A visible-light active catechol–metal oxide carbonaceous polymeric material for enhanced photocatalytic activity

Abstract

Designing new materials for sustainable energy and environmental applications is one of the prime focuses in chemical science. Here, an unprecedented visible-light active catechol–TiO₂ carbonaceous polymer based organic–inorganic hybrid material was synthesized by a photosynthetic route. The visible light induced (>400 nm) photosynthetic polymerization of catechol led to the formation of carbonaceous polymeric deposits on the surface of TiO₂. The band gap energy of hybrids was shifted to the visible region by orbital hybridization between 3d(Ti) of TiO₂ and 2p(O), π(C) of catechol. The Tauc plot clearly revealed that 1.0 wt% catechol–TiO₂ carbonaceous polymer remarkably tailored the optical band gap of TiO₂ from 3.1 eV to 1.9 eV. The synthesized hybrid materials were thoroughly characterized and their photocatalytic activity was evaluated towards toxic Cr(VI) to relatively less toxic Cr(III) reduction under visible light irradiation (>400 nm), and solar light-driven H₂ production through water splitting. Very interestingly, the hybrid material showed 5- and 10-fold enhanced activity for photocatalytic Cr(VI) reduction and solar light-driven H₂ production respectively compared with pure TiO₂. Moreover, the hybrid materials showed enhanced stability during photocatalysis. Thus, the simple photosynthetic strategy for developing light harvesting organic–inorganic hybrid materials can open up potential applications in energy and environmental remediation.

---

1. Introduction

Inorganic–organic composites are promising materials for a variety of applications including photovoltaics, optoelectronics, heterogeneous catalysts, pharmaceutical, biomedical and cosmetic formulations, and photocatalysis.^1–5 In particular, semiconductor metal oxide based organic–inorganic hybrid materials have received extensive interest in the field of photocatalysis and photovoltaics. Although various metal oxide organic–inorganic hybrids including TiO₂, ZnO, CdSe, and PbS have been reported for photocatalytic H₂ production, dye degradation, solar energy conversion etc.,^6–8 TiO₂-based hybrid materials have attracted great attention due to environmental viability with great chemical stability.^9,10

Various synthetic procedures have been developed to synthesise organic–inorganic hybrid materials for several applications.^11–14 Among them, photocatalytic surface induced polymerization of an organic linker represents an ideal strategy for the synthesis of stable organic–inorganic hybrids. A very few reports are known for surface induced polymerization on the metal oxide surface. For example, Wang and coworkers reported that a TiO₂ surface attached styrene and methyl methacrylate monomer can be polymerized under UV-light irradiation.¹⁵ Recently, the conducting polythiophene-TiO₂ organic–inorganic hybrid was developed by photopolymerization for solar cell applications.¹⁶ TiO₂ surface adsorbed benzene and toluene were polymerized in the presence of light radiation.^17,18 The proposed mechanism of photopolymerization involves the interaction of the photogenerated holes and electrons with the organic monomer and solvent molecules, respectively. It was also shown that an anionic initiation through photogenerated electron transfer to the monomer followed by free-radical propagation can lead to polymerization. However, we explored the polymerizable nature of catechol on a TiO₂ surface under visible-light illumination. Previously, Sanchez et al. reported that the catechol polymerization reaction occurred on a Ag surface and formed a stable polymer coating with C–C bonds.¹⁹

Due to high complexation tendency of catechol with TiO₂, it can be utilized as an organic linker for the preparation of a visible light active TiO₂–catechol surface complex. Despite light harvesting, the problem with a surface attached organic linker in the surface complex is that it is highly unstable under the operating conditions for photocatalysis application studies^20,21 and the surface attached or adsorbed organic linker can be leached out during photocatalysis. In this regard, the development of a stable light harvesting organic–inorganic metal oxide surface complex for vital photocatalytic applications is highly challenging. In the present study, we found that visible-light induced photopolymerization behavior of catechol on the TiO₂ surface can effectively alter the band gap energy and also increase the stability of the organic–inorganic hybrid material. The resultant photo-polymerized catechol–TiO₂ organic–inorganic hybrid material exhibits enhanced visible-light photocatalytic activity for the reduction of toxic Cr(VI) to Cr(III) and water splitting. In addition, a tentative mechanism was proposed for the photosynthetic polymerization of catechol on a metal oxide surface and its enhanced visible-light photocatalytic activity. We believe that the photopolymerizable nature of catechol on the TiO₂ surface may find different applications in many other fields.

2. Experimental

2.1 Synthesis of mesoporous TiO₂

The mesoporous TiO₂ was synthesized by the sonochemical method.²² Briefly, 0.032 mol of titanium isopropoxide and 0.016 mol of acetic acid were dissolved in ethanol under stirring and maintained for 1 h. The resultant colloidal solution was added in drops to 100 ml de-ionized water under constant ultrasonication. Subsequently, the resultant suspension was sonicated for 3 h by a low-frequency ultrasonicator (3 s on, 1 s off with amplitude 35%) using a 13 mm high-intensity probe (Sonics and materials VCX 750, 20 kHz). Finally, the product was collected by centrifugation, washed with a series of solvents (ethanol, water) and dried at 100 °C. Finally, the as-obtained TiO₂ powder was calcined at 400 °C for 1 hour.

2.2 Synthesis of the catechol–TiO₂ surface complex

The catechol–TiO₂ surface complex was synthesized by a facile condensation method at ambient temperature. In brief, 400 mg of TiO₂ nanopowder was dispersed in 10 ml of acetone. Subsequently, calculated amounts of catechol dissolved in acetone were added in drops to the TiO₂ solution under stirring conditions and maintained for 1 h. During stirring, the color of the reaction mixture turned to orange indicating the formation of the surface complex. Finally, the colored solid was collected by centrifugation and washed several times with acetone to remove the unreacted catechol and dried at 60 °C.

2.3 Photosynthetic polymerization of the catechol–TiO₂ surface complex

0.5 g of the TiO₂–catechol surface complex was placed in a Petri-dish and irradiated with 150 W visible light (>400 nm) for 6 h in an open atmosphere. A marked color change was observed from orange to dark brown post irradiation.

2.4 Photoelectrochemical studies

Photoelectrochemical studies were carried out with the conventional three-electrode system (CHI608E electrochemical workstation). Pt wire and Ag/AgCl (in saturated KCl) electrodes were used as counter and reference electrodes. A light source of 250 W Xe arc lamp (OSRAM, Germany) was used for photocurrent measurements. An aqueous solution of 0.1 M Na₂SO₄ served as the electrolyte. The working electrode was prepared by mixing 50 mg of the photocatalyst with 150 ml of PEG (mol. wt 400) and 125 ml of ethanol was used to make a slurry. Then it was coated on a 2.5 × 2.5 cm² fluorine-doped tin oxide (FTO) glass substrate with an active area of about 1 × 1 cm² by a doctor-blade method using scotch tape as the spacer and it was dried at 80 °C.²³

2.5 Photocatalytic Cr(VI) reduction activity

The toxic hexavalent chromium was chosen as the pollutant to evaluate the photocatalytic reduction activity of the catechol–TiO₂ carbonaceous polymer. Photocatalytic chromium(VI) reduction was carried out under 150 W visible light (>400 nm). In a typical procedure, 100 mg of the photocatalyst was added to 100 ml of 30 ppm chromium(VI) solution which was prepared from potassium dichromate crystals (K₂Cr₂O₇). After the addition of the photocatalyst to chromium(VI) solution, it was stirred under dark conditions for 30 min to ensure the adsorption–desorption equilibrium. After stirring, the reaction mixture was irradiated with visible light. The light intensity was measured by a Lux meter and the intensity was ∼37 000 Lux. During the photocatalytic reaction, every 5 min, a known volume of the sample was collected and the concentration of chromium(VI) was estimated colorimetrically at 540 nm by using diphenyl carbazide and a UV-Vis spectrometer (Specord 200 plus, Analytik Jena, Germany).

2.6 Diphenyl carbazide assay for Cr(VI) estimation

At each end point, 1 ml of Cr(VI) was collected and mixed with 9 ml of 0.2 M H₂SO₄ in a 10 ml iodine flask. Subsequently, 0.2 ml of newly prepared 0.25% (w/v) DPC in acetone was added to the iodine flask. After 15–30 s of vortexing, the mixture was allowed to stand for 10–15 min to ensure the complete complex formation. After the complexation reaction, the color of Cr(VI) solution changed from red-violet to purple, and then it was measured through a 540 nm filter using a UV-Vis spectrophotometer.

2.7 Solar photocatalytic H₂ production

The photocatalytic H₂ production reaction was carried out under natural solar light irradiation. Typically, 1 mg per ml of catechol carbonaceous polymer–TiO₂ was dispersed in 30 ml of 5% TEOA aqueous solution taken in a 100 ml quartz reactor. The reaction mixture was stirred under dark conditions for 30 min and evacuated for 30 min. Subsequently, N₂ gas was purged for 30 min to remove the dissolved oxygen. The final reaction mixture was exposed to solar light and the produced gas was collected at constant time intervals. The generated H₂ gas was analyzed by an off-line gas chromatograph with a TCD detector (Shimadzu GC-2014 with Molecular Sieve/5 Å column) using N₂ as the carrier gas.

3. Characterization studies

The photosynthesized catechol–TiO₂ carbonaceous polymer was characterized using the following spectroscopic techniques. The X-ray diffraction (XRD) patterns were recorded with a PANalyticalX'pert powder diffractometer using Cu-Ka radiation. The photoluminescence (PL) study was carried out with a HORIBA FLUORO LOG@3. The X-ray photoelectron spectrum (XPS) analysis was performed using a Shimadzu ESCA 3100. The morphological studies of the prepared photocatalysts were carried out by using Field Emission-Scanning Electron Microscopy (FEI Quanta FEG 200 HR-SEM). The bonding nature was analyzed by using Fourier transform infrared (FT-IR) spectroscopy (Perkin-Elmer-USA). The TEM images were obtained from MODEL-TECNIA F30 and Raman analysis was carried out using BRUKER RFS 27. The surface area, pore size, and pore volume of the material were determined by using a Quantachrome Nova-1000 surface analyzer.

4. Results and discussion

The photo-polymerization of the organic linker can be initiated by photo-excitation of TiO₂ under visible light irradiation. It was observed that during the irradiation by UV light the color of the TiO₂–benzene mixture changes from white to brown due to the UV light-induced polymerization.^24,25 Similar to the previous report, we also observed a color change in the catechol–TiO₂ surface complex from orange to brown post irradiation of visible light for 6 h (Fig. 1). This observation clearly suggests that the formation of brownish color is due to the deposition of the carbonaceous polymer on the TiO₂ surface through aromatic ring cation radical polymerization.¹⁸Fig. 1 illustrates the schematic representation of surface complexation followed by visible light induced photopolymerization of the organic linker.

Fig. 1 Schematic illustration of visible light-induced surface photopolymerization of catechol on TiO₂.

It is well known that TiO₂ absorbs photons in the UV region and it can generate electron–hole pairs. However, it is interesting to note that the photopolymerization of catechol–TiO₂ occurred in the visible region. We believe that visible-light induced polymerization is mainly due to ligand to metal charge transfer (LMCT) between catechol and TiO₂. The plausible mechanism of photopolymerization of catechol is proposed based on the LMCT (Fig. 2a). During the visible light illumination, a weak LMCT occurred between the conduction band of TiO₂ and surface attached catechol molecules, which generate electron–hole pairs and directly initiate the radical polymerization.¹⁸ The photo-generated holes further oxidize the phenyl ring of catechol and thereby produce aromatic cation radicals, whereas electrons are scavenged by oxygen and produce superoxide anions (O²⁻).¹⁵ In order to support the proposed mechanism of photopolymerization, the photocurrent measurement was carried out under visible light illumination. Fig. 2b shows the transient photocurrent response of TiO₂ and catechol–TiO₂ (1 wt%) photocatalysts. The catechol loaded TiO₂ complex shows increased photocurrent response compared to pristine TiO₂ under visible light illumination. This could be due to the assembly of catechol on the TiO₂ surface which can act as a light sensitizer and result in the direct injection of electrons from the HOMO level of catechol to the conduction band of TiO₂ leading to enhanced photocurrent response with the effective scavenging of photogenerated electrons by oxygen. In addition, the photoluminescence intensity of the 1 wt% catechol–TiO₂ carbonaceous polymer is significantly quenched compared to others (0.5, 0.75, 2.0 and 3.0 wt%) as depicted in Fig. 2c. This result further strongly supports that the electron–hole pair recombination has been remarkably suppressed by the formation of superoxide anions O₂⁻˙ during visible light illumination. The formation of superoxide anion (O₂⁻˙) radicals was confirmed using nitroblue tetrazolium (NBT) assay (see Fig. S10†).²⁶ The visible light induced photopolymerization reaction of catechol molecules and the TiO₂–catechol surface complex was analyzed by FT-IR spectroscopy. The FT-IR spectra of pure catechol, 20 h visible light irradiated catechol and the TiO₂–catechol surface complex irradiated with visible light for 6 h are displayed in Fig. 3. As shown in Fig. 3, the characteristic peaks of free catechol are as follows: the stretching vibrations of phenolic group ν(C–OH) appeared at 1345 cm⁻¹ and 1300 cm⁻¹ and the corresponding bending vibrations of the phenolic group δ(C–OH) appeared at 1193, 1088, and 1068 cm⁻¹. The peaks appearing at 1615 cm⁻¹ and 1467 cm⁻¹ correspond to aromatic ν(C–H) and ν(C=C) stretching vibrations. Moreover, the FT-IR spectra of 20 h visible light irradiated catechol, exhibits a characteristic peak of ν(C–C) at 1110 cm⁻¹ along with normal stretching and bending vibrations of catechol molecules.^27,28 It is clearly demonstrated that the catechol molecules also undergo the polymerization reaction during visible light irradiation and form a stable carbonaceous polymer. In addition, the two new peaks appearing at 1017 and 926 cm⁻¹ for 20 h visible light irradiated catechol are assigned to the C–H plane bending vibration of aromatic rings.²⁹ On the other hand, no such FT-IR peaks related to the C–H plane bending vibration of aromatic rings are observed for pure catechol. This observation clearly suggested that during the photopolymerization process, the aromatic ring connects via the C–C bond and this bond formation leads to enhance the in-plane C–H bending vibration and reduce its ring strain. Thus, the enhanced in-plane bending vibration further supports the photopolymerization reaction and C–C bond formation. Similar phenomenon has been previously reported for poly aromatic rings.²⁹ Similarly, the 6 h visible light irradiated TiO₂–catechol surface complex also shows the ν(C–C) peak at 1110 cm⁻¹. However, the peak appearing in the range of 1260–1280 cm⁻¹ is mainly attributed to Ti–O–C bonding between TiO₂ and catechol molecules.^30–32 These observations confirmed that the radical polymerization of catechol is initiated by the TiO₂ surface and the resultant carbonaceous polymer was strongly deposited over the TiO₂ surface,^21,33 thus leading to Ti–O–C bonding. The formation of the C–C bond was further analyzed by FT-Raman spectroscopy. Fig. 4 shows the Raman spectra of TiO₂ and after the 6 h visible light irradiated TiO₂–catechol surface complex. As illustrated in Fig. 4, the bands appeared in the range of 1150 to 1300 cm⁻¹ and 930–1077 cm⁻¹ correspond to C–O and C–H ring deformation vibrations of catechol molecules.³⁴ The peaks appearing at 1468, and 1190 cm⁻¹ are related to the aromatic C=C ring vibration, and C–O–C bond, respectively. Importantly, the main peak at 1400 cm⁻¹ is related to the C–C bond vibration.^35,36 This C–C bond vibration peak confirms the formation of the polymeric structure during visible light illumination. Similar C–C bond vibration was observed by Goa et al. for polypyrrole.

Fig. 2 (a) Mechanism of visible light-induced polymerization of catechol, (b) photocurrent response of TiO₂ and the 1 wt% catechol–TiO₂ carbonaceous polymer hybrid, and (c) PL spectra of different wt% catechol loaded TiO₂ carbonaceous polymer hybrids.

Fig. 3 (a) FT-IR spectra of pure catechol and catechol after 20 h of visible light irradiation and (b) TiO₂–catechol surface complex after 6 h of visible light irradiation.

Fig. 4 FT-Raman spectra of TiO₂ and the catechol–TiO₂ surface complex after 6 h of visible light irradiation.

In order to confirm the formation of the C–C bond during the photopolymerization reaction and the chemical interaction between catechol and TiO₂, XPS analysis was carried out. As shown in Fig. 5a, the high resolution C1s spectrum can be deconvoluted into four peaks. The characteristic peak observed at 284.5 eV is mainly attributed to C–C bonds, which clearly revealed that during the photopolymerization the carbonaceous polymer was formed and phenyl rings of catechol molecules are connected through the C–C bond.^17,37–39 Moreover, the additional peaks centered at 284.2, 286.2 and 288.5 eV are assigned to aromatic C=C, C–O and O–C=O bonds, respectively.^40,41 Similarly, the high resolution O1s spectrum (Fig. 5b) is deconvoluted into three main components. The peak identified at 529.5 eV mainly arises from the Ti–O–C bond.⁴² During the condensation reaction, the Ti–O bond from TiO₂ may interact with catechol molecules and form the Ti–O–C bond. Moreover, the surface complex formation between TiO₂ and catechol can occur only via Ti–O–C bond formation and there are no other possibilities for the surface complex formation between TiO₂ and catechol.^43,44 In a previous report, Zhang et al. have synthesized an organic linker functionalized TiO₂ surface complex and obtained the high intense Ti–O–C peak.⁴² This is the clear evidence supporting the TiO₂ and catechol orbital mixing. In addition, the Ti–O–C bond formation in the catechol–TiO₂ surface complex was also confirmed by FT-IR spectra.^30–32 The additional peaks that appeared at 527.1 and 530.5 eV are assigned to Ti–O and Ti–OH bonds of TiO₂.⁴⁵ Moreover, the Ti2p_3/2 and Ti2p_1/2 peaks identified in the high resolution Ti2p spectra confirmed the presence of anatase TiO₂ (Fig. S4†).

Fig. 5 XPS spectra of the catechol–TiO₂ carbonaceous polymer: (a) high-resolution C1s spectrum and (b) high-resolution O1s spectrum.

Fig. 6 shows the N₂ adsorption–desorption isotherms and the corresponding pore size distribution curves of TiO₂, the catechol–TiO₂ surface complex (before photopolymerization) and the catechol–TiO₂ carbonaceous polymer. As shown in Fig. 6, all the prepared photocatalysts exhibit type (IV) isotherms with H3 hysteresis loops. It is well known that type (IV) pattern corresponds to the mesoporous structure of materials.⁴⁶ The average pore size of TiO₂, the catechol–TiO₂ surface complex and the catechol–TiO₂ carbonaceous polymer are 4.8, 3.7 and 2.9 nm, respectively. It can be seen that due to the surface complexation of catechol with TiO₂, the average pore size of TiO₂ decreased to 3.7 nm from 4.8 nm (Table 1). Moreover, after 6 h of visible irradiation the average pore size further decreases (2.9 nm). It clearly indicates that during the visible light illumination, the surface attached catechol molecules undergo the polymerization reaction and the resultant branched polymers block the TiO₂ pores. In contrast to pore volume the surface area of the catechol–TiO₂ surface complex (96.1 m² g⁻¹) increases after the photopolymerization (117.4 m² g⁻¹), which suggests that TiO₂ is strongly intercalated in the branched carbonaceous polymer matrix.⁴⁷

Fig. 6 (a) and (b) Nitrogen adsorption–desorption isotherms and pore distribution curves of TiO₂, the catechol–TiO₂ surface complex (before polymerization) and the catechol–TiO₂ carbonaceous polymer.

Table 1 Surface area and pore size analysis

Entry	Name of samples	S BET (m^2 g^−1)	V p (cm^3 g^−1)	Pore size (nm)
1	TiO2	134.6	0.329	4.8
2	Catechol–TiO2 surface complex	96.1	0.178	3.7
3	Catechol carbonaceous polymer–TiO2 surface complex	117.4	0.170	2.9

---

The influence of polymerization of the organic linker (catechol) on the TiO₂ surface morphology and the elemental composition of catechol–TiO₂ carbonaceous polymer hybrids were studied using FE-SEM and EDX analysis. The bare TiO₂ particles exhibited spherical morphology, whereas the catechol–TiO₂ carbonaceous polymer hybrid showed an irregular morphological pattern with rough embodiments (Fig. S1†). The change in the morphologies of the particles (post-polymerization) clearly demonstrated the presence of catechol groups on the TiO₂ surface. Moreover, FE-SEM-EDX (Fig. S2†) analysis confirmed the presence of a higher carbon content. This is a clear evidence for the deposition of the carbonaceous polymer on the TiO₂ surface.¹⁷ In order to further confirm the structural changes of TiO₂ during the photopolymerization reaction, TEM and HRTEM were performed and the results are illustrated in Fig. 7. It can be clearly seen from Fig. 7 that TiO₂ possesses uniform spherical particles with pores that indicate the formation of mesoporous TiO₂ nanoparticles similar to SEM images. Unlike mesoporous TiO₂ nanoparticles, the photopolymerization of catechol on TiO₂ affects the existing TiO₂ morphology and forms a continuous network with aggregation. This continuous network with high surface area could efficiently facilitate the photoexcited electrons and enhance the photocatalytic performance. Several studies have been reported earlier related to the morphological dependent photocatalytic activity of the photocatalysts.^48–50 The calculated interplanar spacing value for both TiO₂ and the catechol–TiO₂ carbonaceous polymer hybrid was 0.34 nm and it corresponded to the (101) plane of anatase phase TiO₂. This clearly suggests that photopolymerization occurred only on the surface of TiO₂ without affecting the existing crystalline phases of TiO₂, as revealed from XRD (Fig. S3†).

Fig. 7 TEM images of: (a) TiO₂; (b) catechol–TiO₂ carbonaceous polymer and HRTEM images of (c) TiO₂; (d) catechol–TiO₂ carbonaceous polymer.

The optical absorption of different concentrations of the catechol loaded TiO₂ surface complex and the carbonaceous polymer hybrid was investigated by UV-Vis absorption spectra recorded in Diffuse Reflectance Spectroscopy (DRS) mode. As depicted in Fig. 8a, the absorption spectra of the catechol–TiO₂ surface complex shows enhanced absorption in the visible region compared to pristine TiO₂. The increased absorption in the visible region is due to complex formation between catechol and the TiO₂ surface. Moreover, after 6 h of visible light irradiation, the absorbance of the surface complex further shifted to the visible region and a small peak at 630 nm is noted (Fig. 8b). It is mainly because all the aromatic rings of catechol combined through the C–C bond and extended the π–π conjugation. The maximum absorption is observed with the 1 wt% catechol–TiO₂ carbonaceous polymer hybrid and the subsequent decrease in the absorption of TiO₂ was observed irrespective of the loading of catechol (wt%). This is clearly attributed to the higher loading of catechol (2 and 3 wt%) concentration on the TiO₂ surface leading to light penetration. As can be seen from Fig. 8c, the calculated band gap values of pristine TiO₂ and catechol–TiO₂ carbonaceous polymer hybrids are 3.1 eV (TiO₂), 2.9 eV (0.5 wt% catechol–TiO₂), 2.8 eV (0.75 wt% catechol–TiO₂), 1.8 eV (1.0 wt% catechol–TiO₂), 2.2 eV (2.0 wt% catechol–TiO₂) and 2.0 eV (3.0 wt% catechol–TiO₂). It is worthy to mention here that the band gap of TiO₂ can be tuned by changing the concentration of the organic linker. The decrease in the band-gap of the catechol–TiO₂ carbonaceous polymer hybrid may be attributedto the following reasons.

Fig. 8 (a) UV-Vis absorption spectra (DRS mode) of different wt% catechol–TiO₂ surface complexes, (b) different wt% catechol–TiO₂ carbonaceous polymer hybrids (after 6 h of visible light irradiation), and (c) Tauc plot of various catechol–TiO₂ carbonaceous polymer hybrids.

(i) The interfacial charge transfer complex formation between surface hydroxyl groups of TiO₂ and catechol (it was confirmed by FT-IR spectra).

(ii) The effective orbital hybridization between the 2p(O), π(C) orbitals of catechol (ligand) and 3d(Ti) orbital of TiO₂ could lead to the formation of the Ti–O–C bond.⁵¹ It was confirmed from FT-IR and O1s XPS spectra.

It is interesting to note that the mixing of orbitals can be controlled by the organic linker concentration. Upon increasing the catechol concentration from 0.5 wt% to 1 wt%, the effective orbital mixing also increases and the optical band gap energy shifts towards the red region. Whereas, beyond 1 wt% loading, the band gap increases and this could be attributed to the mixing of molecular orbitals of TiO₂ and catechol getting saturated with maximal orbital mixing and further loading of catechol molecules had minimal effect (nominal increase) on the band gap.

4.1 Photocatalytic Cr(VI) reduction activity

The photocatalytic activity of catechol–TiO₂ carbonaceous polymer hybrids was assessed for toxic Cr(VI) to less toxic Cr(III) reduction. The photocatalytic Cr(VI) reduction was performed and monitored through UV-Vis absorption spectra using a DPC assay. Prior to the reaction, Cr(VI) solution was stirred under dark conditions for 30 min to achieve the adsorption–desorption equilibrium. As depicted in Fig. 9a, pure TiO₂ shows only 20% of Cr(VI) reduction in 20 min under visible light irradiation. In contrast, the catechol–TiO₂ carbonaceous polymer hybrids exhibit complete reduction (100%) of Cr(VI) to Cr(III) in 20 min under identical experimental conditions (Fig. 9b). There is complete reduction of Cr(VI) (30 ppm) in the presence of the catechol–TiO₂ carbonaceous polymer hybrid supporting the efficiency of the prepared photocatalyst. Efficient reduction of Cr(VI) to Cr(III) can be explained by the phenomenon that the catechol–TiO₂ carbonaceous polymer hybrid facilitates the interfacial charge transfer between the aromatic rings of catechol and TiO₂via (–Ti–O–C–) linkage as evident from the O1s spectrum (see Fig. 5b). However, another possible reason for high photocatalytic Cr(VI) reduction activity is adsorption. As shown in Fig. 9c, the catechol–TiO₂ carbonaceous polymer showed higher Cr(VI) adsorption efficiency (33%) compared to that of pure TiO₂. The high Cr(VI) adsorption capacity of catechol–TiO₂ carbonaceous polymer may be due to the complexing tendency of oxygen atoms present in the polymeric material with Cr(VI) and the high surface area. In general, the adsorption of Cr(VI) ions on the catalyst's surface plays a vital role in the photoreduction reaction. The high adsorption of Cr(VI) ions on the catechol–TiO₂ carbonaceous polymer can enhance the photoreduction of Cr(VI) to Cr(III). During the photocatalysis, the photogenerated electrons from the photocatalysts can readily migrate to the surface adsorbed Cr(VI) ions and thereby efficiently reduce Cr(VI) to Cr(III).^52–54 In order to evaluate the mechanistic aspects and effect of different catechol loaded carbonaceous polymer hybrids on the enhanced photoreduction activity of photocatalysts, we carried out a set of experiments with different experimental conditions and the results are depicted in Fig. 9c. As can be seen from Fig. 9c, the photocatalytic activity is increased on increasing the catechol loading (up to 1 wt%). Beyond 1 wt%, the photocatalytic activity is gradually decreased which clearly indicates that the higher catechol loading (2 and 3 wt%) greatly affects the photocatalytic activity due to the steric-screening effect of excess aromatic rings adsorbed on the surface of TiO₂.⁵⁵ The maximum Cr(VI) reduction is observed with 1 wt% catechol–TiO₂ carbonaceous polymer hybrids due to their very low band gap energy (1.8 eV) compared to other composites (Fig. 8b), which further supports that there is more absorption of light in the visible region. In addition, the photocatalytic Cr(VI) reduction activity of the catechol–TiO₂ carbonaceous polymer hybrid was compared with that of the already reported photocatalysts and is tabulated in Table 2. It can be seen that the photocatalytic Cr(VI) reduction activity of the catechol–TiO₂ carbonaceous polymer hybrid showed enhanced photocatalytic activity compared to other reported photocatalysts.^56–65

Fig. 9 UV-Vis absorption spectra of photocatalytic Cr(VI) reduction using (a) TiO₂, (b) the catechol–TiO₂ carbonaceous polymer hybrid and (c) C/Co profile of different wt% catechol–TiO₂ carbonaceous polymer hybrids.

Table 2 Comparison of photocatalytic Cr(VI) reduction activity

Catalyst	Concentration of Cr(VI) ions	Irradiation time (min)	Light source	Number of recycles	Cr(VI) estimation	Ref.
Au/N–TiO2	10 ppm	250 min	Visible	—	UV-Vis spectrophotometer	56
TiO2–2-naphthol	30 ppm	180 min	Visible	2	UV-Vis spectrophotometer	21
CdS–rGO	10 ppm	250 min	Visible	—	UV-Vis spectrophotometer	57
Mg, Ag co-impregnated TiO2	5 ppm	100 min	Visible	—	UV-Vis spectrophotometer	58
Bi nanowire networks	40 ppm	60 min	Visible	5	UV-Vis spectrophotometer	59
CuO–ZnO nanocomposite	20 ppm	60 min	Visible	4	UV-Vis spectrophotometer	60
NH2-MIL-88B (Fe)	8 ppm	45	Visible	3	DPC	61
MIL-68(In)-NH2	20 ppm	180	Visible	3	DPC	62
UiO-66(NH2)	10 ppm	100	Visible	3	DPC	63
rGO–UiO-66(NH2)	10 ppm	120	Visible	3	DPC	64
Pd@UiO-66(NH2)	10 ppm	120	Visible	3	DPC	65
Catechol–TiO 2 carbonaceous polymer	30 ppm	20	Visible	5	DPC	This work

---

It is well known that the organic linker-functionalized metal oxide surface complexes are highly unstable due to the formation of surface bonding between the organic linkers and metal-oxides. As a result, their application is greatly limited for the commercialization of organic linker attached metal oxide photocatalysts. Kim et al. reported that the TiO₂–phenolic surface complex was less stable due to the surface interaction between phenolic –OH and TiO₂.²⁰ However, in the present work, the stability of catechol-functionalized TiO₂ surface complexes were achieved by the photosynthetic polymerization approach under visible light (>400 nm) irradiation. The carbonaceous copolymer formation was previously identified by XPS analysis (Fig. 5). Fig. 10a represents the durability of catechol–TiO₂ carbonaceous polymer hybrids for five successive cycles. For all the five cycles the photocatalytic Cr(VI) reduction is consistently noted above 96% without significant loss in the photocatalytic activity. Moreover, the stability of hybrid photocatalysts was further confirmed by FT-IR and TEM analysis. The FT-IR and TEM (Fig. 10b and S8†) analyses were carried out before and after five cycles of the photocatalytic reaction. There is no significant changein both FT-IR and TEM after five consecutive cycles, demonstrating the high chemical stability of catechol–TiO₂ carbonaceous polymer hybrids.

Fig. 10 (a) Reusability experiment for photocatalytic Cr(VI) reduction using the 1 wt% catechol–TiO₂ carbonaceous polymer hybrid, and (b) FT-IR spectra before and after five cycles of reaction.

4.2 Photocatalytic H₂ production

The photocatalytic H₂ production activity was evaluated under solar light irradiation with different amounts of catechol loaded TiO₂ carbonaceous polymer hybrids by using triethanolamine (TEOA) as a hole scavenger. No H₂ production was observed in the absence of either solar light or the photocatalyst. As can be seen from Fig. 11, the descending order of photocatalytic H₂ production activity is 1 wt% catechol–TiO₂ (10 925 μmol h⁻¹ g_cat⁻¹) > 0.75 wt% catechol–TiO₂ (6148 μmol h⁻¹ g_cat⁻¹) > 0.5 wt% catechol–TiO₂ (3082 μmol h⁻¹ g_cat⁻¹) > 2 wt% catechol–TiO₂ (3995 μmol h⁻¹ g_cat⁻¹) > 3 wt% catechol–TiO₂ (2534 μmol h⁻¹ g_cat⁻¹) > TiO₂ (1035 μmol h⁻¹ g_cat⁻¹). The results obviously demonstrated that the photocatalytic activity of pristine TiO₂ is lower than that of different wt% catechol loaded TiO₂ carbonaceous polymer hybrid photocatalysts. Notably, the optimized 1.0 wt% catechol–TiO₂ carbonaceous polymer photocatalyst shows the highest AQY (59.3%) compared to that of other photocatalysts. Thus, it clearly demonstrates that the 1.0 wt% catechol–TiO₂ carbonaceous polymer photocatalyst effectively utilized the photons and thereby enhanced the photocatalytic H₂ production rate (see Table S2†). The enhanced photocatalytic activity of the catechol–TiO₂ carbonaceous polymer hybrid photocatalyst is due to interfacial charge transfer transition between TiO₂ and the catechol carbonaceous polymer. However, the maximum H₂ production was observed with the 1 wt% catechol–TiO₂ carbonaceous polymer hybrid photocatalyst which strongly suggested that interfacial electron transfer has rapidly occurred compared to other photocatalysts as revealed from the photoluminescence (PL) spectra. In addition, the lower band gap energy (Fig. 8) of the 1 wt% catechol–TiO₂ carbonaceous polymer hybrid photocatalyst facilitates more absorption in the visible region and thereby enhances the photocatalytic activity. Moreover, the photocatalytic H₂ production activity of the catechol–TiO₂ carbonaceous polymer hybrid was compared with the already reported organic/inorganic hybrid materials and metal–organic framework (MOF) photocatalysts (Table 3). Compared to known photoactive organic/inorganic hybrids and metal–organic framework (MOF) photocatalysts, catechol–TiO₂ carbonaceous polymer hybrids showed better photocatalytic activity under solar light irradiation.^66–75

Fig. 11 (a) Photocatalytic H₂ production using different wt% catechol–TiO₂ carbonaceous polymers, (b) reusability test, and (c) FT-IR spectra of the catechol–TiO₂ photocatalyst before and after 3 cycles of the photocatalytic H₂ production reaction.

Table 3 photocatalytic H₂ production activity with the existing TiO₂ based organic/inorganic hybrid materials

Catalyst	Reaction condition	H2 production activity	Ref.
TO-SCA[4]/Pt–TiO2	λ ≥ 450 nm, 0.2 M TEOA	∼200 μmol h^−1	66
(PR)-Pt/TiO2	λ ≥ 420 nm, 50 mM Na2S, pH 12.50	∼700 μmol g^−1 in 4 h	67
EY–Pt/TiO2	Visible light, 5% TEOA	2000 μmol in 20 h	68
RhB-Pt/UiO-66(Zr) (MOF)	λ ≥ 420 nm, 10% TEOA, pH 7.00	116 μmol h^−1 g^−1	69
EB–UiO-66 (MOF)	λ ≥ 420 nm, 30 mg ErB, 0.5 wt% Pt	4.6 μmol h^−1	70
ErB–MIL-101 (MOF)-Ni/NiOx	λ ≥ 420 nm, 30 mg ErB	125 μmol h^−1	71
Pt/NH2-MIL-125(Ti)	Xe lamp (TEOA)	33 μmol	72
Co@NH2-MIL-125(Ti)	500 W Xe/Hg lamp	37 μmol	73
Pt@CdS/MIL-101(Cr)	300 W Xe lamp (lactic acid)	150 μmol h^−1	74
Pt@MOF-1 or Pt@MOF-2	450 W Xe-lamp (TEA), visible	3400 or 7000 TON	75
Catechol–TiO 2 carbonaceous polymer	Solar light, 5% TEOA aqueous solution	10 925 μmol h ^−1 g cat ^−1	This work

---

In order to verify the photostability of the catechol–TiO₂ carbonaceous polymer hybrid during the photocatalytic H₂ production, the recyclability experiment was carried out under identical conditions by evacuating the generated gas at certain time intervals for the next cycle and the results are illustrated in Fig. 11b. It can be seen that no significant change was observed in the H₂ production activity for 3 cycles and it indicates the superior stability of the photocatalyst. However, the stability of catechol–TiO₂ carbonaceous polymer hybrids was further confirmed using FT-IR and TEM analyses. As observed from FT-IR and TEM (Fig. 11c and S9†), no notable change was observed in both FT-IR and TEM analyses of fresh catalysts and those used after 3 cycles and it clearly demonstrated the high sustainability of catechol–TiO₂ carbonaceous polymer hybrids as in the case of photocatalytic Cr(VI) reduction reaction.

4.3 Possible mechanism for photocatalytic Cr(VI) reduction and H₂ production

Based on the above discussion, a possible LMCT photocatalytic reaction mechanism has been proposed and illustrated in Fig. 12. In this process, the catechol attached on the TiO₂ surface sensitizes TiO₂ by forming a type-II ligand to metal charge transfer (LMCT) complex.^76–78 In general, a wide band gap semiconductor metal oxide can be activated by visible light by surface adsorption of dyes and with simple organic molecules and complexes as sensitizers other than coupling with narrow band gap semiconductors. In this work, during the visible light irradiation the photoexcited electrons present in the HOMO level of catechol can directly inject into the conduction band of TiO₂ due to the exact match of their energy levels. Therefore, the higher the population of electrons in the conduction band of TiO₂ the more effective is the reduction of Cr(VI) to Cr(III)²¹ within a short span of time. In addition, during LMCT holes are also generated in the valence band of TiO₂. This electron–hole (h⁺) generation is supported by photocurrent studies (Fig. 2b). The photo-generated holes are scavenged by triethanolamine (TEOA) to produce H⁺ and the produced H⁺ ions undergo a reduction reaction with photo-excited electrons and generate H₂.

Fig. 12 Proposed mechanism for photocatalytic Cr(VI) reduction and H₂ production.

5. Conclusion

A visible light active catechol–TiO₂ carbonaceous polymer organic–inorganic hybrid was successfully synthesized by a facile photosynthetic method. The synthesized hybrid materials exhibited superior photocatalytic activity for Cr(VI) reduction and H₂ production from the aqueous TEOA solution through ligand to metal charge transition. In addition, the lowest band gap (1.8 eV), highest photocatalytic Cr(VI) reduction (100% within 20 min) and maximum H₂ production activity were achieved with the 1 wt% catechol-loaded hybrid. The plausible mechanism was proposed for the photosynthetic polymerization of catechol on TiO₂ and photocatalytic Cr(VI) reduction as well as H₂ production activity. The current new finding may provide new insights into the design and development of visible light harvesting stable organic–inorganic hybrid materials for energy and environmental applications.

Acknowledgements

This work was financially supported by the Science and Engineering Research Board (SERB) [File No. EMR/2014/000645], New Delhi, India and Ministry of New and Renewable Energy (MNRE), New Delhi, India [File No. 103/239/2015-NT].

References

1. J. Ge and Y. Yin, Adv. Mater., 2008, 20, 3485–3491.
2. F. Caruso, R. A. Caruso and H. Möhwald, Science, 1998, 282, 1111–1114.
3. H. Lu, B. Fei, J. H. Xin, R. Wang and L. Li, Adv. Colloid Interface Sci., 2006, 300, 111–116.
4. P. Pitukmanorom, T. H. Yong and J. Y. Ying, Adv. Mater., 2008, 20, 3504–3509.
5. N. Zhang, M.-Q. Yang, S. Liu, Y. Sun and Y.-J. Xu, Chem. Rev., 2015, 115, 10307–10377.
6. X.-M. Gao, L.-D. Shu, L.-Y. Yang, Y.-Q. Shen, Y.-J. Zhang and Q.-F. Hu, Bull. Korean Chem. Soc., 2013, 34, 246–248.
7. A. Batool, F. Kanwal, M. Imran, T. Jamil and S. A. Siddiqi, Synth. Met., 2012, 161, 2753–2758.
8. F. Tan, S. Qu, L. Wang, Q. Jiang, W. Zhang and Z. Wang, J. Mater. Chem. A, 2014, 2, 14502–14510.
9. C. Han, N. Zhang and Y.-J. Xu, Nano Today, 2016, 11, 351–372.
10. S. Liu, C. Han, Z.-R. Tang and Y.-J. Xu, Mater. Horiz., 2016, 3, 270–282.
11. A. Schmid, S. P. Armes, C. A. Leite and F. Galembeck, Langmuir, 2009, 25, 2486–2494.
12. A. Schmid, S. Fujii, S. P. Armes, C. A. Leite, F. Galembeck, H. Minami, N. Saito and M. Okubo, Chem. Mater., 2007, 19, 2435–2445.
13. L.-J. Xu, W.-Z. Zhou, L.-Y. Zhang, B. Li, H.-Y. Zang, Y.-H. Wang and Y.-G. Li, CrystEngComm, 2015, 17, 3708–3714.
14. B. Kobin, L. Grubert, S. Blumstengel, F. Henneberger and S. Hecht, J. Mater. Chem., 2012, 22, 4383–4390.
15. X. Wang, Q. Lu, X. Wang, J. Joo, M. Dahl, B. Liu, C. Gao and Y. Yin, ACS Appl. Mater. Interfaces, 2015, 8, 538–546.
16. Y. Otsuka, Y. Okamoto, H. Y. Akiyama, K. Umekita, Y. Tachibana and S. Kuwabata, J. Phys. Chem. C, 2008, 112, 4767–4775.
17. A. Ramakrishnan, S. Neubert, B. Mei, J. Strunk, L. Wang, M. Bledowski, M. Muhler and R. Beranek, Chem. Commun., 2012, 48, 8556–8558.
18. S. Neubert, A. Ramakrishnan, J. Strunk, H. Shi, B. Mei, L. Wang, M. Bledowski, D. A. Guschin, M. Kauer and Y. Wang, ChemPlusChem, 2014, 79, 163–170.
19. S. Sanchez-Cortes, O. Francioso, J. Garcıa-Ramos, C. Ciavatta and C. Gessa, Colloids Surf., A, 2001, 176, 177–184.
20. S. Kim and W. Choi, J. Phys. Chem. B, 2005, 109, 5143–5149.
21. P. Karthik, R. Vinoth, S. G. Babu, M. Wen, T. Kamegawa, H. Yamashita and B. Neppolian, RSC Adv., 2015, 5, 39752–39759.
22. C. Y. Jimmy, L. Zhang and J. Yu, New J. Chem., 2002, 26, 416–420.
23. S. G. Babu, R. Vinoth, D. P. Kumar, M. V. Shankar, H.-L. Chou, K. Vinodgopal and B. Neppolian, Nanoscale, 2015, 7, 7849–7857.
24. O. d'Hennezel, P. Pichat and D. F. Ollis, J. Photochem. Photobiol., A, 1998, 118, 197–204.
25. H. Einaga, S. Futamura and T. Ibusuki, Appl. Catal., B, 2002, 38, 215–225.
26. A. R. Upreti, Y. Li, N. Khadgi, S. Naraginti and C. Zhang, RSC Adv., 2016, 6, 32761–32769.
27. S. Hu, L. Ma, H. Wang, L. Zhang, Y. Zhao and G. Wu, RSC Adv., 2015, 5, 31947–31953.
28. J. Suárez-Guevara, O. Ayyad and P. Gómez-Romero, Nanoscale Res. Lett., 2012, 7, 1.
29. B. C. Smith, Infrared spectral interpretation: a systematic approach, CRC press, 1998.
30. P. Lei, F. Wang, S. Zhang, Y. Ding, J. Zhao and M. Yang, ACS Appl. Mater. Interfaces, 2014, 6, 2370–2376.
31. S. Sakthivel and H. Kisch, Angew. Chem., Int. Ed. Engl., 2003, 42, 4908–4911.
32. G. Lui, J.-Y. Liao, A. Duan, Z. Zhang, M. Fowler and A. Yu, J. Mater. Chem. A, 2013, 1, 12255–12262.
33. T. D. Savić, I. A. Janković, Z. V. Šaponjić, M. I. Čomor, D. Ž. Veljković, S. D. Zarić and J. M. Nedeljković, Nanoscale, 2012, 4, 1612–1619.
34. Y. Gao, Y. Wang, X. Xu, K. Ding and D. Yu, RSC Adv., 2014, 4, 63719–63724.
35. T. Lana-Villarreal, A. Rodes, J. M. Pérez and R. Gómez, J. Am. Chem. Soc., 2005, 127, 12601–12611.
36. P. Deshmukh, S. Patil, R. Bulakhe, S. Pusawale, J.-J. Shim and C. Lokhande, RSC Adv., 2015, 5, 68939–68946.
37. X. H. Yang, J. W. Guo, S. Yang, Y. Hou, B. Zhang and H. G. Yang, J. Mater. Chem. A, 2014, 2, 614–619.
38. S. Mao, K. Yu, S. Cui, Z. Bo, G. Lu and J. Chen, Nanoscale, 2011, 3, 2849–2853.
39. M. H. Ahmed, J. A. Byrne, J. McLaughlin and W. Ahmed, Sci. Res., 2013, 4, 194–203.
40. O. Akhavan and E. Ghaderi, Nanoscale, 2013, 5, 10316–10326.
41. R. Larciprete, S. Fabris, T. Sun, P. Lacovig, A. Baraldi and S. Lizzit, J. Am. Chem. Soc., 2011, 133, 17315–17321.
42. S. Zhang, Y. Peng, W. Jiang, X. Liu, X. Song, B. Pan and H.-Q. Yu, Chem. Commun., 2014, 50, 1086–1088.
43. L. Zhang and J. M. Cole, ACS Appl. Mater. Interfaces, 2015, 7, 3427–3455.
44. G. Zhang, G. Kim and W. Choi, Energy Environ. Sci., 2014, 7, 954–966.
45. Y. Jiang, L. Meng, X. Mu, X. Li, H. Wang, X. Chen, X. Wang, W. Wang, F. Wu and X. Wang, J. Mater. Chem., 2012, 22, 23642–23649.
46. N. C. Strandwitz, Y. Nonoguchi, S. W. Boettcher and G. D. Stucky, Langmuir, 2010, 26, 5319–5322.
47. N. Parveen, M. O. Ansari and M. H. Cho, Ind. Eng. Chem. Res., 2015, 55, 116–124.
48. W. Sun, H. Liu, J. Hu and J. Li, RSC Adv., 2015, 5, 513–520.
49. L. Pan, J.-J. Zou, S. Wang, X.-Y. Liu, X. Zhang and L. Wang, ACS Appl. Mater. Interfaces, 2012, 4, 1650–1655.
50. Y. Wang, X. Li, N. Wang, X. Quan and Y. Chen, Sep. Purif. Technol., 2008, 62, 727–732.
51. J.-I. Fujisawa, R. Muroga and M. Hanaya, Phys. Chem. Chem. Phys., 2015, 17, 29867–29873.
52. B. Neppolian, A. Bruno, C. L. Bianchi and M. Ashokkumar, Ultrason. Sonochem., 2012, 19, 9–15.
53. C. Chenthamarakshan, K. Rajeshwar and E. J. Wolfrum, Langmuir, 2000, 16, 2715–2721.
54. X. Zhang, N. Zhang, Y.-J. Xu and Z.-R. Tang, New J. Chem., 2015, 39, 6756–6764.
55. Y. Fu, D. Sun, Y. Chen, R. Huang, Z. Ding, X. Fu and Z. Li, Angew. Chem., 2012, 124, 3420–3423.
56. X. Liu, L. Pan, T. Lv, Z. Sun and C. Sun, RSC Adv., 2012, 2, 3823–3827.
57. X. Liu, L. Pan, T. Lv, G. Zhu, Z. Sun and C. Sun, Chem. Commun., 2011, 47, 11984–11986.
58. H. Eskandarloo, A. Badiei, M. A. Behnajady and G. M. Ziarani, RSC Adv., 2014, 4, 28587–28596.
59. J. Zhao, Q. Han, J. Zhu, X. Wu and X. Wang, Nanoscale, 2014, 6, 10062–10070.
60. J. Yu, S. Zhuang, X. Xu, W. Zhu, B. Feng and J. Hu, J. Mater. Chem. A, 2015, 3, 1199–1207.
61. L. Shi, T. Wang, H. Zhang, K. Chang, X. Meng, H. Liu and J. Ye, Adv. Sci., 2015, 2, 1500006–1500014.
62. R. Liang, L. Shen, F. Jing, W. Wu, N. Qin, R. Lin and L. Wu, Appl. Catal., B, 2015, 162, 245–251.
63. L. Shen, S. Liang, W. Wu, R. Liang and L. Wu, Dalton Trans., 2013, 42, 13649–13657.
64. L. Shen, L. Huang, S. Liang, R. Liang, N. Qin and L. Wu, RSC Adv., 2014, 4, 2546–2549.
65. L. Shen, W. Wu, R. Liang, R. Lin and L. Wu, Nanoscale, 2013, 5, 9374–9382.
66. T. Kamegawa, S. Matsuura, H. Seto and H. Yamashita, Angew. Chem., Int. Ed., 2013, 52, 916–919.
67. G. Zhang and W. Choi, Chem. Commun., 2012, 48, 10621–10623.
68. D. Chatterjee, Catal. Commun., 2010, 11, 336–339.
69. J. He, J. Wang, Y. Chen, J. Zhang, D. Duan, Y. Wang and Z. Yan, Chem. Commun., 2014, 50, 7063–7066.
70. Y.-P. Yuan, L.-S. Yin, S.-W. Cao, G.-S. Xu, C.-H. Li and C. Xue, Appl. Catal., B, 2015, 168, 572–576.
71. X.-L. Liu, R. Wang, M.-Y. Zhang, Y.-P. Yuan and C. Xue, APL Mater., 2015, 3, 104403.
72. Y. Horiuchi, T. Toyao, M. Saito, K. Mochizuki, M. Iwata, H. Higashimura, M. Anpo and M. Matsuoka, J. Phys. Chem. C, 2012, 116, 20848–20853.
73. M. Nasalevich, R. Becker, E. Ramos-Fernandez, S. Castellanos, S. Veber, M. Fedin, F. Kapteijn, J. Reek, J. Van Der Vlugt and J. Gascon, Energy Environ. Sci., 2015, 8, 364–375.
74. J. He, Z. Yan, J. Wang, J. Xie, L. Jiang, Y. Shi, F. Yuan, F. Yu and Y. Sun, Chem. Commun., 2013, 49, 6761–6763.
75. C. Wang and Z. Xie, J. Am. Chem. Soc., 2012, 134, 7211–7214.
76. V. Houlding and M. Gratzel, J. Am. Chem. Soc., 1983, 105, 5695–5696.
77. T. D. Savić, Z. V. Šaponjić, M. I. Čomor, J. M. Nedeljković, M. D. Dramićanin, M. G. Nikolić, D. Ž. Veljković, S. D. Zarić and I. A. Janković, Nanoscale, 2013, 5, 7601–7612.
78. Z. Tachan, I. Hod and A. Zaban, Adv. Energy Mater., 2014, 4, 1301249.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ta07685h

---