separation efficiency. With increasing the size of the macropore, the reflection peaks shift towards longer wavelengths (Fig. 12e). The red-edge position of CdS@ZnO-csIO-230 and blue-edge position of CdS@ZnO-csIO-290 are located near the electronic band edge of CdS, providing great potential of slow-photon effect for promoting photocatalysis (Fig. 12f). The CdS@ZnO-csIO-290 photocatalyst displayed the enhanced hydrogen evolution rates of 48.7 mmol g^-1 h^-1 under simulated solar light irradiation (Fig. 12g). On account of VB-XPS and Mott Schottky characterization results, the CB and VB positions of ZnO and CdS components can be determined (Fig. 12h). And a type Ⅱ heterojunction charge transfer pathway was existed in the CdS@ZnO-csIO composite photocatalyst for effect separation of photogenerated electrons and holes, endowing the reduction and oxidation reactions at two different reaction positions (Fig. 12i). This work further confirms the synergy of the slow photon effect and heterostructure in the photocatalysis system.

Overall, slow photon photocatalysis has shown many significant advantages in water splitting, including increasing light absorption, prolonging carrier lifetime, facilitating charge transfer, and expanding visible-light response range. It provides an effective way to improve the efficiency of photocatalytic water splitting and achieve efficient utilization of solar energy. However, slow photon photocatalytic water splitting still faces many challenges in practical applications, including the complexity of inverse opal structure design and preparation, the stability issues of macropore, and limitations in cost and scale production.

4.2. Photocatalytic CO₂ conversion

The rapid consumption of fossil fuels has caused a significant rise in CO₂ levels in the atmosphere, resulting in severe environmental issues like an intensified greenhouse effect and rising sea levels, which pose a major threat to human life and industry [101,102]. Converting CO₂ into sustainable energy and clean fuel via photocatalysis technique can not only reduce the amount of CO₂ in the atmosphere, but also alleviate the global energy shortage [103]. Enhancing the adsorption and activation of CO₂ and H₂O molecules, increasing the efficiency of capturing and utilizing incoming photons, and encouraging the dissociation of photogenerated carriers are crucial factors in advancing photocatalytic CO₂ conversion [104,105].

In this context, Zhao's team introduced a photonic crystal composite photocatalyst (PC GDY-Cu/ZnO) that features a graphdiyne (DGY) loaded with a Z-type heterojunction. This catalyst demonstrates highly effective photocatalytic conversion of CO₂ into CH₄, thanks to its distinctive slow photon effect, pore properties, and structural advantages [106].

As shown in Fig. 13a, the PC GDY-Cu/ZnO composite photocatalyst can be prepared by inducing in-situ growth of GDY on the surface of PC Cu/ZnO. The introduction of GDY component and oxygen vacancies in ZnO can enhance the adsorption and activation capability of CO₂ molecules, thereby promoting the entire photoreduction of CO₂. By modifying the dimensions of the large pores in the photonic crystal composite (Fig. 13b and c), the increased absorption of visible light following the addition of GDY aligns closely with the slow photon region at the red edge (Fig. 13d). This alignment enhances the capture and use of visible light, thereby boosting photocatalytic activity. The schematic slow photons effect and electron transfer pathway of PC GDY-Cu/ZnO photocatalyst are depicted in Figs. 13e and f. The GDY component is uniformly dispersed on the PC GDY-Cu/ZnO framework, significantly enhancing the visible-light absorption in the range of 400–550 nm. By adjusting the position of the photonic bandgap, the red edge slow photons region of PC GDY-Cu/ZnO overlaps with the visible-light response range of GDY, greatly improving the refraction of photons on the cavity surface and thus enhancing the absorption and utilization of visible-light. Under the enhanced light harvesting of slow photons effect, the photoinduced electrons on ZnO and GDY are rapidly excited, and the electrons on ZnO are quickly transferred to GDY through the Z-type heterojunction, thus improving the separation efficiency of photoinduced electrons and holes. Moreover, the oxygen vacancies generated by hydrogen reduction on the surface of PC GDY-Cu/ZnO enhance the adsorption and activation ability of H₂O and CO₂ molecules. In summary, the photogenerated carrier separation efficiency of PC GDY-Cu/ZnO composite has been improved by 10.3 times, and the adsorption ability for CO₂ and H₂O molecules has been increased by 32.5 and 2.2 times in comparison to pure ZnO, respectively. Driven by slow photons effect, the increased CH₄ yield of PC GDY-Cu/ZnO photocatalyst is 33.67 µmol g^-1 h^-1, with a desirable product selectivity of 93.3%.

In addition, their group also reported a Ti-F bond bridged IL-CuCQDs-F/TiO₂ inverse opal composite photocatalyst (IO—CFTi) (Fig. 13g) [107], which can promote photoreduction of CO₂ by the synergy of slow photons effect and Ti-F bond. Firstly, EDTA (Cu) and ionic liquid [HOEtMIM] [BF₄] are used to prepare modified carbon quantum dots (IL-CuCQDs-F), which can grow on the surface of TiO₂ photonic crystal substance by virtue of the intense Ti-F bonds. The IL-CuCQDs-F can effectively extend light absorption from UV to the visible light region. Meanwhile, the Ti-F bond between the interface of IL-CuCQDs-F and TiO₂ photonic crystal facilitates the photogenerated electron transfer from TiO₂ to CO₂ molecules, thereby providing enough reducing potential in this composite system. More importantly, the electronic excitation wavelength region of IL-CuCQDs-F is overlapped with the blue-edge of TiO₂ photonic crystal (Fig. 13h), leading to an improved slow photon effect. Thus, the light harvesting is drastically elevated via increasing the effective optical path length for efficient photocatalytic activity.

The photocatalytic CO₂ reduction mechanism of the obtained IO—CFTi composite photocatalyst is depicted in Fig. 13i. Under UV light irradiation, IO CFTi can form photogenerated electron-hole pairs. The CB position of IL-CuCQDs-F (−1.6 eV) is more negative than that of TiO₂ (−0.6 eV), indicating that the photoinduced electrons of IL-CuCQDs-F can be transferred to the CB of TiO₂, thereby achieving effective separation of photogenerated electrons. Firstly, CO₂ molecules are adsorbed by IL-CuCQDs-F and activated into CO₂* intermediates. Then the adsorbed CO₂* is reduced by coupling photogenerated electrons and protons, generating the COOH* intermediates, which undergo further reduction by accepting electrons and protons, releasing CO* and H₂O. Finally, CO* is desorbed from the surface of photocatalyst and converted into CO product. A robust Ti-F bond was established between the interface of IL-CuCQDs-F and TiO₂ inverse opal, accelerating the separation and rapid migration of photogenerated electron-hole pairs. As a result, the synthesized IO CFTi presents an increased CO generation rate of 78.1 µmol g^-1 h^-1 via photoreduction of CO₂, being about 50 times higher than that of the pristine TiO₂ photocatalyst.

In a whole, efficient photocatalytic CO₂ reduction can be achieved by slow photon effect due to its obvious merits in increasing light absorption, prolonging carrier lifetime, facilitating interface charge transfer, and enhancing product selectivity. Nevertheless, some challenges should be considered and addressed before the practical applications of this technique, including the separation of reduced products, stability issues in photocatalysts and limitations in cost and scale production.

4.3. Photothermal catalytic CO₂ conversion

In comparison to single photocatalysis, thermal-assisted photocatalytic reactions (photothermal catalysis) present much higher CO₂ conversion performance. In particular, thermal energy plays a significant role in reducing the activation energy of CO₂ molecule in catalytic reactions [108]. Moreover, the fast separation and transport of photogenerated charges can be achieved with the existence of thermal energy. Photothermal catalysis can combine the advantages of thermochemistry and photochemistry, where the synergistic effect of light energy and heat energy promotes the total reaction progress [109]. In photothermal reactions, heat can be generated by photothermal materials (self-heating) and external auxiliary heating [110]. Recently, photocatalysts with full spectrum absorption to harvest solar energy and achieve photothermal effects under sunlight irradiation has received more and more attention.

Typically, Jiang and co-workers fabricated Pt nanoparticles loaded TiO₂ photonic crystal (Pt@HP-TiO₂) composite photothermal catalyst for efficient CO₂ conversion [111]. This composite catalyst couples the merits of slow photon effect, Pt-Ti hot electron transportation bridge, Pt nanoparticles induced SPR effect, and fast mass diffusion by rich pore structure. As depicted in Figs. 14a–d, the fabricated Pt@HP-TiO₂ composite possesses periodic macropores and interconnected pores in inner space, which can enhance irradiated light utilization due to the slow photon effect and present fast mass diffusion. The higher reaction temperatures of Pt-loaded TiO₂ compared with those of bare TiO₂ sample implied that the Pt nanoparticles triggered improved visible light absorption and elevated the reaction temperature. In comparison to B-TiO₂ and Pt@B-TiO₂ samples, Pt@HP-TiO₂ and HP-TiO₂ presented higher reaction temperatures (Figs. 14e–i), indicating that the 3DOM structured TiO₂ can harvest more UV light. The local heating induced by slow photons and SPR effects can accelerate the conversion of CO₂ molecules.

The X-ray absorption near-edge spectroscopy (XANES) spectra indicated that the intense electronic Pt-TiO₂ interaction existed in Pt@HP-TiO₂ due to electron transport between Pt particles and O atoms (Fig. 14j and k). The Fourier-transformed extended X-ray absorption fine structure (FT EXAFS) characterization (Fig. 14l) illustrated that the Pt-O bonds in Pt@HP-TiO₂ photothermal catalyst could promote charge shift between Pt nanoparticles and TiO₂ through the “Pt-Ti” interaction. Based on the aforementioned analysis, the authors put forth a proposed electron transfer pathway for Pt@HP-TiO₂ in both dark and light conditions (Fig. 14m). In the absence of light, the electrons are transferred from Pt to O via the “Pt-O” bond driven by the robust metal-support interaction between Pt and TiO₂. Meanwhile, the electrons in Ti migrate to Pt due to the difference in work function and Fermi level. Under visible light irradiation, the hot electrons overstep the Schottky barrier and are transferred from Pt to Ti via the efficient “Pt-Ti” hot electron transportation bridge. Additionally, the photogenerated electrons are excited from O to Ti. The efficient “Pt-Ti” hot electron transportation bridge is responsible for the remarkable photothermal catalytic reduction of CO₂ observed in this system. Finally, in-situ diffuse reflectance infrared Fourier transform spectra were performed to elucidate the reaction intermediates involved in the photothermocatalytic CO₂ reduction. The authors provided a comprehensive account of the CO₂ conversion pathway over Pt@HP-TiO₂ driven by the photothermal effect. The conversion of CO₂ follows a pathway: CO₂ → COO⁻ → HCHO⁻ → CH₃O⁻ → CH₄ (Fig. 14n). The high selectivity of CO₂ conversion into CH₄ on Pt@HP-TiO₂ induced by the hot electron transportation bridge and photothermal effect (Fig. 1l), as demonstrated in their work is a reliable and excellent result. The combination of the “Pt-Ti” hot electron transportation bridge, structure advantages, and photothermal effect, resulted in a CH₄ evolution rate of 29.68 µmol h⁻¹ g⁻¹ for Pt@HP-TiO₂ (Fig. 1h), which was significantly higher than that of bulk TiO₂ sample.

In summary, slow photons induced photothermal catalytic CO₂ conversion has significant advantages, which is considered as a highly promising approach for the resource utilization of CO₂. However, this technology also has problems such as complex thermal management, poor catalyst stability, unclear reaction mechanism, and high cost of noble metal utilization. Thus, the in-depth research on the design and preparation of photothermal catalysts with high stability and low cost should be conducted. Meanwhile, the advanced characterizations and theoretical calculation methods to explore the mechanism of the synergistic effect of light and heat is necessary to provide theoretical guidance for the optimization of catalysts and reaction systems.

5. Outlooks and challenges

In this review, the basic principles of the slow photon effect in inverse opal structured materials for photocatalysis and photothermal catalysis are comprehensively summarized. Meanwhile, the preparation methods, structure properties of inverse opal structured materials, as well as their advantages on photocatalytic and photothermal catalytic reactions are discussed. When compared to other types of porous structures, the inverse opal structure offers a greater specific surface area, numerous active surface sites, and superior performance in photoinduced charge migration. Additionally, it provides a distinctive slow photon effect, which is essential for boosting light absorption and enhancing photocatalytic activity. In addition, photocatalytic water splitting and CO₂ reduction under the effective slow photons effect were thoroughly summarized. In recent years, although significant research progress has been achieved in the field of photocatalysis using inverse opal structured materials, there is still a long way to go before commercial applications, and further research is still needed in many aspects.

(1) Photonic crystals with unique periodic structures can promote photon capture and control the interaction between light and materials, thereby improving the light utilization efficiency of semiconductor photocatalyst materials. However, their light response is mainly located at UV and visible-light regions and limits the solar full spectrum photocatalysis. It is meaningful to extend the light-harvesting range of photonic crystal catalysts to the infrared region. The fabrication of a donor-acceptor system to broaden the light absorption of materials should be a good solution to improve the quantum efficiency of solar energy conversion.

(2) As to the fabrication process, although the high activity of inverse opal materials has been prepared in the laboratory, their catalytic performance is still far from meeting industrial standards, and it is indeed very difficult to produce uniform inverse opal structures on an industrial scale. Meanwhile, a series of routes have been reported for the preparation of inverse opal structure, but the template method remains the main approach, thus the operation is cumbersome and the procedure is complex. In this regard, further research should be conducted on manufacturing large-scale and uniform inverse opal structures through facile template-free routes.

(3) Improving the stability of these pore structure for long-term photocatalytic reaction is one of the key challenges in the field of photocatalysis. By introducing stable coatings or functional groups on the surface of porous materials, the active components can be isolated from direct contact with corrosive substances, which is favorable for improving the stability of these pore structure.

(4) Finally, the in-situ spectroscopy techniques are suggested to track the dissociation and shift of photogenerated charges, to better understand the charge migration pathway and reaction mechanism. Meantime, it is suggested that machine learning could be employed to further investigate the relationship among slow photon effect, electronic structure, and charge shift, thereby offering theoretical views for photocatalytic sustainable energy evolution.

In the long run, slow photos induced efficient photocatalytic technology has undeniable potential under the "dual carbon" goal. With the continuous breakthrough and improvement of technology, it is expected to achieve large-scale application in the fields of energy production and environmental remediation, contributing an important force to global sustainable development. We sincerely hope that this critical review can provide researchers with fundamental knowledge and inspire innovative concepts in the synthesis of inverse opal materials for solar energy conversion.

This review presented the recent progress of for slow photons effect in photocatalysts for efficient photo/photothermocatalytic solar fuel production including water splitting and CO₂ conversion.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was financially supported by the National Natural Science Foundation of China (Nos. 22402044 and 22406169), Zhejiang Provincial Natural Science Foundation of China (Nos. LQ24E020011 and LQ24B070001). Zhejiang Education Department of China (No. Y202352478) and Basic Research Expenses of Zhejiang Gongshang University (No. QRK23025) were also thanked.