Solar Driven Photocatalytic Activity of Porphyrin Sensitized TiO2: Experimental and Computational Studies

The absence of a secure long-term sustainable energy supply is recognized as a major worldwide technological challenge. The generation of H2 through photocatalysis is an environmentally friendly alternative that can help solve the energy problem. Thus, the development of semiconductor materials that can absorb solar light is an attractive approach. TiO2 has a wide bandgap that suffers from no activity in the visible spectrum, limiting its use of solar radiation. In this research, the semiconductor absorption profile was extended into the visible region of the solar spectrum by preparing porphyrin-TiO2 (P-TiO2) composites of meso-tetra(4-bromophenyl)porphyrin (PP1) and meso-tetra(5-bromo-2-thienyl)porphyrin (PP2) and their In(III), Zn(II) and Ga(III) metal complexes. Density functional theory (DFT) and time-dependent density functional theory (TD-DFT) calculations were performed on the porphyrins to gain insight into their electron injection capability. The results demonstrate that P-TiO2 systems merit further in-depth study for applications that require efficient photocatalytic H2 generation.


Introduction
Rapid growth in world population and increased industrial activities have led to a steady rise in global energy consumption [1]. According to the BP Statistical Review of World Energy, global energy consumption stands at 13.9 MTOE (Million Tonnes of Oil Equivalent). This figure is expected to nearly double by 2050 [1]. Currently,~85% of global energy running the transport and industrial sector is supplied by fossil fuels. This has led to increased greenhouse gas (GHG) emissions and substantial depletion of the most easily exploited carbon-based energy resources that could alternatively have been used to produce valuable chemicals [2,3]. The key to reducing problematic emissions (NO x , SO x, and GHGs) is to increase the efficiency of energy production and consumption while reducing the fossil fuel component in the energy mix [4,5]. A transition to hydrogen fuel is considered an ideal alternative to fossil fuels [4][5][6]. Hydrogen has potentially higher energy efficiency, high energy density and low generation of pollutants [7,8]. The world in the recent years has seen a steady transition from solid to liquid to gas fuels. Each successive source has increased the ratio of hydrogen to carbon. Hydrogen is an attractive chemical energy carrier that on reaction with oxygen, produces only water, thus a clean energy source.
Advanced oxidation processes (AOPs) such as photocatalytic hydrogen generation from sacrificial electron donors (SEDs) in aqueous media is one of the green sources being  UV-visible absorption spectra are one of the most informative spectroscopic methods in the context of porphyrin chemistry. PP1, PP2 and their metal complexes all exhibit absorption spectra characteristic of etio type porphyrins with an intense Soret (or B) band that lies between 418−434 nm and weaker Q bands in the 500−700 nm region (Figures 1 and 2).
Gouterman's four-orbital model [33] can be used to rationalize trends in the electronic excitation spectra of free base porphyrins and their metal complexes. The main trends in the optical and redox properties can be rationalized through a consideration of four frontier orbitals that are derived from the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) of a C 16 H 16 2− parent hydrocarbon perimeter. Michl introduced an a, s, −a and −s nomenclature ( Figure 3) for these orbitals depending on whether there is a nodal plane (a and -a) or large coefficients (s and -s) aligned with the y-axis [34]. For example, the HOMO−1, HOMO, LUMO and LUMO+1 of PP1 can be assigned as the a, s, −a and −s MOs, respectively ( Figure 3). The MOs derived from the HOMO and LUMO of the C 16 H 16 2− parent perimeter have M L values of ±4 and ±5, respectively. Four spin allowed one-electron transitions are possible within the four-orbital model. It is on this basis that Gouterman predicted the possibility of allowed B transition (∆M L = ±1) at high energy and a forbidden Q transition (∆M L = ±9) at lower energy [33,35]. Free base porphyrins (PP1 and PP2) have D 2h symmetry so there is a splitting of the main Molecules 2021, 26, 3131 4 of 15 electronic Q 00 transition and a Q 01 overtone into xand y-polarized components, so that four Q bands are observed [33,36]. However, metalloporphyrins have D 4h symmetry and hence have only two Q bands due to the main electronic transition and a vibrational overtone [33]. This reduction in the number of the Q bands is a characteristic feature of metalloporphyrins, further confirming the successful insertion of metal ions into the porphyrin core.  Gouterman's four-orbital model [33] can be used to rationalize trends in the electronic excitation spectra of free base porphyrins and their metal complexes. The main trends in the optical and redox properties can be rationalized through a consideration of four frontier orbitals that are derived from the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) of a C16H16 2− parent hydrocar-  Gouterman's four-orbital model [33] can be used to rationalize trends in the electronic excitation spectra of free base porphyrins and their metal complexes. The main trends in the optical and redox properties can be rationalized through a consideration of four frontier orbitals that are derived from the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) of a C16H16 2− parent hydrocar- tion is reflected in bathochromic shift of the Q and B bands [37,41]. Among the metallated porphyrins, indium complexes exhibited the largest shifts of 8−9 nm, a scenario that has been reported previously [42]. Similar observations were observed in previous studies [41,[43][44][45]. It is noteworthy that InPP1 and InPP2 were predicted to have the smallest HOMO−LUMO gap values in the geometry optimization calculations (Figure 3), which is consistent with the observed red shifts of the major spectral bands.

Photo-Physical Characterization of PP1, PP2 and Their Metal Complexes
The synthesized porphyrins exhibit two characteristic porphyrin emission peaks that arise from the Q00 and Q01 transitions [35,46,47]. The emission spectra of PP1, PP2 and their metal conjugates ( Figures S5 and S6) are generally similar but with differing maxima. The spectra of the metalloporphyrins contain less intense emission bands that are blue-shifted compared to those of the free bases. The ΦF values for the metalloporphyrins are lower than those of PP1 and PP2 (Table 2), due to the heavy atom ensuring strong spin-orbit coupling and enhanced intersystem crossing (ISC) [47][48][49]. Improved ISC populates the triplet states lowering fluorescence quantum yield. The ΦF values of ~0.01 for InPP1 and InPP2 are the lowest. The τF values for the metalloporphyrins were shorter than those of free base porphyrins (Table 2) due to the heavy atom effect. The experimental data provided in Table 1 demonstrate that there are only relatively minor shifts in the absorption bands of PP1 relative to those of tetraphenylporphyrin (H 2 TPP), since the only structural difference is the bromine atom at the para-position of the phenyl rings. Phenyl moieties have only a minimal influence on the electronic structure of the porphyrins, due to the unfavorable steric interaction between β hydrogens of the pyrrole ring and ortho hydrogens of the phenyl group. This positions the phenyl substituent orthogonal to the porphyrin core, thus resulting in an increased rotational barrier [32,37].
However, meso-5-bromothien-2-yl-substituted PP2 has spectral bands that are red shifted by 10 nm compared to the spectrum of PP1 (Table 1). Smaller 5-membered ring meso-substituents such as the thienyl groups do not have ortho hydrogens, thus making it easier for the meso-aryl rings to adopt a coplanar orientation with the porphyrin core [38]. This enhances the resonance interactions with the tetrapyrrole ring, which results in a significant red shift of the Q and B bands of PP2 [39,40]. This trend is also observed in the spectra of the metal complexes of PP2 and PP1 (Table 1).
The B bands for the metal derivatives of PP1 and PP2 are red shifted by 2-9 nm compared to those of their respective free base monomers (Table 1). Metal insertion can slightly alter the π electron delocalization of the porphyrin ring. Co-ordination of metals at the porphyrin core improves their structural symmetry since the inner NH protons tilt the pyrrole ring out of plane [34,38]. Therefore, greater resonance interaction with the meso-substituents is observed due to the reduced steric hindrance. The extended conjugation is reflected in bathochromic shift of the Q and B bands [37,41]. Among the metallated porphyrins, indium complexes exhibited the largest shifts of 8−9 nm, a scenario that has been reported previously [42]. Similar observations were observed in previous studies [41,[43][44][45]. It is noteworthy that InPP1 and InPP2 were predicted to have the smallest HOMO−LUMO gap values in the geometry optimization calculations (Figure 3), which is consistent with the observed red shifts of the major spectral bands. Table 1. The UV-visible absorption data for PP1, PP2 and their metal complexes in toluene. Please note: the assignment of the Q bands (Q yo1 , Qy 00 , Q xo1 , Q x00 ) is shown in Figure 2.

Photo-Physical Characterization of PP1, PP2 and Their Metal Complexes
The synthesized porphyrins exhibit two characteristic porphyrin emission peaks that arise from the Q 00 and Q 01 transitions [35,46,47]. The emission spectra of PP1, PP2 and their metal conjugates (Figures S5 and S6) are generally similar but with differing maxima. The spectra of the metalloporphyrins contain less intense emission bands that are blue-shifted compared to those of the free bases. The Φ F values for the metalloporphyrins are lower than those of PP1 and PP2 (Table 2), due to the heavy atom ensuring strong spin-orbit coupling and enhanced intersystem crossing (ISC) [47][48][49]. Improved ISC populates the triplet states lowering fluorescence quantum yield. The Φ F values of~0.01 for InPP1 and InPP2 are the lowest. The τ F values for the metalloporphyrins were shorter than those of free base porphyrins (Table 2) due to the heavy atom effect. Table 2. Photophysical parameters of PP1, PP2 and their metal complexes in toluene. Q yo1 , Qy 00 , Q xo1 , Q x00 ) is shown in Figure 2. The equations for the determination of fluorescence quantum yield (Φ F ) and lifetime (τF), the radiative (k r ) and non-radiative (k nr ) rate constant are provided in Section 3.4.

Density Function Theory Calculations
Density functional theory (DFT) computations were performed to help identify suitable compounds for photosensitization and compute the photophysical properties of the synthesized porphyrins for comparison with experimental values. The spectral bands observed in the visible region of the absorption spectra of PP1, PP2 and their metal complexes can be readily assigned to the forbidden Q and allowed B transitions of Gouterman's 4-orbital model on the basis of the TD-DFT calculations ( Figure 4 and Table 3). Only the Q and B bands are predicted to lie in the visible region when a systematic over-estimation of the band energies is taken into account, since other π-MOs are well separated in energy terms from the a, s, −a, and −s MOs (Figure 3). The calculated Q and B band maxima for meso-5-bromothien-2-yl-substituted PP2, ZnPP2, GaPP2, and InPP2 are red shifted compared to those of meso-phenyl-substituted PP1, ZnPP1, GaPP1, and InPP1 ( Figure 4 and Table 3), thus exhibiting the same trend observed in the experimental values (Figures 1 and 2, Table 1). The major spectral bands of InPP1 and InPP2 are more red shifted than those of Zn(II) and Ga(III) complexes (Table 3). This is also consistent with the trend observed experimentally (Figures 1 and 2, Table 1 Table 3. Table 3. Electronic excitation bands of the synthesized porphyrins at the TD-DFT level of theory using the CAM-B3LYP functional with either SDD or 6-31G(d) basis sets.  Table 3.   Figure 3 compares the relative HOMO and LUMO energies of the PPs studied here with the energies of the valence and conduction band (VB and CB respectively) edges of TiO 2 . It can be observed from the DFT calculations ( Figure 3) that the HOMO and LUMO energy levels of the photosensitizers fit the requirements for an efficient photosensitizer [50], since the LUMO energies of the photosensitizers lie between −2.48 and −3.34 eV, well above the TiO 2 CB (ca. −4.0 eV), implying that efficient electron injection from the photoexcited sensitizer into the TiO 2 CB should be viable [51]. It was anticipated that the stabilization of the frontier π-MOs associated with the incorporation of Ga(III) and In(III) ions ( Figure 3) would better match the TiO 2 CB due to the electron-withdrawing effect of the trivalent ions and that this would be further enhanced by the electron-withdrawing properties of the meso-5-bromothien-2-yl rings of PP2.

Characterization of the Porphyrin-TiO 2 (P-TiO 2 ) Photocatalysts by Diffuse Reflectance Spectroscopy
UV-visible diffuse reflectance spectra ( Figure 5) provide evidence for the presence of porphyrins on the TiO 2 surface. TiO 2 does not exhibit any absorbance in the visible region at >400 nm, while the spectra of the P-TiO 2 photocatalysts contain the characteristic B and Q bands of the porphyrins. The presence of these bands demonstrates that porphyrins were successfully loaded onto the surface of TiO 2 . The photocatalysts are therefore anticipated to exhibit a significantly broader absorption range. To infer the amount of porphyrin loaded in each system, the intensity of the porphyrins Soret bands were compared using the normalized diffuse reflectance spectra (Figures S7 and S8). The analyses suggest that in general PP1 porphyrins attach better to the surface of TiO 2 than their PP2 counter parts. Within the both PP1 and PP2 series, the attachment follows the trend Ga-PP > PP~In-PP >> Zn-PP. porphyrins on the TiO2 surface. TiO2 does not exhibit any absorbance in the visible region at >400 nm, while the spectra of the P-TiO2 photocatalysts contain the characteristic B and Q bands of the porphyrins. The presence of these bands demonstrates that porphyrins were successfully loaded onto the surface of TiO2. The photocatalysts are therefore anticipated to exhibit a significantly broader absorption range. To infer the amount of porphyrin loaded in each system, the intensity of the porphyrins Soret bands were compared using the normalized diffuse reflectance spectra (Figures S7 and S8). The analyses suggest that in general PP1 porphyrins attach better to the surface of TiO2 than their PP2 counter parts. Within the both PP1 and PP2 series, the attachment follows the trend Ga-PP > PP~In-PP >> Zn-PP. Figure 5. Diffuse Reflectance spectra of the P-TiO2 photocatalysts.

Photocatalytic Activity of the Photocatalysts
The activity of the P-TiO2 photocatalysts in the generation of hydrogen in aqueous solution was evaluated for each of the photocatalysts. Eight photocatalysts, namely: ZnPP1-TiO2, GaPP1-TiO2, GaPP2-TiO2, InPP2-TiO2 ZnPP2-TiO2, InPP1-TiO2, PP1-TiO2, and PP2-TiO2, were assessed in 1% methanol solution. All the photocatalysts exhibited the ability to generate H2 (Figure 6), including bare TiO2. PP1-TiO2, PP2-TiO2, and GaPP2-TiO2 showed a superior photocatalytic activity rate for H2 generation relative to bare TiO2. In the case of native PPs, H2 production from PP1 is at least 30% higher than that found for PP2. Although this could be related to slightly higher PP1 loading, differences in the porphyrin loadings cannot explain the decrease of in production of H2 when using metalloporphyrins, particularly Ga-PP and In-PP that show higher to similar loadings compared to their native PPs. Therefore, the variation in the activity of the systems could be a result of different electron dynamics between the surface of the dye and TiO2. Freebase porphyrins (PP1 and PP2) had better photosensitization than their metal complexes due to more electronic coupling with the TiO2 surface. Both PP1 and PP2 being metal free are flexible with fairly non-planar conformations thus can afford multiple adsorption configurations due to variable dihedral angles. The freebase cavity can co-ordinate more effectively with TiO2 by flat attachment favoring electron injection in comparison to the metal Figure 5. Diffuse Reflectance spectra of the P-TiO 2 photocatalysts.

Photocatalytic Activity of the Photocatalysts
The activity of the P-TiO 2 photocatalysts in the generation of hydrogen in aqueous solution was evaluated for each of the photocatalysts. Eight photocatalysts, namely: ZnPP1-TiO 2 , GaPP1-TiO 2 , GaPP2-TiO 2 , InPP2-TiO 2 ZnPP2-TiO 2 , InPP1-TiO 2 , PP1-TiO 2 , and PP2-TiO 2 , were assessed in 1% methanol solution. All the photocatalysts exhibited the ability to generate H 2 (Figure 6), including bare TiO 2 . PP1-TiO 2 , PP2-TiO 2, and GaPP2-TiO 2 showed a superior photocatalytic activity rate for H 2 generation relative to bare TiO 2 . In the case of native PPs, H 2 production from PP1 is at least 30% higher than that found for PP2. Although this could be related to slightly higher PP1 loading, differences in the porphyrin loadings cannot explain the decrease of in production of H 2 when using metalloporphyrins, particularly Ga-PP and In-PP that show higher to similar loadings compared to their native PPs. Therefore, the variation in the activity of the systems could be a result of different electron dynamics between the surface of the dye and TiO 2 . Freebase porphyrins (PP1 and PP2) had better photosensitization than their metal complexes due to more electronic coupling with the TiO 2 surface. Both PP1 and PP2 being metal free are flexible with fairly non-planar conformations thus can afford multiple adsorption configurations due to variable dihedral angles. The freebase cavity can co-ordinate more effectively with TiO 2 by flat attachment favoring electron injection in comparison to the metal complexes [52]. Additionally, metallophorphyrins have greater non-radiative decays compared to their native porphyrins suggesting fast singlet state decay, which can hamper electron injection into TiO 2 .
Molecules 2021, 26, x FOR PEER REVIEW 11 of 17 complexes [52]. Additionally, metallophorphyrins have greater non-radiative decays compared to their native porphyrins suggesting fast singlet state decay, which can hamper electron injection into TiO2.

Mechanism of Photocatalysis
The proposed mechanism for the photochemical reactions of the sacrificial molecules is illustrated in Scheme 2. Methanol was used as a sacrificial electron donor to facilitate the generation of H2, for simplicity Scheme 2 shows only the first oxidation step (formation

Mechanism of Photocatalysis
The proposed mechanism for the photochemical reactions of the sacrificial molecules is illustrated in Scheme 2. Methanol was used as a sacrificial electron donor to facilitate the generation of H 2 , for simplicity Scheme 2 shows only the first oxidation step (formation of hydroxymethyl radical) although further oxidation is expected [53]. Under simulated solar irradiation, UV and visible light energy can be absorbed by the P-TiO 2 composite. In addition to the direct excitation of TiO 2 creating electron-hole pairs, injection of electrons from the LUMO of photoexcited porphyrins into the TiO 2 conduction band is also energetically favorable (Figure 3) for all P-TiO 2 systems studied. This could generate a continuous flux of electrons into the TiO 2 CB that are readily available for the reduction of the H + ion. Significantly, more efficient H 2 production was obtained when PP1-TiO 2 and PP2-TiO 2 composites were used. With the exception of GaPP2-TiO 2 , the composites prepared from metal porphyrin complexes have lower H 2 production efficiencies than that obtained for bare TiO 2 . This could be related to the ability of the heavy metals to favor ISC and hence populate triplet states (laying at lower energy levels). This is likely to hamper rather than enhancing the electron flow from the LUMO of the photoexcited porphyrin to the TiO 2 CB. The presence of the heavy sulfur atoms of the meso-aryl rings of PP2 could explain why the photocatalytic properties of the PP1-TiO 2 composite were superior to those of PP2-TiO 2 . Interestingly, however, the MPP2-TiO 2 composites are consistently more efficient than the analogous MPP1-TiO 2 materials. This is consistent with the higher Φ F and τ F values for these complexes in Table 2, which is consistent with greater population of the S 1 state that can be expected to enhance electron injection from the photoexcited sensitizer into the TiO 2 CB. In a similar manner, the Φ F and τ F values for PP1 are higher than those for PP2 ( Table 2). It is noteworthy that although the InPP1 and InPP2 are predicted to lie closer in energy to the TiO 2 CB (Figure 3), it is the GaPP1-TiO 2 and GaPP2-TiO 2 materials that have superior photocatalytic effects ( Figure 6) among the metallophorphyrins studied.

Instrumentation
The ground state electronic absorption spectra were obtained at room temperature from an Agilent Technologies Cary 60 UV-Vis spectrophotometer in 1.0 × 1.0 cm quartz cuvettes, while corrected emission spectra were obtained with a PTI Quanta Master spectrofluorimeter (Photon Technology International, of London, ON, Canada) and 0.4 × 1.0 cm cuvettes. The IR spectra were recorded on a Nicolet 6700 FT-IR spectrometer (Madison, WI, USA), and centrifugation was carried out on an Eppendorf centrifuge 5804 R. Fluorescence lifetimes were measured using a time-correlated EasyLife fluorescence lifetime fluorimeter (Photon Technology International, of London, ON, Canada). The instrument response function (IRF) was measured at the excitation wavelength (395 nm) using Ludox as the scatterer. Mass spectrometry data were obtained from a Bruker AutoFLEX III Smartbeam MALDI-TOF instrument (Billerica, MA, USA). The elemental analyses were carried out on a Vario EL III Microcube CHNS Analyzer. H NMR spectra were obtained from a Bruker AVANCE II 400 MHz NMR spectrometer (Burlinton, Ontario, Canada). with CDCl 3 as the solvent. Diffuse reflectance spectroscopy (DR) data were collected on a Varian Cary 100 UV-Vis spectrophotometer (Santa Clara, CA, USA) coupled to an integrating sphere accessory. Samples were irradiated using a Luzchem SolSim CCP instrument (Ottawa, Canada) providing simulated solar radiation approximately matching in intensity the 1.5 air mass (AM) solar spectrum. The hydrogen generated was detected and quantified with a Perkin Elmer, Clarus 480 GC-TCD (Madison, WI, USA) using argon as a carrier gas and a 5 A zeolite molecular sieve column.

Synthesis of Porphyrins
The aromatic meso-substituted porphyrins PP1 and PP2 were synthesized by a single step acid-catalyzed condensation reaction of pyrrole with an appropriate aldehyde based on the Adler-Longo method [56,57]. Their metal complexes were synthesized according to literature methods [58,59]. Scheme 1 gives a summary of the synthesis procedure used and the reaction conditions.
In this equation, A and A Std are the optical densities for the sample and standard solutions respectively, while F and F Std are the integrated areas under the emission curves for sample and standard, n and n Std are the refractive indices of the solvents. The radiative (k r ) and non-radiative (k nr ) rate constant data were obtained by calculation according to Equations (2) to (3) [61].
Fluorescence lifetime (τ F ) values were determined using a time-correlated EasyLife X filter fluorescence lifetime fluorimeter. Time-resolved fluorescence decay curves were analyzed by deconvoluting the observed decay with the IRF to obtain the intensity decay function. Reduced chi-squared values were evaluated and considered to determine the best fit for experimental data.

Theoretical Calculations
Quantum calculations were performed using the Gaussian 09 program [62] and the theoretical model of choice was density functional theory (DFT). DFT is useful in providing direct information on electronic structure and strikes a favorable balance between computational cost and accuracy [63]. The gas-phase geometries of the porphyrin monomer structures were optimized by Becke's 3-parameter Lee Yang Parr (B3LYP) exchange-correlation functional. Linear time-dependent DFT (TD-DFT) calculations were performed with the Coulomb Attenuating Method B3LYP (CAM-B3LYP) functional. Mixed basis sets were used for these calculations; 6-31G(d,p) and Stuttgart-Dresden (SDD), the latter being preferred for porphyrin monomers with In(III) or Ga(III) metal ions. Simulation of the spectra was carried out using the Chemcraft software package with fixed bandwidths of 1000 cm −1 .

Supporting Porphyrins on TiO 2 P25
A procedure reported by Duan and co-workers in 2010 [64] was followed to deposit the porphyrins on the surface of TiO 2 . The porphyrins (10 mg) were dissolved in DCM (8 mL) in eight separate vessels. To each solution, TiO 2 (0.4 g) was added, and the mixture was sealed. The resulting suspension was sonicated for 30 min, then stirred at ambient temperature for 12 h. The photocatalyst was collected by centrifugation after washing three times with DCM and left to dry at 60 • C for 12 h.

Photocatalytic Hydrogen Generation
Photocatalytic hydrogen generation studies were carried out according to the procedure initially used by Hainer et al. [11]. These studies were carried out in a 10 mL glass vial sealed with crimp seals and rubber septa. In each experiment, 10 mg of the photocatalyst and 4 mL aqueous methanol solution (1%) was added to the glass vial. The system was degassed under argon for 30 min and sealed in the vacuum degassing chamber. The catalyst suspension was dispersed for 5 min in an ultrasonic bath then irradiated from the top for 4 h under simulated solar light with magnetic stirring. Headspace gas (4 mL) was sampled from each vial by a sample lock syringe through the septum and immediately injected in the GC-TCD. The H 2 signal was detected at~4.0 min. A calibration curve for H 2 gas detection in the GC-TCD instrument was used for quantification.

Conclusions
In conclusion, meso-tetra(4-bromophenyl)porphyrin PP1 and meso-tetra(5-bromo-2thienyl)porphyrin PP2 and their Zn(II), (Cl)In(III) and (Cl)Ga(III) complexes were successfully synthesized and characterized. Diffuse reflectance spectroscopy demonstrated that the porphyrins were successfully loaded onto the surface of TiO 2 P25. PP2-TiO 2 , PP1-TiO 2 and GaPP2-TiO 2 exhibited superior hydrogen generation compared to bare TiO 2 demonstrating their potential utility for photocatalytic H 2 generation. A wide range of porphyrin analogs exist with enhanced absorption in the Q band region, so research will continue in identifying dyes that are best suited for this application by focusing on dyes with long-lived and well-populated S 1 excited states.
Supplementary Materials: The following are available online: details for the synthesis and characterization of the porphyrins as well as their 1H NMR, mass, fluorescence spectra and Diffuse reflectance spectra ( Figures S1-S8).