Monolithic organic/inorganic ternary nanohybrids toward electron transfer cascade for enhanced visible-light photocatalysis

Abstract

A hierarchical electron transfer cascade system composed of monolithic organic/inorganic ternary nanohybrids (electrospun carbon nanofibers, ZnO nanorods, porphyrin) has been facilely prepared. The hybrid system can make concurrent use of efficient organic absorbers, dielectric screening and the electron shuttle provided by the inorganic components. Consequently, photogenerated electron–hole recombination is harnessed by the formed hierarchical electron transfer cascade. The monolithic organic/inorganic hybrid hereby exhibited remarkably high photocurrent generation. Moreover, the photocatalytic studies revealed that the hybrid displayed enhanced visible-light photocatalytic activity toward degrading organic waste compared with other components, and the reproducibility could be guaranteed after successive use. These results pave a way to discover new organic/inorganic assemblies for high-performance optoelectronic applications, as well as for device integration.

---

Introduction

Photocatalysis is considered as one of the emerging green processes, which has been increasingly employed for different purposes such as environmental remediation (airborne and water decontamination), energy generation (H₂ production), or organic synthesis by use of freely available, natural sunlight.^1,2 A photocatalysis event usually involves three key steps: (i) photonic absorption, (ii) photogenerated electrons and holes separation/recombination, and (iii) electron transfer (ET). Key factor to photocatalysis is the multistep ET in the photocatalysts.³ A well-organized multistep ET system allows efficient solar energy conversion by promoting ET to the photocatalyst surface, and then to the reactive species. To form multistep ET, construction of ternary nanohybrids is thereby sought on the principle of visible-light absorption, energy band match, and the more important hierarchical ET cascade.^3–5 However, challenge is still remained to translate the functional ternary nanohybrids into a macroscopic structure, forming an ET cascade system as a monolith. Thus, it is necessary to construct a monolithic ternary nanohybrid toward realizing ET cascade.

Over the past decade, low dimensional (0D or 1D) nano-/microstructured materials have been paid abundant interest because of their potentials as building blocks for miniaturized optoelectronic devices.^6,7 1D carbon nanofibers (CNFs) by electrospinning are thought to be ideal monolithic electron pathways due to their good conductivity (ρ = (3–7) × 10⁻³ Ω cm) and high specific surface area (SSA) rendered by its richly porous structure. It has been demonstrated that CNFs can efficiently capture and transport the photogenerated electrons through highly conductive long CNFs.^8,9 0D or 1D inorganic semiconductors are appealing as photocatalysts for their photostability, high charge-carrier mobility and weak electron–hole binding but suffer from inefficient coupling to light.^10–14 Photoactive organic materials exhibit a wide structural diversity, designing feasibility, and can act as not only a photosensitizer but an alternative photocatalyst, in particular when visible light is used. However one of the major problems associated with the use of organic photocatalysts is their limited stability and notoriously low charge-carrier mobility.^15–17 Regarding the individual properties of the three different material classes mentioned above, organizing the monolithic CNFs, photofunctional inorganic and organic materials into a monolithic system allows for not only combining the best of them but above all going beyond the characteristics of their constituents.¹⁸ So far, examples are still rare on fabricating monolithic organic/inorganic ternary nanohybrid as an ET cascade system that exhibit enhanced visible-light photocatalytic activity.

In this context, we report a first example of monolithic organic/inorganic ternary nanohybrids as an ET cascade system. It utilizes tetra(4-carboxyphenyl)porphyrin (TCPP), ZnO nanorods (NRs) and CNFs as electron donor–acceptor components. They are sequentially organized via fabrication of CNF monolith by electrospinnig (i), electrodeposition of ZnO NRs on CNF monolith (ii), graft of TCPP molecules around ZnO NRs (iii). The obtained monolithic CNF/ZnO/TCPP nanohybrids formed a hierarchical ET cascade that could efficiently suppress the recombination of photogenerated electrons and holes. As a result, the novel monolithic organic/inorganic ternary nanohybrids exhibited high photocurrent generation, and also tested to be an efficient, yet convenient photocatalyst for degrading organic wastes under visible light with excellent stability, easy recovery and recycling. Thus, the resulting monolithic organic/inorganic ternary nanohybrids are promising photocatalysts in environmental remediation.

Experimental

Sample preparation

Preparation of CNF monolith (I). CNFs were produced from precursor phenolic resin (PhR). In a typical experiment as we previously reported,¹⁹ PhR (25 wt%) and poly(vinyl butyral) (PVB, 0.6 wt%, M_w = 303 800) were dissolved in ethanol to form the polymer solution. Addition of PVB was to adjust the viscosity of the solution. Electrospinning of the polymer solution was conducted with a spinning distance of about 25 cm, a flow rate of 1 mL h⁻¹, and an applied voltage of 25 kV to obtain the polymer fibers. The as-spun fibers were then cured at a stepwise heating until 180 °C for 1 h, then carbonized in nitrogen and activated in water, separately, at 800 °C for 30 min.

Preparation of CNF/ZnO monolith (II). Electrodeposition was carried out in a three-electrode cell, in which CNF monolith was employed as the working electrode, and, platinum wire and saturated calomel electrode (SCE) were employed as counter and reference electrodes, respectively. Growth of ZnO NRs on CNF monolith was performed on CHI 600E in an aqueous solution containing 5 mM of ZnAc₂ and 0.1 M of KCl at 85 °C. After applying a high potential (−1.3 V vs. SCE) to the working electrode for 10 s, a shifted potential (−1.0 V vs. SCE) was continuously applied to it for the subsequent growth of ZnO NRs. During the whole process, O₂ was continuously bubbled onto the working electrode surface. After about 3000 s of electrodeposition, a layer of ZnO NRs on CNF monolith was obtained. It is noteworthy here CNFs exhibit hierarchical porous structure and large SSA, enabling an intimate contact between precursor solution and CNFs. Furthermore, CNFs present good conductivity and homogeneous property like flat electrode. Accordingly, well-dispersed ZnO deposits form on CNF framework via strong chemical bond. The detailed growing mechanism can be found in Fig. S1.†

Preparation of CNF/ZnO/TCPP monolith (III). This step aims to have TCPP molecules chemically bind to the surface of ZnO NRs. It is known that porphyrins bearing carboxylic acid anchoring groups have strong absorption coefficients to TiO₂, ZnO, etc. They have been grafted successfully to ZnO nanostructures possibly through unidentate coordination bond between carboxylic acid group and ZnO surface, resulting in an ester type of bond formation between them.²⁰ In our experiment, CNF/ZnO monolith was immersed in 0.2 mM TCPP tetrahydrofuran (THF) solution overnight to have TCPP molecules sufficiently coordinate to ZnO NRs in CNF/ZnO hybrids. The resulting ternary nanohybrids were subsequently washed with THF several times to eliminate nongrafted TCPP molecules, and then dried. Schematic illustration of the ternary nanohybrids sees Fig. 1.

Fig. 1 Schematics of the CNF/ZnO/TCPP ternary nanohybrids.

Characterizations

The morphologies were observed using Hitachi S-4800 scanning electron microscope (SEM), operated at an accelerating voltage of 10 kV. The X-ray diffraction (XRD) pattern was obtained with a Bruker D8 Focus under Cu-Kα radiation with a scanning speed of 5° min⁻¹. The X-ray photoelectron spectroscopy (XPS) measurements were performed on the Thermo Scientific ESCALAB 250Xi spectrophotometer. Raman spectra were recorded using a Renishaw Raman spectrometer (Invia reflex Raman microscope) equipped with a semiconductor-cooled CCD detector and a confocal Leica microscope. The spectrograph uses 1800 g mm⁻¹ gratings and a linearly polarized Nd:YAG laser (532 nm). UV-vis absorption spectra were recorded with a Shimadzu UV-2450 spectrophotometer.

Photoelectrochemical measurement

The photoelectrochemical measurement was carried on a CHI 660E potentiostat/galvanostat electrochemical analyzer in a three-electrode cell, in which monolithic samples were employed as the working electrode under illumination with AM 1.5 light (100 mW cm⁻²), and, platinum wire and SCE were employed as counter and reference electrodes, respectively. The electrolyte was 1 mM NaOH aqueous solution. The photocurrent was recorded by alternately opening and shutting the path of the incident light. The light was produced with Oriel Sol3A solar simulator (Newport).

Photoactivity evaluation

To eliminate the adsorptive influence on photocatalytic activity, the monolithic catalysts were first immersed in a Rhodamine B (RhB) aqueous solution (10 mg L⁻¹, 100 mL) in the dark overnight until an adsorption–desorption equilibrium was reached, and hence the adsorption capacity, which was defined as the maximum adsorption amount of RhB (mg) per mass of one catalyst (g), could be obtained. Through calculation, the adsorption capacity of RhB was determined to be 9.20, 8.25 and 8.13 mg g⁻¹ for CNF, CNF/ZnO and CNF/ZnO/TCPP monolith, respectively. Subsequently, 10 mg of the pre-treated monolithic catalyst was immersed in a diluted RhB (5 mg L⁻¹, 20 mL) aqueous solution and exposed to visible irradiation for the evaluation of the photocatalyst activity. A Xe lamp (500 W) equipped with a cut-off filter (λ > 400 nm) was used as the light source. At a given time interval of irradiation, small aliquots were withdrawn and measured using a spectrophotometer (Shimadzu UV-2450). The photodegradation of RhB pollutant over these as-fabricated samples was monitored by measuring the real-time UV-vis absorption spectra of RhB at 553 nm.

Results and discussion

Structure and morphology of CNF/ZnO monolith

Fig. 2 shows the optical and SEM images of the CNF and CNF/ZnO samples. The pure CNFs (I) and the hierarchically assembled CNF/ZnO hybrids (II) featured a monolithic (paper-like) and robust structure (Fig. 2A and B). Compared with the black CNF monolith, the CNF/ZnO was observed to be white on the surface. Fig. 2C and D portray the SEM images of CNF and CNF/ZnO monoliths, and their corresponding magnified images performed in Fig. 2E and F. It can be seen that pure CNFs have a fiber-like shape with smooth surface and stack randomly to form a network structure (Fig. 2C and E). The diameter of the CNFs ranged from 800 to 850 nm. When treated by electrodeposition, a layer of ZnO NRs were orderly embedded in the CNFs (Fig. 2D and F). It can be observed that the diameter of the embedded ZnO NRs was around 300 nm, and its length ranged from 1.5 to 2.0 μm while CNF diameters kept unchanged after electrodeposition.

Fig. 2 The optical images of the as-obtained CNF and CNF/ZnO monoliths (A and B); the SEM images of CNFs (C) and CNF/ZnO hybrids (D), and their corresponding magnified SEM images (E and F).

Fig. 3 portrays the XRD patterns of CNFs and CNF/ZnO hybrids. For CNFs, the broad peaks located around 22° was assigned to the (002) plane of the carbon structure in CNFs,²¹ and the other weak peaks around 29.4, 35.8, 39.4, 43.2, 47.5, 48.5 and 57.3° could be assigned to calcite CaCO₃ arising from the adhering plasticine.²² While for CNF/ZnO hybrids, beside the peak around 22° that belongs to CNFs, the diffraction peaks of ZnO were clearly observed, and all could be indexed as the hexagonal wurtzite structure (JCPDS 36-1451). It can be also seen that the diffraction peaks of ZnO were sharp and intense, implying the high crystallinity of the ZnO NRs in the CNF/ZnO hybrids. These results confirmed that the hybrids were composed of ZnO and CNFs.

Fig. 3 XRD spectra of CNFs and CNF/ZnO hybrids.

The chemical composition of CNF/ZnO hybrids was further compared with CNFs by XPS analysis. Fig. 4A shows the fully scanned spectra in the range of 0–1200 eV. The overview spectra demonstrated that C, O, and Zn existed in CNF/ZnO hybrids, whereas only C and O were present in pristine CNFs. In order to fully understand the surface chemistry of CNF/ZnO, the high-resolution XPS spectra of C 1s region around 285 eV was shown in Fig. 4B. The peak at 284.6 eV is attributed to graphitized carbon. The peak at 285.2 eV corresponds to C–C bands. Additionally, the two peaks at 286.2 and 290 eV are characteristic of the oxygen bound species C–O and C=O, respectively.^23–25 Fig. 4C presents the O 1s spectrum for CNF/ZnO. The peaks at 530.8, 531.4, 532, 533, and 533.6 eV were assigned to Zn–O in ZnO (lattice O),²¹ surface hydroxyl groups (O–H), oxygen making a single bond with carbon (C–O), adsorbed O₂ and oxygen making a double bond with carbon (C=O), respectively.²⁵ The Zn 2p spectrum in Fig. 4D shows two symmetric peaks, of which, the peaks centered at 1022.1 eV is ascribed to Zn 2p_3/2 and the other at 1045.0 eV corresponds to Zn 2p_1/2, indicating the valence state of Zn²⁺ in CNF/ZnO hybrids. Thus, it can be deduced that the hybrids were composed of Zn(II), C and O elements, in good agreement with the XRD results.

Fig. 4 (A) XPS fully scanned spectra of CNFs and CNF/ZnO hybrids; C 1s (B), O 1s (C), Zn 2p (D) scan for CNF/ZnO hybrids.

The N₂ adsorption/desorption isotherms of CNFs and CNF/ZnO hybrids in Fig. 5A show a typical type-IV isotherm with a clear hysteresis loop between the adsorption and desorption branches, implying the existence of a large amount of mesopores. The Brunauer–Emmett–Teller SSA of the CNFs was up to 406 m² g⁻¹ with a total pore volume of 0.27 m³ g⁻¹ (according to the capacity of the N₂ adsorption isotherm at P/P₀ of 0.99). While for CNF/ZnO hybrids, the SSA reached 153 m² g⁻¹ and the total pore volume was 0.15 m³ g⁻¹. The reduced SSA and pore volume were mainly caused by the blockage of pores during electrodeposition of ZnO NRs on CNFs, but still exceeded most other mesoporous photocatalysts.^26–29 The pore size distribution indicated a mean pore size of 2.7 nm for CNFs but 3.9 nm for CNF/ZnO hybrids, as shown in Fig. 5B. The mesoporous structure of CNF/ZnO hybrids is expected to improve the adsorption of organic pollutants, and the large SSA is of great importance for photocurrent generation in photoelectrochemical or photocatalytic applications.^30,31

Fig. 5 (A) Nitrogen adsorption–desorption isotherms and (B) pore size distributions of CNFs and CNF/ZnO hybrids.

Structure and optical properties of CNF/ZnO/TCPP monolith

To determine the structural composition of the CNF/ZnO/TCPP hybrids, Raman spectra of the as-obtained samples were investigated, as shown in Fig. 6. The CNFs displayed two apparent bands at 1333 and 1590 cm⁻¹, corresponding to the D (the presence of defects in the hexagonal graphitic layers) and G (the presence of sp² carbon-type structures) band in carbon.³² ZnO NRs were dominated by a sharp E₂ mode at 433 cm⁻¹.³³ When coupling, the CNF/ZnO hybrids exhibited the characteristic bands of both carbon and ZnO, indicating that ZnO NRs were successfully deposited on CNFs. For individual TCPP molecules, it is reasonable to assign the band at 1550, 1487, 1443, 1230, and 996 cm⁻¹ to pyrrolidine ring C_b−C_b (N) vibration, pyrrole ring C_b−C_b (NH) vibration, C_a–C_m stretching mode, C_m–Ph mode, and phenyl mode, respectively.^34–37 When TCPP molecules were grafted on CNF/ZnO hybrids, the CNF/ZnO/TCPP hybrids basically maintained the respective bands of each component, though adjacent bands of TCPP (1550 and 1330 cm⁻¹) and CNFs (G and D band) might overlap to form 1557 and 1337 cm⁻¹ in the CNF/ZnO/TCPP hybrids. Meanwhile, the original bands of TCPP at 1487, 1443, and 996 cm⁻¹ were somewhat shifted to higher frequencies, and that of ZnO at 433 cm⁻¹ was shifted to lower frequency to some extent. Such shift is most likely due to the charge transfer mechanism of the CNF/ZnO/TCPP hybrids related to the surface-state energy levels of the constituents.^17,38

Fig. 6 Raman spectra of the as-obtained samples.

The UV-vis absorption spectra of the as-obtained samples were shown in Fig. 7. TCPP displayed a wide absorption in visible light region. It demonstrates a Soret band common for all porphyrins, which appears at ca. 438 nm arising from the a_1u (π) to e^*_g (π) transition, and four less-intensive absorption peaks at 522, 556, 595, and 652 nm, which are attributed to the Q-bands of a_2u (π) to e^*_g (π) transition.^17,39,40 ZnO NRs exhibited a sharp fundamental absorption edge rise at about 400 nm. ZnO/TCPP hybrids involved the characteristic absorption peaks of both ZnO and TCPP, whereas the Soret band gets only visible for high TCPP adsorption, the intensity of the four Q-bands is too weak to be detected.⁴¹ Moreover, the Soret band of TCPP blue-shifted from 438 nm to 426 nm, which was possibly due to the charge transfer between ZnO and TCPP when coupling.¹⁷ CNFs displayed a strong absorption in the UV-visible light region. The absorption spectra of CNF-based hybrids, CNF/ZnO or CNF/ZnO/TCPP hybrids showed great similarities with that of CNFs because the extraordinarily high absorption intensity of CNFs made that of other components difficult to be detected. Nevertheless, the intense and broad absorption of CNF/ZnO/TCPP hybrids in the visible light region could possibly provide rich photocharges needed for the photocatalytic reactions.

Fig. 7 The UV-vis absorption of the as-obtained samples.

Photoelectrochemical properties

Photoelectrochemical measurements were performed to investigate the ET cascade between CNFs, ZnO NRs and TCPP. The photoresponse of the as-obtained samples was measured with light on/off cycles at 0.1 V versus SEC (Fig. 8). Pure ZnO NRs showed a weak anodic photocurrent of 1 μA cm⁻² (Fig. S2†). Contrarily, a cathodic photocurrent was generated for all others. Of them, the CNFs displayed a negligible photocurrent generation. The photocurrent density of ZnO/TCPP hybrids (ca. 3 μA cm⁻²) was higher than that of ZnO (ca. 1 μA cm⁻²) or CNF/ZnO hybrids (ca. 2 μA cm⁻²), implying that organic dye sensitization has a significant impact on the photocurrent generation as established previously.^29,42,43 It is worth noting that the photocurrent density of CNF/ZnO/TCPP hybrids was up to 6 μA cm⁻², about 2 and 3 times as high as that of ZnO/TCPP and CNF/ZnO, respectively. In the ternary hybrid, TCPP has a photosensitization effect that can broaden sunlight absorption of ZnO; ZnO has been proved to show surface plasmon resonance (SPR) excitation that might improve the light absorption of TCPP;^44,45 The presence of CNFs here influences the overall performance by (i) accepting electrons from the ZnO/TCPP hybrids and (ii) providing a medium to store and shuttle electrons within the hybrid film, resembling the ZnO–C, GR–TiO₂ or CNT–TiO₂ composite systems, etc.^46–50 These results herein highlight the advantages of TCPP, ZnO and CNFs for enhanced light absorption, and the more important occurrence of a multistep ET cascade in the ternary hybrid electrode that improves separation efficiency of photogenerated electron and hole pairs.

Fig. 8 Photoresponse curves of the as-obtained samples.

On the basis of the monolithic structure and the photoelectrochemical properties of CNF/ZnO/TCPP systems, the photocurrent generation diagram can be proposed as illustrated in Fig. 9. ZnO has the conduction band (CB) edge (E_CB = −4.2 eV) that lies below the LUMO band of TCPP (E_LUMO = −3.6 eV) and above the Fermi level of CNFs (∼−4.50 eV).²⁰ Therefore, cascading band positions are formed in the ternary hybrid. When the electron–hole pairs are generated in TCPP, electrons can be efficiently separated from holes by migrating to ZnO and then to CNFs in a cascading way along the potential gradient⁵¹ as Fig. 9 illustrates. In detail, when the ternary hybrids are photoirradiated, a high-energy photon first excites an electron from HOMO to the LUMO band of TCPP. Second, electron injection from the LUMO band of TCPP to the CB of ZnO NRs occurs, in addition to minor electron transfer from the LUMO band of TCPP to the Fermi level of the CNFs. In the former case, the collected electrons in the CB of ZnO NRs are then transferred into the CNFs network. Consequently, the generated electron–hole pairs are effectively separated, and the energy-wasteful recombination is largely minimized. Last, the active TCPP undergoes ET reaction with H₂O in the electrolyte solution, resulting in the photocurrent generation. Thus, the hierarchical ET cascade system of CNF/ZnO/TCPP hybrids contributes greatly to the high photocurrent generation. Meanwhile, such multiple ET system provides higher possibility of photogenerated carrier separation and promises for higher photocatalytic activity.

Fig. 9 Schematics of the energy diagram of CNF/ZnO/TCPP ternary hybrids.

Photocatalytic activities

The photocatalytic performance of the monolithic catalysts was evaluated in degrading RhB under visible light irradiation, and the results were shown in Fig. 10A. It revealed that no obvious RhB degradation occurred in the dark, even in the presence of the CNF/ZnO/TCPP system. Under visible light irradiation, RhB itself displayed negligible decrease on absorption, suggesting that the self-sensitized photodegradation of RhB could hardly take place in the absence of any photocatalysts. Also, there was no appreciable degradation of RB over the CNFs as a consequence of the very weak photocatalytic activity of pure CNFs. In comparison, 50% of RhB was degraded over CNF/ZnO catalysts within 320 min. Here, the exhibited photocatalytic activity by CNF/ZnO under visible light is possibly due to the carbon dope in ZnO during ZnO growth on CNFs. It was reported that carbon-doped ZnO could extend its optical response from UV to visible region caused by the substitution of lattice oxygen by carbon in ZnO,^52,53 although it is difficult to be detected in the UV-vis absorption spectra because of the high absorption background from CNFs. Moreover, a significant increase in the rate of degradation was observed when CNF/ZnO/TCPP system was employed with degradation efficiency up to 95%. The UV-Vis absorption spectra of RhB as a function of photoreaction time over CNF/ZnO/TCPP system (Fig. S3†) showed weak red-shift on the absorption wavelength with time. Furthermore, the photodegradation process is fit to pseudo-first-order kinetics, and a pseudo-first-order linear relationship was revealed as plots shown in Fig. S4.† From it, the photodegradation apparent rate constant (k), the degradation rate (r), and the photonic efficiency (η) of RhB in different systems were derived as listed in Table S1.† As it showed, the photonic efficiency of the CNF/ZnO/TCPP was much higher than CNF/ZnO and CNFs. The enhanced photocatalytic activity by the order of CNFs < binary hybrid < ternary hybrid showed great similarity with that of photocurrent generation. That is, in both photocurrent collection and photodegradation, the photosensitization effect of TCPP and the SPR excitation of ZnO enable great enhancement on visible-light absorption; the preferable mesoporous structure and electron transport property of CNFs can improve RhB adsorption and the electron transfer efficiency; moreover, the cascading band positioning formed in ternary hybrid is mostly important to maximize the electron transfer efficiency by facilitating electron migration.^51,54,55 All together contribute to the enhanced photocatalytic activity.

Fig. 10 (A) Photocatalytic degradation of RhB with different monolithic catalysts and (B) recycling experiment using the monolithic CNF/ZnO/TCPP hybrids for RhB degradation. Experiment condition: λ_irradiation > 400 nm.

The stability of CNF/ZnO/TCPP system was further examined for the concern of practical water treatment. As shown in Fig. 10B, seven repeated runs were conducted over CNF/ZnO/TCPP system to photodegrade RhB. It was found that the photocatalytic efficiency of CNF/ZnO/TCPP system displayed only a slight decrease without significant loss of activities when consecutively conducted 7 times. The results demonstrated that the CNF/ZnO/TCPP system could work as stable, yet efficient photocatalysts for degrading organic dyes under visible light irradiation. More importantly, such monolithic photocatalyst can be easily separated and recovered, and will make it possible to achieve substantial advancements in the remediation of real wastewaters.

Conclusions

In conclusion, we have developed a hierarchical ET cascade system by using CNF monolith as a versatile scaffold to anchor organic/inorganic hybrid materials. The obtained monolithic organic/inorganic ternary nanohybrid exhibited enhanced photocurrent generation, and hence enhanced photocatalytic efficiency toward degrading organic dyes under visible light irradiation, compared with any other component. Furthermore, it could be easily separated and recycled without apparent decrease of the photocatalytic activity possibly due to the geometrical restraint by the nanostructured monolith. These results open a perspective toward constructing new organic/inorganic hybrid materials for optoelectronic applications.

Acknowledgements

Authors are grateful for the financial support from the cooperative project JST-MOST (no. 2011DFA50430), the Beijing Higher Education Young Elite Teacher Project (no. YETP0419), and the National High Technology Research and Development Program of China (863 Program-no. 2010AA064907).

Notes and references

1. D. Pile, Nat. Photonics, 2012, 6, 70–70.
2. T. P. Yoon, M. A. Ischay and J. Du, Nat. Chem., 2010, 2, 527–532.
3. H. Hayashi, I. V. Lightcap, M. Tsujimoto, M. Takano, T. Umeyama, P. V. Kamat and H. Imahori, J. Am. Chem. Soc., 2011, 133, 7684–7687.
4. X. Liu, L. Pan, T. Lv, Z. Sun and C. Sun, J. Mol. Catal. A: Chem., 2012, 363, 417–422.
5. T. Lv, L. Pan, X. Liu, T. Lu, G. Zhu, Z. Sun and C. Q. Sun, Catal. Sci. Technol., 2012, 2, 754–758.
6. A. S. Aricò, P. Bruce, B. Scrosati, J.-M. Tarascon and W. Van Schalkwijk, Nat. Mater., 2005, 4, 366–377.
7. G. Konstantatos and E. H. Sargent, Nat. Nanotechnol., 2010, 5, 391–400.
8. H. E. Unalan, D. Wei, K. Suzuki, S. Dalal, P. Hiralal, H. Matsumoto, S. Imaizumi, M. Minagawa, A. Tanioka and A. J. Flewitt, Appl. Phys. Lett., 2008, 93, 133116.
9. J. Liu, J. Li, A. Sedhain, J. Lin and H. Jiang, J. Phys. Chem. C, 2008, 112, 17127–17132.
10. X. Wang, Z. Li, J. Shi and Y. Yu, Chem. Rev., 2014, 114, 9346–9384.
11. P. Zhang, B. Li, Z. Zhao, C. Yu, C. Hu, S. Wu and J. Qiu, ACS Appl. Mater. Interfaces, 2014, 6, 8560–8566.
12. B. Li, Z. Zhao, Q. Zhou, B. Meng, X. Meng and J. Qiu, Chem.–Eur. J., 2014, 20, 14763–14770.
13. F. Liu, Y. Jin, H. Liao, L. Cai, M. Tong and Y. Hou, J. Mater. Chem. A, 2013, 1, 805–813.
14. J. Jiang, Y. Feng, N. Mahmood, F. Liu and Y. Hou, Sci. Adv. Mater., 2013, 5, 1667–1675.
15. M. L. Marin, L. Santos-Juanes, A. Arques, A. M. Amat and M. A. Miranda, Chem. Rev., 2011, 112, 1710–1750.
16. Y. Chen, C. Zhang, X. Zhang, X. Ou and X. Zhang, Chem. Commun., 2013, 49, 9200–9202.
17. Y. Chen, Z.-H. Huang, M. Yue and F. Kang, Nanoscale, 2014, 6, 978–985.
18. C. Draxl, D. Nabok and K. Hannewald, Acc. Chem. Res., 2014, 47, 3225–3232.
19. Y. Chen, M. Yue, Z.-H. Huang and F. Kang, Chem. Eng. J., 2014, 252, 30–37.
20. A. Said, G. Poize, C. Martini, D. Ferry, W. Marine, S. Giorgio, F. Fages, J. Hocq, J. Boucle and J. Nelson, J. Phys. Chem. C, 2010, 114, 11273–11278.
21. J. Mu, C. Shao, Z. Guo, Z. Zhang, M. Zhang, P. Zhang, B. Chen and Y. Liu, ACS Appl. Mater. Interfaces, 2011, 3, 590–596.
22. C. G. Kontoyannis and N. V. Vagenas, Analyst, 2000, 125, 251–255.
23. J.-H. Zhou, Z.-J. Sui, J. Zhu, P. Li, D. Chen, Y.-C. Dai and W.-K. Yuan, Carbon, 2007, 45, 785–796.
24. Z. Dai, F. Shi, B. Zhang, M. Li and Z. Zhang, Appl. Surf. Sci., 2011, 257, 6980–6985.
25. P. Zhang, C. Shao, Z. Zhang, M. Zhang, J. Mu, Z. Guo, Y. Sun and Y. Liu, J. Mater. Chem., 2011, 21, 17746–17753.
26. L. Shang, C. Zhou, T. Bian, H. Yu, L.-Z. Wu, C.-H. Tung and T. Zhang, J. Mater. Chem. A, 2013, 1, 4552–4558.
27. J. Cao, L. Ren, N. Li, C. Hu and M. Cao, Chem.–Eur. J., 2013, 19, 12619–12623.
28. Q. Li, B. Guo, J. Yu, J. Ran, B. Zhang, H. Yan and J. R. Gong, J. Am. Chem. Soc., 2011, 133, 10878–10884.
29. Y. Shi, C. Hua, B. Li, X. Fang, C. Yao, Y. Zhang, Y. S. Hu, Z. Wang, L. Chen and D. Zhao, Adv. Funct. Mater., 2013, 23, 1832–1838.
30. Z. Wu and D. Zhao, Chem. Commun., 2011, 47, 3332–3338.
31. J. Du, X. Lai, N. Yang, J. Zhai, D. Kisailus, F. Su, D. Wang and L. Jiang, ACS Nano, 2010, 5, 590–596.
32. C. Kim, S. H. Park, J. I. Cho, D. Y. Lee, T. J. Park, W. J. Lee and K. S. Yang, J. Raman Spectrosc., 2004, 35, 928–933.
33. R. Cuscó, E. Alarcón-Lladó, J. Ibanez, L. Artús, J. Jiménez, B. Wang and M. J. Callahan, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 75, 165202.
34. D. Akins, C. Guo and H. Zhu, J. Phys. Chem., 1993, 97, 3974–3977.
35. B. Vlckova, P. Matejka, J. Simonova, K. Cermakova, P. Pancoska and V. Baumruk, J. Phys. Chem., 1993, 97, 9719–9729.
36. B. Vlckova, P. Matejka, P. Pancoska, V. Baumruk and V. Kral, Inorg. Chem., 1991, 30, 4103–4105.
37. T. M. Cotton, S. G. Schultz and R. P. Van Duyne, J. Am. Chem. Soc., 1982, 104, 6528–6532.
38. W. Tu, Y. Dong, J. Lei and H. Ju, Anal. Chem., 2010, 82, 8711–8716.
39. M. Zhu, Z. Li, B. Xiao, Y. Lu, Y. Du, P. Yang and X. Wang, ACS Appl. Mater. Interfaces, 2013, 5, 1732–1740.
40. J. Sun, D. Meng, S. Jiang, G. Wu, S. Yan, J. Geng and Y. Huang, J. Mater. Chem., 2012, 22, 18879–18886.
41. A. Said, G. Poize, C. Martini, D. Ferry, W. Marine, S. Giorgio, F. Fages, J. Hocq, J. Bouclé and J. Nelson, J. Phys. Chem. C, 2010, 114, 11273–11278.
42. A. L. Sisson, N. Sakai, N. Banerji, A. Fürstenberg, E. Vauthey and S. Matile, Angew. Chem., 2008, 120, 3787–3789.
43. A. Kira, T. Umeyama, Y. Matano, K. Yoshida, S. Isoda, J. K. Park, D. Kim and H. Imahori, J. Am. Chem. Soc., 2009, 131, 3198–3200.
44. J. R. Lombardi and R. L. Birke, J. Phys. Chem. C, 2014, 118, 11120–11130.
45. X. Wang, W. Shi, G. She and L. Mu, Phys. Chem. Chem. Phys., 2012, 14, 5891–5901.
46. J. Qu, L. Shi, C. He, F. Gao, B. Li, Q. Zhou, H. Hu, G. Shao, X. Wang and J. Qiu, Carbon, 2014, 66, 485–492.
47. A. Kongkanand and P. V. Kamat, ACS Nano, 2007, 1, 13–21.
48. Y. Zhang, Z.-R. Tang, X. Fu and Y.-J. Xu, ACS Nano, 2010, 4, 7303–7314.
49. A. Du, Y. H. Ng, N. J. Bell, Z. Zhu, R. Amal and S. C. Smith, J. Phys. Chem. Lett., 2011, 2, 894–899.
50. A. Du, S. Sanvito, Z. Li, D. Wang, Y. Jiao, T. Liao, Q. Sun, Y. H. Ng, Z. Zhu and R. Amal, J. Am. Chem. Soc., 2012, 134, 4393–4397.
51. H.-I. Kim, J. Kim, W. Kim and W. Choi, J. Phys. Chem. C, 2011, 115, 9797–9805.
52. Y.-G. Lin, Y.-K. Hsu, Y.-C. Chen, L.-C. Chen, S.-Y. Chen and K.-H. Chen, Nanoscale, 2012, 4, 6515–6519.
53. X. Zhou, Y. Li, T. Peng, W. Xie and X. Zhao, Mater. Lett., 2009, 63, 1747–1749.
54. F.-X. Xiao, J. Miao and B. Liu, Mater. Horiz., 2014, 1, 259–263.
55. H. Park, W. Choi and M. R. Hoffmann, J. Mater. Chem., 2008, 18, 2379–2385.

---

Footnote

† Electronic supplementary information (ESI) available: Additional photoelectrochemical and photocatalytic activity measurements. See DOI: 10.1039/c5ra01219h

---