α-Fe₂O₃ nanospheres: facile synthesis and highly efficient photo-degradation of organic dyes and surface activation by nano-Pt for enhanced methanol sensing

Abstract

We report a simplified electrochemical route to synthesize thin films of nanosphere α-Fe₂O₃ from a suitable electrolytic solution. X-ray diffraction studies revealed the formation of pure hematite phase (hexagonal structure) α-Fe₂O₃ films. Field emission scanning electron microscopy revealed a highly compact surface morphology with evenly distributed almost spherical grains. Raman, electron paramagnetic resonance and Fourier transform infrared spectroscopic analyses confirmed the presence of α phase Fe₂O₃ (hematite). Optical analysis revealed a band gap energy of 2.15 eV; this is most suitable for visible light driven photocatalysis towards the degradation of Indigo Carmine (IC) and Rhodamine B (Rh B) dyes, which are widely used in the textile industry and were taken as model organic compounds. About 90% photodegradation was achieved at rates of 0.0188 min⁻¹ for IC and 0.0133 min⁻¹ for Rh B. The synthesized films were used as modified electrodes, and their catalytic activity towards methanol oxidation was investigated. A comparison was also made between Pt modified FTO/Fe₂O₃ and unmodified FTO/Fe₂O₃ electrodes towards dye degradation and methanol oxidation, and it was found that the Pt modified FTO/Fe₂O₃ electrode yielded superior results.

---

1. Introduction

Iron oxide in the α phase, i.e. α-Fe₂O₃ (hematite), is an attractive semiconductor for many cutting edge applications; among these, photoelectrochemical and photocatalytic applications are of major importance due to the stability, abundance, and environmental compatibility of α-Fe₂O₃, as well as its suitable band gap and the position of its valence band edge.^1–8 In general, hematite exhibits an n-type character, which can be due to the tendency of α-Fe₂O₃ to become oxygen deficient, irrespective of the preparation method. α-Fe₂O₃ is a promising candidate for photocatalytic applications due to its narrow band gap energy of about 2.0 to 2.2 eV, which leads to absorption mainly in the visible region. It absorbs light up to 600 nm and collects up to 40% of solar spectrum energy. α-Fe₂O₃ is cost effective, with low toxicity and high photo/heat-stability; it also has a suitable valence band edge position. This makes hematite a promising anode material for photochemical, photoelectrochemical and electrochemical applications, such as methanol electrooxidation.⁹ Nanoparticles of α-Fe₂O₃ perform photocatalytic reactions through the generation of electron–hole pairs upon visible light irradiation, which leads to a better charge transfer process from (LUMO)_dye to (HOMO)_catalyst. Various techniques have been investigated for the synthesis of hematite nanocrystalline thin films, viz. thermal evaporation,¹⁰ chemical vapor deposition,¹⁰ aqueous chemical growth,¹⁰ spray-pyrolysis,¹¹ ultrasonic spray-pyrolysis,¹¹ the sol–gel method¹² and plasma enhanced atomic layer deposition.¹³ Electrodeposition is a technique that is well suited to the preparation of nanostructures.¹⁴ In fact, through proper parameter control, electrodeposited materials can form nanostructured to even epitaxial films with superior properties.¹⁵ Moreover, the electrodeposition technique is currently emerging as an important low-cost and low temperature method for preparing semiconducting thin films.¹⁶

In recent years, several reports have appeared concerning the electrodeposition of iron oxide thin films onto foreign substrates, such as Fe₃O₄ (magnetite), α-FeOOH (goethite), and γ-FeOOH (lepidocrocite).¹⁷ Among the various oxides of iron, such as hematite (α-Fe₂O₃), maghemite (γ-Fe₂O₃) and wustite (Fe_1−xO), α-Fe₂O₃ is of special interest due to its unique catalytic properties; however, γ-Fe₂O₃ nanoparticles are excellent candidates for nanoscale magnetism.¹⁸ Owing to its suitable electronic properties, such as flatband potential, surface current density and doping density, α-Fe₂O₃ is a very important material for the electrooxidation of various toxic chemicals, such as methanol.^9,19 It also serves as a good photocatalyst for aniline²⁰ and promotes the photopolymerization reaction of acrylamide through a radical mechanism.²¹ α-Fe₂O₃ with different nanostructures can act as an efficient photocatalyst for the degradation of phenolic compounds and other dyes, such as methylene blue (MB),²² Rose Bengal (RB)²³ and Methyl Orange (MO).²⁴ Xu et al.²⁵ and Tang et al.²⁶ have also investigated the UV-light driven photocatalytic performance of hollow spindle²⁵ and flowerlike²⁶ α-Fe₂O₃ towards phenol degradation. α-Fe₂O₃ is also used to improve the photocatalytic efficiency of other semiconductors, such as TiO₂, by forming composites, because Fe³⁺ can act as a better electron scavenger than O₂, and there is significant matching of band structures between Fe₂O₃ and TiO₂.²⁷ Hematite is also very important for the synthesis of composite heterostructures, such as Fe₂O₃/SnO₂ (ref. 28) and Fe₂O₃/CdS,²⁹ for superior photocatalytic performance through effective charge carrier separation.

Over the past few decades, our society has become increasingly sensitive towards the protection of the environment against adverse industrial effects. The effluents of the dyeing, textile, and pulp and paper industries contain various dye pigments and methanol, which should be removed before discharging the effluents to the environment to avoid health hazards and protect ecosystems. Due to the low biodegradability of dyes, conventional biological wastewater treatment pathways are not very efficient. To solve this problem, scientists have developed semiconductor photocatalysis. The main advantage of this technology is its environmental compatibility, as its main reagent, electrons, is a clean reagent. Also, semiconductor photocatalysis has a faster degradation rate than biodegradation. In this regard, efficient, stable, low-cost and environmentally friendly photocatalysts must be developed. These are the motivations behind this work. In this work, the deposition of nanocrystalline thin films of α-Fe₂O₃ on fluorine-doped tin oxide (FTO) coated glass was carried out by a simple electrochemical method using the chemicals FeCl₂·4H₂O, KCl and H₂O₂. The method reported here is cheaper, simpler and less time consuming for large scale production. In the electrochemical synthesis of iron oxides, accurate control of growth potential and pH is necessary, which has been addressed in this work. The catalytic behavior of the deposited α-Fe₂O₃ films for the photo-degradation of hazardous organic dyes, such as Indigo Carmine (IC) and Rhodamine B (Rh B), has been explored. A comparison was also made between the Pt modified FTO/Fe₂O₃ and unmodified FTO/Fe₂O₃ electrodes in methanol electrooxidation, and the former was found to give superior performance.

2. Experimental procedure

2.1. Materials

All chemicals and solvents were of analytical reagent grade and were used as received, without any further purification. Commercially available FTO substrates with ∼10 ohm per sq. surface resistivity were cleaned properly and were used as the cathode for the deposition.

2.2. Deposition of α-Fe₂O₃ thin films

The working solution was prepared by thoroughly mixing 1 mL of 0.1 M FeCl₂·4H₂O and 1 mL 0.1 M KCl in 100 mL double distilled water. KCl was taken as the supporting electrolyte to stabilize the redox system. Next, 0.5 mL 0.02 M H₂O₂ was added dropwise with stirring at 500 rpm for 15 minutes. The pH of the working solution was maintained at around 2.5 by adding dilute hydrochloric acid (1 : 30) dropwise. The working solution was brownish yellow in color.

The electrochemical cell was constructed by immersing the FTO glass and a Zn rod (99.8% pure) in the solution, which were short-circuited externally through a copper wire. The Zn rod served as the self-decaying anode to produce Zn²⁺ ions and the FTO glass served as the cathode (the schematic is reported elsewhere³⁰). As soon as the two electrodes were short-circuited, the deposition started very slowly and was continued for around 24 hours at room temperature without stirring to obtain almost pinhole-free compact films with good adherence. After deposition, the films were washed thoroughly with double distilled water and dried in a hot air oven at 60 °C for 15 minutes.

The mechanism for α-Fe₂O₃ deposition involves two reactions that take place simultaneously on the FTO cathode surface when the electrodes are short circuited externally.

FeCl₂·4H₂O dissolves in water at low pH (ca. 2.5) and dissociates as:

FeCl₂·4H₂O → Fe²⁺ + Cl⁻ + 4H₂O (i)

When a Zn rod is used as the sacrificial anode and is dipped into a low pH (ca. 2.5) medium, metallic Zn will dissociate in the following manner:

Zn → Zn²⁺ + 2e⁻, E⁰ = +0.76 V (ii)

The Zn²⁺ ions will enter the solution, and the two electrons that were released on the Zn surface will flow to the FTO cathode through the external path; these, in turn, will reduce the Fe²⁺ ions present in the vicinity of the cathode, as the E⁰ value of Fe²⁺/Fe⁰ is much lower (−0.41 V) than that of Zn²⁺/Zn⁰ (−0.76 V).

Fe²⁺ + 2e⁻ → Fe⁰ (iii)

In the presence of H₂O₂, the Fe⁰ formed on the FTO surface will undergo the following series of reactions to finally produce Fe₂O₃, as described earlier by Schrebler-Guzman et al. and Kelly et al.^31,32 According to their study, H₂O₂ first reacts with the dissolved electrons (Photo-Fenton reaction) in the following manner to produce active species such as OH⁻ and OH*:

H₂O₂ + e⁻ (aq) → OH* + OH⁻ (iv)

Fe⁰ + OH⁻ ⇄ [FeOH] + e⁻ (v)

[FeOH] ⇄ [FeOH]⁺ + e⁻ (vi)

[FeOH]⁺ + OH⁻ ⇄ Fe(OH)₂ (vii)

Fe(OH)₂ + OH⁻ → [FeOOH] + H₂O + e⁻ (viii)

2[FeOOH] ⇄ Fe₂O₃·H₂O (ix)

The Cl⁻ ions released by Fe²⁺ ions during reduction capture the free Zn²⁺ ions present in the solution, thus restricting their deposition along with Fe and enhancing the purity of the deposited films.

To avoid ‘poisoning’ of the catalyst, i.e. the deposited Fe₂O₃ films in this case, Pt nanoparticles were deposited on the film surface by dipping the film in a beaker containing 1% H₂PtCl₆·H₂O solution for 2 minutes. After removal, the films were washed thoroughly with distilled water and dried at 200 °C.

2.3. Characterization techniques

The crystalline structures and phases of the deposited films were determined by X-ray diffraction (XRD) using a Bruker D8 advance X-ray diffractometer with Bragg–Brentano goniometer geometry and a Cu-K_α X-radiation source (λ = 1.5418 Å). Morphological analyses of the films were carried out using both field emission scanning electron microscopy (FESEM) (JEOL, JSM-6700F) and atomic force microscopy (AFM) (NT-MDT Solver Next). The AFM measurements were performed in semicontact mode, and a silicon probe (length = 95 μm, width = 30 μm, thickness = 2 μm, resonance frequency = 240 kHz, force constant = 18 N m⁻²) was used for scanning purposes. The scans were performed with a scanning speed of 0.5 Hz. Electron paramagnetic resonance (EPR) spectra were recorded using an EPR spectrometer (JEOL-JES FA200) with X-Band frequency: 8.75 to 9.65 GHz, sensitivity: 7 × 10⁹ spins/0.1 mT, and resolution: 2.35 μT. UV-Vis spectroscopic measurements were carried out using a Perkin-Elmer Lambda 25 UV-Vis spectrophotometer to evaluate the optical properties of the deposited films. Infrared spectra were recorded using a Perkin-Elmer Spectrum Two with ATR Mode. Raman analysis (NXR FT-RAMAN MODULE, Thermo Electron Corporation, USA, operating with a 1064 nm Nd:YVO4 excitation laser source at 0.1 W power) was also carried out to establish the phase of the deposited material. A Perkin Elmer LS-55 fluorescence spectrophotometer was used to study the ˙OH radical trapping phenomenon. The electrochemical measurements were carried out using a CHI 620D (USA) electrochemical analyzer.

2.4. Photocatalytic measurement

The photocatalytic behavior of the deposited thin films was measured by noting the gradual decrease of the characteristic absorption peak of the test dye in the presence of light. Standard aqueous solutions of Indigo Carmine (IC) (100 ppm, pH ca. 6) and Rhodamine B (RhB) (100 ppm, pH ca. 5.5) were prepared and allowed to attain equilibrium in the dark for 30 minutes to study the photocatalytic performance of the films. Two prototype films, each with areas of 2 cm × 3 cm, were placed at the bottom of two 100 mL beakers containing 50 mL of 100 ppm IC and RhB dye solutions, respectively. The same technique was also applied for the Pt activated films. As Fe₂O₃ is a semiconductor whose main electronic transitions are in the visible region, a 200 W tungsten filament lamp (Philips) was used as the source of visible light; it was placed 2.0 cm above the beaker to produce an intensity of 1 Sun. The entire experimental setup was maintained in a water bath at 30 °C with a constant flow of air to the beaker to eliminate any heating effects that might be caused by the light source. At certain time intervals, specific amounts of the solution were withdrawn, and the changes in the concentration of the dye were observed using a UV-Vis spectrophotometer by noting the decrease in intensity of the characteristic absorption peak of the respective dye.

2.5. Electrocatalytic activity measurement

The electrochemical oxidation and chronoamperometric detection of CH₃OH by α-Fe₂O₃ and Pt modified α-Fe₂O₃ thin films (FTO/α-Fe₂O₃ and FTO/Pt-α-Fe₂O₃), each with 1.0 cm² active areas, were monitored with an electrochemical analyzer using a conventional three electrode system with Ag/AgCl as the reference electrode. Cyclic voltamograms (CV) were recorded in a cell containing 10 mL of 0.1 M phosphate buffer solution (PBS) (pH ca. 7.4, 32 °C) with a scan rate of 0.1 V s⁻¹. For chronoamperometric measurements, a similar cell setup was used with an applied potential of −1.0 V (vs. Ag/AgCl).

3. Results and discussion

3.1. Structural and morphological properties of α-Fe₂O₃ and Pt activated α-Fe₂O₃ thin films

The compositions and purities of the deposited films were examined by X-ray diffraction (XRD), and the resulting diffractograms are displayed in Fig. 1. The diffraction pattern of the as deposited film (Fig. 1a) showed characteristic peaks of the (104), (202), (116), (214), (300), (119) and (223) planes, which can be indexed to the pure hematite phase (hexagonal structure), i.e. α-Fe₂O₃ (JCPDS ID: 33-0664). The sharpness of the peak indicates the crystallinity of the deposited material. The diffraction peaks marked “*” correspond to SnO₂ from the bottom of the FTO coated glass substrate (JCPDS ID 21-1250). However, two diffraction peaks of α-Fe₂O₃, viz. (104) and (116), were found to overlap with the diffractions from SnO₂. The crystallite size corresponding to the most intense diffraction was estimated using the Debye–Scherrer equation:³² D = 0.9λ/β cos θ, where D is the crystallite size, λ is the wavelength of the Cu-K_α radiation used (λ = 1.5418 Å), β is the experimentally observed diffraction peak width at half maximum (FWHM), and θ is the Bragg's angle; the average crystallite size was estimated to be around 10 nm. The dislocation density (δ), which is the length of the dislocation lines per unit volume, was also estimated using the equation³³ δ = 1/D² and was found to be significantly low (0.0 nm⁻²), indicating fewer crystal defects and good crystallinity of the deposited films. The high value (3.83 × 10⁻²) of microstrain (ε) obtained using the relation ε = (β cos θ)/4 also supports the formation of crystallites with low dimensions.

Fig. 1 X-ray diffraction patterns of (a) the as deposited α-Fe₂O₃ thin film and (b) the Pt-activated α-Fe₂O₃ thin film.

After activating the surface of the deposited Fe₂O₃ films with Pt, the samples were also subjected to XRD analysis, and the result is given in Fig. 1b; this figure reveals the presence of diffractions from the (111), (200) and (220) planes of metallic Pt (JCPDS ID: 04-0802) corresponding to the 2θ values of 39.76, 46.24 and 67.45°, respectively. The deposition of metallic Pt leads to the suppression of some peaks corresponding to α-Fe₂O₃. The standard diffraction ID indicates the presence of quasi-spherical metallic platinum on the surface of the deposited α-Fe₂O₃ film. The small intensity of the diffraction peaks indicates the presence of metallic platinum in small amounts.

The 2D and 3D surface morphologies and the average particle sizes of the as deposited α-Fe₂O₃ films were investigated using field emission scanning electron microscopy (FESEM) and atomic force microscopy (AFM); the results are shown in Fig. 2. One can see from the FESEM image (Fig. 2a) that the FTO surface is fully covered with uniformly distributed spherical grains of α-Fe₂O₃ with diameters in the range of 40 to 60 nm. The grains are distinct, and the grain boundaries are prominent. To study the nature of an individual grain, a high magnification image (Fig. 2a inset) was captured, and it has been found that this grain is composed of crystallites with an average diameter of 15 nm. This value of the crystallite diameter agrees well with that obtained from the XRD analysis. However, the average grain size as obtained from low magnification FESEM is somewhat larger than that calculated from XRD analysis, which is attributed to the formation of larger grains from smaller crystallites, which is essential for spontaneous lowering of Gibbs free energy in the system to gain stability. The total free energy of a system containing nanoparticles (NPs) is given by:

G_total = μ_NP × N_NP + γ × S

where μ_NP is the chemical potential of the nanoparticles and N_NP is the total number of nanoparticles. γ is the interfacial tension (or the interfacial energy per unit area) and S is the surface area. Thus, γ × S denotes the total surface energy of the system. As we know, NPs have high surface-to-volume ratios; therefore, the total surface energy of the system is significantly high. As every system has a tendency to decrease its free energy spontaneously, the γ × S part of the free energy for a system consisting of nanoparticles also has a tendency to decrease by means of forming larger grains (which reduces the value of S) from smaller crystallites by agglomeration. For this reason, from FESEM image analysis, we see grains that are somewhat larger than the crystals.

Fig. 2 Microstructural analyses. (a) FESEM and (b) AFM images of the as deposited α-Fe₂O₃ thin film. (c) FESEM and (d) AFM images of the Pt-activated α-Fe₂O₃ thin film.

AFM studies were also carried out to find the salient topographical and surface features of the as deposited films, and the typical 3D AFM image is shown in Fig. 2b. From this image also, we can see the average grain size is about 50 nm, which is in good agreement with the value obtained from the FESEM analysis. AFM revealed a significantly undulated topography along with the presence of numerous finer crystallites in the grains of the as deposited α-Fe₂O₃ films.

On the other hand, the FESEM image of the Pt activated α-Fe₂O₃ film (Fig. 2c) clearly reveals the presence of numerous almost-spherical Pt nanoparticles with an average diameter of 50 nm. This is in good agreement with the X-ray diffraction pattern of quasi-spherical Pt nanoparticles (JCPDS ID: 04-0802), as explained previously. The spikes in the 3D AFM image of the Pt activated α-Fe₂O₃ sample (Fig. 2d) also indicate the growth of distinct nanoparticles on the surface of the deposited semiconductor. Various parameters obtained from XRD, FESEM and AFM analyses relating to the structure of the deposited α-Fe₂O₃ thin films before and after Pt activation are summarized in Table 1. It has been observed from the AFM measurements and can be seen from Table 1 that the both the root mean square (RMS) and average (R_a) roughness increased significantly after Pt activation. This increase in surface roughness indicates that the Pt activated α-Fe₂O₃ thin films may serve as superior candidates for photochemical and electrochemical applications.

Table 1 Comparison of the parameters obtained from XRD, FESEM and AFM analyses

Name of the sample	Crystallite size (nm) calculated from XRD	Grain size (nm) obtained from FESEM	Grain size (nm) obtained from AFM	Root mean square (RMS) roughness (nm) obtained from AFM	Average (Ra) roughness (nm) obtained from AFM
α-Fe2O3 thin film	10	40–60	50	8.055	6.208
Pt activated α-Fe2O3 thin film	20 for Pt	50 for Pt	70 for Pt	14.648	9.148

---

3.2. Electron paramagnetic resonance studies (EPR)

The room temperature electron paramagnetic resonance (EPR) spectrum of α-Fe₂O₃ is shown in Fig. 3; it exhibits an intense resonance signal at g ≈ 2.30 within 300 to 325 mT. This confirms the oxidation state of +3 in the deposited material, as the Fe³⁺ ions belong to the d⁵ configuration with ⁶S as the ground state, and there is no spin–orbit coupling. The theoretical g value for the free Fe³⁺ ion is 2.0023. Our experimentally observed value is in good agreement with the theoretical value. The slightly high experimentally obtained g value can be attributed to the formation of clusters of Fe³⁺ ions, which causes exchange interactions between the coupled ions.³⁴ A previous report³⁴ indicates that if the Fe₂O₃ crystallites remain within a few nanometers of the domain, as is the case here (Section 3.2), a broad spectrum of g (≈2.3) may be observed. Therefore, the observed resonance signal at g ≈ 2.3 in this study justifies that the oxidation state of Fe in the deposited iron oxide is +3, which is in good agreement with the XRD analysis, confirming the presence of α-Fe₂O₃. The broad nature and slightly higher value of g also indicate the formation of small clusters of Fe₂O₃ with nano-dimensions, which in turn supports the findings from the FESEM and AFM studies.

Fig. 3 EPR analyses. Plots for (a) Lande g-factor vs. intensity and (b) magnetic field strength vs. intensity.

3.3. Fourier transform infrared (FTIR) and Raman spectroscopy

For further detection of the phase and purity of the deposited iron oxide films, FTIR analysis was carried out within the range of 500 to 4000 cm⁻¹ (Fig. 4). The absorption band at ∼3340 cm⁻¹ is assigned to the presence of adsorbed water.³⁵ The band around ∼1626 cm⁻¹ can be attributed to the O–H bending vibrations of the adsorbed H₂O molecules (moisture) and the band around 1100 cm⁻¹ is attributed to the Fe–OH vibrational mode of iron oxide in the presence of moisture. The bands at ∼540 and ∼462 cm⁻¹ correspond to the characteristic metal–oxygen (Fe–O) vibrational modes of spinel iron oxide. The small intensity of the Fe–O bands agrees well with the previous report.³⁶ The observed shift is gradual and depends on the particle size. The main band has been attributed to the Fe–O or Fe–O–Fe bindings of hematite. This confirms the formation of Fe₂O₃ films with α phase.

Fig. 4 FTIR spectrum of the as deposited α-Fe₂O₃ thin film.

The Raman spectra of the various phases of iron oxides (viz. α-Fe₂O₃, Fe₃O₄ and γ-Fe₂O₃) are distinctly different from each other, and this technique can be used effectively to determine the phase of the deposited iron oxide.³⁷ Iron oxide belongs to the D₃d⁶ space group; typically, seven phonon lines are expected in the Raman spectrum, viz. two A_1g and five E_g phonon modes.³⁸ The spectrum is normally dominated by the overtones of the longitudinal optical (LO) phonons. First, a background scan within the range of 200 to 700 cm⁻¹ was performed; a blank was taken with FTO glass, and then the spectrum was acquired for the iron oxide sample deposited on FTO glass. The spectrum (Fig. 5) reveals the presence of five strong resonant Raman peaks at 220, 285, 408, 491 and 608 cm⁻¹. These values are typical for α-Fe₂O₃ and are in good agreement with previous reports. The Raman lines are broader and are found to be shifted to low wavenumbers with respect to the values obtained for commercial hematite (α-Fe₂O₃). The broader line width and the shifting of wavenumbers of the Raman peaks are due to the phonon confinement effect in the nanocrystals, as was described previously. Bersani et al.³⁹ proved that the peak that appears at 660 cm⁻¹ is typical for hematite iron (α-Fe₂O₃) and seems to be related to disorder effects and/or to the presence of nanocrystals. Xu et al.⁴⁰ recorded the Raman spectrum of α-Fe₂O₃ nanoleaves synthesized by oxygenating pure iron; they observed A_g at 225 and 498 cm⁻¹ and E_g at 252, 293, 411, 612 cm⁻¹. Therefore, in our case, Raman analysis confirms the presence of α-Fe₂O₃ in the deposited films.

Fig. 5 Raman spectrum of the as deposited α-Fe₂O₃ thin film.

3.4. UV-Vis analysis

As displayed in Fig. 6, the absorption spectrum showed a moderately sharp onset in absorbance at around 580 nm, which represents the typical band to band transition for α-Fe₂O₃. The band gap energy for the α-Fe₂O₃ thin film was calculated using Tauc's relation: [(αhν)^1/n = A(hν − E_g)], where A is a constant related to the effective masses of the charge carriers, h is the Planck constant, E_g is the band gap energy, hν is the incident photon energy and 1/n is an exponent that depends on the nature of the optical transition (n = 0.5 and 2 for direct and indirect transitions, respectively). The Tauc's plot (Fig. 6 inset) indicates a direct band gap energy of 2.15 eV for the deposited α-Fe₂O₃ films. The higher band gap energy compared to that of the bulk material (∼1.8 eV) is due to the presence of smaller particles (∼50 nm) in our case. The band structure of the semiconductors changes from a continuous to a discrete pattern as the crystallite size decreases, and the lowest transition energy (E_l) is blue shifted with respect to the original band gap energy (E_g). The quantum confined or blue shifted energy (ΔE) can be expressed by the standard equation as:

ΔE = E_l − E_g = (ħ²π²/2μR²) − (1.8e²/4πεR)

where E_l is the lowest transition level, E_g is the original band gap energy, ħ is the reduced Planck's constant, R is the radius of the nanocrystal, μ is the effective mass of an electron, ε is the permittivity of the material and e is the electronic charge. Since (ΔE) is inversely proportional to R, the reduction of the crystallite size leads to an increase in the band gap energy, which we have also observed in this case.

Fig. 6 Absorption spectrum of the as deposited α-Fe₂O₃ thin film (inset: Tauc's plot for band gap determination).

3.5. Photochemical activities toward toxic organic dyes

The photocatalytic activities of the α-Fe₂O₃ thin films were evaluated by observing the photodegradation of Indigo Carmine (IC) and Rhodamine B (RhB) dyes under simulated irradiation, as mentioned in Section 2.4. The photocatalytic degradations were evaluated by measuring the successive decrease of the characteristic absorbance peaks for IC and RhB at different time intervals in the presence of light and the catalyst. Significant decreases in the absorption intensity with increasing irradiation time were observed for IC and RhB, as shown in Fig. 7a and 8a, respectively. Two sets of blank experiments, taking the dye solution in the absence of α-Fe₂O₃ films but in the presence of light and the dye solution in the presence of α-Fe₂O₃ but in the absence of light, were also carried out for each dye to confirm the roles of both the catalyst and light in the degradation process.

Fig. 7 For Indigo Carmine: (a) spectrophotometric degradation curves in the presence of light and catalyst; (b) degradation efficacy with respect to time; (c) relative change in concentration with time; and (d) rate kinetics curves.

Fig. 8 For Rhodamine B: (a) spectrophotometric degradation curves in the presence of light and catalyst; (b) degradation efficacy with respect to time; (c) relative change in concentration with time; and (d) rate kinetics curves.

No new absorption peaks appeared during the whole process, which indicates complete photolysis in the presence of the proposed catalyst. No degradation of these dyes was observed in the dark, and very slow degradations were observed in the absence of the catalyst, i.e. α-Fe₂O₃ thin films. About 90% degradation of the dye IC was observed after 150 minutes, and about 95% degradation of RhB was achieved after 225 minutes, which is evident from Fig. 7b and 8b. Fig. 7c and 8c show the extent of the photocatalytic degradation of IC and RhB, respectively, by the α-Fe₂O₃ thin films in the form of the relative concentration (C_t/C₀) change with irradiation time under three different conditions, viz. (a) in the absence of both light and catalyst, (b) in the presence of light but in the absence of catalyst, and (c) in the presence of both light and catalyst. Here, C₀ is the initial concentration of the dye and C_t is the concentration of the dye at time ‘t’. To determine the nature of the degradation kinetics, ln(C₀/C_t) was plotted against irradiation time (Fig. 7d and 8d), and the negative slopes in each case indicate pseudo-first order kinetics. The rate constant ‘k’ for IC was found to be about 0.0188 min⁻¹, and the same for RhB was about 0.0133 min⁻¹. These two values are notably higher than many other previous reports (Table 2) on the degradation of IC and RhB. Significant enhancements of the degradation rate in the presence of both catalyst and light, as seen from Fig. 7c and 8c, indicate that the α-Fe₂O₃ thin film can act as an excellent photocatalyst towards the degradation of Indigo Carmine and Rhodamine B dyes under visible light irradiation.

Table 2 Comparison of the photocatalytic performances of different systems towards RhB and IC

Sl. No.	Type of catalyst	Dye	Degradation time (min)	Degradation rate (min^−1)	Reference
1	Titania–silica nanosol		240	0.01719	41
	SiO2@TiO2 (590 nm)		720	0.00348
	SiO2@TiO2 (470 nm)		480	0.00465
	SiO2@TiO2 (220 nm)		480	0.00473
2	Peptide I xerogel	RhB	1440	Not mentioned	42
3	50 at% Bi2WO6/BiOBr	RhB	150	Not mentioned	43
4	LaFeO3	RhB	180	0.01237	44
5	Bi2WO6	RhB			45
	1 × 10^−4 mL^−1		60	0.0146
	5 × 10^−5 mL^−1		60	0.0218
	1 × 10^−5 mL^−1		60	0.0517
	5 × 10^−6 mL^−1		60	0.0592
	1 × 10^−6 mL^−1		60	0.0714
6	TiO2–zeolite (TZ) composite	RhB	360	0.0155	46
7	Nb2O5	IC	90	Not mentioned	47
8	Ti/TiO2	IC	90	0.33	48
9	α-Fe2O3	IC	150	0.0188	This work
10	α-Fe2O3	RhB	225	0.0133	This work

---

Fig. 9 depicts a plausible schematic of the band structure and charge transfer process for the generation of active species leading to the photodegradation of organic dye molecules on the surface of α-Fe₂O₃ nanoparticles. Firstly, electron–hole pairs are generated in the α-Fe₂O₃ nanoparticles by light irradiation;^49,50 the electrons occupy the conduction band (CB) and the holes occupy the valence band (VB). The photogenerated electrons (e⁻) thus produced in the conduction band of α-Fe₂O₃ are scavenged by O₂ to produce the O₂˙⁻ anion radical, which on protonation yields HOO˙ in the presence of water. On the other hand, the holes (h⁺) which are produced in the valence band of α-Fe₂O₃ react with H₂O or –OH ions originating from H₂O to produce a highly active species, ˙OH radical. The hydroxyl radicals thus produced are responsible for the complete mineralization of the dye, thereby producing products such as CO₂ and H₂O. The degradation processes normally follow both oxidative (using holes) and reductive (using electrons) pathways which may be summarized as follows:^51–53

α-Fe₂O₃ + hν → e_CB⁻ + h_VB⁺

e_CB⁻ + O₂ → O₂˙⁻

O₂˙⁻ + H⁺ → HOO˙

HOO˙ → H₂O₂ + O₂

H₂O₂ + e⁻ → ˙OH + –OH

h_VB⁺ + OH⁻/H₂O → ˙OH

˙OH + dye-intermediates → CO₂ + H₂O

Fig. 9 Schematic of active species formation by the absorption of light in α-Fe₂O₃ nanoparticles.

3.5.1. Evaluation of active oxidative species. In photocatalysis studies, there are two common pathways for the degradation of organic pollutants: through a direct oxidation reaction by photogenerated holes, and through oxidative radicals. It can also be concluded that the activity of a particular catalyst is expected to be higher if it generates reactive species, such as O₂˙⁻ and ˙OH, at higher rates.^54,55 Here, the hydroxyl radicals perform the key function for the decomposition reactions mentioned above. Therefore, it is necessary to investigate the presence of hydroxyl radicals in the medium produced by the catalyst, to support the above mechanism. We used a fluorescence probe technique to establish the formation of ˙OH radicals by the deposited α-Fe₂O₃ films; terephthalic acid (TPA) was used as the fluorescent probe. TPA was illuminated with the same light source used for the photocatalysis in slightly alkaline medium (pH ≈ 8). The fluorescence intensity was measured by a fluorescence spectrophotometer, and the emission spectrum upon excitation at 320 nm was measured at 10 min. intervals during irradiation (Fig. 10). TPA is directly attacked by the ˙OH radicals to form 2-hydroxy-terephthalic acid (TPAOH), giving a fluorescence signal at ∼425 nm. The fluorescence intensity is directly related to the number of hydroxyl radicals formed by the photocatalyst, i.e. α-Fe₂O₃ in this case. This means that the higher the rate of hydroxyl radical formation, the greater the yield of TPAOH and the greater the intensity of the fluorescence peak. An intense fluorescence signal, on the other hand, indicates enrichment of the degradation efficiency (Table 3).

Fig. 10 Fluorescence spectrum upon excitation at 320 nm showing ˙OH radical trapping in a 2 × 10⁻⁵ M alkaline solution of terephthalic acid.

Table 3 Comparison of methanol electrooxidation properties

Sl. No.	Composition of the electrode material	Sensitivity	Reference
1	Pt–CeO2	0.45 mA cm^−2	57
2	10% Pt/TiO2-NFs	0.110 mA cm^−2	58
	16% Pt/TiO2-NFs	0.161 mA cm^−2
	21% Pt/TiO2-NFs	0.336 mA cm^−2
	27% Pt/TiO2-NFs	0.277 mA cm^−2
	Pt/C	0.259 mA cm^−2
3	Pt–CePO4	0.1 mA cm^−2	59
5	Pt–Fe2O3	0.748 mA cm^−2	This work

---

3.5.2. Reactive oxygen species (ROS) scavenging studies. In order to obtain insight into the mechanism of photocatalytic dye degradation by the α-Fe₂O₃ thin films, specific ROS scavengers, e.g., tertiary butyl alcohol (TBA) as a hydroxyl radical scavenger, were added to the reaction medium. As the ROS generation study was based on scavenging action, it is necessary to confirm the evidence for the photocatalytic mechanism; therefore, TBA was chosen for the detection and quantification of ˙OH radicals. The effects of different concentrations of TBA (0.01 M, 0.1 M and 1 M) on dye degradation were observed, and the results showed quenching of the degradation efficiency of both IC and Rh B with increasing concentration of the scavenger (Fig. 11). Therefore, it can be concluded that the drastic inhibition of dye degradation is due to the scavenging of hydroxyl species (˙OH) which were produced in the reaction medium by the α-Fe₂O₃ films. Mitigating hydroxyl radical generation towards dye degradation with radical scavengers in various concentrations affects the rate of the reaction. This provides evidence to support the proposed photocatalytic mechanism.

Fig. 11 Relative concentration profiles of the photocatalytic degradation of (a) IC and (b) Rh B as a function of irradiation time in the presence of α-Fe₂O₃ thin films and different concentrations of the scavenger.

3.6. Comparative study on methanol oxidation by the as deposited and Pt-activated α-Fe₂O₃ thin films

The electrocatalytic activities of the bare FTO/α-Fe₂O₃ and Pt-activated α-Fe₂O₃/FTO electrodes (designated as FTO/α-Fe₂O₃–Pt hereafter) towards methanol were also studied. These measurements were carried out using conventional three electrode systems, where a platinum wire acted as the counter electrode and an Ag/AgCl electrode acted as the reference electrode. The FTO/α-Fe₂O₃ and FTO/Pt–α-Fe₂O₃ electrodes were taken as the working electrodes in each case. The reaction bath was prepared by placing 10 mL 0.1 M phosphate buffer solution (PBS, pH = 7.4) and 10 mL of 1 M CH₃OH in a 25 mL beaker in which the electrodes were dipped, and the solution was degassed by purging with pure N₂ for 15 min Fig. 12a–c represent the cyclic voltammograms (CV) of blank FTO, FTO/α-Fe₂O₃ and FTO/α-Fe₂O₃–Pt electrodes in this working solution. The characteristic CV curve for the FTO-PBS system shows a maximum current in the nano-ampere range (Fig. 12a), whereas, when the FTO electrode was modified with α-Fe₂O₃, it was found to show a moderate enhancement in current to the micro-ampere range (Fig. 12b). A notable increase in peak current was observed (Fig. 12c) when the FTO/α-Fe₂O₃ electrode was further modified with nano-Pt to produce FTO/α-Fe₂O₃–Pt electrode. From the CV studies, it is clear that the catalytic activity of FTO/Pt–α-Fe₂O₃ electrode towards methanol is the highest among the three. The lower reactivity of the FTO/α-Fe₂O₃ electrode towards methanol may be attributed to the restricted movement of electrons from the surface to the bulk and may also be due to poisoning of the electrode in the presence of the oxidized byproducts.

Fig. 12 Cyclic voltammograms in 10 mL 0.1 M PBS and 10 mL 1.0 M CH₃OH for (a) blank FTO (b) FTO/α-Fe₂O₃ (c) FTO/α-Fe₂O₃–Pt electrode. (d) Cyclic voltammograms of FTO/α-Fe₂O₃–Pt electrode with successive addition of CH₃OH with different concentrations.

Fig. 12d represents the cyclic voltammograms of FTO/α-Fe₂O₃–Pt in the same working solution with the gradual addition of 10, 30, 50, 100, 200, 300, 400, 500, 1000, 1500 and 2000 micromolar methanol solutions, respectively. A steady increase in current occurs with increasing methanol concentration, which indicates the stability of the FTO/α-Fe₂O₃–Pt electrode even with gradual addition of methanol. In this process, the anodic current (i_pc) increases sharply due to the fast electro-oxidation of methanol on the electrode surface and reaches the anodic current height (i_pa) at a peak potential of around −1.0 V (vs. Ag/AgCl). The electro-oxidation of methanol was significantly fast within the potential range of −0.85 to −1.0 V (vs. Ag/AgCl). The CV curves also indicate the quasi-reversibility of the electro-oxidation process using FTO/α-Fe₂O₃–Pt electrodes with a progressive increase in analyte concentration, where the anodic peak current (i_pa) increases with a concomitant increase in the cathodic peak current (i_pc) to achieve quasi-reversibility. Thus, the Pt nanostructures on α-Fe₂O₃ serve as foreign domains for modifying the physicochemical properties and consequently enhance the methanol electro-oxidation reactions.

3.6.1. Explanation for enhanced electro-oxidation by Pt-activation. The activity of the FTO/α-Fe₂O₃ electrode may be enhanced by modifying the electrode surface with noble metal nanoparticles such as Pt, which aids the chemisorption of oxidized byproducts such as carbon monoxide (CO) on the metal surface by the following reactions,^9,19,56 thus minimizing catalyst (α-Fe₂O₃) poisoning.

Pt + CH₃OH → Pt–CO_ads + 4H⁺ + 4e⁻

α-Fe₂O₃ + H₂O → α-Fe₂O₃–OH + H⁺ + e⁻

Coupling Pt with α-Fe₂O₃ increases the rate of methanol oxidation by decreasing the cathodic peak current and increasing the anodic peak current. Continuous removal of adsorbed CO from the Pt active sites also favors the oxidation. During the electrochemical process, reactions between the chemisorbed CO and OH⁻ take place to produce CO₂, which is removed from the electrode surface according to the following scheme, and the Pt active sites return to their initial positions:

Pt–CO + α-Fe₂O₃–OH → Pt + α-Fe₂O₃ + CO₂ + H⁺ + e⁻

From the AFM analyses (Section 3.2), it has been observed that the activation of the α-Fe₂O₃ film surface by Pt nanoparticles increases both the RMS and average roughness significantly. Enhanced surface roughness in turn aids the adsorption of the analyte molecules on the electrode surface, leading to better electrochemical performance.

On the other hand, the decoration of the α-Fe₂O₃ surface with noble metals, in this case Pt nanoparticles, can enhance the electro-oxidation performance due to the difference in Fermi levels (E_f) of α-Fe₂O₃ and the metal nanoparticles, where the energy difference at the semiconductor/metal interface drives the electrons from the CB of the α-Fe₂O₃ into the metal nanoparticles (Fig. 13). In other words, the metal acts as an electron trap, promoting interfacial charge transfer and therefore minimizing recombination of the e⁻/h⁺ pairs, as shown in Fig. 13. The effective band alignment between the semiconductor (Fe₂O₃) and the metal (Pt) in the presence of the redox system improves the electrocatalytic activity towards methanol oxidation.

Fig. 13 Schematic for energy level alignment and electron trapping by Pt nanoparticles towards enhanced methanol electrooxidation.

4. Conclusion

Polycrystalline α-Fe₂O₃ thin films with hematite (hexagonal) phase were deposited on FTO coated glass substrates by a modified electrochemical technique. The morphologies of the deposited films were highly dense with a regular distribution of spherical nanoparticles. The material showed a direct band gap energy of about 2.2 eV. These films were found to efficiently degrade toxic dyes, viz. Indigo Carmine and Rhodamine B, in the presence of visible light. The degradation followed pseudo first order kinetics, and the measured rate coefficient was found to be 0.0188 min⁻¹ for Indigo Carmine and 0.0133 min⁻¹ for Rhodamine B, respectively. After activating the surface of the FTO/α-Fe₂O₃ electrodes with Pt nanoparticles, significant enhancement in methanol oxidation has been observed. This indicates that the deposited thin films can act as good photocatalysts for the degradation of Indigo Carmine and Rhodamine B and are also good candidates for methanol detection and electro-oxidation.

Acknowledgements

The author BS would like to acknowledge the Department of Chemistry, IIEST, Shibpur and the Department of Chemistry, Jadavpur University, for infrastructural support.

References

1. P. S. Shinde, G. H. Go and W. J. Lee, J. Mater. Chem., 2012, 22, 10469–10471.
2. A. Duret and M. Gratzel, J. Phys. Chem. B, 2005, 109, 17184–17191.
3. B. Scrosati, Nature, 1995, 373, 557–558.
4. M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. X. Mi, E. A. Santori and N. S. Lewis, Chem. Rev., 2010, 110, 6446–6473.
5. Z. S. Li, W. J. Luo, M. L. Zhang, J. Y. Feng and Z. G. Zou, Energy Environ. Sci., 2013, 6, 347–370.
6. T. Hisatomi, J. Kubota and K. Domen, Chem. Soc. Rev., 2014, 43, 7520–7535.
7. K. Sivula, F. L. Formal and M. Graetzel, ChemSusChem, 2011, 4, 432–449.
8. A. Kay, I. Cesar and M. Gratzel, J. Am. Chem. Soc., 2006, 128, 15714–15721.
9. S. J. Hoseini, M. Bahramia and M. Roushanib, RSC Adv., 2014, 4, 46992–46999.
10. Z. R. Dai, Z. W. Pan and Z. L. Wang, Adv. Funct. Mater., 2003, 13, 9–24.
11. J. D. Desai, H. M. Pathan, S.-K. Min, K.-D. Jung and O. S. Joo, Appl. Surf. Sci., 2005, 252, 1870–1875.
12. L. Armelao, R. Bertoncello, L. Crociani, G. Depaoli, G. Granozzi, E. Tondello and M. Bettinelli, J. Mater. Chem., 1995, 5(1), 79–83.
13. R. K. Ramachandran, J. Dendooven and C. Detavernier, J. Mater. Chem. A, 2014, 2, 10662–10667.
14. G. Hodes, Electrochem. of Nanomater, Wiley-VCH Verlag Gmb, Weinheim, 2001, pp. 201–228.
15. T. Pauporte and D. Lincot, Appl. Phys. Lett., 1999, 75, 3817–3819.
16. M. Kosa, H. N. Barad, V. Singh, D. A. Keller, K. Shimanovich, S. Ruhle, A. Y. Anderson, A. Zaban and D. T. Major, Phys. Chem. Chem. Phys., 2016, 18, 781–791.
17. L. Martinez, D. Leinen, F. Martin, M. Gabas, J. R. Ramos-Barrado, E. Quagliata and E. A. Dalchiele, J. Electrochem. Soc., 2007, 154, D126–D133.
18. L. S. Zhong, J. S. Hu, H. P. Liang, A. M. Cao, W. G. Song and L. J. Wan, Adv. Mater., 2006, 18, 2426–2431.
19. F. W. Chang, H. Y. Yu, L. S. Roseline, H. C. Yang and T. C. Ou, Appl. Catal., A, 2006, 302, 157–167.
20. C. Karunakaran and S. Senthilvelan, Electrochem. Commun., 2006, 8, 95–101.
21. A. L. Stroyuk, I. V. Sobran and S. Ya Kuchmiy, J. Photochem. Photobiol., A, 2007, 192, 98–104.
22. A. Qurashi, Z. Zhong and M. W. Alam, Solid State Sci., 2010, 12, 1516–1519.
23. S. K. Maji, N. Mukherjee, A. Mondal and B. Adhikary, Polyhedron, 2012, 33, 145–149.
24. X. Xie, H. Yang, F. Zhang, L. Li, J. Ma, H. Jiao and J. Zhang, J. Alloys Compd., 2009, 477, 90–99.
25. X. Li, X. Yu, J. H. He and Z. Xu, J. Phys. Chem. C, 2009, 113, 2837–2845.
26. S. Y. Zeng, K. B. Tang, T. W. Li, Z. H. Liang, D. Wang, Y. K. Wang, Y. X. Qi and W. W. Zhou, J. Phys. Chem. C, 2008, 112, 4836–4843.
27. Q. Sun, W. Leng, Z. Li and Y. Xu, J. Hazard. Mater., 2012, 229–230, 224–232.
28. M. T. Niu, F. Huang, L. F. Cui, P. Huang, Y. L. Yu and Y. S. Wang, ACS Nano, 2010, 4, 681–688.
29. L. Wang, H. W. Wei, Y. J. Fan, X. Gu and J. H. Zhan, J. Phys. Chem. C, 2009, 113, 14119–14125.
30. A. Mondal, N. Mukherjee and S. K. Bhar, Mater. Lett., 2006, 60, 1748–1752.
31. R. S. Schrebler-Guzman, J. R. Vilche and A. J. Arvia, Electrochim. Acta, 1979, 24, 395–403.
32. E. J. Kelly, J. Electrochem. Soc., 1965, 112(2), 124–131.
33. G. B. Williamson and R. C. Smallmam, Philos. Mag., 1956, 1, 34–45.
34. G. Piazzesi, D. Nicosia, M. Devadas, O. Krocher, M. Elsener and A. Wokaun, Catal. Lett., 2007, 115, 33–39.
35. E. Darezereshki, Mater. Lett., 2011, 65, 642–645.
36. H. M. Pathan and C. D. Lokhande, Bull. Mater. Sci., 2004, 27, 85–111.
37. D. L. A. de Faria, S. V. Silva and M. T. de Olliveria, J. Raman Spectrosc., 1997, 28, 873–878.
38. M. Chakrabarti, A. Banerjee, D. Sanyal, M. Sutradhar and A. Chakrabarti, J. Mater. Sci., 2008, 43, 4175–4181.
39. D. Bersani, P. P. Lottici and A. Montenero, J. Raman Spectrosc., 1999, 30, 355–360.
40. Y. Y. Xu, D. Zhao, X. J. Zhang, W. T. Jin, P. Kashkarov and H. Zhang, Phys. E, 2009, 41, 806–811.
41. P. Wilhelm and D. Stephan, J. Photochem. Photobiol., A, 2007, 185, 19–25.
42. P. Koley and A. Pramanik, Soft Matter, 2012, 8, 5364–5374.
43. J. Xia, J. Di, S. Yin, H. Xu, J. Zhang, Y. Xu, L. Xu, H. Li and M. Ji, RSC Adv., 2014, 4, 82–90.
44. S. Thirumalairajan, K. Girija, V. R. Mastelaro and N. Ponpandian, New J. Chem., 2014, 38, 5480–5490.
45. H. Fu, C. Pan, W. Yao and Y. Zhu, J. Phys. Chem. B, 2005, 109, 22432–22439.
46. L. You-ji and C. Wei, Catal. Sci. Technol., 2011, 1, 802–809.
47. A. G. S. Prado, L. B. Bolzon, C. P. Pedroso, A. O. Moura and L. L. Costa, Appl. Catal., B, 2008, 82, 219–224.
48. T. T. Guaraldo, S. H. Pulcinelli and M. V. B. Zanoni, J. Photochem. Photobiol., A, 2011, 217, 259–266.
49. S. Mohan, B. Subramanian and G. Sarveswaran, J. Mater. Chem. C, 2014, 2, 6835–6842.
50. Q. Zeng, J. Bai, J. Li, L. Xia, K. Huang, X. Li and B. Zhou, J. Mater. Chem. A, 2015, 3, 4345–4353.
51. J. Zhao, P. Yang, H.-S. Chen, J. Li, Q. Che, Y. Zhu and R. Shi, J. Mater. Chem. C, 2015, 3, 2539–2547.
52. S. Chabri, A. Dhara, B. Show, D. Adak, A. Sinha and N. Mukherjee, Catal. Sci. Technol., 2016, 2016(6), 3238–3252.
53. K. D. Malviya, H. Dotan, D. Shlenkevich, A. Tsyganok, H. Mor and A. Rothschild, J. Mater. Chem. A, 2016, 4, 3091–3099.
54. T. Wang, J. Lang, Y. Zhao, Y. Su, Y. Zhao and X. Wang, CrystEngComm, 2015, 17, 6651–6660.
55. A. Sharmaa and R. K. Dutta, RSC Adv., 2015, 5, 43815–43823.
56. Q. G. He, W. Chen, S. Mukherjee, S. W. Chen and F. Laufek, J. Power Sources, 2009, 187, 298–304.
57. X. M. Qu, L. X. You, X. C. Tian, B. W. Zhang, G. D. Mahadevan, Y. X. Jiang and S. G. Sun, Electrochim. Acta, 2015, 182, 1078–1084 . mailto: E-mail: sgsun@xmu.edu.cn.
58. Y. Zheng, H. Chen, Y. Dai, N. Zhang, W. Zhao, S. Wang, Y. Lou, Y. Li and Y. Sun, Electrochim. Acta, 2015, 178, 74–79.
59. J. Park, Y. Oh, Y. Park, S. Nam, J. Moon, J. Kang, D.-R. jung, S. Byun and B. Park, Curr. Appl. Phys., 2011, 11, S2–S5.

---