Investigation of photoinduced electron transfer on TiO₂ nanowire arrays/porphyrin composite via scanning electrochemical microscopy

Abstract

In the present study, vertically aligned single-crystalline TiO₂ nanowire arrays with various lengths grown on transparent substrate have been successfully prepared using a facile hydrothermal method. They were then sensitized with a photoactive compound, 5,10,15,20-tetrakis(4-carboxylphenyl)porphyrin. The material was fully characterized by scanning electron microscopy (SEM), X-ray diffraction (XRD) and ultraviolet-visible absorption spectroscopy (UV-vis). A UV-vis/scanning electrochemical microscope (SECM) platform was developed by utilizing TiO₂ nanowire arrays, and the influence of wire length as well as wavelength and intensity of light on the electron transfer rate was discussed in detail through the generated probe approach curves. The results showed that the electron transfer rate at the surface of the TiO₂/porphyrin composite was accelerated by increasing the length of the nanowire, probably due to more porphyrin loading and the match of the conduction band edge of TiO₂ with the LUMO energy level of the porphyrin molecules, as evidenced by the DFT theoretical calculations. The maximum rate of electron transfer was reached at a wavelength of 546 nm and an illumination intensity of 90%. The efficient PET system constructed in this study will, we believe, provide useful information to help people investigate the mechanism of the charge transfer process of artificial photosynthesis.

---

Introduction

In the initial steps of photosynthesis, photoinduced electron transfer (PET) is one of the most important processes in the conversion of solar energy into chemical energy, and its has attracted significant scientific attention aimed at preparing artificial photosynthetic systems, energy harvesting photovoltaic devices and opto-electronic molecular switches.^1–16 The presence of porphyrins in natural photosynthetic systems ensures a source of electron donors. Moreover, porphyrin, known as “the pigment of life”,¹⁷ and its derivatives are excellent organic semiconductor materials. In this context, many porphyrins and their derivatives linked with an acceptor have been designed to emulate the photosensitizing, redox-active components, charge separation, antenna junctions of reaction centers and light harvesting complexes of natural photosynthetic membranes in the model studies of artificial photosynthesis.^1,18–23 For all those applications, light-harvesting porphyrin-based supramolecular assemblies are essential for the fabrication of supramolecular photo-functional devices.^24–26

Titanium dioxide (TiO₂) has been studied extensively because of its strong optical absorption, high chemical stability, environmental benignity and low cost.²⁷ In particular, TiO₂ nanowire (NW)²⁸ synthesized by a hydrothermal method has become a hot research material for photoelectrochemical (PEC) conversion due to its large internal surface area, good electrical transport, low charge carrier recombination losses and short diffusion distance for photo-generated carriers in the recent years.²⁹ Many groups have been working on developing TiO₂ NWs in the fields of dye-sensitized solar cells (DSSCs), PEC water oxidation, renewable energy, functional building blocks and photoinduced electron transfer.^30–34 Compared to the formerly used nanoparticles (NPs), NWs as a 1D nanostructure possesses the advantages of high length-to-diameter ratio and low number of grain boundaries and surface defects, as well as rapid electron transport along the NW, resulting in good PEC properties.³⁵ It was reported that the electron mobility in a single crystalline NW was about two orders of magnitude higher than that in a NC film.^36,37 A few studies, however, considered the effect of TiO₂ NW length on PET and used a TiO₂ nanowire arrays/porphyrin composite as a platform to investigate the PET process. It is, therefore, considerably interesting to study the charge transfer process using this novel PET system, in which oriented TiO₂ nanowire arrays grown on a transparent substrate are used as supporting material and porphyrin molecules are used as the sensitizer. For this purpose, many optical microscopy techniques, such as fluorescence microscopy and electron microscopy, are commonly utilized approaches, which can provide some information on chemical fluxes from individual cells.³⁸ Contrasting with the above mentioned methods, scanning electrochemical microscopy (SECM) based on the micro- and nano-interface is particularly suitable to determine the kinetics of electron transfer in a wide range of redox-active species with high spatial and architectonics.

A few pioneering works for researching the PET process of a porphyrin-based system by a new technique of combining SECM and UV-visible light illumination have been reported by our group.^39,40 For example, an electron donor-acceptor linked system of a porphyrin as the donor and benzoquinone (BQ) as the acceptor was developed as a sensor platform to detect benzoquinone via a PET process. However, this model was not actually efficient to annotate the electron transfer process in photosynthesis and relative kinetic information such as electron transfer rate was not given. To offer deeper insight into the mechanism of photosynthesis, we constructed here a simple light harvesting model by chemical pathways and investigated the multiple electron-transfer reaction. The relationship between the electron transfer efficiency and length of TiO₂ nanowire was examined carefully via a UV-vis/SECM technique in this study. In addition, it was found that the charge transfer rate heavily depends on the wavelength and intensity of light (Scheme 1).

Scheme 1 Schematic illustration of (A) TCPP deposited on TiO₂ NW array electrode, photo-induced charge transfer in the nanocomposite structures, and (B) energy band in the composite structures.

Experimental

Reagents

NaCl (AR, Xilong Chemical Ind., Co., Ltd), K₄Fe(CN)₆ (AR, Kaixin Chemical Ltd, Tianjin), tetrabutyl titanate (TNBT) (AR, Sinopharm Chemical Reagent Co., Ltd, Shanghai) and hydrochloric acid (HCl) (AR, Sigma-Aldrich Trading Co., Ltd, Shanghai) were used as received. 5,10,15,20-Tetrakis(4-carboxylphenyl)porphyrin (TCPP) was synthesized in our laboratory by the previous reported methodology. Unless otherwise stated, reagents were of analytical grade and were used as received. All aqueous solutions were prepared by ultrapure water with the resistivity of 18.25 MΩ cm.

Apparatus

UV-vis spectra measurements were conducted with a UV-1102 spectrometer (Shanghai, China). The morphologies of the TiO₂ nanowire were observed through SEM on an ULTRA plus. XRD patterns were recorded using a Shimadzu XD-3A (Japan). TiO₂ sample annealing was conducted in the tube furnace of Tianjin Central Experimental Furnace Co., Ltd., model SK-G05123K. Photoelectrochemical (PEC) measurements were performed with a home-built PEC system. A 150 W Xe lamp light resource was used as the irradiation source. PEC measurements were conducted on a CHI900B electrochemical workstation (CH Instruments, Austin, TX). A custom electrolytic cell was made from a Teflon base with cone-shaped lining and a small opening on the back side, which was used for light irradiation. All experiments were carried out at room temperature using a conventional three-electrode system of a Pt ultramicroelectrode (UME), a platinum wire and a saturated calomel electrode as the working electrode, counter electrode and reference electrode, respectively. A 25 μm diameter Pt UME (RG = 10, RG is the ratio of the overall electrode radius over the platinum disk radius) sealed in a borosilicate glass capillary under vacuum was used as the SECM tip electrode. The UME was characterized by cyclic voltammetry and optical microscopy imaging. When a graceful S-model cyclic voltammetry curve (as shown in S Fig. 2A) was obtained, it meant that the Pt UME was successfully prepared. Fluorine-doped tin oxide (FTO) glass pieces (with a specific surface conductivity of ca. 10 Ω cm⁻²) were purchased from Zhuhai Kaivo Electronic Components Co. (Guangdong, China). The FTO glass pieces were cut into 1 cm × 2.5 cm sized slides and ultrasonically cleaned in turn in absolute ethanol, acetone, and deionized water for 10 min. They were finally dried with nitrogen gas. EIS experiments were performed on a Multi-potentiostat (VMP2, Princeton Applied Research, USA) with a superimposed 5 mV sinusoidal voltage in the frequency range of 10 mHz to 100 kHz.

Preparation of TiO₂ NWs and sensitizer modification

The preparation of TiO₂ NWs was similar to that in the reference,²⁸ and the detailed procedure is as follows. A mixture of 10 mL of deionized water and 5 mL of concentrated hydrochloric acid was stirred in a sample vial. After adding 0.5 mL of tetrabutyl titanate to the mixture, the reaction mixture was stirred for 15 min until a transparent solution was obtained. Then, one piece of FTO substrate was placed at an angle against the wall of a 25 mL Teflon liner with the conducing side facing down. The hydrothermal synthesis was proceeded at 180 °C for 4, 5, 6, 7, 8, and 9 h, respectively, in an electric oven. After synthesis, the autoclave was cooled to room temperature in a fume cupboard for half an hour. The TiO₂ films were flushed with deionized water and ethanol. After annealing at 450 °C for 2 h, the TiO₂ films were immersed in a 0.1 mM ethanol solution of the carboxyl porphyrin for 24 h.^41,42 The prepared TiO₂/porphyrin electrodes were stored in a watchglass for use under dark conditions.

Photoelectrochemical behavior

Photoelectrochemical measurements were performed at a UV-vis/SECM platform. SECM experimentation records approach curves, wherein the normalized current I_T = i_T/i_T,∞ is plotted versus the normalized distance L = d/a. In the SECM operation, the feedback mode was employed to study the kinetics in order to obtain the kinetics of the heterogeneous ET with a good resolution. The SECM apparatus for photoelectrochemical experiments is described elsewhere. In the experiment, all approach curves are obtained with the same zero origin for a defined sample. I₃⁻/I⁻ in acetonitrile was used as the redox couple in this section of the experiment.

Theoretical calculations

The molecular structure of TCPP was optimized with density functional theory (DFT) at the B3LYP/6-31G(d) levels, and the HOMO and LUMO energy levels were calculated. All calculations were performed with the Gaussian 03 software package.

Results and discussion

Characterization of TiO₂ NWs and TiO₂–TCPP composite

The morphologies and thickness of the TiO₂ NWs were observed by SEM. Fig. 1A presents the top-surface images of a typical synthesized TiO₂ NW array sample. It can be seen that a highly uniform and densely packed array of TiO₂ NWs was obtained with an average wire diameter of approximately 30 nm. The 4.7, 6.3, and 7.2 μm thick TiO₂ nanowire films were achieved after reacting for 5 h, 7 h and 8 h, as shown from the cross-sectional view of the SEM images in Fig. 1B–D, respectively. Fig. 1 clearly indicates that the TiO₂ NWs grew almost perpendicularly from the substrate. The XRD pattern in Fig. 2 demonstrates a remarkably dominating (002) peak, suggesting the oriented growth of the TiO₂ nanowire along the [001] direction. The XRD data revealed that the films deposited on the FTO substrates are tetragonal rutile TiO₂,⁴² which is in good agreement with the published experimental data.²⁸ In addition, the XRD patterns (Fig. 2) indicated that the TiO₂ NWs were not only aligned, but also were single crystalline. It was reported that the length of TiO₂ NWs could be varied by changing the growth time.²⁸ According to this, TiO₂ NWs with different lengths have been fabricated in this work by adjusting the reaction time, and the results are listed in Table 1.

Fig. 1 SEM images of TiO₂ nanowire arrays after two-step growth: (A) top view; (B), (C) and (D) cross-sectional view of samples with the reaction time of 5, 7 and 8 h, respectively.

Fig. 2 XRD patterns of TiO₂ film on the FTO substrate prepared by the hydrothermal method after annealing at 450 °C for 2 h.

Table 1 Electron transfer rate constants (k_eff) of TiO₂–TCPP with different lengths of TiO₂ nanowires. D is the diffusion coefficient for I₃⁻ (D = 1.37 × 10⁻⁵ cm s⁻¹)

Sample	S1	S2	S3	S4	S5	S6
Time (h)	4	5	6	7	8	9
Length (μm)	2.5	4.7	5.3	6.3	7.2	9.2
keff (10^−2 cm s^−1)	2.74	3.288	3.836	4.384	5.151	5.918

---

TCPP was introduced into the system for the efficient absorption of light energy. Carboxylic acid groups are known to be spontaneously adsorbed onto TiO₂ surfaces. They form bidentate chelating, bidentate bridging, and/or ester-like binding with TiO₂ surfaces, depending on the experimental conditions.⁴¹ The UV-vis absorption spectra of TiO₂ NWs and TiO₂–TCPP nanocomposites are shown in Fig. 3. The TiO₂ NWs displayed a remarkable peak from 350 to 400 nm. The cut-off absorption at 400 nm is consistent with the typical band gap of rutile TiO₂ (3.0–3.2 eV). The TiO₂–TCPP exhibited a typical Soret band at 419 nm and four weak Q-band absorptions at 522, 560, 599 and 653 nm, illustrating the successful anchoring of TCPP onto the surface of TiO₂ NWs.

Fig. 3 UV-vis absorption spectra of the TiO₂ NWs (black line) and TiO₂–TCPP (red line).

Effect of the light wavelength and intensity

To examine the effects of the wavelength of light on the photoelectrochemical response, the PET process of the TiO₂–TCPP system was studied under the illumination of different wavelengths of light with an intensity of 100%. In this section of the experiment, I₃⁻ in acetonitrile was used as the sacrificial electron donor. LiI and I₂ were dissolved in the conventional solvent, acetonitrile.^43,44 Fig. 4A shows a series of probe approach curves (PACs) at the wavelengths of 469, 487, 515, 531, 546 and 570 nm. As expected, the fit of the experimental probe approach curves to the theoretical ones is very good and the apparent heterogeneous electron transfer rate constant, k_eff, which is extracted from the fitting of an experimental prove approach curve to the theoretical one, is determined according to k_eff = κD/r_T in cm s⁻¹. From the fitting, positive feedback was obtained when the tip approached the TiO₂ NWs due to reaction of I⁻ with excited TCPP, which indicates that a bimolecular reaction occurs. Porphyrins are able to absorb a certain wavelength of light. Herein, positive feedback was observed when the wavelength of the light was 515, 531, 546 and 570 nm. In contrast, negative feedback was obtained under illumination with wavelength of 469 and 487 nm due to the weak absorption by the porphyrin molecules. According to these studies,^45,46 the most likely charge transfer mechanism is as follows:

TiO₂ − TCPP + hυ → TiO₂ − TCPP* (1)

TiO₂ − TCPP* → TiO₂ − TCPP⁺ + e_CB⁻(TiO₂) (2)

2TiO₂ − TCPP⁺ + 3I⁻ → 2TiO₂ − TCPP + I₃⁻ (3)

I₃⁻ + 2e⁻ → 3I⁻ (at tip electrode) (4)

Fig. 4 (A) SECM PACs on an FTO substrate with TiO₂ NWs/TCPP film under different wavelengths of light illumination. (B) Dependence of k_eff on the light wavelength with SD definition (n = 5).

In addition, we also investigated the effect of light intensity on the regular probe approach curve. The wavelength of light was fixed, while the illumination intensity was varied. The effect of illumination intensity on k_eff is shown in Fig. 5. A slowly increasing trend of the current and k_eff vs. the light intensity was observed when the illumination intensity ranged from 10% to 90%, whereas both current and k_eff decreased once light intensity was increased to 90%. The results demonstrated that light intensity was not the only factor that influenced the k_eff and blind strengthening of the illumination intensity did not always work to accelerate electron transfer rate.

Fig. 5 (A) SECM PACs of FTO substrate developed TiO₂ NWs–TCPP under different light intensity at 546 nm. (B) Dependence of k_eff on the light intensity with SD definition (n = 5).

Effect of TiO₂ NWs

It is not uncommon to obtain longer TiO₂ NWs by extending the reaction time or repeating growth in a fresh precursor solution. However, these ways would lead to the diameter widening of the TiO₂ NWs and the decrease of the surface area due to gap filling caused by the following growth. In order to address this issue, NaCl was added in the second or third growth cycle, because it has been reported that Cl⁻ could selectively adsorb onto the surface of rutile TiO₂ nanowire and lead to wire growth along the [001] direction. The results show that the length of the TiO₂ NWs increased significantly with prolonged reaction time and presence of NaCl but the diameter of the wire remained almost unchanged after reported growth (Table 1). This ensured an increase in the surface area.³¹ On the basis of the above mentioned results, we investigated the electron transfer process with different lengths of TiO₂ NWs. The wavelength and illumination intensity were fixed. Fig. 6 shows the length profile of normalized PACs acquired from sample S1–6 in the solution of I₃⁻/I⁻ with light wavelength of 546 nm and intensity of 100%. The PACs of all the samples in Fig. 6A correspond to positive feedback, and k_eff slightly increased with increasing length of the wire. We believe that the enhanced k_eff for longer wire under visible light is likely due to the high loading amount of TCPP on the surface of TiO₂ NWs. It has been reported that the porous nanocrystalline TiO₂ layer with high surface area serves as an excellent platform for efficient light harvesting.^47,48 In other words, the larger the surface area TiO₂ NW possesses, the more the TCPP anchoring onto TiO₂ occurs.

Fig. 6 SECM PACs of TiO₂ NWs–TCPP on FTO substrate with different lengths of TiO₂ NWs: (S1) 2.5 μm, (S2) 4.7 μm, (S3) 5.3 μm, (S4) 6.3 μm, (S5) 7.2 μm, and (S6) 9.2 μm at 546 nm.

Moreover, the charge transfer process that occurs in the TiO₂ NWs–TCPP was studied by AC impedance spectroscopy to identify the results obtained from PACs, since electrochemical impedance spectroscopy (EIS) was considered to be a useful tool to characterize the charge-carrier migration.⁴⁹ The EIS were measured with the frequency ranging from 100 kHz to 0.01 Hz under illumination. The Nyquist plots for the TiO₂ NWs and TiO₂ NWs–TCPP electrodes are shown in Fig. 7A. Two semicircles at high- and medium-frequency ranges as well as a line tilted at an approximately 45° angle to the real axis at low frequency were observed. The equivalent circuit that modelled the TiO₂ NWs and TiO₂ NWs–TCPP electrode is shown in the lower right of Fig. 7A. In particular, the simulated equivalent circuit consisted of a series of resistances: the ohmic resistance of the electrolyte, membrane and electrode (R_s, starting point of the first semicircle in Nyquist plot); the resistance of the solid–electrolyte interface layer (R_SEI, first semicircle in Nyquist plot); charge transfer resistance (R_ct, second semicircle in Nyquist plot); constant phase element (CPE), which can be replaced by the electrochemical double-layer capacitance at the electrode–electrolyte interface; and the mass-transfer impedance (Z_W, the sloping line in Nyquist plot), which indicated Warburg impedance and represented the diffusion of K₃[Fe(CN)₆] ions.⁵⁰ The reaction between the electrolyte and the surface of the electrode caused the CPE and resistance R_SEI. Furthermore, the second semicircle was attributed to charge transfer kinetics.⁵¹ In the Nyquist plots, it is observed that the semicircle diameter of the TiO₂ NWs–TCPP electrode was smaller than that of the bare TiO₂ NWs electrode, implying that the charge transfer resistance (R_ct) of TiO₂ NWs–TCPP is smaller than that of TiO₂ NWs. As depicted in Table S1,† R_ct reduced in turn for the sample of S1 to S6. These results revealed that the charge transfer rate in TiO₂ NWs–TCPP was higher than that in plain TiO₂ NWs, and the rate for S6 was higher than that for S5, S4, S3, S2 and S1, consistent with the results gained from PACs.

Fig. 7 (A) Typical EIS, presented as Nyquist plots for the TiO₂ and TiO₂–TCPP electrodes under 546 light illumination and the equivalent circuit used to fit the EIS. (B) Typical EIS, presented as Nyquist plots for varying the length of TiO₂–TCPP electrodes under 546 light illumination.

Theoretical calculations

In order to affirm the electron transfer pathway, we performed density functional theory (DFT) calculations at the B3LYP/6-31G level to gain further insight into the geometries and electronic properties of TCPP. Six typical orbitals of TCPP were exhibited as follows: HOMO-2, HOMO-1, HOMO, LUMO, LUMO+1 and LUMO+2; their values were −6.63, −5.76, −5.38, −2.74, −2.73 and −1.85 eV, respectively. All unoccupied orbital values were higher than the lower bound level of the conduction band (CB) of TiO₂, indicating that the efficiency of charge injection from the excited sensitizer molecule to the TiO₂ CB is viable.⁵²

Transfer of electrons from the singlet excited-state of TCPP to the TiO₂ NWs must be downhill in energy. In this measurement, if the singlet excited-state of TCPP is more positive than that of the CB of TiO₂ NWs, then electrons should transfer easily from TCPP to TiO₂. As shown in Scheme 2, the photoexcitation of TCPP can give rise to electrons transferring from the HOMO to the LUMO. Afterwards, since the CB edge of TiO₂ is well matched with the energy level of the LUMO of TCPP and electronic coupling between the TCPP and the TiO₂ surface through the carboxyl group is fairly strong, efficient electron injection from the TCPP singlet excited-state to the CB of TiO₂ occurs. Moreover, a dye regeneration reaction happens via direct electron transfer from I-to the HOMO of TCPP.

Scheme 2 Schematic energy level diagram and electron transfer path.

Conclusions

In summary, a photo-induced electron transfer (PET) system was constructed by the combination of vertically aligned single-crystal TiO₂ nanowire arrays and porphyrin to investigate the mechanism of the charge transfer process in artificial photosynthesis. The reaction mechanism of the PET process was proposed, and the important kinetic information was evaluated using electrochemical impedance spectra and a unique technique of UV-vis/scanning electrochemical microscopy (SECM). It was found that the electron transfer rate was significantly accelerated once longer nanowire arrays were used due to more porphyrin loading and the match of the conduction band edge of TiO₂ with the LUMO energy level of the porphyrin molecules, as evidenced by DFT calculations. It was found that the rate constants of electron transfer reach the maximum at a wavelength of 546 nm and an illumination intensity of 90%. In a word, our results show that the integrated system in this study is an effective platform to gain deep insight into the mechanism of the PET process, which is known to be a critical part of artificial photosynthesis.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (Grant No. 21327005, 21175108); the Program for Chang Jiang Scholars and Innovative Research Team, Ministry of Education, China. (Grant No. IRT1283); the Program for Innovative Research Group of Gansu Province, China (Grant No. 1210RJIA001); the Program of Innovation and Entrepreneurial for Talent, Lan Zhou, Gansu Province, China (Grant No. 2014-RC-39).

Notes and references

1. M. R. Wasielewski, Chem. Rev., 1992, 92, 435.
2. M. P. Debreczeny, W. A. Svec and M. R. Wasielewski, Science, 1996, 274, 584.
3. S. Wallin, C. Monnereau, E. Blart, J. R. Gankou, F. Odobel and L. Hammarstrom, J. Phys. Chem. A, 2010, 114, 1709.
4. M. Berberich, A. M. Krause, M. Orlandi, F. Scandola and F. Wurthner, Angew. Chem., 2008, 120, 6718.
5. H. Imahori and S. Fukuzumi, Adv. Funct. Mater., 2004, 14, 525.
6. U. Bach, D. Lupo, P. Comte, J. E. Moser, F. Werssortel, J. Salbeck, H. Spreitzer and M. Gratzel, Nature, 1998, 395, 583.
7. M. Gratzel, Nature, 2001, 414, 338.
8. A. Hagfeldt and M. Gratzel, Acc. Chem. Res., 2000, 33, 269.
9. N. Aratani, D. Kim and A. Osuka, Acc. Chem. Res., 2009, 42, 1922.
10. A. Uetomio, M. Kzaki, S. Suzuki, K. I. Yamanaka, O. Ito and K. Okada, J. Am. Chem. Soc., 2011, 133, 13276.
11. V. Garg, G. Kodis, P. A. Liddell, Y. Terazono, T. A. Moore, A. L. Moore and D. Gust, J. Phys. Chem. B, 2013, 117, 11299.
12. Y. Leroux, D. Schaming, L. Ruhlmann and P. Hapiot, Langmuir, 2010, 26, 14983.
13. Y. Terazono, G. Kodis, K. Bhushan, J. Zaks, C. Madden, A. L. Moore, T. A. Moore, G. R. Fleming and D. Gust, J. Am. Chem. Soc., 2011, 133, 2916.
14. M. Gratzel, Inorg. Chem., 2005, 44, 6841.
15. Y. Y. Liang, Y. Wu, D. Q. Feng, S. T. Tsai, H. J. Son, G. Li and L. P. Yu, J. Am. Chem. Soc., 2009, 131, 56.
16. K. S. Leschkies, R. Divakar, J. Basu, E. E. Pommer, J. E. Boercker, C. B. Carter, U. R. Kortshagen, D. J. Norris and E. S. Aydil, Nano Lett., 2007, 7, 1793.
17. P. D. Frischmann, K. Mahata and F. Wurthner, Chem. Soc. Rev., 2013, 42, 1847.
18. M. R. Wasielewski, Acc. Chem. Res., 2009, 42, 1910.
19. (a) H. Imahori, K. Tamaki, D. M. Guldi, C. Luo, M. Fujitsuka, O. Ito, Y. Sakata and S. Fukuzumi, J. Am. Chem. Soc., 2001, 123, 2607; (b) H. Imahori, D. M. Guldi, K. Tamaki, Y. Yoshida, C. Luo, Y. Sakata and S. Fukuzumi, J. Am. Chem. Soc., 2001, 123, 6617; (c) F. Spanig, M. Ruppert, J. Dannhauser, A. Hirsch and D. M. Guldi, J. Am. Chem. Soc., 2009, 131, 9378; (d) F. D. Souza, N. K. Subbaiyan, Y. S. Xie, J. P. Hill, K. Ariga, K. Ohkubo and S. Fukuzumi, J. Am. Chem. Soc., 2009, 131, 16138.
20. (a) F. Wessendorf, J. F. Gnichwitz, G. H. Sarova, K. Hager, U. Hartnage, D. M. Guldi and A. Hirsch, J. Am. Chem. Soc., 2007, 129, 16057; (b) H. Lemmetyinen, N. V. Tkachenko, A. Efimov and M. Niemi, J. Phys. Chem. C, 2009, 113, 11475.
21. (a) H. Imahori, J. Phys. Chem. B, 2004, 108, 6130; (b) F. D. Souza, E. Maligaspe, K. Ohkubo, M. E. Zandler, N. K. Subbaiyan and S. Fukuzumi, J. Am. Chem. Soc., 2009, 131, 8787.
22. J. H. Kim, M. Lee, J. S. Lee and C. B. Park, Angew. Chem., Int. Ed., 2012, 51, 517.
23. P. D. Frischmann, K. Mahata and F. Wurthner, Chem. Soc. Rev., 2013, 42, 1847.
24. T. Honda, T. Nakanishi, K. Ohkubo, T. Kojima and S. Fukuzumi, J. Am. Chem. Soc., 2010, 132, 10155.
25. C. B. KC, K. Stranius, P. D'Souza, N. K. Subbaiyan, H. Lemmetyinen, N. V. Tkachenko and F. D'Souza, J. Phys. Chem. C, 2013, 117, 763.
26. D. Bonifazi, O. Enger and F. Diederich, Chem. Soc. Rev., 2007, 36, 390.
27. M. Xu, P. M. Da, H. Y. Wu, D. Y. Zhao and G. F. Zheng, Nano Lett., 2012, 12, 1503.
28. B. Liu and E. S. Aydil, J. Am. Chem. Soc., 2009, 131, 3985.
29. K. Shankar, J. I. Basham, N. K. Allam, O. K. Varghese, G. K. Mor, X. J. Feng, M. Paulose, J. A. Seabold, K. S. Choi and C. A. Grimes, J. Phys. Chem. C, 2009, 113, 6327.
30. C. Y. Zha, L. M. Shen, X. Y. Zhang, Y. F. Wang, B. A. Korgel, A. Gupta and N. Z. Bao, ACS Appl. Mater. Interfaces, 2014, 6, 112.
31. Z. J. Zhou, J. Q. Fan, X. Wang, W. H. Zhou, Z. L. Du and S. X. Wu, ACS Appl. Mater. Interfaces, 2011, 3, 4349.
32. (a) F. M. Pesci, G. M. Wang, D. R. Klug, Y. Li and A. J. Cowan, J. Phys. Chem. C, 2013, 117, 25837; (b) Y. C. Pu, Y. C. Ling, K. D. Chang, C. M. Liu, J. Z. Zhang, Y. J. Hsu and Y. Li, J. Phys. Chem. C, 2014, 118, 15086.
33. I. S. Cho, M. Logar, C. H. Lee, F. B. Prinz and X. L. Zheng, Nano Lett., 2014, 14, 24.
34. A. Wolcott, W. A. Smith, T. R. Kuykendall, Y. P. Zhao and J. Z. Zhang, Small, 2009, 5, 104.
35. H. Wang, Y. S. Bai, H. Zhang, Z. H. Zhang, J. H. Li and L. Guo, J. Phys. Chem. C, 2010, 114, 16451.
36. L. Forro, O. Chauvet, D. Emin, L. Zuppiroli, H. Berger and F. Levy, J. Appl. Phys., 1994, 75, 633.
37. M. Law, L. E. Greene, J. C. Johnson, R. Saykally and P. D. Yang, Nat. Mater., 2005, 4, 455.
38. K. Mckelvey, S. Martin, C. Robinson and P. R. Unwin, J. Phys. Chem. B, 2013, 117, 7878.
39. X. Q. Lu, Y. Q. Hu, W. T. Wang, J. Du, H. X. He, R. X. Ai and X. H. Liu, Colloids Surf., B, 2012, 103, 608.
40. W. T. Wang, Y. Q. Hu, C. M. Wang and X. Q. Lu, Electrochim. Acta, 2012, 65, 244.
41. S. Eu, S. Hayashi, T. Umeyama, Y. Matano, Y. Araki and H. Imahori, J. Phys. Chem. C, 2008, 112, 4396.
42. S. Hayashi, M. Tanaka, H. Hayashi, S. Eu, T. Umeyama, Y. Matano, Y. Araki and H. Imahori, J. Phys. Chem. C, 2008, 112, 15576.
43. K. Hosomizu, H. Imahori, U. Hahn, J. F. Nierengarten, A. Listorti, N. Armaroli, T. Nemoto and S. Isoda, J. Phys. Chem. C, 2007, 111, 2777.
44. Y. Shen, K. Nonomura, D. Schlettwein, C. Zhao and G. Wittstock, Chem.–Eur. J., 2006, 12, 5832.
45. U. M. Tafashe, K. Nonomura, N. Vlachpoulos, A. Hagfeldt and G. Wittstock, J. Phys. Chem. C, 2012, 116, 4316.
46. J. S. Lee, K. H. You and C. B. Park, Adv. Mater., 2012, 24, 1084.
47. J. R. Jennings, A. Ghicov, L. M. Peter, P. Schmuki and A. B. Walker, J. Am. Chem. Soc., 2008, 130, 13364.
48. E. Galoppini, J. Rochford, H. H. Chen, G. Saraf, Y. C. Lu, A. Hagfeldt and G. Boschloo, J. Phys. Chem. B, 2006, 110, 16159.
49. (a) P. Diao and Z. F. Liu, J. Phys. Chem. B, 2005, 109, 20906; (b) X. Pan, Y. Zhao, S. Liu, C. L. Korzeniewski, S. Wang and Z. Y. Fan, ACS Appl. Mater. Interfaces, 2012, 4, 3944.
50. Y. H. Jang, X. K. Xin, M. Byun, Y. J. Jiang, Z. Q. Liu and D. H. Kim, Nano Lett., 2012, 12, 479.
51. M. M. Rahman, J. Z. Wang, D. Wexler, Y. Y. Zhang, X. J. Li, S. L. Chou and H. K. J. Liu, Solid State Electrochem., 2010, 14, 571.
52. S. Karthikeyan and J. Y. Lee, J. Phys. Chem. A, 2013, 117, 10973.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra08485g

---