Supercapacitor and dye-sensitized solar cell (DSSC) applications of shape-selective TiO₂ nanostructures

Abstract

Shape-selective TiO₂ nanomaterials with different morphology, namely wire-like, flake-like and flower-like, have been synthesized by utilizing a simple wet chemical route by the reaction of titanium isopropoxide with ethanol and water mixture in the presence of cetyltrimethylammonium bromide. The shape of the particles can be easily tuned by altering the concentration of surfactant relative to the metal salt and changing the other reaction parameters. The formation mechanism of different shapes has been elaborated in detail. The shape-selective TiO₂ nanomaterials have been utilized for electrochemical supercapacitor and DSSC applications. It was observed that TiO₂ nanomaterials with various shapes showed different specific capacitance (C_s) values, and the order of C_s values is as follows: wire-like > flower-like > flake-like. The highest C_s of 3.16 F g⁻¹ and better cycling stability was observed for TiO₂ nanomaterials having wire-like shapes. At a high scan rate 150 mV s⁻¹, the capacitance retention of wire-like TiO₂ electrode remains at about 90% after 5000 cycles. DSSC study results show that all the differently shaped TiO₂ nanomaterials can be used as potential anode materials, and, among the different shapes, the flower-like morphology shows better photo-conversion efficiency. The presented synthesis process is fast, cost-effective and environmentally friendly and could be utilized for other applications like gas sensors, photo-catalysts or elimination of pollutants from contaminated soils.

---

Introduction

The controllable synthesis of nanocrystalline materials has received tremendous attention worldwide in the last couple of years due to their uncommon optical, electronic, catalytic, transport and mechanical properties compared to their bulk-phase counterparts. All these uncommon properties of materials at the nanoscale are mainly governed by their size, shape, crystallinity, arrangement and structure. When the size of a material is comparable to or smaller than its bulk Bohr exciton diameter, its optical property becomes strongly dependent on size due to quantum confinement of electrons and holes.^1,2 Recently, many efforts have been made to develop metals inside semiconductor materials with typical optical properties such as strong photoluminescence, electroluminescence or nonlinear optical behaviour which may lead to new optoelectronic devices with superior performance.^3,4 So exploring the ‘bottom up’ method for the facile synthesis of nanomaterials with controllable size, shape, and external properties is dictated by these structural parameters. So it is very important to develop new routes for the size- and shape-selective synthesis of nanoscale materials with reasonably short reaction time and in a controlled fashion.

Titanium dioxide (TiO₂) is an n-type semiconductor and a typical photocatalyst, attracting much attention from both fundamental and practical viewpoints. At ambient pressure TiO₂ exists in three polymorph forms: rutile, anatase and brookite. Antase and rutile are the two main crystalline phases of TiO₂ with band energies at 3.2 eV and 3.0 eV respectively. TiO₂ is well known for its various applications in the fields of photocatalysis, photo-electrochemistry, gas sensing, water purification, pigments and cosmetics.^5–7 Moreover, it is also used as a promising electrode material for dye-sensitized solar cells (DSSCs).⁸ Recently, the use of anisotropic TiO₂ nanomaterials instead of nanocrystalline films has been considered.^8,9 In addition, a more than twofold increase in maximum photo-conversion efficiency for water splitting has been observed by replacing TiO₂ nanocrystalline films with TiO₂ nanowires.⁹

A large number of preparation methods of TiO₂ nanomaterials have been investigated and reported in the literature.^11–29 These include the sol–gel method,¹⁰ direct deposition from aqueous solution,¹¹ sputtering technology,¹² ultrasonic spray pyrolysis¹³ and hydrothermal crystallization.¹⁴ Nanocrystalline anatase is generally synthesized by the sol–gel method and hydrothermal method using titanium alkoxides.^10,14 Ultrafine powder of TiO₂ was prepared by the citrate gel method¹⁵ and digestion method.¹⁶ Anatase TiO₂ thin films have been prepared by the sol–gel method in the presence of surfactant.¹⁷ Ravi et al. prepared nanocrystalline TiO₂ by the tartrate gel method.¹⁸ Monodispersed porous TiO₂ was prepared by Wang et al. using a Teflon-lined stainless steel autoclave.¹⁹ Rutile type TiO₂ rods were grown over glass substrate at low temperature (80 °C) under hydrothermal condition.²⁰ Size-controlled synthesis of TiO₂ nanorods was achieved by non-hydrolytic sol–gel reaction of continuously delivered two titanium precursors using two separate syringe pumps.²¹ Nanowhiskers of rutile TiO₂ were synthesized by annealing at high temperature.²² Cozzoli et al. demonstrated the controlled growth of TiO₂ nanorods in oleic acid surfactant at 80 °C.²³ The growth of TiO₂ pillars by chemisorbed nanotitania was realized recently by a high-temperature annealing process.²⁴ Apart from these examples, there are several other reports of TiO₂ nanomaterials.^25–29 Although there are several reports on TiO₂ in different forms, the preparation of size- and shape-controlled TiO₂ nanomaterials at low temperature and in short reaction time is not much highlighted. Among the various ways of synthesis, the size- and shape-controlled synthesis of TiO₂ nanomaterials in solution phase would be one of the best synthetic routes for obtaining stable colloidal suspensions.

Apart from the different applications discussed before, the specific application of TiO₂ nanomaterials is focused more on electrochemical supercapacitors,³⁰ lithium-ion batteries³¹ and DSSC.^32–34 TiO₂ has been used as a promising electrode material for Li-ion batteries and supercapacitors due to its safety against overcharging and stable voltage plateau at 1.78 V. Between ZnO and TiO₂, it was observed that TiO₂ is a more promising material in DSSC application. To the best of our knowledge, there is little information available for the formation of shape-selective TiO₂ nanomaterials in aqueous solution at low temperature and in shorter reaction time. Moreover, there is also less information available for the application of shape-selective anatase TiO₂ nanomaterials in electrochemical supercapacitor and DSSC applications.

In the present study, we report the synthesis of shape-selective TiO₂ nanomaterials in aqueous solution within an hour by conventional heating at ∼60 °C. Different morphologies, namely wire-like, flake-like and flower-like, have been synthesized by the reaction of titanium isopropoxide with ethanol in cetyltrimethylammonium bromide (CTAB) surfactant media. The size and shape of the nanomaterials can be tuned by varying the concentrations of surfactant and metal salt and the other reaction parameters. The shape-selective TiO₂ nanomaterials have been utilized as potential anode materials in electrochemical supercapacitor application and in DSSC studies. From study of supercapacitance, it was observed that the TiO₂ nanomaterials showed different specific capacitance (C_s) values for the various shapes, and the order of C_s values is as follows: wire-like > flower-like > flake-like. The highest C_s of 3.16 F g⁻¹ was observed for TiO₂ having wire-like shapes. Moreover, the shape-selective TiO₂ nanomaterials were also tested for DSSC applications and the observed photo-conversion efficiency was found to be highest in the case of flower-like shape compared to the other shapes. The proposed synthesis procedure is simple, cost-effective, less time consuming and does not require any harsh reduction conditions or any toxic chemicals.

Experimental section

Reagents

CTAB (99%) was purchased from Sigma-Aldrich and used as received. Ti(IV) isopropoxide, ethanol, lithium iodide (LiI), iodine(I), hexachloroplatinic acid (H₂PtCl₆), di-tetrabutylammonium-cis-bis(isothiocyanato)-bis(2,2′-bipyridyl-4,4′-dicarboxylato) ruthenium(II) (also called N719 dye), N-methyl-2-pyrrolidone (NMP), acetylacetone and fluorine-doped tin oxide (FTO; resistivity 7 Ω per square) were purchased from Sigma-Aldrich. Acetonitrile was purchased from Fisher Scientific and used as received. Triton X-100 was obtained from Himedia and used as received. Sodium sulfate (Na₂SO₄), carbon black, and polyvinylidene fluoride (PVDF) were obtained from Alfa Aesar. Absolute ethyl alcohol (CH₃CH₂OH) was obtained from ChangshuYangyuan Chemical, China. De-ionized (DI) water was used for all synthesis and application purposes.

Instruments

The synthesized shape-selective CTAB–TiO₂ nanomaterials were characterized using several spectroscopic techniques. UV-visible (UV-Vis) absorption spectra were recorded with a Unico (model 4802) UV-Vis-NIR spectrophotometer equipped with a 1 cm quartz cuvette holder for liquid samples. Transmission electron microscopy (TEM) analysis was done with a Tecnai model TEM instrument (Tecnai™ G2 F20, FEI) with an accelerating voltage of 200 kV. Brunauer–Emmett–Teller (BET) analysis was performed with a Quantachrome® ASiQwin™, Quantachrome Instruments v2.02 and nitrogen (N₂) gas was used as an adsorptive for the determination of the surface area. Energy dispersive X-ray spectroscopy (EDS) analysis was done with a field emission scanning electron microscopy (FE-SEM) instrument (Zeiss ultra FE-SEM) with a separate EDS detector (INCA) connected to the instrument. X-ray diffraction (XRD) analysis was done using a PAN analytical Advanced Bragg-Brentano X-ray powder diffractometer with Cu K_α radiation (λ = 0.154178 nm) with a scanning rate of 0.020 s⁻¹ in the 2θ range 10–90°. Laser Raman measurements were carried out with a Renishaw inVia Raman microscope using an excitation wavelength of 632.8 nm (He–Ne laser). The excitation light intensity in front of the objective was ∼10 mW with a spectral collection time of 1 s for Raman experiments. The integration time for our measurement was set to 10 s. The photoluminescence (PL) study was done with a Varian (Cary Eclipse Winflr) fluorescence spectrophotometer (serial number el02045776) in both excitation and emission mode using a xenon pump lamp. Fourier transform infrared (FT-IR) spectroscopy analysis was done with a Nexus 670 (FTIR), Centaurms 10× (microscope) having a spectral range of 4000–400 cm⁻¹ with an MCT-B detector. Thermal analysis was done with a simultaneous thermal analyser TGA/DTA instrument (SDT Q600) and the analysis was performed in air. A hot air oven (temperature up to 1000 °C) was used to anneal the samples at specific temperature. All the electrochemical experiments were carried out using an AUTOLAB PGSTAT302N electrochemical work station in 1 M Na₂SO₄ aqueous solution. For DSSC, the thickness of working electrode was measured by an SJ-301 surface roughness tester. The conductivity (current–voltage, I–V) measurement was performed by using a solar simulator (SS80AAA) under light illumination of 1000 W m⁻². A spin coater (Spektron Company, Chennai) was used to coat the samples on glass substrate. The rotation speed was 1000 rpm and the samples were coated for 100 s on FTO plate for DSSC study.

Preparation of shape-selective CTAB–TiO₂ nanomaterials

Shape-selective TiO₂ nanomaterials were synthesized by the reaction of Ti isopropoxide with ethanol in CTAB micellar media. In a typical synthesis, 30 mL of ethanol was mixed with 2 mL of CTAB solution (10⁻¹ M) and stirred to give a homogeneous mixture. Then 1 mL of Ti isopropoxide solution was added and the solution mixture was heated to nearly 60 °C with continuous stirring. After heating and stirring for 1–2 min, a white precipitate appeared and with increasing time, the intensity of the color increased. After ∼40 minutes the solutions become fully white in color. Then 50 mL of DI water was added with continuous heating and stirring for another 20 min. The white solution containing the precipitate settled down at the bottom of the solution. The white precipitate was collected and washed with DI water and ethanol several times and finally centrifuged at 6000 rpm 3–4 times using DI water. The solid white mass was collected and dried. The dried sample was annealed at 600 °C for 1 hour using a hot air oven. The sample typically contains anatase TiO₂ having wire-like morphology. The other samples were prepared by changing the reagent concentrations. The final concentration of all the reagents used for synthesis, and the size and shape of TiO₂ particles are summarized in Table 1. Moreover, the overall preparation procedure is schematically shown in Scheme 1. The dried solid mass was used for different characterization and application purposes.

Table 1 The final concentration of all reagents used for synthesis, and the size, shape and BET surface area of TiO₂ nanomaterials

Set no.	Final conc. of CTAB (M)	Final conc. of Ti isopropoxide (M)	Volume of ethanol added (mL)	Total time of heating (min)	Shape of TiO2 particles	BET surface area (m^2 g^−1)	Approx. size of TiO2 nanomaterials	Shape distribution
1	2.4 × 10^−3	4.2 × 10^−2	30	60	Wire-like	31.89	L ∼ 0.7–1 μm; D ∼ 35 ± 5 nm	100% wire-like
2	2.4 × 10^−4	4.2 × 10^−2	30	60	Flake-like	91.90	∼40 ± 10 nm	∼95% flake-like
3	2.4 × 10^−5	4.2 × 10^−2	30	60	Flower-like	198.57	∼450 ± 100 nm	100% flower-like

---

Scheme 1 The overall preparation procedure for the TiO₂ nanomaterials.

Electrode fabrication and electrochemical characterization of as-prepared samples

The working electrodes were prepared by mixing 80 wt% of electro-active materials, 10 wt% of carbon black and 10 wt% of PVDF in the presence of NMP, and the resulting slurry was pasted on nickel foam (1 cm × 1 cm). Then, the electrodes were dried at 100 °C for 12 hours in a vacuum oven. A typical three-electrode experimental cell, equipped with a working electrode, platinum foil as a counter electrode, and an Ag/AgCl reference electrode, was used for measuring the electrochemical properties of the working electrode. The cyclic voltammetry (CV) measurements were obtained at various scan rates (5, 10, 25, 50, 75, 100, and 125 mV s⁻¹) over the potential range of 0–0.8 V. The specific capacitance (C_s) of the as-prepared electrodes was calculated from the CV curves according to the following equation:³⁵where I is the oxidation or reduction current, Δt is the time differential, m is the mass of the active material, v is the scan rate, and ΔV indicates the voltage range of one sweep segment. The galvanostatic charge–discharge curves were obtained over the potential range of 0–0.8 V at different current densities (60, 70, 80, 90, and 100 μA cm⁻²). Specific capacitance of the electrode material could also be calculated from the galvanostatic charge and discharge curves, using the following equation:³⁶where I is the discharge current, Δt is the time for a full discharge, m is the mass of active material excluding the binder and conductive carbon black, and ΔV is the voltage change of a full discharge excluding IR drop. Electrochemical impedance spectroscopy (EIS) measurements were carried out using a direct current bias of 0.1 V with a signal of 10 mV over the frequency range of 0.1 Hz to 100 kHz.

Electrode fabrication for DSSC studies using CTAB–TiO₂ nanomaterials

TiO₂ working electrode and Pt-coated counter electrode were designed and fabricated for use in DSSC applications. The synthesized CTAB–TiO₂ powder was ground in a porcelain mortar with addition of 5 mL of DI water and 3 mL of acetylacetone solution. Then, a few drops of Triton X-100 were added to the TiO₂ dispersion to make a paste. The TiO₂ paste was coated on the conducting layer of FTO plate by using the screen printing method. The working area of the TiO₂ films was 0.16 cm². Then the TiO₂ films were annealed with a programmable heating process: at 100 °C for 60 min, 200 °C for 20 min, 300 °C for 20 min, 400 °C for 15 min. While electrodes were cooled to 80 °C, they were immersed in 0.3 mM N719 dye in ethanol for 4 hours. Conventional counter electrodes were prepared using a thermal reduction method. 5 mg of hexachloroplatinic acid (H₂PtCl₆) was mixed with 1 mL of isopropyl alcohol (C₃H₈O) solution and the solution mixture was deposited via spin coating onto cleaned FTO glass with a pre-drilled hole and heated at 400 °C for 60 min. Then the dye-absorbed TiO₂ working electrode and Pt-coated counter electrode were fixed together with less than 1 mm spacing using Parafilm paper. After sealing, the liquid electrolyte, which was composed of 0.3 M lithium iodide (LiI) and 0.03 M iodine(I) in acetonitrile, was injected into the cell through a pre-drilled hole on the prepared cell assembly. Finally the efficiency of the fabricated materials was calculated by using a solar simulator, under xenon light illumination with intensity of 1000 W m⁻².

Preparation of samples for other characterizations

The shape-selective CTAB–TiO₂ nanostructures were characterized using UV-Vis, TEM, EDS, XRD, PL, laser Raman, FT-IR, and thermal analyses. The TiO₂ nanostructure solution after successive centrifugation and annealing was used for the measurement of UV-Vis spectra after dispersing in aqueous solution. The solid TiO₂ powder was mixed with DI water, sonicated for 30 min, and used for TEM sample preparation and other thin-film preparation. The dispersed TiO₂ nanostructures in aqueous solution were used for the PL measurement. The samples for TEM were prepared by placing a drop of the corresponding TiO₂ nanostructure solution onto a carbon-coated Cu grid followed by slow evaporation of solvent at ambient conditions. For EDS, XRD, laser Raman, and FT-IR analysis, glass slides were used as substrates for thin-film preparation. The slides were cleaned thoroughly in acetone and sonicated for about 30 min. The cleaned substrates were covered with the TiO₂ nanostructure solution and then dried in air. After the first layer was deposited, subsequent layers were deposited by repeatedly adding more TiO₂ nanostructure solution and drying. Final samples were obtained after 4–5 depositions and then analyzed using the above techniques. For TGA/DTA analysis, the as-prepared TiO₂ nanopowders (before annealing) are directly used for the measurement. The sample preparation for supercapacitor and DSSC studies has already been discussed above.

Results and discussion

UV-Vis spectroscopic analysis

Fig. 1 shows the room temperature UV-Vis spectra of the different solution mixtures for the shape-selective synthesis of TiO₂ nanomaterials. Curve A shows the absorption spectra of only CTAB solution which has no specific absorption band in the visible region. Curve B shows the absorption spectrum of the mixture of Ti isopropoxide and CTAB. The spectrum shows a small hump at ∼310–320 nm due to ligand-to-metal charge transfer (LMCT) or due to interaction of Ti-salt solution with the surfactant.³⁷ Similar types of LMCT bands were also observed for gold chloride solution.³⁸ Curves C, D and E show the excitonic absorption spectra of the TiO₂ nanomaterials having wire-like, flake-like and flower-like morphology respectively. The bulk excitonic absorption spectra of TiO₂ having exciton binding energy at 10 meV have been observed at Bohr radius ∼2.35 nm.³⁹ The absorption bands for all the different structures show similar types of absorption features although the band position is shifted due to different morphology of the particles. The wire-like TiO₂ nanomaterial shows a small hump at 259 nm and a broad peak in the range 260–350 nm (curve C, Fig. 1). The flake-like TiO₂ nanomaterial shows a small hump at 262 nm and a small intense peak at 326 nm (curve D, Fig. 1). Similarly, the flower-like TiO₂ nanomaterial shows two small humps at 266 nm and 327 nm with a broad peak ranging from 257 to 383 nm (curve E, Fig. 1). For all three morphologies, the absorption band shifted a few nanometers among themselves which signifies the different size and shape of the particles. The existence of a band near 300 nm for all the morphologies indicates the existence of TiO₂ which is a typical absorption band of bulk TiO₂. All the excitonic features we observed for the shape-selective TiO₂ nanomaterials match with earlier reports.^23,40,41

Fig. 1 UV-visible (UV-Vis) spectra of the different solution mixtures for the shape-selective TiO₂ nanomaterial synthesis. Curve A shows the absorption spectrum of only CTAB solution; curve B shows the absorption spectrum of the mixture of Ti isopropoxide with CTAB solution; curves C, D and E show the excitonic absorption spectra of the TiO₂ nanomaterials having wire-like, flake-like and flower-like morphology respectively.

TEM and BET analysis

The TEM images of the shape-selective TiO₂ nanomaterials are shown in Fig. 2. Fig. 2A–C shows the low- and high-magnification TEM images of TiO₂ nanomaterials having wire-like morphology. The typical lengths of the wires are ∼0.7–1 μm and the diameters are ∼35 ± 5 nm. From the high-magnification image in Fig. 2C, the lattice spacing between two individual planes is ∼0.43 nm. The inset of Fig. 2C shows the corresponding selected area electron diffraction (SAED) pattern which confirms that the particles are crystalline in nature. Fig. 2D–F shows the TEM images at low and high magnification of the TiO₂ nanomaterials having flake-like morphology. The average sizes of the flake-like TiO₂ particles are ∼40 ± 10 nm. The lattice spacing between two individual crystal planes observed from Fig. 2F is nearly ∼0.25 nm. The inset of Fig. 2E shows the corresponding SAED pattern which indicates that the nano-flakes are crystalline in nature. Fig. 2G–I shows the low- and high-magnification images of the TiO₂ nanomaterials having flower-like morphology. From the images, the average diameters of the nano-flower are found to be ∼450 ± 100 nm. The high-magnification image in Fig. 2F shows that the lattice spacing between two individual planes is ∼0.20 nm. Careful observation of the high-magnification image of the nano-flower and the FE-SEM analysis (not shown here) indicate that the nano-flowers are formed with the aggregation of huge number of small spherical particles. The average diameters of the spherical particles are ∼20 ± 5 nm. The inset of Fig. 2I shows the corresponding SAED pattern which indicates that the nano-flowers are poly-crystalline which might be due to the presence of fewer small spherical particles at the beam spot during obtaining of the electron diffraction pattern. From the above TEM analysis, it is confirmed that different sized and shaped TiO₂ nanomaterials can be prepared by changing the reagent concentrations that are given in Table 1. Further, we also carried out BET analysis to get the specific surface area of the synthesized nanomaterials. The observed BET surface areas for wire-like, flake-like and flower-like TiO₂ nanostructures are 31.89, 91.90 and 198.57 m² g⁻¹ respectively as given in Table 1.

Fig. 2 Transmission electron microscopy (TEM) images of the shape-selective TiO₂ nanomaterials. (A–C) Show the low- and high-magnification TEM images of TiO₂ nanomaterial having wire-like morphology; (D–F) show the low- and high-magnification TEM images of the TiO₂ nanomaterial having flake-like morphology; (G–I) show the low- and high-magnification images of the TiO₂ nanomaterial having flower-like morphology. The insets of C, F and I show the corresponding SAED patterns.

EDS analysis

Fig. 3 shows the EDS analysis of the CTAB-stabilized TiO₂ nanomaterials. All three morphologies of shape-selective TiO₂ give similar types of spectral patterns and we analyzed the EDS spectrum taking flower-like morphology. The spectrum consists of peaks for the elements Ti, O, C and Br. The highly intense Ti and O peaks came from the TiO₂ nanomaterial samples and the C and Br came from the substrate (we used glass substrate with carbon film) and from the CTAB stabilizing agent respectively. The intensity of the Br peak is negligible which means that our sample is almost pure and free from excess surfactant.

Fig. 3 The energy dispersive X-ray spectroscopic (EDS) analysis of the CTAB-stabilized TiO₂ nanomaterials.

XRD analysis

Fig. 4 shows the XRD patterns of the three different morphologies of the shape-selective TiO₂ nanomaterials. Spectra A–C are the XRD patterns of wire-like, flake-like and flower-like morphologies respectively. All the diffraction patterns consist of the characteristic peaks originating from the lattice planes of (101) at 2θ value of 25.3°, (004) at 2θ value of 37.8°, (200) at 2θ value of 48°, (105) at 2θ value of 54°, (211) at 2θ value of 55.1°, (204) at 2θ value of 62.8°, (116) at 2θ value of 68.9°, (220) at 2θ value of 70.3° and (215) at 2θ value of 75.1°. All the diffraction peaks match properly with anatase TiO₂ nanomaterials having JCPDS file number 21-1272.⁴² We have not observed any peak from other phases which confirmed that our material is pure anatase phase. Moreover, we have not observed any peak below 25° which implies that our samples do not contain any impurities due to Ti(OH)₄. The different XRD patterns nicely match with other reports for the formation of anatase TiO₂ nanomaterials.^17,18,23

Fig. 4 The X-ray diffraction (XRD) patterns of the shape-selective TiO₂ nanomaterials. Spectra A, B and C are the patterns of wire-like, flake-like and flower-like morphologies respectively.

Laser Raman study

Fig. 5 shows the laser Raman spectra of shape-controlled TiO₂ nanomaterials. For the laser Raman study we have used a 633 nm He–Ne laser as excitation source. Laser Raman spectra can sensitively identify the anatase and rutile phase of TiO₂ on the basis of their characteristic Raman bands. Curves A, B and C are the Raman spectra for wire-like, flake-like and flower-like TiO₂ nanomaterials respectively. The Raman spectral features for three different morphologies are very similar. For all three morphologies, dominant sharp peaks appear at 175 cm⁻¹, 201 cm⁻¹, 403 cm⁻¹, 521 cm⁻¹ and 643 cm⁻¹ which are attributed to the symmetries of E_g, E_g, B_1g, A_1g and E_g respectively. The original Raman bands for anatase phase are reported at 144, 197, 399, 519 and 639 cm⁻¹ respectively⁴³ which matches nicely with our experimental data. It is reported that TiO₂ anatase phase might transform to rutile phase after heating above 700 °C as observed from both XRD and Raman studies.⁴⁴

Fig. 5 Laser Raman spectra of shape-controlled TiO₂ nanomaterials. Curves A, B and C are the spectra for wire-like, flake-like and flower-like morphologies respectively.

PL study

Fig. 6A and B shows the PL excitation and emission spectra of shape-selective TiO₂ nanomaterials having different morphologies. All the excitation spectra have similar spectral features and as an example we plot the excitation spectra obtained from wire-like morphology. As TiO₂ has a broadband absorption, the excitation spectrum is very significant for finding the excitation wavelength at which it has maximum emission. Fig. 6A shows the excitation spectrum at 274 nm where the peak wavelength appears at high absorption range. Fig. 6B shows the combined PL emission spectra of the three different shaped TiO₂ nanomaterials for an excitation wavelength at 274 nm. From the spectra, the strong emission bands appear at 377 nm, 428 nm, 487 nm and 529 nm. Two small weak emission bands also appeared at 460 nm and 600 nm. The emission bands observed above 377 nm are attributed to the quasi-free recombination at the absorption band edge, the shallow-trap state near the absorption band edge, the deep-trap band far below the band edge and a combination of these effects called surface state emission. The emission at 377 nm is due the excitonic emission. The emission bands at 428 nm, 487 nm and 529 nm are due to the surface state emission and recombination of the trapped electron–hole arising from dangling bonds on the TiO₂ nanomaterials.⁴⁵ The emission band intensity at 377 nm and 428 nm is similar although the intensity for 487 nm and 529 nm is less compared to 377 nm and 428 nm peaks. In the literature it is suggested that the 529 nm peak is generally observed due to O²⁻ vacancies, although as PL is mostly a surface phenomenon, the change in surface environment would also affect the specific position of the PL emission bands. As in our case we have CTAB surfactant at the surfaces as capping agent, so the original emission band for TiO₂ has been shifted to higher wavelength. The observed emission bands in our experiment show similarities with earlier reports.^46,47

Fig. 6 The photoluminescence excitation (A) and emission (B) spectra of shape-selective TiO₂ nanomaterials having different morphology.

FT-IR spectroscopic analysis

Fig. 7 shows the FT-IR spectra of pure CTAB and CTAB-capped TiO₂ nanomaterials. All three morphologies of CTAB–TiO₂ nanomaterials generate similar types of spectral pattern and as an example we show the spectrum of the wire-like morphology here. Curve A shows the FT-IR spectrum for only CTAB and curve B for CTAB–TiO₂ nanomaterials in the wavenumber region from 4000 cm⁻¹ to 400 cm⁻¹. A detailed comparison between the spectra not only supports the presence of CTAB on the TiO₂ surface but also indicates a probable interaction between them. The N–H stretching vibration of only CTAB appears at ∼3436 cm⁻¹ but in the case of CTAB-bound TiO₂, it is shifted and a broad band appears at 3390 to 3456 cm⁻¹ which indicates the binding of CTAB with the TiO₂ nanomaterial surface. We assumed that the surfactant might bind to the TiO₂ surface via the –N–H group and give rise to an unfavorable interaction by extending the hydrophobic tails towards the aqueous environment and generating a double layer arrangement. Three distinct and strong peaks for only CTAB appear at 2842 cm⁻¹, 2918 cm⁻¹ and 3015 cm⁻¹ due to stretching vibrations of –CH₃ and –CH₂ groups. All the strong peaks are not very prominent for the CTAB-bound TiO₂ sample which indicates the interaction with TiO₂ nanomaterials. Two intense peaks in the range 1350 to 1500 cm⁻¹ are observed for pure CTAB samples. The peaks at 1385 cm⁻¹ and 1468 cm⁻¹ are due to the deformation of –CH and –CH₂ groups which are absent or of very low intensity in the case of the CTAB-bound TiO₂ sample. A small intense but broad peak near 3600 cm⁻¹ is observed for both samples indicating the presence of moisture, the band arising due to stretching vibration of –OH group which generally appears between 3400 and 3600 cm⁻¹. In the case of the pure CTAB sample, there are three other distinct peaks at lower wavenumber region at 721 cm⁻¹, 910 cm⁻¹ and 957 cm⁻¹ which appear due to different structural modes specific to the surfactant which are absent in the spectrum of the CTAB-capped TiO₂ sample. In the case of CTAB–TiO₂ samples, different peaks in the range 460–680 cm⁻¹ are observed that are absent in the case of pure CTAB due to characteristic absorption of Ti–O bonds in the TiO₂ lattice that matches nicely with earlier reports.^48–50 Gao et al. also observed the strong band centered at 500–600 cm⁻¹ due to the vibration of Ti–O bonds.⁵¹ Wang et al. reported that TiO₂ submicrospheres showed Ti–O stretching mode at 568 cm⁻¹.⁵² So the above FT-IR analysis clearly confirmed the formation of TiO₂ nanomaterials and their specific binding with the CTAB surfactant.

Fig. 7 The Fourier-transform infrared (FT-IR) spectra of pure CTAB (curve A) and CTAB-capped TiO₂ nanomaterials (curve B).

Thermal analysis study

Fig. 8 shows the combined plot for thermogravimetric analysis (TGA) and differential thermal analysis (DTA) of the CTAB-bound TiO₂ nanomaterials which are carried out for determining the crystalline condition and thermal stability of the synthesized nanomaterials. The sample was heated up to 1000 °C with a rate of 10 °C min⁻¹ in air. Plot A shows the TGA curve and plot B shows the DTA curve. From the TGA curve, the first weight loss takes place at a temperature below 100 °C indicating that ethanol and water started to evaporate from the TiO₂ powder. At a temperature below 300 °C, another weight loss takes place which could be attributed to the removal of unhydrolyzed isopropoxide ligand or excess surfactant from the sample. The broad and smooth peak observed in the range 400–900 °C might be due to crystallization or phase change for TiO₂. In the DTA plot of curve B, a sharp exothermic peak appeared at 316.1 °C indicating the phase change of the TiO₂ powder from amorphous to anatase phase. The overall percentage weight loss calculated from the TGA curve was 29.76% while heating up to 1000 °C.

Fig. 8 (A) Thermogravimetric analysis (TGA) and (B) differential thermal analysis (DTA) of the CTAB-capped TiO₂ nanomaterials.

Mechanisms of formation of shape-selective TiO₂ nanomaterials

Shape-selective TiO₂ nanomaterials are formed by the reaction of Ti isopropoxide, ethanol, water and CTAB under conventional heating and stirring under 60 °C. In our proposed reaction, the presence of all the reagents is extremely important for the formation of specific morphology of TiO₂. In absence of CTAB and keeping other reaction parameters fixed, TiO₂ nanomaterials are formed but they aggregated to micron-size particles due to the absence of any specific stabilizing agent in the solution. In the reaction we used Ti isopropoxide as Ti salt precursor which is a light yellowish liquid having melting point of 17 °C and boiling point ∼232 °C and is soluble in several organic solvents like benzene, chloroform, ether and ethanol. The formation mechanism of the crystalline TiO₂ can be expressed as follows. The ethanolic solution of Ti isopropoxide reacts with water to form the TiO₂. The overall reaction is:

Ti{OCH(CH₃)₂}₄ + 2H₂O → TiO₂ + 4(CH₃)₂CHOH

It is given in the literature that an ethanolic solution of Ti salt in the presence of water initially forms titanium polycations like [Ti(OH)(H₂O)₅]³⁺ or [TiO(H₂O)₅]²⁺ and after a stipulated time the reaction generates Ti(OH)₄ which is white in color.⁵³ Then during the annealing process, the amorphous Ti(OH)₄ breaks down and generates the crystalline TiO₂. The annealing temperature is a major factor for the generation of specific phases. At low temperature, ∼200–300 °C, it becomes amorphous TiO₂ but above that temperature and up to 600 °C, it converts to anatase TiO₂. When annealing above 700 °C, the anatase TiO₂ gradually converts to rutile TiO₂. The initially formed Ti(OH)₄ species during annealing is converted into TiO₂. In our proposed reaction, at higher CTAB concentration, wire-like TiO₂ nanomaterials are formed, whereas, at medium or low CTAB concentration, flake-like or flower-like TiO₂ nanomaterials are formed as shown in Table 1. It was reported earlier that during shape-controlled synthesis, the formation of specific shape depends upon a few important factors like the faceting tendency of the stabilizing agent and the rate of supply of metal ions to the different crystallographic planes of the metal. Normally, the formation of nanostructures in a solution-based route takes place via three steps, namely nucleation, growth and stabilization. Other than these three factors, several other internal and external factors sometimes influence the formation of specific shapes. It was reported earlier that at relatively high CTAB concentration like 0.1 M or more, the surfactant itself formed a ‘wire-like’ or ‘worm-like’ micellar template, and while the particles grow on those templates, they generate those specific shapes.^54,55 However, this mechanism is reported for metal nanoparticles where Au nanorods are formed at higher CTAB concentration due to generation of ‘rod-shaped’ micellar template.⁵⁴ So in the present study, it is quite reasonable to understand that at higher CTAB concentrations, the wire-like TiO₂ nanomaterials are formed. Here, CTAB acts as a shape-directing agent, adsorbing on the surface of oxide nuclei and restricting their further aggregation and keeping their shape uniformity. At a moderate concentration, the process results in flake-like TiO₂ nanostructures, where the ‘templating’ effect of CTAB does not play any major role. At a comparatively low CTAB concentration, the flower-like TiO₂ nanomaterials are formed. Careful observation of FE-SEM images (not shown here) shows that the nano-flower structures are composed of very small-size aggregates of spherical particles. During continuous heating and stirring, the initially formed small spherical particles aggregated and fused together to form the flower-like morphology. So at low CTAB concentration, the nucleation and growth of the particles are not restricted like at high concentration and the TiO₂ nuclei grow in all possible directions to generate spherical particles and later they aggregate together to generate the flower-like morphology. At this point we have not fully confirmed the exact growth mechanism for the different shapes and further study necessary to get clear insight about the formation of shape-controlled TiO₂ in a single reaction. The potential of the three different morphologies of TiO₂ nanomaterials is examined for electrochemical supercapacitor and DSSC applications as described below.

Shape-selective TiO₂ nanomaterials in electrochemical supercapacitor applications

The electrochemical behavior of as-prepared electrodes was investigated by CV, galvanostatic charge–discharge, and EIS in a three-electrode system with 1 M Na₂SO₄ as the aqueous electrolyte. The ion transport behaviors of the as-prepared electrodes were characterized using the EIS technique with a frequency range of 0.1 Hz–100 kHz. Fig. 9A displays the Nyquist plots of flake-, flower- and wire-like TiO₂ electrodes. The Nyquist plots are composed of a semi-circular arc in the high-frequency region, followed by a linear part in the low-frequency region. The 45° sloped portion of the Nyquist plots indicates a typical Warburg impedance, which results from the frequency dependence of ion transport in the electrolyte.^56,57 The distinct Warburg region for the electrode suggests a higher ion diffusion resistance. The semi-circular arc in the high-frequency region corresponds to the charge transfer resistance (R_ct) of the electrodes and electrolyte interface.⁵⁶ The R_ct value of the flake-, flower- and wire-like TiO₂ electrodes was 437, 321, and 155 Ω, respectively.

Fig. 9 (A) Nyquist plots of flake-like, flower-like and wire-like TiO₂ electrodes and (B) the cyclic voltammetry (CV) measurements for flake-like, flower-like and wire-like TiO₂ electrodes measured at a scan rate of 25 mV s⁻¹ in 1 M Na₂SO₄ solution.

Fig. 9B shows the CV measurements for flake-, flower- and wire-like TiO₂ electrodes measured at a scan rate of 25 mV s⁻¹ in 1 M Na₂SO₄ solution. The shape of the CV curves slightly deviated from the ideal rectangular behavior, which indicates the pseudo-capacitance behavior of the TiO₂ electrodes. As seen from the CV curves, the wire-like TiO₂ electrode exhibited slightly larger area of the applied voltage and output current, indicating higher charge storage compared to flake- and flower-like TiO₂ electrodes. The detailed electrochemical processes of the flake-, flower- and wire-like TiO₂ electrodes are illustrated by CV curves at various scan rates as shown in Fig. 10A–C. It can be clearly observed that all the CV curves are close to a rectangular shape even at higher scan rate, indicating higher rate capability of the electrodes. It is observed from Fig. 10A–C that with an increase of the scan rate, the integrated area of the CV curves varies. From the CV curves, the specific capacitance values for the flake-, flower- and wire-like TiO₂ electrodes are calculated to be 1.84, 2.62 and 3.16 F g⁻¹ at a scan rate of 5 mV s⁻¹, respectively. Further, a comparison of scan rate dependence of specific capacitance of the TiO₂ electrodes is shown in Fig. 10D. The results show that the specific capacitance of all the electrodes gradually decreases with an increase of scan rate, which is due to insufficient time being available for ion transport inside the small pores at high scan rates.³⁵ The wire-like TiO₂ electrode showed better capacitive performance compared to flower- and flake-like electrodes because of the one-dimensional (1D) nanostructure morphology that increased the electroactive sites. In addition, the 1D nanostructure can reduce the diffusion path of active species and facilitates the fast ion diffusion into the inner region of the electrode, which finally leads to the improved capacitance and rate capability.⁵⁸

Fig. 10 (A–C) The cyclic voltammetry (CV) process of the flake-like, flower-like and wire-like TiO₂ electrodes at various scan rates. (D) Comparison of the scan rate dependence of specific capacitance of the TiO₂ electrodes.

Galvanostatic charge–discharge curves of flake-, flower- and wire-like TiO₂ electrodes measured at 0.1 mA cm⁻² at a working potential of 0 to 0.8 V are shown in Fig. 11A. It can be observed that the shapes of the curves slightly deviate from the typical triangular symmetric behavior which indicates the pseudocapacitance behavior of the electrodes. Further, an excellent rate performance of the electrodes at various current densities is a prerequisite for practical applications of supercapacitors. Fig. 11B–D depicts the galvanostatic charge–discharge behavior of the as-prepared flake- (Fig. 11B), flower- (Fig. 11C) and wire-like (Fig. 11D) TiO₂ electrodes at different current densities. The results show that while the current density increases, the discharge time reduces. According to the results, the specific capacitance of the flake-, flower- and wire-like TiO₂ electrodes can thus be calculated from the discharge curves to be 0.3, 0.4 and 0.45 F g⁻¹ at 60 μA cm⁻², respectively. Electrochemical stability with long cycle life of supercapacitors at high scan rate is essential for their practical application. Therefore, the as-prepared wire-like TiO₂ electrode was subjected again to CV measurement at a scan rate of 150 mV s⁻¹ for 5000 cycles, and the capacitance retention as a function of cycle number is displayed in Fig. 11E. It could be observed that for the first 2000 cycles, the capacitance increased instead of decreasing as in most cycle life measurements which indicates that the electroactive material is fully activated and exposed in the electrolyte and finally reached the optimum condition.^59,60 At the beginning, the electroactive materials are not fully used and after monotonic cycling, the active materials inside the nickel foam electrode will be fully exposed to the electrolyte. Hence, an increase in specific capacitance was observed in the cyclic tests. After 2500 cycles, the capacitance only decreases by about 10% of the initial capacitance over 5000 cycles, demonstrating a good cycling performance of the electrode.

Fig. 11 (A) The galvanostatic charge–discharge curves of flake-like, flower-like and wire-like TiO₂ electrodes measured at 0.1 mA cm⁻² at a working potential of 0–0.8 V. The galvanostatic charge–discharge behavior of the as-prepared flake-like (B), flower-like (C) and wire-like (D) TiO₂ electrodes at different current densities. Cycling performance of wire-like TiO₂ electrode at high scan rate of 150 mV s⁻¹ in 1 M Na₂SO₄ electrolyte solution (E).

DSSC studies of the shape-selective TiO₂ nanomaterials

The first idea of the DSSC was proposed in the late 1970s, although the first full report of DSSC came around 20 years later. In 1991, O'Regan and Grätzel first published a paper in Nature where they reported 7% efficiency when using oxide nanomaterials as anode.³² A DSSC is composed of a porous layer of oxide nanomaterials like ZnO, TiO₂ or SnO₂ which is covered with a molecular dye that absorbs sunlight similar to chlorophyll in green plant leaves. In a conventional DSSC, oxide nanomaterials are used as anode, platinum as a cathode and a liquid electrolyte is placed between them. An important issue for designing an efficient solar cell is that the electrons and holes generated in the absorber layer reach the membranes. Once the sunlight passes via the transparent electrode into the dye layer it can excite electrons that then flow into the metal oxide. Finally, the electrons travel through the wire from the anode towards the cathode and create an electrical current. DSSCs can work under low-light conditions like a cloudy sky or even for non-direct sunlight which is not possible for commercially available silicon-based costly solar cells. In the literature, there are several reports of ZnO and TiO₂ nanomaterials for DSSC applications, although TiO₂ shows better performance compared to ZnO.^32–34 The TiO₂ or ZnO nanocrystallites provide the films with the necessary large internal surface area which can maximize the uptake of the dye molecules, giving rise to DSSCs with large current density and high photon to current conversion efficiency. For example, ZnO aggregates,⁶¹ ZnO hierarchical structure,⁶² and TiO₂ beads⁶³ have been demonstrated to be efficient structures for improved photo-conversion efficiency when used in DSSC applications. The incident-photon-to-current conversion efficiency depends on several parameters like thickness and shape of the working electrode material, dye adsorption process, types of electrolyte and counter electrode materials. The short-circuit current in DSSC is mainly dominated by two factors: one is the dye loading on TiO₂ photoelectrodes and the other is charge recombination at photoanode.^64–66 The synthesized wire-like, flake-like and flower-like TiO₂ nanomaterials have been tested for their performance in photovoltaic conversion efficiency.

Fig. 12 shows the I–V characteristic curves for the different morphologies of CTAB–TiO₂ material. From the curves, the fill factor (FF) and the power conversion efficiency (η) can be calculated by using the following equations:

where V_max and J_max are the maximum cell voltage and current respectively at the maximum power point; J_sc is the short-circuit photocurrent density; V_oc is the open-circuit voltage; FF is the fill factor; and P_in is the input power density. Curves A–C of Fig. 12 are the I–V results for the flower-like, flake-like and wire-like morphologies of TiO₂ nanomaterials respectively. The solar cell efficiency of different morphologies of TiO₂ thin film with the same thickness is shown in Table 2. From Table 2, we can see that the efficiency is highest in the case of the flower-like morphology compared to the other morphologies. It was reported earlier that the anatase crystal structure of TiO₂ can be used as best semiconductor oxide for DSSCs, because an anatase film has a larger surface area per unit volume than a rutile film. So anatase TiO₂ is better able to absorb dye, has a larger electron diffusion coefficient, and has a shorter electron transit time.^67,68 In general, larger contact area improves the adherence between the TiO₂ layer and FTO surface which facilitates better electron transfer between them. According to the Lambert–Beer law, a higher absorbance means a higher concentration of the dye molecules. In our study, the flower-like structure having higher surface to volume ratio than flake-like and wire-like structures as confirmed from BET analysis also generates better adsorption of the dye molecules. It is well known that the photocurrent of DSSCs is correlated directly with the number of dye molecules adsorbed, and therefore an increase of adsorbed dye molecules results in an increase of incident light being harvested and consequently a larger photocurrent (J_sc). The change in TiO₂ morphologies affects electrical conductivity, meaning whether the conductivity is increasing or decreasing is influenced by the TiO₂ nanostructure. As discussed before and given in Table 1, the BET surface area of the different shaped nanomaterials decreases in the following order: flower-like > flake-like > wire-like. The flower-like structure gives a higher efficiency than flake-like and wire-like structures with the same thickness due to the larger surface area as confirmed from the BET analysis which in turn means it can adsorb more dye molecules. The better dye adsorption by the flower-like structures was further confirmed by a de-adsorption experiment using UV-Vis spectrophotometry where the amounts of dye adsorbed can be calculated. The adsorbed dye in the working electrode was de-adsorbed while immersed in ethanolic NaOH solution for two hours. From the absorbance data, the amounts of dye adsorbed were calculated, being 1.1 × 10⁻⁷ mol cm⁻² for flower-like, 0.72 × 10⁻⁷ mol cm⁻² for flake-like and 0.65 × 10⁻⁷ mol cm⁻² for wire-like TiO₂ samples. Increasing the dye adsorption in the TiO₂ layer is also useful for the electron generation processes to produce more electrons.⁶⁹ In general, in the fabrication of DSSCs, the dye sensitization plays a crucial role in obtaining good solar conversion efficiency. The highest efficiency we observed for the flower-like shape is ∼3.16% which is promising for a preliminary study. To achieve much higher photovoltaic efficiency, we are currently optimizing the testing conditions by altering several physical parameters of the nanomaterials such as increase in surface area, morphology and particle size and shape distribution and altering the electrolyte, which will be discussed in the near future.

Fig. 12 The current (I)–voltage (V) characteristics for the DSSC using CTAB–TiO₂ nanomaterials having different morphologies of wire-like, flake-like and flower-like structures.

Table 2 Summary of the different parameters used to calculate the efficiency of DSSC

Shapes of the CTAB–TiO2 nanomaterials	Jsc (mA cm^−2)	Voc (V)	FF	η (%)
Flower-like	13.75	0.460	0.503	3.16
Flake-like	12.25	0.451	0.447	2.45
Wire-like	12.00	0.451	0.397	2.14

---

Conclusion

In conclusion, shape-selective TiO₂ nanomaterials with different morphology, namely wire-like, flake-like and flower-like, have been synthesized by utilizing a simple wet chemical route by the reaction of Ti isopropoxide with ethanol and water in the presence of CTAB. The shape of the particles can be easily tuned by altering the concentration of surfactant and metal salt and by changing the other reaction parameters. The characterization of the different morphologies was done by different spectroscopic techniques. The mechanism of the formation of the different shapes has been elaborated in detail. The shape-selective TiO₂ nanomaterials have been used in electrochemical supercapacitor and DSSC studies. It was found that the TiO₂ nanomaterials showed different specific capacitance (C_s) values depending on the various shapes and the order of C_s values are as follows: wire-like > flower-like > flake-like. The highest C_s of 3.16 F g⁻¹ was observed for TiO₂ having wire-like shape. Furthermore, the wire-like TiO₂ electrode exhibited an excellent cycling stability with 90% specific capacitance being retained after 5000 cycles. From DSSC studies it was observed that the different shaped TiO₂ nanomaterials can be used as potential anode materials. Among the three different shapes tested, the flower-like morphology shows better efficiency compared to the other shapes due to its larger BET surface area. This larger surface area leads to better adsorption of the dye molecules which in turn enhances the conversion efficiency. The presented synthesis process is fast, cost-effective and environmentally friendly. The synthesized TiO₂ nanomaterials can be utilized in other applications like gas sensing, photocatalysis or pollutant elimination from contaminated soils.

Acknowledgements

The research funding from DST, SERB, New Delhi (DST Fast Track Project number SR/FT/CS-98/2011, GAP 7/12), support from the Central Instrumental Facility (CIF) and help from Mr A. Rathishkumar (TEM in-charge, CIF), Mr R. Ravishankar and Mr J. Kennedy (SEM in-charge, CIF), CSIR-CECRI, Karaikudi are greatly appreciated. U. Nithiyanantham wishes to thank CSIR-CECRI for a research internship fellowship. S. Kundu wishes to acknowledge Dr Vijayamohanan K. Pillai, Director and Dr M. Jayachandran, HOD, ECMS Division, CSIR-CECRI for their continuous support and encouragement.

References

1. A. P. Alivisatos, Science, 1996, 271, 933.
2. L. E. Brus, J. Chem. Phys., 1983, 79, 5566.
3. S. Ito, Y. Yamada, M. Kuze, K. Tabata and T. Yashima, J. Mater. Sci., 2004, 39, 5853.
4. Y. Nakato, A. Tsumura and H. Tsubomura, Chem. Phys. Lett., 1982, 85, 387.
5. G. Ramakrishna and H. N. Ghosh, Langmuir, 2003, 19, 505.
6. M. M. Rahman, K. M. Krishna, T. Soga, T. Jimbo and M. Umeno, J. Phys. Chem. Solids, 1999, 60, 201.
7. E. Pelizzetti and C. Minero, Elecrochim. Acta, 1993, 38, 47.
8. W. U. Huynh, J. J. Dittmer and A. P. Alivisatos, Science, 2002, 295, 2425.
9. S. U. M. Khan and T. Sultana, Sol. Energy Mater. Sol. Cells, 2003, 76, 211.
10. H. S. Yun, K. C. Miyazawa, I. Honma, H. S. Zhou and M. Kuwabara, Mater. Sci. Eng., C, 2003, 23, 487.
11. K. Shimizu, H. Imai, H. Hirashima and K. Tsukuma, Thin Solid Films, 1999, 351, 220.
12. M. R. Teresa Viseu and M. I. C. Ferreira, Vacuum, 1999, 52, 115.
13. M. D. J. Blesic, Z. V. Saponjic, J. M. Nedeljkovic and D. P. Uskokovic, Mater. Lett., 2002, 54, 298.
14. D. S. Zhang, T. Yoshida and H. Minoura, Adv. Mater., 2003, 15, 814.
15. S. R. Dhage, R. Pasricha and V. Ravi, Mater. Res. Bull., 2003, 38, 1623.
16. S. R. Dhage, V. Choube, V. Samuel and V. Ravi, Mater. Lett., 2004, 58, 2310.
17. G. Q. Liu, Z. G. Jin, X. X. Liu, T. Wang and Z. F. Liu, J. Sol-Gel Sci. Technol., 2007, 41, 49.
18. S. R. Dhage, S. P. Gaikwad and V. Ravi, Bull. Mater. Sci., 2004, 27, 487.
19. Y. Wang, H. Xu, X. Wang, X. Zhang, H. Jia, L. Zhang and J. Qiu, J. Phys. Chem. B, 2006, 110, 13835.
20. K. Kakiuchi, E. Hosono, H. Imai, T. Kimura and S. Fujihara, J. Cryst. Growth, 2006, 293, 541.
21. B. Koo, J. Park, Y. Kim, S. H. Choi, Y. E. Sung and T. Hyeon, J. Phys. Chem. B, 2006, 110, 24318.
22. G. L. Li, G. H. Wang and J. M. Hong, J. Mater. Res., 1999, 14, 3346.
23. P. D. Cozzoli, A. Kornowski and H. Weller, J. Am. Chem. Soc., 2003, 125, 14539.
24. V. G. Pol and A. Zaban, J. Phys. Chem. C, 2007, 111, 13313.
25. J. J. Wu and C. C. Yu, J. Phys. Chem. B, 2004, 108, 3379.
26. L. Miao, S. Tanemura, S. Toh, K. Kaneko and M. Tanemura, J. Cryst. Growth, 2004, 264, 246.
27. B. B. Lakshmi, P. K. Dorhout and C. R. Martin, Chem. Mater., 1997, 9, 857.
28. T. Kasuga, M. Hiramatsu, A. Hoson, T. Sekino and K. Niihara, Langmuir, 1998, 14, 3160.
29. X. Y. Zhang, L. D. Zhang, W. Chen, G. W. Meng, M. J. Zheng and L. X. Zhao, Chem. Mater., 2001, 13, 2511.
30. G. Sudant, E. Baudrin, D. Larcher and J.-M. Tarascon, J. Mater. Chem., 2005, 15, 1263.
31. M. Wagemaker, W. J. H. Borghois and F. M. Mulder, J. Am. Chem. Soc., 2007, 129, 4323.
32. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737.
33. B. Weintraub, Y. Wei and Z. L. Wang, Angew. Chem., Int. Ed., 2009, 48, 8981.
34. L. Kavan, B. O'Regan, A. Kay and M. Gratzel, J. Electroanal. Chem., 1993, 346, 291.
35. A. Ramadoss and S. J. Kim, Carbon, 2013, 63, 434.
36. A. Ramadoss, G.-S. Kim and S. J. Kim, CrystEngComm, 2013, 15, 10222.
37. M. Gratzel, Pure Appl. Chem., 2001, 73, 459.
38. W. R. Mason III and H. B. Gray, Inorg. Chem., 1968, 7, 55.
39. U. Hörmann, U. Kaiser, M. Albrecht, J. Geserick and N. Hüsing, J. Phys.: Conf. Ser., 2010, 209, 012039.
40. L. S. Qiang, Z. Hong, Z. R. Rong, S. X. Yu, Y. S. De and W. S. Long, Sci. China, Ser. B: Chem., 2008, 51, 367.
41. R. Vijayalakshmi and V. Rajendran, rch. Appl. Sci. Res., 2012, 4, 1183.
42. F.-F. Cao, S. Xin, Y.-G. Guo and L.-J. Wan, Phys. Chem. Chem. Phys., 2011, 13, 2014.
43. T. Oshaka, F. Izumi and Y. Fujiki, J. Raman Spectrosc., 1978, 7, 321.
44. J. Shi, J. Chen, Z. Feng, T. Chen, Y. Lian, X. Wang and C. Li, J. Phys. Chem. C, 2007, 111, 693.
45. Y. X. Zhang, G. H. Li, Y. X. Jin, Y. Zhang, J. Zhang and L. D. Zhang, Chem. Phys. Lett., 2002, 365, 300.
46. S. Mochizuki, Physica B, 2003, 340–344, 944.
47. D. Pan, N. Zhao, Q. Wang, S. Jiang, X. Ji and L. An, Adv. Mater., 2005, 17, 1991.
48. G. De, A. A. Soler-Illia, A. Louis and C. Sanchez, Chem. Mater., 2002, 14, 750.
49. J. M. Yu, L. Z. Zhang, Z. Zheng and J. C. Zhao, Chem. Mater., 2003, 15, 2280.
50. E. L. Crepaldi, G. J. De, A. A. Soler-Illia, D. Grosso, F. Cagnol and F. Ribot, J. Am. Chem. Soc., 2003, 125, 9770.
51. Y. Gao, Y. Masuda, Z. Peng, T. Yonezawa and K. Koumoto, J. Mater. Chem., 2003, 13, 608.
52. H. Wang, B. Li, Z. Yan, Z. Lu, R. Cheng and D. Qian, Rare Met., 2008, 27, 1.
53. D. A. H. Hanaor, I. Chironi, I. Karatchevtseva, G. Triani and C. C. Sorrell, Adv. Appl. Ceram., 2012, 111, 149.
54. N. R. Jana, L. Gearheart and C. J. Murphy, J. Phys. Chem. B, 2001, 105, 4065.
55. Y. Song, R. M. Garcia, R. M. Dorin, H. Wang, Y. Qiu, E. N. Coker, W. A. Steen, J. E. Miller and J. A. Shelnutt, Nano Lett., 2007, 7, 3650.
56. J. Yan, T. Wei, W. Qiao, B. Shao, Q. Zhao, L. Zhang and Z. Fan, Electrochim. Acta, 2010, 55, 6973.
57. R. Kotz and M. Carlen, Electrochim. Acta, 2000, 45, 2483.
58. J. Jiang, Y. Li, J. Liu and X. Huang, Nanoscale, 2011, 3, 45.
59. J. Yan, J. Liu, Z. Fan, T. Wei and L. Zhang, Carbon, 2012, 50, 2179.
60. H. Wang, Q. Hao, X. Yang, L. Lu and X. Wang, Nanoscale, 2010, 2, 2164.
61. Q. F. Zhang, T. P. Chou, B. Russo, S. A. Jenekhe and G. Z. Cao, Angew. Chem., 2008, 120, 2436.
62. Q. F. Zhang, T. P. Chou, B. Russo, S. A. Jenekhe and G. Z. Cao, Angew. Chem., Int. Ed., 2008, 47, 2402.
63. D. Chen, F. Huang, Y. B. Cheng and R. A. Caruso, Adv. Mater., 2009, 21, 2206.
64. B. C. O'Regan, J. R. Durrant, P. M. Sommeling and N. J. Bakker, J. Phys. Chem. C, 2007, 111, 14001.
65. M. Berginc, U. O. Krasovec, M. Jankovec and M. Topic, Sol. Energy Mater. Sol. Cells, 2007, 91, 821.
66. A. N. M. Green, E. Palomares, S. A. Haque, J. M. Kroon and J. R. Durrant, J. Phys. Chem. B, 2005, 109, 12525.
67. G. K. Mor, K. Shankar, M. Paulose, O. K. Varghese and C. A. Grimes, Nano Lett., 2006, 6, 215.
68. K. P. Wang and H. Teng, Appl. Phys. Lett., 2007, 91, 173102.
69. Z. Lan, J. Wu, J. Lin and M. Huang, J. Mater. Chem., 2011, 21, 15552.

---