Hydrogen energy (H ₂ ) is a clean energy with high energy density, high conversion efficiency, wide range of application, green and pollution-free, which can be used as an efficient energy storage carrier and a potential support ^[97] for promoting China's green and low-carbon energy transformation and achieving the "double carbon" goal 。 In recent years, semiconductor photocatalytic hydrogen production has attracted extensive attention due to its simple working principle, eco-friendliness and cost-effectiveness, but it is still a great challenge to develop photocatalysts with suitable band structure, visible light response and efficient charge separation efficiency for hydrogen production. As an excellent semiconductor material, halide perovskites have the advantages of large absorption coefficient, adjustable band gap and long carrier lifetime, so they have attracted extensive research interest in the field of photocatalytic hydrogen production ^{[98, 99]} .

Photocatalytic cracking of hydroiodic acid (HI) is an ₂ effective method for the production of H, which requires only two electrons to decompose HI relative to the four-electron process of splitting water, so the overpotential of decomposing HI is lower, the reaction is easier to proceed, and has also been extensively studied in the iodine-sulfur cycle ^[100] . In the past few years, the H-production ₂ reaction of HI cleavage using perovskite materials as photocatalysts has been rapidly developed. For example, MAPbI ₃ (MA = CH ₃ NH ₃ ⁺ ) has also been widely studied as a hydrogen evolution photocatalyst, and our group has also reported that the bandgap funnel structure of the mixed halide pertitanium _3-x MAPbBrI _x is conducive to the photogenerated carriers from the inside to the surface, indicating that _3-x MAPbBrI _x The sample has excellent photocatalytic activity of hydrogen evolution ^[88] ; and Pt single-atom supported _3-x FAPbBrI _x (FA = CH ₂ (NH ₂ ) ₂ ⁺ ) with extremely high photocatalytic hydrogen production activity in saturated hydrohalic acid solution ^[73] . However, the perovskite powder prepared by the commonly used co-precipitation method is still an irregular large-size particle with a diameter of tens of microns, with a very low specific surface area and a limited active site, which is not conducive to photocatalytic hydrogen evolution reaction ^[101] 。 Therefore, the preparation of high-efficiency small-size halide perovskites has become a follow-up research idea. However, traditional perovskite nanoparticles and colloidal quantum dots are unstable and tend to aggregate in solution, limiting their reaction efficiency and stability. Recently, some researchers have reported photocatalytic applications that can improve the stability of perovskite nanocrystals in aqueous solution and maintain their photoelectronic properties by embedding perovskite nanoparticles into different types of ^{[102, 103]} matrices, such as CsPbBr ₃ @MOFs and FeO _{2 3} /rGO/CsPbBr ₃ Z-type heterojunctions have been reported to be used to enhance the separation of photogenerated carriers and thus improve photocatalytic performance ^{[104, 105]} . Among the many matrices, MCM-41 zeolite zeolite has a uniform pore structure, large pore volume, high specific surface area, and excellent thermal/chemical stability, and has been proven to be an ideal matrix for the dispersion and stabilization of catalytically active functional nanomaterials under harsh reaction conditions ^[106] . The small-sized perovskite nanoparticles are confined to the pore size of zeolite zeolite and can also have a high specific surface area, which enables the carriers to transport and diffuse rapidly. These properties make MCM-41 zeolite a good candidate matrix for the preparation of pyrolysis HI composite photocatalysts.

In this chapter, the bulk FAPbBr _3-x I _x (FPBI) was prepared by ion exchange method, and it was verified that the bandgap funnel of Br and I ions could promote the production of H ₂ by photocatalytic cracking of HI 。 On this basis, small-sized FPBI nanoparticles confined to MCM-41 zeolite were prepared by impregnation method. High-power transmission electron microscopy (HRTEM) images confirmed that FPBI nanoparticles of 2~4 nm were successfully confined to the pores of MCM-41. Under the condition of supporting Pt as a co-catalyst, the rate of hydrogen production by photocatalytic cracking of HI with the optimal activity of 40% MCM-41@FPBI was as high as 22.6 mmol g ⁻¹ h ⁻¹ . Therefore, small-size FPBI perovskite nanoparticles were prepared in the pore size of MCM-41 zeolite by spatially confined growth method, and efficient photocatalytic cleavage of HI to produce H was realized ₂ .

2.2 Experimental part

2.2.1 Experimental reagents

Details of all chemical reagents used in the experimental section of this chapter are listed in Table 2-1.

Table 2-1 Details of the chemical reagents used in the experimental part of this chapter

The experimental water is deionized water, and all reagent purity specifications are analytically pure, which are obtained by direct purchase, and have not been further purified and treated before use.

2.2.2 Sample preparation

Synthesis of FABr: Weigh 10 g CH ₄ N ₂ · ₄ CHo _{2 2} was dissolved in 14 mL of HBr, stirred until completely dissolved, and then evaporated in a sand bath at 100°C. The obtained yellow-white powder was washed with anhydrous ether, and then recrystallized with absolute ethanol, and the obtained white powder was dried in a vacuum drying oven at 60 °C for 12 h, and the FABr powder was obtained.

Preparation of FAPbBr ₃ block and preparation of saturated solution: 16 mL of HBr solution was measured, 4 mL of H ₃ PO ₂ was added as stabilizer, 2.25 g of FABr was added, 6.6 g of PbBr was added after stirring until ₂ it was completely dissolved, and 1 h of vigorous stirring was stirred to achieve equilibrium between the precipitated precipitate and the solution. After centrifugation, the orange-red precipitate and the solution were separated, and the pellet was dried in a vacuum drying oven at 60°C for 12 h to obtain bulk FAPbBr ₃ (FPB), and the colorless supernatant was ₃ the saturated solution of FAPbBr.

Preparation of FAPbBr _{3- x} I _x block and preparation of saturated solution: 16 mL of HBr solution was measured, 4 mL of H ₃ PO ₂ was added as stabilizer, a certain volume of HI (6%, 8%, 10%) was added, 2.25 g of FABr was added, and 6.6 g of PbBr ₂ was added after stirring until it was completely dissolved , stir vigorously for 1 h to achieve equilibrium between the precipitated precipitate and the solution. After centrifugation, the dark red precipitate and the solution were separated, and the pellet was dried in a vacuum drying oven at 60°C for 12 h to obtain bulk FAPbBr _3-x I _x (FPBI), and the colorless supernatant was the saturated solution of FAPbBr _3-x I _x .

Preparation of MCM-41@FPBI: Samples of MCM-41@FPBI were prepared by immersion method and ion exchange method, as detailed below. 0.1875 g FABr and 0.55 g PbBr ₂ were dissolved in 1 mL of DMF under ultrasound-assisted conditions. The above precursor solution was added dropwise to a mortar filled with MCM-41 powder, ground and mixed for 30 min, and the mortar was placed in an oven at 150°C for 12 h. By adjusting the amount of MCM-41, a series of MCM-41@FPB samples with different mass ratios of MCM-41 can be obtained. The obtained orange powder was added to the saturated solution of FPBI, stirred for 20 min, centrifuged, and washed with anhydrous ether to obtain MCM-41@FPBI samples, and the preparation method of FPBI photocatalyst supported by other zeolites was similar to the above method, that is, MCM-41 was replaced by other zeolites.

Preparation of Pt/FPBI and Pt/MCM-41@FPBI: Pt/FPBI (Pt/MCM-41@FPBI) was prepared by photoreduction method, as shown below. 100 mg of FPBI (MCM-41@FPBI) sample was taken in saturated solution, different amounts of H ₂ PtCl ₆ aqueous solution with a concentration of 0.1 mol L ⁻¹ were added dropwise and stirred evenly, placed under xenon lamp for 30 min, centrifuged to separate the precipitate and solution, and the pellet was dried in a vacuum drying oven at 60°C for 12 h to obtain Pt/FPBI (Pt/MCM-41@FPBI).

2.2.3 Characterization Testing

Crystal structure characterization was performed on a powder X-ray diffractometer (Bruker AXS D8) with Cu Kα as the radiation. The microscopic topography of the sample was characterized by a scanning electron microscope (SEM, Hitachi S-4800) equipped with an energy dispersive spectrometer (EDS). Focused ion beam (FIB)-assisted transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) were tested by field emission transmission electron microscopy (Thermo Scientific Talos F200X). X-ray photoelectron spectroscopy (XPS) was measured on an energy spectrometer (Thermo ESCALAB 250XI) and the binding energy was calibrated by the C1s peak (284.8 eV). The light absorption of the sample was characterized by a UV-Vis spectrophotometer (Shimadzu UV 2600i) with BaSO ₄ as a reference. The photoluminescence spectra (PL) of the sample were measured by spectrometer (FluoTime 300, PicoQuant) with an excitation wavelength of 468 nm. The nitrogen adsorption/desorption isotherm of the sample was characterized on the Kubo-X1000 high-performance micropore analyzer, and the Brunauer-Emmett-Teller surface area of the sample was calculated by formula. The amount of hydrogen generated was detected by GC-7806 gas chromatograph, and the chromatography was carried out with Ar as the carrier gas.

2.2.4 Sample performance test

2.2.4.1 Photocatalytic testing

Photocatalytic experiments were carried out in a self-made quartz amphora reactor for the photocatalytic decomposition of HI in the sample. Through the circulating cooling water system, the temperature of the whole reaction system is maintained at 15°C. In the photocatalytic decomposition HI experiment, a 300 W xenon lamp with AM 1.5 filter (CEL-HXF300, Beijing CEAULight, China) was used as the light source for photocatalysis. In a specific photocatalytic experiment, 100 mg of the sample was first evenly dispersed in a saturated solution of 20 mL. Before the reaction, the saturated solution was bubbled with high-purity Ar for 15 min. 0.5 mL of gas was periodically drawn from the reactor with a sampler and analyzed by an in-line gas chromatograph (GC-7806) equipped with a flame ionization detector (TCD).

2.2.4.2 Photoelectrochemical testing

The photoelectrochemical test was performed on a CHI660E electrochemical system that illuminated a 300 W xenon lamp (Perfectlight, PLS-SXE300D) in a solution of 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF ₆ ). Dichloromethane solution. Tests were performed using a standard three-electrode system, including a catalyst-coated FTO glass working electrode, an Ag/AgCl reference electrode, and a platinum sheet counter electrode. First, bulk FPBI and MCM-41@FPBI working electrodes were prepared by drip coating, and 30 mg of samples were sonicated and dispersed in a mixed solution containing 2 mL of absolute ethanol, 1 mL of isopropanol, and 20 μL of Nafion. The suspension was dropped on a cleaned FTO glass, dried at 80 °C to obtain a working electrode, and the transient photocurrent response and electrochemical impedance spectra of the samples were subsequently determined.

2.3 Results and Discussion

2.3.1 Catalyst characterization

Firstly, a series of FPBI samples with different Br and I contents (HI/HBr = 6%, 8% and 10%) were synthesized by simple co-precipitation method and ion exchange method, and after the optimal ratio of Br and I was determined by photocatalytic hydrogen evolution test (the optimal ratio is HI/HBr = 8%, unless otherwise specified, the following research objects are all samples of this ratio), a series of MCM-41@FPBI samples with different mass proportions were synthesized by immersion method and ion exchange method. The specific steps are in the sample preparation section. It can be seen from Figure 2-1a that the FPB powder material prepared by co-precipitation method has good crystallinity and no impurity peaks, and the position of the diffraction peaks is consistent with the literature report ^[107] . However, there were no other diffraction peaks in FPBI after halogen ion exchange, but the diffraction peaks shifted to a low angle with the increase of the volume ratio of HI/HBr in the saturated solution, which was caused by ^[74] the radius of iodine ions being larger than the radius of the replaced bromine ions , conforming to the Bragg equation. As shown in Figure 2-1b, there is only one broad peak of MCM-41, indicating that the crystallinity of MCM-41 is poor, and the XRD patterns of a series of MCM-41@FPBI with different MCM-41 contents prepared by immersion method and ion exchange method are consistent with the diffraction peak position of block FPBI, and no other diffraction peaks appear, indicating the successful preparation of the composite materials.

Ultraviolet-Vis diffuse reflectance spectroscopy (UV-Vis DRS) was used to further investigate the optical properties of the prepared photocatalysts. As can be seen from Figure 2-2a, the pure phase FPB has typical direct bandgap semiconductor light absorption characteristics, and its absorption edge is around 580 nm, which indicates that the material has good crystallinity and fewer defects ^[108] . After ion exchange, with the increase of iodine ion content, the absorption edge gradually redshifted, and tailing appeared after 600 nm, which can be attributed to the fact that after ion exchange, the substitution of iodine ions for bromine ions gradually increased, and at the same time, iodine ions will also be enriched on the particle surface to a certain extent, resulting in photoabsorption tailing, and a similar phenomenon ^[88] has also been reported in mixed halogen perovskite thin film materials 。 In addition, the light absorption capacity of FPBI is enhanced compared with that of FPB pure phase. And with the change of iodine ion content, the position of the absorption edge corresponding to FPBI also changes. The bandgap Eg of semiconductor materials can be calculated by diffuse reflection spectroscopy through Trafig's formula, that is, the light absorption energy hv is used as the abscissa, (ahv)1/m is the ordinate to plot, and the intercept of the corresponding linear tangent in the x-axis is the semiconductor band gap, because the perovskite materials are direct bandgap semiconductors, the m value is 1/2 ^[109] 。 The absorption edges of FPB and FPBI are around 580 nm and 630 nm, respectively, so the corresponding bandgap widths (Eg) are 2.18 eV and 1.98 eV, respectively (Figures 2-2b). In conclusion, the displacement of halogen ions can broaden the light absorption range of materials, which is conducive to the further utilization of light energy.

As shown in Figure 2-2c, the pure MCM-41 zeolite has no absorption in the range of 350~800 nm, after loading FPBI in the MCM-41 zeolite, and with the increase of the MCM-41 loading ratio, the absorption edge is gradually blueshifted, which can be attributed to the quantum size effect ^[95] caused to the smaller size nanoparticles 。 The absorption edge of 40% MCM-41@FPBI is located at 612 nm, and its corresponding band gap is 2.11 eV. Subsequently, VB-XPS spectroscopy was performed on 40% MCM-41@FPBI, and the results are shown in Figure 2-3, and the contact potential obtained by extrapolating the curve is 1.27 eV. According to the formula ^[110] : E _NHE = Φ + X − 4.44 (X: extrapolated contact potential; Φ: instrumental work function, here 4.2 eV), the valence band potential (E _VB ) of 40% MCM-41@FPBI can be calculated to be 1.03 eV (vs. NHE）。 From the equation E _CB = E _VB − Eg, the conduction band position (E _CB ) of 40% MCM-41@FPBI is −1.08 eV.

Next, the morphology of the prepared photocatalyst was characterized, and it can be seen from the SEM image that the FPB is an irregular block with a size of about 30 μm and a smooth particle surface (Fig. 2-4a), while the surface of the particles becomes significantly rough after the FPBI generated by halogen ion exchange (Fig. 2-4b). In order to further investigate the loading state of FPBI in the composite MCM-41@FPBI, 30% MCM-41@FPBI, 40% MCM-41@FPBI and 50% MCM-41@FPBI samples were analyzed by transmission electron microscopy (TEM). As can be seen from the scanning transmission microscopy (STEM) images under the high-angle annular darkfield mode (HAADF) in Figure 2-5a-c, the regular pore structure of the MCM-41 zeolite with a pore size of about 3 nm is about 3 nm, and the vast majority of FPBI in these three samples exist in the form of small particles of about 2~4 nm and are uniformly dispersed on MCM-41. In addition, it can be seen that the regular pore structure of the zeolite can still be maintained after the FPBI is loaded. The above results show that the FPBI perovskite particles in the MCM-41@FPBI composites are much smaller in size than the bulk FPBI (average particle size of about 30 μm) (Fig. 2-4b). However, with the increase of FPBI load, the number of perovskite particles on the surface of MCM-41 also increases. As shown in Figure 2-5a, the high content of FPBI in 30% MCM-41@FPBI causes some FPBI to aggregate on the surface of MCM-41 to form some larger particles. In Figures 2-5b and c, the small size of FPBI nanoparticles can be well confined to the MCM-41 channel when the zeolite content is high (40% MCM-41@FPBI and 50% MCM-41@FPBI). The STEM element distribution plot in Figure 2-5e further shows that the Pb, Br, and I elements are evenly distributed on MCM-41, indicating that FPBI nanoparticles are evenly distributed in MCM-41. The HRTEM image in Figure 2-6 clearly shows the lattice fringes of FPBI in MCM-41@FPBI, with the plane spacing of the (210) plane of FPBI being slightly larger than that of FPB (0.27 nm) at 0.28 nm, due to the partial substitution of Br ions by the I ion with a larger ionic radius ^[111] .

In order to further explore the binding between MCM-41 and FPBI, we investigated the chemical states of surface elements in the prepared FPBI and 40% MCM-41@FPBI by X-ray photoelectron spectroscopy (XPS). Figure 2-7a shows the XPS spectra of FPBI and 40% MCM-41@FPBI samples, with characteristic peaks of C 1s, N 1s, Pb 4f, Br 3d, and I 3d detected in FPBI, and characteristic peaks of C 1s, N 1s, Pb 4f, Br 3d, I 3d, Si 2p, and O 1s in 40% MCM-41@FPBI. Figure 2-7b shows the high-resolution XPS spectra of Pb 4f, with the two peaks observed at 138.2 eV and 143.1 eV belonging to Pb 4f _7/2 and Pb 4f _5/2 , respectively. The high-resolution XPS spectra of Br 3d are shown in Figure 2-7c, with the two peaks at 68.1 eV and 69.1 eV corresponding to the characteristic peaks of Br 3d _5/2 and Br 3d _3/2 , respectively. In addition, Figures 2-7d show the high-resolution XPS spectra of I 3d, where the two peaks at 618.7 eV and 630.3 eV belong to I 3d _5/2 and I 3d _3/2 ^[88] 。 As can be seen from Figure 2-7, there is no significant shift in the peak positions of Pb 4f, Br 3d, and I 3d in the XPS spectra of 40% MCM-41@FPBI samples compared to the XPS spectra of bulk FPBI. This result indicates that there is no chemical interaction between MCM-41 and FPBI, and that in 40% of MCM-41@FPBI, MCM-41 only serves as a template for the spatially confined growth of FPBI particles.

In addition, the specific surface area and pore size distribution of bulk FPBI, MCM-41 zeolite and a series of MCM-41@FPBI photocatalysts were characterized to further study the confined growth of FPBI in zeolite. The N ₂ adsorption/desorption isotherm of each sample is shown in Figure 2-8a, and the specific surface area of the MCM-41 zeolite is as high as 849.4 mg ^{2 −1} . After loading FPBI, the specific surface area of the MCM-41@FPBI begins to decrease. When the loading of FPBI was increased from 50 wt% to 70 wt%, the specific surface area of the sample decreased from 373.1 to 158.1 mg ^{2 −1} , correspondingly (Tables 2-2). Despite this, the specific surface area of 40% MCM-41@FPBI (272.4 mg ^{2 −1} ) was still higher than that of FPBI (2.2 mg ^{2 −1} ) large, which is conducive to photocatalytic reactions. Figure 2-8b shows the pore size distribution of the above samples, and the Barret-Joyner-Halenda model analysis shows that the pore size distribution of a series of MCM-41@FPBI catalysts prepared is 2~10 nm. This result also demonstrated that the loading process had little effect on the pore size distribution of MCM-41 zeolite. The average pore size of the prepared sample is about 2~4 nm, which is consistent with the results observed in the above TEM images (Figures 2-5).

2.3.2 Photocatalytic performance test

The photocatalytic hydrogen production performance of FPB, FPBI, and MCM-41@FPBI composites was tested in a pre-formulated saturated solution under simulated sunlight (AM 1.5) irradiation conditions. As can be seen from Figure 2-9a, the photocatalytic hydrogen evolution activity of the ion-exchanged bulk FPBI is higher than that of the pure-phase FPB under the same loading of 1% Pt as a co-catalyst. Among them, the HI/HBr = 8% 1% Pt/FPBI (unless otherwise specified) with the best catalytic activity had an H ₂ generation rate of 7.4 mmol g ⁻¹ h ⁻¹ and a pure phase block 1% Pt/FPB (2.7 mmol g ⁻¹ h ⁻¹ ), which is due to the gradual band gap caused by the gradient of Br and I ion concentration gradients from the inside to the surface of the FPBI perovskite particles, which creates a funnel-shaped channel ^[88] that accelerates the transport of carriers to the catalyst surface 。 In order to further improve the photocatalytic hydrogen evolution activity of FPBI, small-size FPBI nanoparticles were prepared by spatially confined growth of FPBI in the pore size of MCM-41 zeolite by immersion method and ion exchange method. As shown in Figure 2-9b, the pure MCM-41 zeolite does not have photocatalytic hydrogen production activity, and the prepared MCM-41@FPBI samples have better photocatalytic hydrogen production activity than bulk FPBI, and the 1% Pt/40% MCM-41@FPBI sample has the best H ₂ generation rate (22.6 mmol g ⁻¹ h ⁻¹ ) compared to the bulk 1% Pt/FPBI (7.4 mmol g). ⁻¹ h ⁻¹ ), because the spatially confined growth of FPBI in MCM-41 can effectively reduce the size of FPBI, increase its specific surface area, and provide more reactive sites for photocatalytic hydrogen evolution reaction. After finding the optimal ratio of HI/HBr to zeolite, we then optimized the loading of the cocatalyst Pt. As shown in Figure 2-9c, the optimal cocatalyst Pt loading for a 40% MCM-41@FPBI catalyst is 1%. Next, we used the same preparation method and test conditions to prepare and test the hydrogen evolution activity of HI after photocatalytic cracking after FPBI loading with other zeolites, as shown in Fig. 2-9d, the photocatalytic activities of different zeolite @FPBI were 8.8, 13.1, and 12.3 mmol g ⁻¹ h ⁻¹ , respectively, which were better than those of bulk FPBI (7.4 mmol g ⁻¹ h ⁻¹ These results indicate that the spatially confined growth of FPBI in molecular sieves is a general method to improve the photocatalytic hydrogen evolution capacity. The 40% MCM-41@FPBI was compared with the recently reported photocatalytic cracking HI hydrogen production activity (Table 2-3), and its photocatalytic hydrogen production ability was excellent. Next, we tested the cycling stability of the composites, as shown in Figure 2-10a, and after five cycles (2 h/time), the hydrogen evolution activity of the 1% Pt/40% MCM-41@FPBI sample was slightly reduced. Subsequently, XRD, SEM, TEM and XPS were used to characterize the photocatalyst after cycling. The results show that the position of the characteristic peaks in the XRD spectrum (Fig. 2-10b) and XPS spectra (Fig. 2-11) does not change, indicating that the material has good stability. The transmission electron microscopy images (Figs. 2-10c and d) show that some bulk FPBI precipitates after cyclic testing, but FPBI nanoparticles still remain in the zeolite, because some of the FPBI nanoparticles on the surface of the zeolite are not tightly bonded, and the FPBI nanoparticles fall off from the zeolite during ion exchange with the saturated solution during the reaction, and after losing the restriction of the zeolite, the dynamic dissolution crystallization equilibrium in the saturated solution leads to the increase in the size of the FPBI particles.

2.3.3 Discussion on photocatalytic mechanism

Steady-state photoluminescence (PL) spectroscopy was used to test the internal separation of the carriers in the sample. As shown in Figure 2-12a, the bulk FPBI has a distinct emission peak at 630 nm, which is due to the recombination of electrons and holes. The PL intensity of 40% MCM-41@FPBI is significantly lower than that of FPBI, which confirms that the small size of FPBI nanoparticles can shorten the carrier transmission distance and achieve efficient photogenerated carrier separation, thereby reducing the recombination luminescence of carriers and enhancing the performance ^[116] of photocatalytic hydrogen evolution 。 In addition, the emission peak of the 40% MCM-41@FPBI shifted to a shorter wavelength compared to the bulk FPBI, due to the reduction in particle size, consistent with the blue-shifted absorption band observed in the UV-visible diffuse reflection results (Figures 2-2c). In addition, the photogenerated carrier separation performance of the prepared catalyst was further evaluated by photoelectrochemical testing. In Figure 2-12b, the transient photocurrent response of 40% MCM-41@FPBI is greater than that of bulk FPBI, indicating that photogenerated carriers with 40% MCM-41@FPBI can achieve efficient separation. In addition, Figure 2-12c shows the electrochemical impedance curves for both samples. In general, the diameter of the semicircular arc is positively correlated with the value of the charge transfer resistance. Compared to FPBI, the radius of the electrochemical impedance curve of the 40% MCM-41@FPBI sample is smaller, indicating that it has good conductivity ^[117] . These results show that the presence of MCM-41 in 40% of the MCM-41@FPBI samples results in a significant reduction in the size of the FPBI, which means a decrease in the charge migration distance from the inside of the catalyst to the surface. The short charge transfer distance is conducive to the successful arrival of more photogenerated electrons and holes on the surface of the catalyst, which is conducive to improving the activity of photocatalytic hydrogen production.

Based on the above experimental results, we proposed the possible reasons for the increase of the activity of the 40% MCM-41@FPBI catalyst and the corresponding reaction mechanism. Compared with pure-phase FPB, the FPBI after ion exchange has excellent light absorption capacity, and the gradual band gap caused by the gradient change of halogen ion concentration gradient from the inside to the surface of the FPBI catalyst constructs a funnel-shaped channel suitable for the transport of photogenerated carriers from the inside to the outside ^[88] , which promotes the transport of carriers 。 At the same time, the small-sized FPBI nanoparticles in the pore size of MCM-41 have higher specific surface area and more active sites than bulk FPBI, and have a shorter carrier transport distance, which further improves the carrier transport efficiency and promotes the photocatalytic hydrogen evolution reaction. Under light conditions, the photogenerated electrons are excited from the valence band of the FPBI nanoparticles in 40% MCM-41@FPBI and transported to the surface of the catalyst, and the photogenerated electrons are immediately captured by Pt, and use this as the reaction site to ⁺ produce hydrogen gas by proton reduction with H in the saturated solution. In order to achieve electroneutral equilibrium, the photogenerated holes generated in the conduction band are transferred to the surface to participate in the oxidation reaction with I ⁻ in the saturated solution (3I ⁻ + 2h ⁺ → I ₃ ⁻ ). At the same time, I ₃ ⁻ is ₂ further reduced to I ⁻ by H ₃ PO, which in turn participates in the subsequent photocatalytic reaction. Therefore, 1% Pt/40% MCM-41@FPBI exhibited excellent photocatalytic hydrogen evolution reactivity.

2.4 Summary of this chapter

In this study, small-sized FPBI nanoparticles confined in the pore channels of MCM-41 zeolite were successfully fabricated by immersion and ion exchange methods. First, the concentration gradient of Br ions and I ions in perovskite particles forms a funnel-shaped channel that facilitates the transport of photogenerated carriers from the inside to the surface. At the same time, in the 40% MCM-41@FPBI composites, the small size of FPBI nanoparticles is confined to the pores of MCM-41, and the reduction of the size greatly increases the specific surface area of the catalyst, which provides more reaction sites for the photocatalytic hydrogen evolution reaction, and also shortens the transport distance of the carriers, further promotes the transport of photogenerated carriers, greatly improves the photocatalytic hydrogen evolution activity, and further improves the photocatalytic hydrogen evolution activity through the loading of Pt cocatalyst, and further improves the photocatalytic activity by 1% The hydrogen production activity of Pt/40% MCM-41@FPBI (22.6 mmol g ⁻¹ h ⁻¹ ) was 8.4 times higher than that of bulk 1% Pt/FPB (2.7 mmol g ⁻¹ h ⁻¹ ). The ion exchange method and confined growth method proposed in this chapter are of great significance for the preparation of small-size perovskite photocatalysts.