Electrochemical reduction of hydrogen peroxide by nanostructured hematite modified electrodes

Abstract

In this study, various nanostructured hematites (α-Fe₂O₃), including nanorods (α-Fe₂O_3NR), nanoparticles (α-Fe₂O_3Np), and nanosheets (α-Fe₂O_3NS), were synthesized and their electrocatalytic properties towards the reduction of H₂O₂ were investigated. All nanostructured α-Fe₂O₃ hematites were synthesized using chemical bath deposition (CBD) under mild conditions, followed by thermal treatment at 500 °C. The nanostructure was controlled simply by adjusting the composition of precursor solution and reaction duration for the CBD process. It was found that iron phosphate (FePO₄) was deposited in situ onto the surface of these nanostructured α-Fe₂O₃ hematites during the electrochemical pretreatment in the phosphate electrolyte, and both FePO₄ and α-Fe₂O₃ showed activity in catalysing the electrochemical reduction of H₂O₂. In addition, the interaction/compatibility between deposited FePO₄ and α-Fe₂O₃ has a decisive effect on the overall electrocatalytic activity of the resultant electrodes; FePO₄ only showed a synergetic effect on the overall electrocatalytic activity with α-Fe₂O_3NR and α-Fe₂O_3NS. The rate constant is highest for the electro-reduction of H₂O₂ on FePO₄ modified α-Fe₂O_3NR (α-Fe₂O_3NR|FePO₄), but FePO₄ modified α-Fe₂O_3NS (α-Fe₂O_3NS|FePO₄) showed the best overall electrocatalytic activity due to its relatively higher surface area. Furthermore, dissolved oxygen showed negligible interference on the activity of α-Fe₂O_3NR|FePO₄ and α-Fe₂O_3NS|FePO₄, which makes them promising sensing materials in oxidase-based electrochemical sensors.

---

Introduction

The development of a highly sensitive and reliable hydrogen peroxide (H₂O₂) electrochemical sensor is of great importance not only because H₂O₂ has been identified as a chemical threat to the environment and a major factor causing some diseases,¹ but also it a frequent intermediate involved in many important oxidase-catalyzed chemical processes, such as glucose oxidation catalysed by glucose oxidase (GOD) (eqn (1) and (2)):²

Glucose + GOD–FAD ⇄ gluconolactone + GOD–FADH₂ (1)

O₂ + GOD–FADH₂ → H₂O₂ + GOD–FAD (2)

H₂O₂ is an electroactive species that can be oxidized or reduced electrochemically, and therefore, the electrochemical detection of chemicals involved in oxidase-catalyzed chemical processes can be achieved by detection of H₂O₂. Some oxidase-based electrochemical sensors that utilize the anodic current from the electrooxidation of H₂O₂ as the output signal have been developed, but these kinds of sensors often suffer interference from some common electro-oxidizable species, such as ascorbic acid and uric acid existing in biological samples.³ As a result, to minimize the interference and enhance the selectivity of oxidase-based electrochemical sensors, the development of oxidase-based electrochemical sensors that use the cathodic current from the electroreduction of H₂O₂ as the output signal is preferred. However, in this case, dissolved oxygen, required to re-oxidize the oxidase (e.g., eqn (2)), becomes a potential interfering species as oxygen can also be reduced electrochemically. Therefore, an electrocatalyst that can selectively catalyse the reduction of H₂O₂ against the reduction of O₂ is highly required for constructing oxidase-based electrochemical sensors that operate in a cathodic regime.

Many materials have been explored as active species to catalyze the electrochemical reduction of H₂O₂, including prussian blue,⁴ iron oxides,⁵ silver,⁶ manganese oxides,⁷ copper oxides,⁸ etc. Among them, iron oxides have received much attention not only because they are robust, earth abundant, and can be easily synthesized in a cheap way, but also it exhibited peroxidase-like activity.⁹ Nevertheless, the mechanism for the electrocatalysis of H₂O₂ by iron oxides is still not well-understood. For example, in a previous report,¹⁰ iron oxide nanorods, including β-FeOOH, α-Fe₂O₃, γ-Fe₂O₃, were found to be active for the electrochemical reduction of H₂O₂ in non-phosphate buffer, but only α-Fe₂O₃ was found to be active in the phosphate buffer. The interaction/compatibility between the surface modifier (iron phosphate) and the iron oxide matrix played an important role in determining the overall activity of the iron oxide based material. On the other hand, iron oxides of various nanostructures, such as nanorods,^5f nanoparticles,^5b–d,f,11 and nanotubes,^5e have been synthesized, and these nanostructured iron oxides have been shown to exhibit enhanced apparent electrocatalytic activity compared with the bulk counter-parts. Crystallinity, crystal size, structure, and exposed surface facets, have been shown to have decisive effects on the overall activity of these nanostructured iron oxides.^11c,12 Nevertheless, most nanostructured iron oxides were synthesized in powder form, and rarely directly deposited onto an electrode surface, which would not only cause irreproducibility due to the uncontrollable aggregation of these nano-sized iron oxides, but also complicate the following electrode preparation process.

In this work, we report the direct growth of hematite (α-Fe₂O₃) with different nanostructures, including nanorods, nanosheets, and nanoparticles, onto a fluorine-doped tin oxide coated glass substrate (FTO) using chemical bath deposition under mild conditions with follow-up thermal treatment. The effects of nanostructure and the interplay of FePO₄ with different nanostructured α-Fe₂O₃ on the overall electrocatalytic properties towards the reduction of H₂O₂ were thoroughly investigated. It was found that synergetic effects of FePO₄ with α-Fe₂O₃ greatly enhanced the overall electrocatalytic activity, in terms of overpotential and catalytic current, compared with FePO₄ or α-Fe₂O₃ alone, but this effect occurs only for α-Fe₂O₃ nanorods and nanosheets. In addition, the detection of H₂O₂ by Fe₂O_3NR|FePO₄ and Fe₂O_3NS|FePO₄ is insensitive to the dissolved oxygen, which allows their application towards electrochemical detection of key biomolecules involving in the oxidase-catalyzed chemical processes.

Experimental section

General consideration

The starting materials for the synthetic part of the work were purchased from commercial suppliers and are of the highest available purity for the analytical work. Fluorine-doped tin oxide (FTO) coated glass (sheet resistance 7 ohm sq⁻¹, TEC GlassTM 7) substrates (1.0 × 3.0 cm²) were cleaned with an ammonia–hydrogen peroxide–deionized water mixture (volume ratio: 1 : 1 : 5) at 70 °C for 30 min, after which the FTO substrates were dried at room temperature under a nitrogen purge. A hydrogen peroxide stock solution (0.5 M) was prepared before each experiment by direct dilution of hydrogen peroxide (H₂O₂, 30 wt%) with electrolyte solutions, of different pHs, either containing (i) sodium phosphate (0.1 M) and sodium sulfate (0.1 M), or (ii) sodium sulfate (0.1 M). Deionized water (DIW) was used throughout the work.

Preparation of the FTO|α-Fe₂O_3NR, FTO|α-Fe₂O_3NS and FTO|α-Fe₂O_3NP electrodes

The FTO|α-Fe₂O_3NR electrode was prepared by first growing akaganeite nanorods (NR) onto the FTO substrate using chemical bath deposition (CBD) in an aqueous solution containing 1.0 M urea and 0.15 M iron chloride at 90 °C for 4 h, followed by thermal conversion of akaganeite NRs to α-Fe₂O_3NR at 500 °C for 1 h. The α-Fe₂O₃ nanosheets (α-Fe₂O_3NS) were grown onto the FTO substrate, designated as FTO|α-Fe₂O_3NS, by CBD in an aqueous solution containing 0.75 M urea and 0.15 M iron nitrate at 90 °C for 4 h and follow-up thermal treatment at 500 °C for 1 h. The α-Fe₂O₃ nanoparticles (α-Fe₂O_3NP) were grown onto the FTO substrate, designated as FTO|α-Fe₂O_3NP, by CBD in an aqueous solution containing 2.0 M urea and 0.15 M iron nitrate at 90 °C for 24 h and follow-up thermal treatment at 500 °C for 1 h. The exposed area of the FTO substrate for growing nanostructured hematite was kept at 2.0 cm².

Physical characterization

The surface morphology of the electrodes was characterized using scanning electron microscopy (SEM, Hitachi SU-8010). X-Ray diffraction (XRD) analyses were carried out using an Ultima IV (Rigaku Co., Japan) X-ray diffractometer. The surface composition of the films was verified by X-ray photoelectron spectroscopy (XPS, PHI 5000 VersaProbe system, ULVAC-PHI, Chigasaki, Japan), using a microfocused (100 μm, 25 W) Al X-ray beam, with a photoelectron take off angle of 45°. The Ar⁺ ion source for XPS (FIG-5CE) was controlled using a floating voltage of 0.2 kV. The binding energies obtained in the XPS analyses were corrected for specimen charging, by referencing the C 1s peak to 285.0 eV.

Electrochemical characterization

Electrochemical characterizations of the electrocatalytic properties of the nanostructured hematite modified electrodes were performed with a CHI 660 electrochemical workstation (CH Instruments, Inc., USA) at room temperature and all potentials are reported against Ag/AgCl (saturated KCl). A conventional three-electrode electrochemical cell was employed with nanostructure hematite modified electrodes (exposed area of ∼1.0 cm²) as the working electrode, Pt foil (exposed area 4.0 cm²) as counter electrode, and Ag/AgCl as a reference electrode. Prior to experiments, all hematite modified electrodes were pretreated either in (i) phosphate buffer solution (PBS, pH 7) containing sodium phosphate (0.1 M) and sodium sulfate (0.1 M), or in (ii) sodium sulfate (0.1 M, pH 7), using cyclic voltammetry (CV) at a scan rate of 50 mV s⁻¹ in the potential window between −0.7 V and +0.4 V (vs. Ag/AgCl) until the CV curves became stabilised. The sensitivities of all the hematite modified electrodes determined by CV were the slopes of the curves of cathodic peak current density vs. H₂O₂ concentration. A suitable operating potential in the limiting current plateau region for the amperometric detection of H₂O₂ was determined using linear sweep voltammetry (LSV) at a scan rate of 0.1 mV s⁻¹ in PBS (pH 6) containing 0 mM and 4.95 mM H₂O₂. After obtaining the operating potential, which is −0.3 V vs. Ag/AgCl, the amperometric detection of H₂O₂ was carried out in PBS (pH 6) under constant magnetic stirring. The current density responses to the change in H₂O₂ concentration were collected, and the calibration curve for the detection of H₂O₂ was then constructed. All the electrochemical measurements were repeated at least three times.

Results and discussion

Synthesis of the nanostructured hematite electrodes

All nanostructured hematites (α-Fe₂O₃) were directly grown on to FTO by chemical bath deposition under mild conditions and follow-up thermal treatment at 500 °C for 1 h (see ESI† for details). The XRD analyses (Fig. S1†) show that all deposited materials were converted into hematite after thermal treatment. As revealed in the SEM images, shown in Fig. 1, the nanostructure of α-Fe₂O₃ can be controlled by tuning the composition of the bath solution and the reaction times. For example, a nanorod array (α-Fe₂O_3NR) can be grown in a FeCl₃–urea bath solution system, whereas nanosheets (α-Fe₂O_3NS) and nanoparticles (α-Fe₂O_3NP) can be grown in a Fe(NO₃)₃–urea bath solution system. The adsorption of anions (Cl⁻, CO₃²⁻, NO₃⁻) to specific crystal faces and their relative concentrations influence the preferential growth direction of crystals, and therefore, different nanostructures are created. The detailed growth mechanism will be submitted elsewhere soon. The relative effective surface area for these nanostructured α-Fe₂O₃ hematites was determined by measuring double-layer capacitance using cyclic voltammetry,¹³ and the results (Fig. S2†) reveal that the relative effective surface area (α-Fe₂O_3NR : α-Fe₂O_3NS : α-Fe₂O_3NP) is 1.00 : 1.58 : 1.74.

Fig. 1 SEM images of (a, d) α-Fe₂O_3NR, (b, e) α-Fe₂O_3NS, (c, f) α-Fe₂O_3NP. Scale bar: 1 μm.

Electrochemical characterization

α-Fe₂O₃ is studied in this work as we found that only α-Fe₂O₃ is compatible with in situ deposited iron phosphate (FePO₄) in phosphate buffer.¹⁰ As the formation of FePO₄ on an α-Fe₂O₃ surface during the electrochemical detection of H₂O₂ in phosphate buffer solution (PBS) is inevitable, and to ensure the surface of all the nanostructured α-Fe₂O₃ modified electrodes are fully covered with FePO₄, all the nanostructured α-Fe₂O₃ modified electrodes were pre-treated in 0.1 M PBS solution (pH 7) using cyclic voltammetry (CV) from +0.4 to −0.7 V vs. Ag/AgCl at a scan rate of 50 mV s⁻¹ until the redox peaks of FePO₄ were saturated. Fig. 2 shows the XPS spectra of Fe 2p, O 1s, and P 2p for all three nanostructured α-Fe₂O₃ electrodes before and after CV pre-treatment. A positive shift (from ∼710.9 to 711.3 eV) in binding energy (BE) of the Fe 2p peak after the pre-treatment was noticed for all the electrodes (Fig. 2a–c). In addition, the appearance of an additional shoulder in the O 1s spectra at a BE of 531.5 eV (Fig. 2d–f) along with a peak in the P 2p spectra at a BE of 133.2 eV (Fig. 2g–i) after the pre-treatment was also observed for all the electrodes.

Fig. 2 XPS spectra of the pretreated (a, d, g) α-Fe₂O_3NR, (b, e, h) α-Fe₂O_3NS, and (c, f, i) α-Fe₂O_3NP (i) before and (ii) after CV-pretreatment. (a–c) Fe 2p region, (d–f) O 1s region, and (g–i) P 2p region.

Fig. 3 shows the CV curves of all the pre-treated nanostructured α-Fe₂O₃ modified electrodes in PBS (pH 7) at different scan rates (v) and the corresponding plots of the peak current density (J_p) vs. v are shown in Fig. S3.† It can be found that the all the pre-treated electrodes exhibited reversible redox peaks which are characteristic of FePO₄,¹⁴ and the relationship between J_p with v is linear (see Fig. S3†), which suggests that the deposited species strongly absorbed onto the electrode surface after the pre-treatment. The above observations (Fig. 2 and 3) suggest that FePO₄ formed during the pre-treatment process. In addition, the slopes of the plot J_pc vs. v for α-Fe₂O_3NR|FePO₄, α-Fe₂O_3NS|FePO₄, and α-Fe₂O_3NP|FePO₄ are found to be −8.38, −18.45, and −2.72, respectively, and the ratio of the relative amount of the deposited FePO₄ on the pre-treated α-Fe₂O_3NR, α-Fe₂O_3NS, and α-Fe₂O_3NP can be deduced according to eqn (3), which is 1.00 : 2.20 : 0.32. Nevertheless, from the XPS analyses, it was found that the elemental ratios of P/Fe for α-Fe₂O_3NR|FePO₄, α-Fe₂O_3NS|FePO₄, and α-Fe₂O_3NP|FePO₄ are 0.26, 0.47, and 0.85, respectively. These findings suggest that the surface of α-Fe₂O_3NP prefers the adsorption of phosphate ions over the deposition of FePO₄.

where J_p is peak current density, n is number of electron transfer, F is the Faraday constant, R is the gas constant, T is temperature, and Γ^*_O is the amount of absorbed electroactive species. For clarification, the pre-treated α-Fe₂O_3NR, α-Fe₂O_3NS, and α-Fe₂O_3NP in PBS are designated as α-Fe₂O_3NR|FePO₄, α-Fe₂O_3NS|FePO₄, and α-Fe₂O_3NP|FePO₄, respectively, in the following discussion.

Fig. 3 Cyclic voltammetry of the pretreated (a) α-Fe₂O_3NR, (b) α-Fe₂O_3NS, and (c) α-Fe₂O_3NP in 0.1 M PBS (pH 7) at various scan rates. Scan rates for (i), (ii), (iii), (iv), (v), (vi), (vii), and (viii) are 5, 10, 15, 20, 40, 60, 80, and 100 mV s⁻¹, respectively.

Fig. 4 shows the CV curves of FTO, α-Fe₂O_3NR|FePO₄, α-Fe₂O_3NS|FePO₄, and α-Fe₂O_3NP|FePO₄ at a scan rate of 20 mV s⁻¹ in PBS solution (pH 7) containing H₂O₂ of various concentrations. The cathodic peak current densities (J_pc) and corresponding peak potentials (E_pc) of all the electrodes in the presence of 4.95 mM H₂O₂ are also summarized in Table 1. It can be found that all FePO₄ modified α-Fe₂O₃ electrodes exhibited better electrocatalytic activity, in terms of J_pc and E_pc, than the FTO substrate. In addition, J_pc at E = −0.30 V vs. Ag/AgCl (r₁) increased and J_pa at E = −0.22 V vs. Ag/AgCl (o₁) decreased upon the addition of H₂O₂ for α-Fe₂O_3NR|FePO₄, α-Fe₂O_3NS|FePO₄, and α-Fe₂O_3NP|FePO₄, which indicates that the electrochemical process at peak r₁ involves an electrocatalytic EC′ mechanism, and the deposited FePO₄ is the active species responsible for the electrochemical process. In addition, an additional cathodic peak (r₂) at E = −0.18 V vs. Ag/AgCl appeared upon the addition of H₂O₂ to α-Fe₂O_3NR|FePO₄ and α-Fe₂O_3NS|FePO₄, and as this peak is more sensitive to H₂O₂ than peak r₁, peak r₂ outpaced peak r₁ at a H₂O₂ concentration higher than 3.31 mM. Fig. 5a–c show the CV curves of α-Fe₂O_3NR|FePO₄, α-Fe₂O_3NS|FePO₄, and α-Fe₂O_3NP|FePO₄, respectively, in PBS containing 1.66 mM H₂O₂ at various pHs ranging from 4 to 7. It can be found that the E_pc for peak r₁ for all the α-Fe₂O₃ modified electrodes shifted to the more negative side as the solution pH was increased, which reflects the redox behavior of FePO₄, whereas the E_pc of peak r₂ (only for α-Fe₂O_3NR|FePO₄ and α-Fe₂O_3NS|FePO₄) was insensitive to the change in solution pH. Fig. 5d shows the sensitivities, i.e., the slope of the curve of J_pc vs. H₂O₂ concentration, of α-Fe₂O_3NR|FePO₄, α-Fe₂O_3NS|FePO₄, and α-Fe₂O_3NP|FePO₄, towards the electrochemical reduction of H₂O₂ at various solution pHs. It can be found that all the electrodes showed their best sensitivity at pH 6, and α-Fe₂O_3NS|FePO₄ exhibited the highest sensitivity among the three α-Fe₂O₃ modified electrodes. Fig. S4† shows the chronoamperometric response of α-Fe₂O_3NS|FePO₄, at an applied potential of −0.3 V vs. Ag/AgCl, after successive addition of the H₂O₂ solution of various concentrations into deaerated 0.1 M PBS (pH 6). It can be found that the current response increased linearly with the increase in H₂O₂ concentration. The sensitivity, i.e., the slope of the calibration curve (shown in the inset of Fig. S4†) was found to be 225.0 ± 19.9 μA cm⁻² mM⁻¹. Besides, the sensor response reached 95% of the steady-state value within 10 s upon the addition of H₂O₂. Furthermore, a limit of detection (signal to noise ratio = 3) of 3.4 ± 0.5 μM can be achieved.

Fig. 4 Cyclic voltammetry, recorded at a scan rate of 20 mV s⁻¹, of (a) FTO, (b) α-Fe₂O_3NR|FePO₄, (c) α-Fe₂O_3NS|FePO₄, and (d) α-Fe₂O_3NP|FePO₄ in 0.1 M PBS solution (pH 7) containing H₂O₂ of various concentrations. Concentrations of H₂O₂ used for curve (i), (ii), (iii), and (iv) are 0, 1.66, 3.31, and 4.95 mM, respectively.

Table 1 Summary of values of J_pc and E_pc of the FTO and nanostructured FePO₄ modified α-Fe₂O₃ electrodes

	Sample
	FTO	α-Fe2O3NR|FePO4	α-Fe2O3NS|FePO4	α-Fe2O3NP|FePO4
a Cathodic peak current density.b Cathodic peak potential. All parameters are determined in 0.1 M PBS solution (pH 7) containing 4.95 mM H2O2.
Jpc^a (mA cm^−2)	−0.996 ± 0.099	−1.025 ± 0.035	−1.236 ± 0.074	−1.008 ± 0.035
Epc^b (V vs. Ag/AgCl)	−0.373 ± 0.026	−0.190 ± 0.003	−0.179 ± 0.003	−0.345 ± 0.005

---

Fig. 5 (a)–(c) Cyclic voltammetry, recorded at a scan rate of 20 mV s⁻¹, of α-Fe₂O_3NR|FePO₄ (a), α-Fe₂O_3NS|FePO₄ (b), and α-Fe₂O_3NP|FePO₄ (c) in 0.1 M PBS containing 1.66 mM H₂O₂ at various pHs ranging from 4 to 7. (d) The sensitivities of α-Fe₂O_3NR|FePO₄, Fe₂O_3NS|FePO₄, and Fe₂O_3NP|FePO₄ in 0.1 M PBS at various pHs.

In a previous report,¹⁰ we proposed the EC′ mechanisms for peak r₁ (eqn (4) and (5)) and peak r₂ (eqn (6) and (7)):

FePO_4(s) + e⁻ + H⁺ ⇄ Fe²⁺_(ad) + HPO₄²⁻ (4)

Fe²⁺_(ad) + H₂O₂ + HPO₄²⁻ ⇄ FePO₄ + H₂O + OH˙ (5)

OH˙ + e⁻ → OH⁻ (6)

OH⁻ + H⁺ ⇄ H₂O (7)

The pH-insensitive reduction peak r₂ was found to be related to the intrinsic catalytic properties of α-Fe₂O₃ itself, and an electron probably comes from the active site, that is, Fe(II) species in the electro-reduced α-Fe₂O₃ under cathodic conditions.¹⁵ The lack of this peak for the case of α-Fe₂O_3NP|FePO₄ implies that this intrinsic catalytic property is structure-dependent or suppressed by other factors, such as the electrolyte or the deposited FePO₄. It is worth noting that although α-Fe₂O_3NP|FePO₄ has the highest surface area (Fig. S2†), without synergetic effects with this activity, it showed the least overall activity, in terms of I_pc and E_pc, among the three α-Fe₂O_3NP modified electrodes. The higher overpotential required for α-Fe₂O_3NP|FePO₄ to reduce H₂O₂ is in agreement with previous reports.^5c,d,12a

Fig. 6a–c show the chronoamperograms for all the α-Fe₂O₃ modified electrodes at an applied potential of −0.3 V vs. Ag/AgCl in 0.1 M PBS solution containing H₂O₂ of various concentration. It can be found that all the nanostructured FePO₄ modified α-Fe₂O₃ modified electrodes exhibited catalytic current densities that are linearly proportional to the H₂O₂ concentration. The rate constants of the reduction of H₂O₂ on the nanostructured FePO₄ modified α-Fe₂O₃ modified electrodes can be derived from Fig. 6a–c and eqn (8) and (9):¹⁶

where J_cat is the catalytic current density in the presence of H₂O₂, J_L is the current density in the absence of H₂O₂, and λ = k_sC*t, where k_s, C*, and t are the apparent rate constant, bulk H₂O₂ concentration, and the elapsed time, respectively. When the value of λ is larger than 1.5, the value of erf(λ^0.5) approaches 1, and the exp(−λ) term can be ignored, and as a result, eqn (8) can be simplified to eqn (9):

Fig. 6 (a)–(c) Chronoamperograms of α-Fe₂O_3NR|FePO₄ (a), α-Fe₂O_3NS|FePO₄ (b), and α-Fe₂O_3NP|FePO₄ (c) recorded at an applied potential of −0.3 V vs. Ag/AgCl in 0.1 M PBS (pH 6) containing H₂O₂ of various concentration. H₂O₂ concentrations used for curves (i), (ii), (iii), (iv), (v), and (vi) are 0, 0.66, 1.32, 1.98, 2.64, and 3.30 mM, respectively. (d) Plots of J_cat/J_L vs. t^1/2 for (i) α-Fe₂O_3NR|FePO₄, (ii) α-Fe₂O_3NS|FePO₄, and (iii) and α-Fe₂O_3NP|FePO₄ using the curves obtained with 0 mM H₂O₂ and 3.30 mM H₂O₂ in (a)–(c).

The values of k_s for the reduction of H₂O₂ on α-Fe₂O_3NR|FePO₄, α-Fe₂O_3NS|FePO₄, and α-Fe₂O_3NP|FePO₄, determined from the slopes of plots of J_cat/J_L versus t^0.5 in Fig. 6d, are found to be 18 253.9, 15 242.7, and 2037.9 L mol⁻¹ s⁻¹, respectively, which further indicates that the kinetics of the H₂O₂ reduction process can be further facilitated with the active Fe(II) species in α-Fe₂O₃. Note that although the value of k_s for Fe₂O_3NR|FePO₄ is higher than that for α-Fe₂O_3NS|FePO₄, α-Fe₂O_3NS|FePO₄ exhibited higher overall electrocatalytic activity over Fe₂O_3NR|FePO₄ as α-Fe₂O_3NS|FePO₄ has a higher surface area.

Fig. S5†shows CV curves of the FTO, α-Fe₂O_3NR, α-Fe₂O_3NS, and α-Fe₂O_3NP at a scan rate of 20 mV s⁻¹ in 0.1 M Na₂SO₄ solution (pH 7) containing H₂O₂ of various concentration. Note that the deposition of FePO₄ is impossible in this electrolyte, and the observed activity from all the α-Fe₂O₃ reflects their intrinsic activity. Values of I_pc and E_pc of all the electrodes in the presence of 4.95 mM H₂O₂ are also summarized in Table S1.† As revealed, α-Fe₂O_3NP showed the highest apparent electrocatalytic activity, in terms of I_pc and E_pc, among the four electrodes, which could be attributed to its high surface area. In addition, as shown in Fig. S6,† all the three nanostructured α-Fe₂O₃ modified electrodes exhibited a pH-independent current response to H₂O₂, which suggests the reaction catalysed by these nanostructured α-Fe₂O₃ should be the same, and therefore, the suppressed activity of α-Fe₂O_3NP|FePO₄ in phosphate buffer could be attributed to the unfavourable interaction between α-Fe₂O_3NP and deposited FePO₄ and/or phosphate ions. It has been reported that the adsorbed phosphate ions would inhibit the reduction of α-Fe₂O₃,¹⁷ which in turn inhibits the formation of the Fe(II) species responsible for the reduction of H₂O₂.

Fig. 7 shows the CV curves of α-Fe₂O_3NR|FePO₄, α-Fe₂O_3NS|FePO₄, and α-Fe₂O_3NP|FePO₄ in 0.1 M PBS (pH 6) at various H₂O₂ concentrations under N₂ and air atmospheres. The sensitivities of α-Fe₂O_3NR|FePO₄, α-Fe₂O_3NS|FePO₄, and α-Fe₂O_3NP|FePO₄ obtained from the data in Fig. 7 are shown in Fig. S7.† When comparing the CV responses in the absence of H₂O₂ under different atmospheres, it can be found that the I_pc of peak r₁ increased and the I_pa of peak o₁ decreased for all the FePO₄ modified α-Fe₂O₃ electrodes, which indicates that the deposited FePO₄ is active in catalysing the electrochemical reduction of dissolved oxygen in PBS. In addition, peak r₂ appeared only after the addition of H₂O₂ regardless of background atmosphere, which suggests that the Fe(II) sites in electro-reduced α-Fe₂O₃ is active for the reduction of H₂O₂ but not for the reduction of dissolved oxygen. As a result, the dissolved oxygen showed little effect on the sensitivity of α-Fe₂O_3NR|FePO₄ and α-Fe₂O_3NS|FePO₄ towards the electrochemical reduction of H₂O₂; the sensitivities of α-Fe₂O_3NR|FePO₄ and α-Fe₂O_3NS|FePO₄ towards the electrochemical reduction of H₂O₂ under air atmosphere remained at 96.5 ± 4.3% and 94.7 ± 4.7%, respectively, of those under N₂ atmosphere. In contrast, the sensitivity of α-Fe₂O_3NP|FePO₄ towards the electrochemical reduction of H₂O₂ under air atmosphere only remained at 41.4 ± 3.9% of that at N₂ atmosphere. The significant influence of dissolved oxygen on α-Fe₂O_3NP|FePO₄ can be attributed to the fact that α-Fe₂O_3NP|FePO₄ lacks Fe(II) sites and the interaction of FePO₄ with dissolved oxygen suppresses the reaction between FePO₄ and H₂O₂. The low interference from oxygen for α-Fe₂O_3NR|FePO₄ and α-Fe₂O_3NS|FePO₄ makes them a potential candidate material for the detection of biomolecules involving oxidase-catalysed chemical processes.

Fig. 7 Cyclic voltammetry, recorded at a scan rate of 20 mV s⁻¹, of (a, a′) α-Fe₂O_3NR|FePO₄, (b, b′) Fe₂O_3NS|FePO₄, (c, c′) Fe₂O_3NP|FePO₄ in 0.1 M PBS (pH 6) containing various H₂O₂ concentrations under (a–c) N₂, and (a′–c′) air atmospheres. H₂O₂ concentrations used for curves (i), (ii), (iii), and (iv) are 0, 1.66, 3.31, and 4.95 mM, respectively.

Conclusions

α-Fe₂O₃ nanorods, nanosheets, and nanoparticles were successfully synthesized using chemical bath deposition and their electrocatalytic properties were thoroughly examined in phosphate buffer and non-phosphate electrolyte. In phosphate buffer, the in situ deposited FePO₄ exhibited synergetic effects on the activity of α-Fe₂O₃ nanorods and nanosheets only, and the surface modification of FePO₄ on these two nanostructured α-Fe₂O₃ greatly enhanced their overall electrocatalytic activity compared with FePO₄ or α-Fe₂O₃ alone. The active sites, Fe(II), in α-Fe₂O₃ nanorods and nanosheets are insensitive to the dissolved oxygen, and makes them potential electrode materials for the fabrication of selective and sensitive H₂O₂ or related electrochemical sensors.

Acknowledgements

This research was, in part, supported by the Ministry of Education, Taiwan, R.O.C. The Aim for the Top University Project to the National Cheng Kung University (NCKU). Financial support from the Ministry of Science and Technology of Taiwan (MOST 104-2221-E-006-235- and 104-ET-E-006-004-ET) are also gratefully acknowledged.

Notes and references

1. E. Nossol and A. J. Gorgatti Zarbin, J. Mater. Chem., 2012, 22, 1824–1833.
2. J. Wang, Chem. Rev., 2008, 108, 814–825.
3. (a) J. Hrbac and R. Kohen, Drug Dev. Res., 2000, 50, 516–527; (b) R. Kohen, E. Vellaichamy, J. Hrbac, I. Gati and O. Tirosh, Free Radical Biol. Med., 2000, 28, 871–879.
4. (a) A. A. Karyakin and E. E. Karyakina, Sens. Actuators, B, 1999, 57, 268–273; (b) A. A. Karyakin, E. A. Puganova, I. A. Budashov, I. N. Kurochkin, E. E. Karyakina, V. A. Levchenko, V. N. Matveyenko and S. D. Varfolomeyev, Anal. Chem., 2004, 76, 474–478.
5. (a) Z. L. Liu, B. Zhao, Y. Shi, C. L. Guo, H. B. Yang and Z. A. Li, Talanta, 2010, 81, 1650–1654; (b) L. Zhang, Y. Zhai, N. Gao, D. Wen and S. Dong, Electrochem. Commun., 2008, 10, 1524–1526; (c) J. Hrbac, V. Halouzka, R. Zboril, K. Papadopoulos and T. Triantis, Electroanalysis, 2007, 19, 1850–1854; (d) L. Zhang, Y. H. Ni, X. H. Wang and G. C. Zhao, Talanta, 2010, 82, 196–201; (e) J. Gong, L. Wang, K. Zhao and D. Song, Electrochem. Commun., 2008, 10, 123–126; (f) J. Z. Marinho, R. H. O. Montes, A. P. de Moura, E. Longo, J. A. Varela, R. A. A. Munoz and R. C. Lima, Mater. Res. Bull., 2014, 49, 572–576.
6. (a) C.-Y. Lin, Y.-H. Lai, A. Balamurugan, R. Vittal, C.-W. Lin and K.-C. Ho, Talanta, 2010, 82, 340–347; (b) S. Wu, H. Zhao, H. Ju, C. Shi and J. Zhao, Electrochem. Commun., 2006, 8, 1197–1203.
7. (a) K. Schachl, H. Alemu, K. Kalcher, J. Jezkova, I. Svancara and K. Vytras, Anal. Lett., 1997, 30, 2655–2673; (b) J.-H. Lee and H.-G. Hong, J. Appl. Electrochem., 2015, 45, 1153–1162.
8. (a) X.-M. Miao, R. Yuan, Y.-Q. Chai, Y.-T. Shi and Y.-Y. Yuan, J. Electroanal. Chem., 2008, 612, 157–163; (b) J. M. Zen, H. H. Chung and A. S. Kumar, Analyst, 2000, 125, 1633–1637.
9. L. Gao, J. Zhuang, L. Nie, J. Zhang, Y. Zhang, N. Gu, T. Wang, J. Feng, D. Yang, S. Perrett and X. Yan, Nat. Nanotechnol., 2007, 2, 577–583.
10. C.-Y. Lin and C.-T. Chang, Sens. Actuators, B, 2015, 220, 695–704.
11. (a) M. Magro, D. Baratella, G. Salviulo, K. Polakova, G. Zoppellaro, J. Tucek, J. Kaslik, R. Zboril and F. Vianello, Biosens. Bioelectron., 2014, 52, 159–165; (b) M. Magro, D. Baratella, N. Pianca, A. Toninello, S. Grancara, R. Zboril and F. Vianello, Sens. Actuators, B, 2013, 176, 315–322; (c) D. Baratella, M. Magro, G. Sinigaglia, R. Zboril, G. Salviulo and F. Vianello, Biosens. Bioelectron., 2013, 45, 13–18; (d) A. K. Dutta, S. K. Maji, D. N. Srivastava, A. Mondal, P. Biswas, P. Paul and B. Adhikary, J. Mol. Catal. A: Chem., 2012, 360, 71–77.
12. (a) J. M. Gong, L. Y. Wang, K. Zhao and D. D. Song, Electrochem. Commun., 2008, 10, 123–126; (b) S. Nath, C. Kaittanis, V. Ramachandran, N. S. Dalal and J. M. Perez, Chem. Mater., 2009, 21, 1761–1767; (c) S. H. Liu, F. Lu, R. M. Xing and J. J. Zhu, Chem.–Eur. J., 2011, 17, 620–625; (d) N. Puvvada, P. K. Panigrahi, D. Mandal and A. Pathak, RSC Adv., 2012, 2, 3270–3273.
13. S. Trasatti and O. A. Petrii, Pure Appl. Chem., 1991, 63, 711–734.
14. (a) F. Marken, D. Patel, C. E. Madden, R. C. Millward and S. Fletcher, New J. Chem., 2002, 26, 259–263; (b) K. J. McKenzie and F. Marken, Pure Appl. Chem., 2001, 73, 1885–1894.
15. M. S. Lin and H. J. Len, Electroanalysis, 2005, 17, 2068–2073.
16. F. Pariente, E. Lorenzo, F. Tobalina and H. D. Abruna, Anal. Chem., 1995, 67, 3936–3944.
17. C. Liu, J. M. Zachara, N. S. Foster and J. Strickland, Environ. Sci. Technol., 2007, 41, 7730–7735.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details. See DOI: 10.1039/c6ra07267d

---