Porous Hybrid Materials Based on Mesotetrakis(Hydroxyphenyl) Porphyrins and TiO 2 for E ﬃ cient Visible-Light-Driven Hydrogen Production

: A series of highly robust nano-micro hybrid materials based on meso-tetra(4-hydroxyphenyl) porphyrins(M = H,Pd,Zn)andtitaniumdioxide(denotedasTHPP-TiO 2 ,THPP-Pd-TiO 2 ,andTHPP-Zn-TiO 2 ) have been prepared by a facile sol-gel method for the first time. When Pt nanoparticles are incorporated in these hybrids, Pt / THPP-Pd-TiO 2 achieves good H 2 production activity (2025.4 µ mol g − 1 h − 1 and 12.03 µ mol m − 2 h − 1 ), higher than that of Pt / THPP-Zn-TiO 2 (1239.8 µ mol g − 1 h − 1 and 7.46 µ mol m − 2 h − 1 ) and Pt / THPP-TiO 2 (576.8 µ mol g − 1 h − 1 and 4.02 µ mol m − 2 h − 1 ), owing to the different central metal ions in porphyrins. The best activity of Pt / THPP-Pd-TiO 2 would be attributed to the two-center catalysis from coordination Pd metal ions and Pt nanoparticles, while the higher activity of Pt / THPP-Zn-TiO 2 than Pt / THPP-TiO 2 could be ascribed to the more e ﬀ ective light harvesting and electron transfer between THPP-Zn and TiO 2 . In addition, the hybridized Pt / THPP-Pd-TiO 2 catalyst exhibits unattenuated hydrogen production stability even after recycling the experiment 10 times (cumulative turnover number of 5111 after 50 h), far superior to that of the surface-sensitized Pt / THPP-Pd / TiO 2 catalyst with analogous components, due to the more stable Ti-O bonds between four phenols in porphyrins and TiO 2 for the hybrid system. The present study provides a promising approach for constructing stable organic–inorganic hybrid systems with unique hierarchical structures for e ﬃ cient light absorption and electron transfer.


Introduction
Photocatalytic hydrogen generation over the semiconductor-based photocatalyst under solar light irradiation is a promising approach to solving current energy and environmental issues [1][2][3][4]. The active and stable photocatalysts are vital for highly efficient hydrogen production. As a good candidate photocatalyst, TiO 2 has been widely used owing to its low toxicity, abundance and stability against the light irradiation [5][6][7][8][9][10]. The photoactivity of TiO 2 is closely relevant to particle size, crystal form, surface morphology, and pore structure, and therefore the proper engineering of TiO 2 in photocatalytic system can greatly improve its activity. TiO 2 with a large surface area and micro/mesoporous structures can facilitate reactant/product transfer and enhance optical absorption efficiency, which shows a perfect performance in dye-sensitized TiO 2 photocatalysis [11][12][13]. However, pure TiO 2 has poor

Results and Discussion
The synthesis of the three porphyrins, THPP-TiO2, THPP-Pd-TiO2, THPP-Zn-TiO2 and their materials loaded with Pt NPs, and comparative Pt/THPP-Pd/TiO2 was included in the experimental section. The X-ray diffraction (XRD) patterns of the samples are shown in Figure 1

Results and Discussion
The synthesis of the three porphyrins, THPP-TiO 2 , THPP-Pd-TiO 2 , THPP-Zn-TiO 2 and their materials loaded with Pt NPs, and comparative Pt/THPP-Pd/TiO 2 was included in the experimental section. The X-ray diffraction (XRD) patterns of the samples are shown in Figure

Results and Discussion
The synthesis of the three porphyrins, THPP-TiO2, THPP-Pd-TiO2, THPP-Zn-TiO2 and their materials loaded with Pt NPs, and comparative Pt/THPP-Pd/TiO2 was included in the experimental section. The X-ray diffraction (XRD) patterns of the samples are shown in Figure 1. Five diffraction peaks at 25.53°, 38.12°, 48.22°, 54.18°, and 62.94° of Gel-TiO2, THPP-TiO2, THPP-Pd-TiO2 and THPP-Zn-TiO2 samples corresponded to (101), (103), (200), (105), and (204) planes of the crystalline anatase structure of titanium dioxide. The diffraction peaks of THPP-TiO2, THPP-Pd-TiO2, and THPP-Zn-TiO2 matched well to those of pure Gel-TiO2, indicating that the presence of porphyrins in the hybrid did not change the crystallinity of TiO2. The relatively regular diffraction peaks without the peaks of other impurities showed the high purity of these samples. The surface areas of the photocatalyst play an important role in the hydrogen production in the photocatalytic system. Nitrogen adsorption-desorption isotherms were measured at 77 K to obtain the Brunauer-Emmett-Teller (BET) surface areas and pore size distributions of the samples (Figures  S1 and S2 and Table S1). The BET surface areas of THPP-TiO2, THPP-Pd-TiO2, THPP-Zn-TiO2, Gel-TiO2, and, for comparison, P25-TiO2 were evaluated to be 143.4, 166.1, 168.3, 162.9, and 55.4 m 2 g −1 , respectively. The results suggested that the sol-gel method could increase the BET surface areas of TiO2. In addition, the micro/mesoporous structure, comprising both micropores (1~2 nm) and mesopores (3~5 nm), could be observed in porphyrin-TiO2 and Gel-TiO2 materials, in contrast to that (29 nm) of mesoporous P25-TiO2, which shortened the transport length and contributed to the faster diffusion of the electrons from the bulk to the surface, thus inhibiting the charge recombination and increasing the catalytic efficiency.
The morphologies of the as-prepared samples have been observed by the scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The SEM image in Figure 2A reveals The surface areas of the photocatalyst play an important role in the hydrogen production in the photocatalytic system. Nitrogen adsorption-desorption isotherms were measured at 77 K to obtain the Brunauer-Emmett-Teller (BET) surface areas and pore size distributions of the samples (Figures S1 and S2 and Table S1). The BET surface areas of THPP-TiO 2 , THPP-Pd-TiO 2 , THPP-Zn-TiO 2 , Gel-TiO 2 , and, for comparison, P25-TiO 2 were evaluated to be 143.4, 166.1, 168.3, 162.9, and 55.4 m 2 g −1 , respectively. The results suggested that the sol-gel method could increase the BET surface areas of TiO 2 . In addition, the micro/mesoporous structure, comprising both micropores (1~2 nm) and mesopores (3~5 nm), could be observed in porphyrin-TiO 2 and Gel-TiO 2 materials, in contrast to that (29 nm) of mesoporous P25-TiO 2 , which shortened the transport length and contributed to the faster diffusion of the electrons from the bulk to the surface, thus inhibiting the charge recombination and increasing the catalytic efficiency.
The morphologies of the as-prepared samples have been observed by the scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The SEM image in Figure 2A reveals that the porphyrin-TiO 2 presented a random aggregation of sheets which were closely associated with each other. The SEM elemental mapping of THPP-TiO 2 , THPP-Pd-TiO 2 , THPP-Zn-TiO 2 , indicated that the elements C and N for THPP-TiO 2 , C, N, and Pd for THPP-Pd-TiO 2 , and C, N, and Zn for THPP-Zn-TiO 2 dispersed homogenously throughout the whole hybrid material, respectively ( Figure 2B-D).
The TEM image of the porphyrin-TiO2 samples ( Figure 3A-C) showed that the hybrid materials were composed of nanospheres with an average diameter of 10 nm. The lattice fringe spacing was estimated to be 0.357 nm, in accordance with the (101) plane of anatase TiO2 ( Figure 3D-F). The selected area electron diffraction (SAED) results of the porphyrin-TiO2 samples confirmed the formation of the anatase phase in TiO2 ( Figure 3G-I).  UV-vis absorption spectra of three porphyrins in DMF solution and in solid-state, Gel-TiO 2 , and the hybrid materials based on porphyrins and TiO 2 are shown in Figure 4. All the porphyrins exhibited strong absorption in the visible region, ranging from 400 to 700 nm. The peak centered at about 420 nm was ascribed to the Soret absorption band originating from the a 1u (π)-e g (π) transition (S0-S2), while the absorption bands ranging from 500 to 700 nm were attributed to the weaker Q bands of a 2u (π)-e g (π) transitions (S0-S1). The absorption intensities of the three porphyrins were measured to be close to the values reported in the literature [38]. Compared with THPP, the absorption peak numbers of THPP-Pd and THPP-Zn in the region above 500 nm were reduced from 4 to 2, owing to the increased porphyrin ring symmetry after coordination with metal, which reduced the degree of molecular orbital splitting and improved the degeneracy. Although the maxmium absorption wavelengths (λ max ) of the Soret absorption bands were similar, the Q-band absorption wavelengths increased in the order of THPP-Pd < THPP < THPP-Zn. Particularly, THPP-Zn had the highest extinction coefficient and most obvious red-shift among those porphyrins, because it was easier for ZnPyd to occur the metal-to-ligand charge transfer (MLCT) owing to the weaker interaction between Zn 2+ (3d 10 ) and Py macrocycle ion without any single electron [39]. This better absorption and charge transfer properties in relation to Zn 2+ ions were favorable for the H 2 production reaction.  UV-vis absorption spectra of three porphyrins in DMF solution and in solid-state, Gel-TiO2, and the hybrid materials based on porphyrins and TiO2 are shown in Figure 4. All the porphyrins exhibited strong absorption in the visible region, ranging from 400 to 700 nm. The peak centered at about 420 nm was ascribed to the Soret absorption band originating from the a1u(π)-eg(π) transition (S0-S2), while the absorption bands ranging from 500 to 700 nm were attributed to the weaker Q bands of a2u(π)-eg(π) transitions (S0-S1). The absorption intensities of the three porphyrins were measured to be close to the values reported in the literature [38]. Compared with THPP, the absorption peak numbers of THPP-Pd and THPP-Zn in the region above 500 nm were reduced from 4 to 2, owing to the increased porphyrin ring symmetry after coordination with metal, which reduced the degree of molecular orbital splitting and improved the degeneracy. Although the maxmium absorption wavelengths (λmax) of the Soret absorption bands were similar, the Q-band absorption wavelengths increased in the order of THPP-Pd < THPP < THPP-Zn. Particularly, THPP-Zn had the highest extinction coefficient and most obvious red-shift among those porphyrins, because it was easier for ZnPyd to occur the metal-to-ligand charge transfer (MLCT) owing to the weaker interaction between Zn 2+ (3d10) and Py macrocycle ion without any single electron [39]. This better absorption and charge transfer properties in relation to Zn 2+ ions were favorable for the H2 production reaction.
When the porphyrins was doped into TiO2 by the sol-gel method, the hybrids exhibited a redshift and significantly extended the absorption profile, which most probably originated from the strong interaction between the -OH moieties of porphyrins and -Ti atoms of the TiO2, in agreement with our previous work [40]. For the Gel-TiO2, a strong absorption peak at 325 nm with an absorption edge of 450 nm corresponded to a 3.25 eV band gap. In contrast to Gel-TiO2 with a single optical band gap in Tauc plots ( Figure S3), the hybrid materials based on porphyrins and TiO2 showed multiple band gaps. These narrow band gaps (2.6-2.8 eV) could be caused by the doping of the porphyrins, while the wide band gap of 3.0-3.3 eV for the three porphyrin-based hybrids should come from the inorganic component TiO2, respectively.  To prove the feasibility of conducting photocatalytic H2 production, the first oxidation potentials of THPP, THPP-Pd, and THPP-Zn were measured by cyclic voltammograms (CVs) to obtain HOMO value (1.12, 1.08, and 1.10 V vs. NHE) ( Figure S4 and Table S2). The zero-zero excitation energies were estimated from the intersection of the absorption and emission spectra as 1.90, 2.19, and 2.05 eV for THPP, THPP-Pd, and THPP-Zn, respectively ( Figure S5 and Table S2). The LUMO levels of THPP, THPP-Pd, and THPP-Zn (−0.78, −1.11, and −0.95 V vs. NHE) were calculated from the HOMO values of the porphyrins and the zero-zero excitation energies. The valence-band potential of Gel-TiO2 was reported to be −0.5 V vs. NHE in our previous literature [41], which was more positive than the LUMO values of the three porphyrins and more negative than the redox potential of H + /H2, and thus facilitated electron injection from the porphyrins to TiO2 and favoured the proton reduction to realize photocatalytic H2 generation. Among the three porphyrins, the higher LUMO level of THPP-Pd than those of THPP and THPP-Zn implied that there is a stronger driving force to promote electron transfer from THPP-Pd to TiO2, which could accelerate the rate of hydrogen production. On the other When the porphyrins was doped into TiO 2 by the sol-gel method, the hybrids exhibited a red-shift and significantly extended the absorption profile, which most probably originated from the strong interaction between the -OH moieties of porphyrins and -Ti atoms of the TiO 2 , in agreement with our previous work [40]. For the Gel-TiO 2 , a strong absorption peak at 325 nm with an absorption edge of 450 nm corresponded to a 3.25 eV band gap. In contrast to Gel-TiO 2 with a single optical band gap in Tauc plots ( Figure S3), the hybrid materials based on porphyrins and TiO 2 showed multiple band gaps. These narrow band gaps (2.6-2.8 eV) could be caused by the doping of the porphyrins, while the wide band gap of 3.0-3.3 eV for the three porphyrin-based hybrids should come from the inorganic component TiO 2 , respectively.
To prove the feasibility of conducting photocatalytic H 2 production, the first oxidation potentials of THPP, THPP-Pd, and THPP-Zn were measured by cyclic voltammograms (CVs) to obtain HOMO value (1.12, 1.08, and 1.10 V vs. NHE) ( Figure S4 and Table S2). The zero-zero excitation energies were estimated from the intersection of the absorption and emission spectra as 1.90, 2.19, and 2.05 eV for THPP, THPP-Pd, and THPP-Zn, respectively ( Figure S5 and Table S2). The LUMO levels of THPP, THPP-Pd, and THPP-Zn (−0.78, −1.11, and −0.95 V vs. NHE) were calculated from the HOMO values of the porphyrins and the zero-zero excitation energies. The valence-band potential of Gel-TiO 2 was reported to be −0.5 V vs. NHE in our previous literature [41], which was more positive than the LUMO values of the three porphyrins and more negative than the redox potential of H + /H 2 , and thus facilitated electron injection from the porphyrins to TiO 2 and favoured the proton reduction to realize photocatalytic H 2 generation. Among the three porphyrins, the higher LUMO level of THPP-Pd than those of THPP and THPP-Zn implied that there is a stronger driving force to promote electron transfer from THPP-Pd to TiO 2 , which could accelerate the rate of hydrogen production. On the other hand, the first oxidation potentials of the porphyrins were more positive than that of the redox couple (0.84 ± 0.12 V vs. NHE) of triethanolamine as a sacrificial agent [42], beneficial to promoting dye regeneration and suppressing the recombination between the injected electrons and the dye cation radical ( Figure 5). For evaluating the influence of coordination metal ions on the photoactivity of the as-prepared materials, the photocatalytic experiments were performed in the visible light region with the Pt/THPP-TiO2, Pt/THPP-Pd-TiO2 and Pt/THPP-Zn-TiO2 catalysts using 0.5 wt% Pt particles as cocatalyst and TEOA as sacrificial electron donor. Figure 6 illustrated the H2 evolution plots of these hybrid systems and the corresponding data were summarized in Table S3. Pt/THPP-Pd-TiO2 catalyst exhibited higher hydrogen production activity (2025.4 μmol g −1 h −1 based on the catalyst mass and 12.03 μmol m −2 h −1 based on the surface area) than those of Pt/THPP-Zn-TiO2 (1239.8 μmol g −1 h −1 and 7.46 μmol m −2 h −1 ) and Pt/THPP-TiO2 (576.8 μmol g −1 h −1 and 4.02 μmol m −2 h −1 ). The result indicated that coordinated Pd metal ions play an essential role in the photoactivity of Pt/THPP-Pd-TiO2. For the known photochemical molecular devices (PMDs), the N^N-chelated NnMX2 (M = Pd 2+ , Pt 2+ or Rh 2+ , X = Cl − , Br − or I − , n = 2 or 4) coordination polyhedron was commonly employed as a catalyst centre, in which the dissociation of the terminal X − anions during the photocatalytic process was found [43,44]. It was also reported that Pd(Py)4 motifs possessing a PdN4 environment with four monodentate pyridyl ligands could be used as the catalytic centres for photocatalytic H2 evolution, where the breaking of the Pd-N bond and the formation of hydride intermediate may happen [45,46]. Thus, we deduced that the higher amount of H2 exhibited by the Pt/THPP-Pd-TiO2 hybrid may be attributed to the two-center catalysis from coordination Pd metal ions of porphyrin dye and Pt nanoparticles. In addition, the better catalytic activity of Pt/THPP-Zn-TiO2 than Pt/THPP-TiO2 could be attributed to the more effective light harvesting and charge transfer between THPP-Zn and TiO2 mentioned above.The apparent quantum yields (AQYs) of corresponding samples were measured and followed the order of Pt/THPP-Pd-TiO2 > Pt/THPP-Zn-TiO2 > Pt/THPP-TiO2 (Table S4), in consistance with the results of their photocatalytic H2 production.
In order to confirm the efficiencies of the photocatalysts and their stability under UV light irradiation, control experiments of the three catalysts under complete UV-vis irradiation have been For evaluating the influence of coordination metal ions on the photoactivity of the as-prepared materials, the photocatalytic experiments were performed in the visible light region with the Pt/THPP-TiO 2 , Pt/THPP-Pd-TiO 2 and Pt/THPP-Zn-TiO 2 catalysts using 0.5 wt% Pt particles as co-catalyst and TEOA as sacrificial electron donor. Figure 6 illustrated the H 2 evolution plots of these hybrid systems and the corresponding data were summarized in Table S3. Pt/THPP-Pd-TiO 2 catalyst exhibited higher hydrogen production activity (2025.4 µmol g −1 h −1 based on the catalyst mass and 12.03 µmol m −2 h −1 based on the surface area) than those of Pt/THPP-Zn-TiO 2 (1239.8 µmol g −1 h −1 and 7.46 µmol m −2 h −1 ) and Pt/THPP-TiO 2 (576.8 µmol g −1 h −1 and 4.02 µmol m −2 h −1 ). The result indicated that coordinated Pd metal ions play an essential role in the photoactivity of Pt/THPP-Pd-TiO 2 . For the known photochemical molecular devices (PMDs), the NˆN-chelated N n MX 2 (M = Pd 2+ , Pt 2+ or Rh 2+ , X = Cl − , Br − or I − , n = 2 or 4) coordination polyhedron was commonly employed as a catalyst centre, in which the dissociation of the terminal X − anions during the photocatalytic process was found [43,44]. It was also reported that Pd(Py) 4 motifs possessing a PdN 4 environment with four monodentate pyridyl ligands could be used as the catalytic centres for photocatalytic H 2 evolution, where the breaking of the Pd-N bond and the formation of hydride intermediate may happen [45,46]. Thus, we deduced that the higher amount of H 2 exhibited by the Pt/THPP-Pd-TiO 2 hybrid may be attributed to the two-center catalysis from coordination Pd metal ions of porphyrin dye and Pt nanoparticles. In addition, the better catalytic activity of Pt/THPP-Zn-TiO 2 than Pt/THPP-TiO 2 could be attributed to the more effective light harvesting and charge transfer between THPP-Zn and TiO 2 mentioned above.The apparent quantum yields (AQYs) of corresponding samples were measured and followed the order of Pt/THPP-Pd-TiO 2 > Pt/THPP-Zn-TiO 2 > Pt/THPP-TiO 2 (Table S4), in consistance with the results of their photocatalytic H 2 production. To check the contribution of the Pd coordination metal ions to H2 production, the control experiments with THPP-TiO2, THPP-Pd-TiO2 and THPP-Zn-TiO2 without Pt NPs for photocatalysis have been performed (Figure 7 and Table S3). It was found that the H2 generation rate of THPP-Pd-TiO2 without Pt NPs was 230.1 μmol g −1 h −1 (1.37 μmol m −2 h −1 ), while no obvious amount of H2 was detected for THPP-TiO2 and THPP-Zn-TiO2. These results supported that Pd coordination metal ions were conducive to the photocatalytic H2 evolution activity of THPP-Pd-TiO2. To study the difference of catalytic activity of the hybrid system Pt/THPP-Pd-TiO2 and the surface sensitization system Pt/THPP-Pd/TiO2, the control experiments of two catalysts were carried out ( Figure 8 and Table S5). In the initial 5 h light irradiation, two catalysts with 0.5 wt% Pt loading and approximate porphyrin amounts exhibited the same H2 evolution activity. However, in second and third run, H2-production activity of hybrid Pt/THPP-Pd-TiO2 were rising from initial 2025.4 to 2755.9 μmol g −1 h −1 (12.03 to 16.38 μmol m −2 h −1 ) in contrast with the continuous decline in the activity of surface sensitization Pt/THPP-Pd/TiO2 to 1110.3 μmol g −1 h −1 (6.81 μmol m −2 h −1 ) after the third run. During the photocatalytic reaction process, the THPP-Pd dye seriously detached from the Pt/THPP-Pd/TiO2 material under visible light irradiation, but the detaching phenomena were not contributed to the Pt/THPP-Pd-TiO2 catalysts even after they were used for 10 times and recycled. As shown in Figure 9 and Table S5, the Pt/THPP-Pd-TiO2 catalyst showed an outstanding stability without a decrease in photocatalysis. The TONH2 calculated as the number of moles of H2 divided by the number of moles of Pt was 5111 after 50 h, indicating the higher photocatalytic activity and stability of the In order to confirm the efficiencies of the photocatalysts and their stability under UV light irradiation, control experiments of the three catalysts under complete UV-vis irradiation have been performed ( Figure S6) and all catalysts exhibited higher H 2 production activities than those under visible light, owing to the effective UV-responses of the porphyrins and TiO 2 substrate. It was noteworthy that Pt/THPP-Zn-TiO 2 showed the highest activity under complete UV-vis irradiation. This result could be ascribed to the better absorption of THPP-Zn compared to THPP-Pd and THPP in the UV band, which could be observed in the UV-vis absorption spectra of THPP, THPP-Pd, and THPP-Zn (Figure 4).
To check the contribution of the Pd coordination metal ions to H 2 production, the control experiments with THPP-TiO 2 , THPP-Pd-TiO 2 and THPP-Zn-TiO 2 without Pt NPs for photocatalysis have been performed (Figure 7 and Table S3). It was found that the H 2 generation rate of THPP-Pd-TiO 2 without Pt NPs was 230.1 µmol g −1 h −1 (1.37 µmol m −2 h −1 ), while no obvious amount of H 2 was detected for THPP-TiO 2 and THPP-Zn-TiO 2 . These results supported that Pd coordination metal ions were conducive to the photocatalytic H 2 evolution activity of THPP-Pd-TiO 2 . To check the contribution of the Pd coordination metal ions to H2 production, the control experiments with THPP-TiO2, THPP-Pd-TiO2 and THPP-Zn-TiO2 without Pt NPs for photocatalysis have been performed (Figure 7 and Table S3). It was found that the H2 generation rate of THPP-Pd-TiO2 without Pt NPs was 230.1 μmol g −1 h −1 (1.37 μmol m −2 h −1 ), while no obvious amount of H2 was detected for THPP-TiO2 and THPP-Zn-TiO2. These results supported that Pd coordination metal ions were conducive to the photocatalytic H2 evolution activity of THPP-Pd-TiO2. To study the difference of catalytic activity of the hybrid system Pt/THPP-Pd-TiO2 and the surface sensitization system Pt/THPP-Pd/TiO2, the control experiments of two catalysts were carried out ( Figure 8 and Table S5). In the initial 5 h light irradiation, two catalysts with 0.5 wt% Pt loading and approximate porphyrin amounts exhibited the same H2 evolution activity. However, in second and third run, H2-production activity of hybrid Pt/THPP-Pd-TiO2 were rising from initial 2025.4 to 2755.9 μmol g −1 h −1 (12.03 to 16.38 μmol m −2 h −1 ) in contrast with the continuous decline in the activity of surface sensitization Pt/THPP-Pd/TiO2 to 1110.3 μmol g −1 h −1 (6.81 μmol m −2 h −1 ) after the third run. During the photocatalytic reaction process, the THPP-Pd dye seriously detached from the Pt/THPP- To study the difference of catalytic activity of the hybrid system Pt/THPP-Pd-TiO 2 and the surface sensitization system Pt/THPP-Pd/TiO 2 , the control experiments of two catalysts were carried out ( Figure 8 and Table S5). In the initial 5 h light irradiation, two catalysts with 0.5 wt% Pt loading and approximate porphyrin amounts exhibited the same H 2 evolution activity. However, in second and third run, H 2 -production activity of hybrid Pt/THPP-Pd-TiO 2 were rising from initial 2025.4 to Catalysts 2020, 10, 656 9 of 12 2755.9 µmol g −1 h −1 (12.03 to 16.38 µmol m −2 h −1 ) in contrast with the continuous decline in the activity of surface sensitization Pt/THPP-Pd/TiO 2 to 1110.3 µmol g −1 h −1 (6.81 µmol m −2 h −1 ) after the third run. During the photocatalytic reaction process, the THPP-Pd dye seriously detached from the Pt/THPP-Pd/TiO 2 material under visible light irradiation, but the detaching phenomena were not contributed to the Pt/THPP-Pd-TiO 2 catalysts even after they were used for 10 times and recycled. As shown in Figure 9 and Table S5, the Pt/THPP-Pd-TiO 2 catalyst showed an outstanding stability without a decrease in photocatalysis. The TON H2 calculated as the number of moles of H 2 divided by the number of moles of Pt was 5111 after 50 h, indicating the higher photocatalytic activity and stability of the hybrid materials than those of the common surface dye-sensitized system. The inorganic component TiO 2 in the hybrid materials was expected to be able to protect the THPP-Pd, and the stable linkages between THPP-Pd and TiO 2 were beneficial for efficient electron transfer from THPP-Pd* to TiO 2 . Thus, organic-inorganic components in the hybrid system, in other words, have a synergistic effect on the photocatalytic reaction.
Catalysts 2020, 10, x FOR PEER REVIEW 9 of 12 TiO2 in the hybrid materials was expected to be able to protect the THPP-Pd, and the stable linkages between THPP-Pd and TiO2 were beneficial for efficient electron transfer from THPP-Pd* to TiO2. Thus, organic-inorganic components in the hybrid system, in other words, have a synergistic effect on the photocatalytic reaction.

Conclusions
In summary, we have, for the first time, developed a series of unique hierarchical hybrid materials (THPP-TiO2, THPP-Pd-TiO2, and THPP-Zn-TiO2) by combining TiO2 and porphyrins through a sol-gel method, which possessed a highly accessible surface area as well as delivered excellent charge separation and transport efficiency due to the stable Ti-O linkage, resulting in efficient and persistent photocatalytic water splitting activities when loaded with Pt nanoparticles. The absorption spectra, electrochemical properties and charge transfer processes of those porphyrins were investigated to vary along with the central metal ions. The effects of central metal ions led to the catalytic properties of the hybrids for H2 production indicating a rising trend in the sequence of Pt/THPP-Pd-TiO2 > Pt/THPP-Zn-TiO2 > Pt/THPP-TiO2. Especially, Pt/THPP-Pd-TiO2 showed the best H2 production activity (2025.4 μmol g −1 h −1 and 12.03 μmol m −2 h −1 ) among the three hybrids under Catalysts 2020, 10, x FOR PEER REVIEW 9 of 12 TiO2 in the hybrid materials was expected to be able to protect the THPP-Pd, and the stable linkages between THPP-Pd and TiO2 were beneficial for efficient electron transfer from THPP-Pd* to TiO2. Thus, organic-inorganic components in the hybrid system, in other words, have a synergistic effect on the photocatalytic reaction.

Conclusions
In summary, we have, for the first time, developed a series of unique hierarchical hybrid materials (THPP-TiO2, THPP-Pd-TiO2, and THPP-Zn-TiO2) by combining TiO2 and porphyrins through a sol-gel method, which possessed a highly accessible surface area as well as delivered excellent charge separation and transport efficiency due to the stable Ti-O linkage, resulting in efficient and persistent photocatalytic water splitting activities when loaded with Pt nanoparticles. The absorption spectra, electrochemical properties and charge transfer processes of those porphyrins were investigated to vary along with the central metal ions. The effects of central metal ions led to the catalytic properties of the hybrids for H2 production indicating a rising trend in the sequence of

Conclusions
In summary, we have, for the first time, developed a series of unique hierarchical hybrid materials (THPP-TiO 2 , THPP-Pd-TiO 2 , and THPP-Zn-TiO 2 ) by combining TiO 2 and porphyrins through a sol-gel method, which possessed a highly accessible surface area as well as delivered excellent charge separation and transport efficiency due to the stable Ti-O linkage, resulting in efficient and persistent photocatalytic water splitting activities when loaded with Pt nanoparticles. The absorption spectra, electrochemical properties and charge transfer processes of those porphyrins were investigated to vary along with the central metal ions. The effects of central metal ions led to the catalytic properties of the hybrids for H 2 production indicating a rising trend in the sequence of Pt/THPP-Pd-TiO 2 > Pt/THPP-Zn-TiO 2 > Pt/THPP-TiO 2 . Especially, Pt/THPP-Pd-TiO 2 showed the best H 2 production activity (2025.4 µmol g −1 h −1 and 12.03 µmol m −2 h −1 ) among the three hybrids under visible light irradiation owing to two catalytic centers from coordination Pd metal ions of porphyrin dye and Pt nanoparticles. The superior activity of Pt/THPP-Zn-TiO 2 to Pt/THPP-TiO 2 could be attributed to the more effective optical absorption of THPP-Zn and electron transfer between THPP-Zn and TiO 2 . Comparatively, the hybrid Pt/THPP-Pd-TiO 2 exhibited superior stability to the analogous surface dye-sensitized Pt/THPP-Pd/TiO 2 system under similar photocatalytic conditions and an impressive TON of 5111 has been achieved through a ten-cycle photoreaction without a decrease for 50 h. The present results reveal that varying the central metal ions in porphyrins or conducting interface and microstructure modulation is a facile strategy to optimize the photogenerated charge transfer pathway and utilize the visible light of sunlight to significantly enhance the efficiency of the solar-to hydrogen conversion efficiency.

Conflicts of Interest:
The authors declare no conflict of interest.