Preparation and characterization of a novel porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode for the anodic oxidation of phenol wastewater

Abstract

Porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrodes were successfully fabricated using a thermal decomposition technique and electro-deposition technologies. Characterization experiments including scanning electron microscopy (SEM), energy-dispersive spectroscopy (EDS), X-ray diffraction (XRD), cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS) and an accelerated lifetime test were performed to evaluate the effect of the CNT-doped SnO₂–Sb₂O₃ intermediate layer on the PbO₂ electrode. The results showed that CNT could be doped into the SnO₂–Sb₂O₃ intermediate layer by thermal decomposition. Compared with the porous Ti/SnO₂–Sb₂O₃ substrate, CNT-doping induced the substrate surface to form a fibrous structure, which suggested that the porous Ti/SnO₂–Sb₂O₃–CNT substrate would provide more active sites for PbO₂ deposition and could make a compact and fine surface coating. Moreover, the CNT-modified electrode had a higher active surface area and higher electrochemical activity than that without CNT doping. The life of the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode (296 h) was 1.38 times as much as that of the porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode (214 h). The electro-catalytic oxidation of phenol in an aqueous solution was studied to evaluate the electrochemical oxidation ability in environmental applications. The porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode displayed not only excellent electro-catalytic performance but also a low energy consumption using phenol as a model organic pollutant. The porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode has a higher kinetic rate constant and chemical oxygen demand (COD), which is 1.73 and 1.09 times those of the porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode, respectively. Moreover, CNT-doping can further increase the hydroxyl radical (˙OH) generation capacity. All these results illustrated that the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode could be used for pollutant degradation and had a great application potential.

---

1 Introduction

In recent years, the electrochemical oxidation of aqueous wastes containing non-biodegradable organics, such as phenol,¹ lignin,² 4-chlorophenol,³ pentachlorophenol,⁴ and perfluorocarboxylic acids,⁵ has been extensively studied due to its many distinctive advantages,⁶ including environmental compatibility, versatile energy efficiency, low-volume application, and amenability to automation.^6–8 For electrochemical oxidation, it has been found that the anode material plays a critical role in the degradation of organic pollutants.⁹ The reason is that electrode materials are conclusive for optimizing the electrochemical oxidation process because of their impact on the effectiveness of the mechanisms and reaction pathways.¹⁰ Hence, the current research hotspot is mainly focused on the development of stable anodes for the removal of persistent organic pollutants by electrochemical oxidation.^7,11

To date, literature results have shown that there are two main research focuses of electrochemical oxidation. The first research focus is the exploration of novel anode materials. The nature of the electrode material strongly affects both process selectivity and efficiency. In particular, anodes with low oxygen evolution overpotential, such as graphite,¹² RuO₂,¹³ and Pt,¹⁴ permit only the partial oxidation of organics, while anodes with high oxygen evolution overpotential, such as SnO₂,¹⁵ PbO₂,¹⁶ and BDD,¹⁷ favor high potential for the anodic oxidation of organic compounds. Among these, BDD anodes show the highest removal rate and stability, but their high cost and especially the difficulties of finding an appropriate substrate for the deposition of the diamond layer limit their large-scale application. The second research focus is to dope some metal elements into the oxide layer or intermediate layer, which can enhance the electro-catalytic activity and chemical or mechanical stability of oxide electrodes,^18,19 such as doping with rare earth (Ce, Pr, La, etc.)^20–23 or some metals (Au, Fe, Bi, etc.).^24–26 Although the abovementioned methods show high effectiveness for the electro-catalytic oxidation of organic pollution, some problems still restrict their practical application such as short service life and complex preparation process.

In order to gain more reaction active sites and more contact between the coating catalyst and reactant to promote electro-catalytic efficiency, we introduced a porous Ti substrate and carbon nanotubes (CNTs) into an SnO₂–Sb₂O₃ intermediate layer to prepare a novel porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode. As a novel titanium matrix, porous Ti has the advantages of good corrosion resistance, high porosity, large surface area, and good biocompatibility, and is thus widely used in aerospace, and in the medical and chemical fields.^27,28 Recent studies have revealed the significance of porous Ti for improving the performance of electrodes. Zhao et al. focused on systematically studying the effect of a porous Ti substrate on the surface structure and electrochemical properties of lead dioxide electrodes.²⁸ Zhang et al. found that the PbO₂/AC asymmetric electrochemical capacitor (AEC) based on porous-Ti/PbO₂ has a higher energy density over traditional electrochemical capacitors and a longer lifespan over traditional lead acid batteries.²⁹ In addition, carbon nanotubes (CNT), discovered by Iijima in 1991, are a novel material with good electrical properties and high chemical stability.³⁰ CNTs have been proposed as the ideal metal catalyst support for sensing and electro-catalytic applications.^15,31–33 The literature has mainly focused on CNT doped into the oxide layer to improve the catalytic ability of electrodes by an electro-deposition method;³⁴ however, slight attention has been paid to CNT-doping into the intermediate layer of electrode by other methods. It is also expected that the CNT-doping will not only improve the catalytic performance of electrodes, but also increase the specific surface area and enhance the number of contacts to increase an electrode's service life.

In this study, we propose a novel modification method for the PbO₂ electrode with the aim of improving the performance of the PbO₂ electrode. To the best of our knowledge, this is the first time that CNTs have been introduced into a SnO₂–Sb₂O₃ intermediate layer by thermal deposition. In addition, a systematic study was carried out to investigate the preparation and characterization of porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ compared with porous Ti/SnO₂–Sb₂O₃/PbO₂ without CNT-doping. The modified electrode was characterized, including its morphology, crystalline structure, electrochemical performance and accelerated service life. The oxidation performance and utilization of hydroxyl radicals on the electrodes were also evaluated. Phenol, as one of the most widely used pollutants in industrial wastewater, was selected as the target pollutant.

2 Experimental

2.1 Materials

The CNTs were purchased from Chengdu Organic Chemicals Co., Ltd. (China) with an outer diameter of 50 nm, and length of 10–20 μm. Porous titanium (purity 99.9%, 20 mm × 10 mm × 1 mm) was purchased from Baoji Jinkai Industrial Technology Co., Ltd. All the chemicals used in the experiments were received without further purification. All the solutions used in this work were prepared with deionized water.

2.2 Electrode preparation

2.2.1 Titanium surface treatment. In order to prepare a good adhesive metal oxide film material, the porous titanium substrate was pretreated according to the following procedure. First, a porous titanium substrate (20 mm × 20 mm × 1 mm) was mechanically polished with 600-grid abrasive paper. Then, the porous titanium substrate was cleaned with deionized water and acetone to remove solid particles and grease. It was subsequently immersed in sodium hydroxide (15% m m⁻¹) at the temperature of 60 °C for 30 min and then was etched in boiling hydrochloric acid (30% v/v) for about 60 min to produce a gray surface with uniform roughness. Finally, it was washed by ultrasonic cleaning in ultrapure water and stored in deionized water.

2.2.2 Coating of SnO₂–Sb₂O₃–CNT. The SnO₂–Sb₂O₃–CNT intermediate layer was prepared on porous titanium substrates by thermal decomposition. The coating solution was prepared by first mixing 1 mL of concentrated HCL in 25 mL isopropanol and then increasing the solution temperature to 80 °C for dissolving 6.65 g SnCl₄·5H₂O, 0.475 g SbCl₃, and 0.1 g CNT.

The treated titanium substrate was dipped in the solution for 5 min, and then dried at about 130 °C for 10 min, with the excess solvent being evaporated by hot air. Then, the treated titanium substrate was calcined at 500 °C for 15 min in a muffle furnace. All the above-mentioned processes were repeated twelve times. Finally, the electrodes were annealed at 500 °C for 60 min.

2.2.3 Electrochemical deposition of PbO₂. PbO₂ was deposited onto Ti/SnO₂–Sb₂O₃–CNT (geometric area: 1.0 × 1.0 cm²) under the current density of 20 mA cm⁻² at 65 °C for 1 h with magnetic stirring. The coating solution consisted of 0.1 M HNO₃ acidic media containing 0.5 M Pb(NO₃)₂ and 0.04 M NaF. After electro-deposition, the modified electrode was rinsed with deionized water. The porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode without CNT doping was made by the same abovementioned method. In the electro-deposition system, a copper foil electrode (20 mm × 20 mm) was used as the cathode.

More details concerning the electrode preparation are given in our previous study.³⁵

2.3 Electrode characterization

2.3.1 Physicochemical characterizations. A Japan RigaKu D/MAX2200PC diffractometer with Cu-Kα radiation (0.15418 nm) incident radiation was employed to analyze the crystal structure of the electrodes. The surface morphologies of the different electrodes were characterized by a HITACHI S-3400N scanning electron microscope. For the elemental analysis and spectral mapping, a Thermo Noran EDS spectrometer equipped with SYSTEM SIX was used.

2.3.2 Electrochemical measurements. Cyclic voltammetry (CV) was executed using a CHI760D electrochemical workstation (Chenhua Instrument Shanghai Co., Ltd., China) with a conventional three-electrode cell. The fabricated PbO₂ electrodes were used as the working electrode, a saturated calomel electrode (SCE) was used as the reference electrode and a platinum sheet electrode was used as the counter electrode. The anti-corrosion performance of the electrodes was investigated using an accelerated lifetime test with a current density of 500 mA cm⁻² in 3.0 mol L⁻¹ H₂SO₄ solution recorded simultaneously. The service life of the electrodes was considered to be terminated when the cell voltage reached 10 V.

All the electrochemical experiments were carried out at room temperature (25 ± 2 °C). All the solutions were prepared with deionized water, and the all reagents used in the experiments were of analytic grade.

2.4 Electrochemical degradation tests

Phenol was selected as a sample of aromatic contamination for studying the CNT-modified electrodes. The electrochemical degradation was carried out in a cylindrical single compartment cell equipped with a magnetic stirrer and a jacketed cooler to maintain a constant temperature. The fabricated PbO₂ electrode was used as an anode and stainless copper with the same dimension was used as the cathode with a distance of 1.5 cm between the two electrodes.

The current density was controlled to be a constant 30 mA cm⁻² by a direct current power supply (RXN-605D, China). The stirring rate was about 800 rpm. The experiments were carried out at 25 °C for 300 min. The volume of phenol solution was 80 mL with an initial concentration of 100 mg L⁻¹, and 0.1 mol L⁻¹ Na₂SO₄ was added to the solution. During the experiments, liquid samples were withdrawn from the electrolytic cell at fixed time intervals to determine the variation of phenol concentration and COD and to evaluate the reaction kinetics. The concentration of phenol was analyzed using a Varian high performance liquid-chromatography (HPLC) with a UV detector set at 270 nm. A C-18 column was used to separate the organics, while the mobile phase made of 30% CH₃CN and 70% water was at a flow rate of 1 mL min⁻¹. COD was digested by potassium dichromate using a 5B-1F Speed Digester (Lianhua Tech Co., China) at 150 °C for 10 min and determined by ultraviolet spectrophotometry at 610 nm using a TU-1810 UV/visible analysis spectrophotometer (Beijing Puxi Instrument Co. Ltd.). Each experiment was conducted three times in order to keep the relative standard deviations (RSD) less than 5%.

The general current efficiency (GCE), representing an average value of current efficiency between the initial time t = 0 and t, was calculated based on the results of the COD test and the following equation:⁷

where COD₀ and COD_t are the experimental values of phenol at times t = 0 (initial) and t (mg L⁻¹), respectively; I is the electrolysis current (A); F is Faraday's constant (96 485 C mol⁻¹); V is the volume of the solution (L); and 8 is the value of the oxygen equivalent mass (g eq.⁻¹).

The specific energy consumption (SEC) was the energy consumption for the removal of one kg of COD and was calculated as follows (expressed in kW h):⁸

where t is the electrolysis time (h); U and I are the average cell potential (V) and current (A); V is the sample volume (L); and COD₀ and COD_t are the difference in COD (mg L⁻¹).

In addition, we used the fluorescence method based on terephthalic acid, as this is a well-known ˙OH scavenger that can be used for estimating the amount of ˙OH radicals generated under various conditions by a fluorescence spectrophotometer (Perkin-Elmer LS-50, American). An aqueous solution of a 200 mL volume containing 0.5 mol L⁻¹ terephthalic acid, 0.5 g L⁻¹ NaOH and 0.25 mol L⁻¹ Na₂SO₄ was used as the electrolyte solution. The anode was a prepared PbO₂ electrode and the cathode was a stainless steel sheet. Hydroxyl radical production was performed at a current density of 30 mA cm⁻² at 30 °C. During the experiments, samples were drawn from the reactor every 5 min and diluted 10 times with deionized water, then analyzed with a fluorescence spectrophotometer. The fluorescence spectra were recorded in the range of 380 to 520 nm, using a 315 nm excitation wavelength.

3 Results and discussion

3.1 Surface morphology and crystal structure

3.1.1 Morphological analysis and film composition. The SEM cross-section morphologies of different electrodes are shown in Fig. 1(a)–(d). It can be clearly seen that the surface or internal parts of the porous titanium substrate were very rough and had irregular pores with average sizes of 30 μm, which can thus provide a larger specific surface area than the traditional plate titanium substrate. In addition, the results of EDS in Fig. 1(e) and (f) show that the SnO₂–Sb₂O₃ intermediate layer and CNT can be distributed into the internal parts of the porous titanium by thermal decomposition (the Pt peak comes from the platinum coating for SEM). However, the CNT-doped SnO₂–Sb₂O₃ films do not reveal major changes compared to the undoped electrode through the SEM cross-section morphology.

Fig. 1 SEM micrograph of (a) porous Ti substrate surface (×50), cross-sections for (b) porous Ti substrate (×100), (c) porous Ti/SnO₂–Sb₂O₃ electrode (×100), (d) porous Ti/SnO₂–Sb₂O₃–CNT electrode (×100), (e) EDS of the zone in (c), and (f) EDS of the zone in (d).

Fig. 2(a)–(d) shows the morphologies of the interlayer and surface layer of different electrodes. It is clearly seen from Fig. 2(a) and (b) that the introduction of CNT-doping into the SnO₂–Sb₂O₃ interlayer film significantly affected the film morphology. In contrast with the porous Ti/SnO₂–Sb₂O₃ substrate, CNT-doping induced the substrate surface to form a fibrous structure; hence, porous Ti/SnO₂–Sb₂O₃–CNT had a larger specific surface area, which could provide more active sites for electro-catalytic oxidation. In addition, the fibrous structure can make the PbO₂ to deposit more firmly and tightly. The CNT-modified electrode was expected to have better electrochemical properties and physico-chemical properties, which was further proved by the following experiments. To further check the impact of CNT-doping on the formation of PbO₂, Fig. 2(c) and (d) display the electrode crystal structure and appearance of porous Ti/SnO₂–Sb₂O₃/PbO₂ and porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ with a magnification of 100× and 4000× in the insets, respectively. It is noticeable that the particles sizes of porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ become smaller than those of porous Ti/SnO₂–Sb₂O₃/PbO₂, and the coating particles grown on the SnO₂–Sb₂O₃–CNT intermediate layer did not present fissures or laminations and appeared to be more compact and uniformly distributed.

Fig. 2 SEM micrograph of the electrodes. (a) Porous Ti/SnO₂–Sb₂O₃ electrode (×10 000), (b) porous Ti/SnO₂–Sb₂O₃–CNT electrode (×10 000), (c) porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode (×100), and (d) porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode (×100); insets in (c) and (d) are SEM images with high magnification (×4000) corresponding to the electrode. (g) EDS of the zone in (a); (h) EDS of the zone in (b).

Fig. 2(g) and (h) show the EDS spectrum of the coating of porous Ti/SnO₂–Sb₂O₃ and porous Ti/SnO₂–Sb₂O₃–CNT, respectively (Pt peak comes from the platinum coating for SEM). Sn, Sb, Ti and O were clearly observed, indicating the formation of the oxide mixture of Sn and Sb. Due to porous Ti with larger surface areas, the oxide layer may not completely cover porous Ti substrate surface. The C peak was detected in EDS (Fig. 2(h)), showing that CNT was successfully doped into the middle layer (SnO₂–Sb₂O₃) by thermal decomposition method and not oxidized during the calcination process. This result was not consistent with the result reported in literature.¹⁵ Based on the results from the SEM and EDS analyses, we conclude that the CNT was successfully doped into SnO₂–Sb₂O₃ intermediate layer by calcination process and would be beneficial to electrochemical properties. Based on the results from the abovementioned SEM-EDS analyses, the formation process of porous Ti/SnO₂–Sb₂O₃–CNT is described in Fig. 3. In addition, we use the gravimetric method to calculate the growing rate of SnO₂–Sb₂O₃–CNT/PbO₂ and the value is 1.979 mg cm⁻².

Fig. 3 Schematic of the fabrication of the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode.

3.1.2 Structural analysis by XRD. The crystal structure of the PbO₂ coating deposited on the porous Ti/SnO₂–Sb₂O₃–CNT substrate was determined by X-ray diffraction. The corresponding patterns of the PbO₂ coatings with different substrate are shown in Fig. 4. It can be seen that the XRD of PbO₂ coatings based on different substrates are quite similar, but the reflection intensities of the CNT-doped coating increased, while the half-widths of the reflection peaks decreased, meaning that the degree of crystallization increased and the crystallite size decreased owing to the CNT-modification. In addition, Fig. 4 displays the main diffraction peaks at 2θ = 25.4°, 32.0°, 36.2°, 49.2°, 52.1°, 59.2°, 62.3°, 74.4°, 84.3°, and 85.9°, which are assigned to the (110), (101), (200), (211), (220), (310), (301), (321), (312), and (411) planes of β-PbO₂, and the strong main crystal plane of β-PbO₂ is assigned to the (101), (211), and (301) planes. At the same time, some weak peaks (2θ = 67.8° and 76.9°) infer the presence of α-PbO₂. However, no diffraction peaks of tin and antimony oxides are observed, indicating that the active layer of PbO₂ was uniform and thus tin and antimony oxides were undetected by XRD. This was also confirmed by the abovementioned SEM results.

Fig. 4 XRD patterns of (a) porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode and (b) porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode.

3.2 Electrochemical characterization of the electrodes

3.2.1 Electrochemical active surface area. The electrochemical active surface area means that the active sites are accessible to the electrolyte when the electrochemical reaction occurs. The voltammetric charge quantity (q*), which is related to the real surface area and the number of active sites, can reflect the electrochemical activity of an electrode. A larger q* indicates a higher electrode activity. We employed the method reported in literature^36–38 to calculate q* for estimating the electrode activity. The equation can be expressed as follows:

(q*)⁻¹ = (q^*_T)⁻¹ + kv^1/2 (1)

The total charge quantity q^*_T stands for the quantity of theoretically electrochemical active sites of the electrode surface, and v stands for the scan rate of voltage, while k is a constant. In order to determine the electrochemical active surface areas of porous Ti/SnO₂–Sb₂O₃/PbO₂ and porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂, q* was tested in Na₂SO₄ solution. Special adsorption occurs between sulfate ions and PbO₂, forming insoluble PbSO₄, which results in pseudo-capacitance (eqn (2) and (3)):

H⁺ + SO_4ads²⁻ ⇔ HSO_4ads⁻ (2)

4HSO_4ads⁻ + PbO₂ + 2e⁻ = PbSO_4ads + 3SO_4ads²⁻ + H₂O (3)

Therefore, the q* in the pseudo-capacitance region of PbO₂ in Na₂SO₄ solution can be defined as the electrochemical active surface area of PbO₂ in the sulfate electrolyte. Fig. 5(A) shows the cyclic voltammograms of porous Ti/SnO₂–Sb₂O₃/PbO₂ and porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂, q* was integrated from the cyclic voltammetric curves over the whole potential range from 0.3 V to 0.8 V. The relationship of the reciprocal of q* versus the square root of the scan rate is shown in Fig. 5(B). For PbO₂ electrodes, the q* at low scan rate reflects the deep discharge performance; the q* at high scan rate reflects the power performance. The q* of PbO₂ electrode at a scan rate of 10 and 50 mV s⁻¹ are listed in Table 1. The k constant is the slope of our straight lines in Fig. 5(B). k constant represents the change rate of the electrochemical active surface area with increasing scan rate. The smaller the k constant is, the slower the electrochemical active surface area declines with increasing the scan rate. Therefore, comparing with the value of q* and k, the result indicated that the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode had the highest active surface area.

Fig. 5 (A) Cyclic voltammograms of (a) porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode and (b) porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode in 0.5 mol L⁻¹ Na₂SO₄ solution at a scan rate of 50 mV s⁻¹. (B) Relationship of the reciprocal of the voltammetric charge quantity versus the square root of the scan rate: (a) porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode and (b) porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode.

Table 1 Voltammetric charge information of the different electrodes

Electrode	k constant	q* at 10 mV s^−1	q* at 50 mV s^−1
Porous Ti/SnO2–Sb2O3/PbO2	33.5588	0.0093	0.0042
Porous Ti/SnO2–Sb2O3–CNT/PbO2	14.3733	0.0186	0.0091

---

3.2.2 Cyclic voltammetry. Fig. 6 shows the cyclic voltammograms of different electrodes in the potential range between 0.8 V and 2.0 V in 0.5 mol L⁻¹ H₂SO₄ solution. As can be observed, the oxidation–reduction properties of the electrodes appear to be more or less similar. An oxidation peak appeared in the anodic branch of the curve, which indicated that Pb²⁺ (PbSO₄) could be oxidized to Pb⁴⁺ (PbO₂) between 1.7 and 1.8 V (vs. SCE). In the cathodic branch curve, a reduction peak might be attributed to the generation of Pb²⁺ (PbSO₄) from the following reaction^29,39 between 1.0 and 1.2 V (vs. SCE):

PbO₂ + H₂SO₄ + 2H⁺ + 2e⁻ ⇔ PbSO₄ + 2H₂O (4)

Fig. 6 Cyclic voltammogram curves of different electrodes: (a) porous Ti/SnO₂–Sb₂O₃/PbO₂ and (b) porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ in 0.5 mol L⁻¹ H₂SO₄ solution at a scan rate of 50 mV s⁻¹.

In addition, The CNT-modified electrode presented higher oxidation–reduction peak currents than the other electrode without CNT-doping, meaning that many more substances are involved in the oxidation reaction. The reason for this mainly lies in the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode having a larger active surface area, which provides more active site for the electrochemical reaction. The result also further proved the above conclusion. In conclusion, the introduction of a CNT-doped interlayer is expected to increase the oxidation ability and catalytic activity of porous Ti/SnO₂–Sb₂O₃/PbO₂.

3.2.3 Electrochemical impedance. EIS is a powerful technique to study porous electrodes. To understand the effect of the CNT-doping, we did further EIS studies. Fig. 7 presents the electrochemical impedance of porous Ti/SnO₂–Sb₂O₃/PbO₂ (a) and porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode (b) in 0.5 mol L⁻¹ H₂SO₄ solution in the oxygen evolution region (1.8 V vs. SCE). The equivalent circuit shown in Fig. 8 was used to fit the EIS data. The simulated data of each parameter in Fig. 7 is listed in Table 2. In this R_s(R_ctQ_dl) circuit, R_s represents the ohmic resistance, including the resistance of electrolyte and active material. R_ct stands for the charge-transfer resistance, reflecting the oxygen evolution reaction activity. Q_dl is introduced to replace the electric double layer capacitor.

Fig. 7 EIS plots in the 0.5 M acidic solution: (a) porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode and (b) porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode. Electrode potential: 1.8 V vs. SCE.

Fig. 8 Equivalent circuit used in the analysis of the experimental EIS data.

Table 2 Simulated values of each electrical element

Electrode	Rs/Ω cm^2	Qdl/Ω cm^−2 s^n	Rct/Ω/Ω cm^2	n
Porous Ti/SnO2–Sb2O3/PbO2	0.3962	0.0067	1.315	0.9591
Porous Ti/SnO2–Sb2O3–CNT/PbO2	0.2847	0.0079	1.187	0.9531

---

Two evident semicircles appear in the electrochemical impedance spectra in Fig. 7. The diameter of the semicircle size reflects R_ct and the resistance values were 1.315 Ω cm² and 1.187 Ω cm² in Table 2, respectively, indicating that the oxygen evolution reaction activity of the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode was slightly higher. According to the reaction mechanism of electrode oxygen evolution, the oxygen evolution activity depends on the active sites of active coating, whereby the more the active sites, the greater the reaction activity. Hence CNT-doping into the intermediate layer also increases the active sites, which can provide a larger specific surface area. The result was also further proven by the R_s values in Table 2 showing that the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode prepared had the smallest R_s, indicating the largest electrochemical active surface area. From the abovementioned results, it can be concluded that CNT-doping can reduce the resistance of the active material.

3.2.4 Electrode stability. The service life is another important factor to be considered for the electrode quality and practical application. As shown in Fig. 9, the service life of porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ was 291 h, as much as 1.35 times that of porous Ti/SnO₂–Sb₂O₃/PbO₂ (214 h). This result reveals that doping the CNT into the SnO₂–Sb₂O₃ interlayer film can improve the electrochemical stability. The improved service lifetime can be attributed to the following reasons. First, it results from the decreased PbO₂ particle size, which makes a compact and fine surface layer, as observed from the SEM images in Fig. 2. The compact surface can baffle the penetration of the supporting electrolyte toward the titanium substrate through the cracks and can thus delay the formation of the non-conductive TiO₂ layer. Second, the fibrous structure of the SnO₂–Sb₂O₃–CNT interlayer itself can make PbO₂ deposit more tightly and reduce the film detachment.

Fig. 9 Accelerated life test of (a) porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode and (b) porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode in 3 mol L⁻¹ H₂SO₄ solution under 0.5 A cm⁻².

3.3 Electrochemical degradation test

3.3.1 Electrochemical degradation of phenol. To further investigate the influence of CNT-doping on the electro-catalytic degradation activity of the electrodes, degradation experiments were carried on different electrodes. The variations of removal efficiency for phenol with electrolysis time are shown in Fig. 10. It can be clearly seen that the removal efficiency rates are up to 94% within 300 min for the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode, but only 80% for the porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode. Therefore, the modified electrode with CNT-doping significantly improves the phenol removal performance.

Fig. 10 Phenol removal efficiency as a function of degradation time for different electrodes (A). Kinetic analysis of the curves (B). Operating conditions: Na₂SO₄ concentration, 0.1 mol L⁻¹; initial concentration, 100 mg L⁻¹; current density: 30 mA cm⁻²; stirring rate: 800 rpm.

In addition, the processes of degradation were found to well fit the pseudo-first-order model for all electrodes as shown in eqn (5). The fitting results are shown in Fig. 10(B).

where C₀ is the initial concentration of phenol, C is the concentration of phenol at given time t, and k is the kinetic rate constant.⁴⁰ The k values for the two electrodes were 0.0088 min⁻¹ and 0.0051 min⁻¹. The reaction rate constant of the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode was 1.73 times greater than that of the porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode. That suggested that phenol could be degraded more rapidly on the CNT-doped electrode.

The COD reduction is very important to monitor the performance of the different electrodes. Hence, COD removals of phenol by the two different electrodes were compared to further evaluate the effect of the CNT-doping for the electro-catalytic oxidation. The variations of COD with electrolysis time are shown in Fig. 11. Clearly, the COD was removed more rapidly on the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ than that on the porous Ti/SnO₂–Sb₂O₃/PbO₂. About 95% and 87% of COD removal were achieved in 300 min, respectively. Hence, the organic compounds were more effectively degraded and oxidized on the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode. The high phenol removal rate can be ascribed to the larger active surface area.²² In addition, the general current efficiency (GCE) and specific energy consumption (SEC) were also calculated and are listed in Table 3.

Table 3 The general current efficiency (GCE) and specific energy consumption (SEC) for the different PbO₂ electrodes

Electrode sample	GCE	SEC/kW h (kg COD)^−1
Porous Ti/SnO2–Sb2O3/PbO2	61.3%	289.45
Porous Ti/SnO2–Sb2O3–CNT/PbO2	67.1%	250.62

---

Evidently, all the above-mentioned results indicate that CNT can effectively enhance the surface area of the electrode, which improves the performance of the electrode. This agrees well with the experimental results presented above.

Fig. 11 Variations of COD with electrolysis time for the different anodes in 80 mL 100 mg L⁻¹ phenol + 0.25 mol L⁻¹ Na₂SO₄ solution with a current density of 30 mA cm⁻² at 25 °C.

3.3.2 Voltammetric characteristics of the electrodes. The two electrodes were characterized electrochemically by cyclic voltammogram tests in a 0.5 mol L⁻¹ Na₂SO₄ solution in the absence and in the presence of 100 mg L⁻¹ phenol. The CV curves for the two electrodes were recorded between 0 V and 2.0 V (vs. SCE) at a scan rate of 10 mV s⁻¹. As show in Fig. 12, when 100 mg L⁻¹ phenol was added to the supporting electrolyte, no additional peaks were found compared to the CV curve recorded in the reference solution, indicating that the direct electron transfer did not occur. Thus, it can be concluded that the phenol degradation in this system should be achieved via indirect electrochemical oxidation and mediated by hydroxyl radicals.

Fig. 12 Cyclic voltammogram curves of different electrodes: (A) porous Ti/SnO₂–Sb₂O₃/PbO₂ and (B) porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ in the absence and in the presence of phenol.

3.3.3 Hydroxyl radical (˙OH) generation capacity of the electrodes. Generally, organic pollutants are mainly degraded by the indirect electrochemical oxidation mediated by ˙OH radicals in the electro-catalytic oxidation process.^8,41 Hence, quantitative determination of the generation ability of hydroxyl radicals in the electrochemical degradation process were necessary to be measured. During electrochemical treatment, terephthalic acid, as a type of ˙OH radical capture agent, can readily react with ˙OH radicals to produce the highly fluorescent product 2-hydroxyterephthalic acid. The amount of ˙OH radicals formed was approximately equal to the amount of 2-hydroxyterephthalic acid, which was represented as a fluorescence intensity.²⁸ As can be seen in Fig. 13, the fluorescence intensity of 2-hydroxyterephthalic acid around 425 nm for two electrodes increased with increasing the reaction time, indicating that ˙OH radicals were indeed formed on the anodes and played an important role in the electrochemical degradation test. Comparing the fluorescence intensity for the different electrodes at different reaction times, it was found that the fluorescence intensity for the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode was higher than that for the porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode, which reveals the excellent electro-catalytic activity of the porous PbO₂ electrode. Therefore, the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode could oxidize pollutants more effectively than the porous Ti/SnO₂–Sb₂O₃/PbO₂.

Fig. 13 Fluorescence spectrum changes observed during the electro-catalytic oxidation process in a 0.5 mol L⁻¹ aqueous solution of terephthalic acid using the prepared PbO₂: (A) porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode and (B) porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode.

4 Conclusions

A highly effective porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode was successfully prepared on a porous titanium substrate by thermal decomposition and electro-deposition. The electrode modified with CNT versus without CNT has a higher specific surface area, which can improve the interlayer coating structure effectively and favor the formation of PbO₂ during electro-deposition. The voltammetric charge quantity indicated that the electrode modified with CNT enhanced the mass transfer rate on a PbO₂ electrode. The results of the accelerated life tests showed that the service life of the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ electrode was longer than that of electrodes without CNT-doping, which was 1.35 times that of the porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode. The electrode modified with CNT was then applied in phenol wastewater treatment and presented excellent ability for organic pollutant oxidation and degradation compared with the electrode without CNT modification. After 5 h, the phenol was almost completely decomposed using porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂, and the COD removal reached 94.81%, while the value for porous Ti/SnO₂–Sb₂O₃/PbO₂ was just 87.01%. Moreover, the general current efficiency (GCE) and specific energy consumption (SCE) were superior to that of the porous Ti/SnO₂–Sb₂O₃/PbO₂ electrode (67.1%, 250.62 kg⁻¹ COD⁻¹ versus 61.3%, 289.45 kg⁻¹ COD⁻¹). In addition, CNT-doping can further increase the hydroxyl radical (˙OH) generation capacity. Considering the improved electro-catalytic oxidation performance and service life in practice, the porous Ti/SnO₂–Sb₂O₃–CNT/PbO₂ anodes can be prepared on a large scale and applied in industry.

Acknowledgements

The authors are grateful for the financial support provided by the Innovative Program of Activities for University in Shanghai (no. DCX2015064). Wenli Zhang (Jilin University) and Quansheng Zhang (Shanghai Institute of Technology) are also gratefully acknowledged for supplying experimental guidance and the electrochemical workstation, respectively.

References

1. H. Li, Y. Chen, Y. Zhang, W. Han, X. Sun, J. Li and L. Wang, J. Electroanal. Chem., 2013, 689, 193–200.
2. D. Shao, J. Liang, X. Cui, H. Xu and W. Yan, Chem. Eng. J., 2014, 244, 288–295.
3. X. Duan, L. Tian, W. Liu and L. Chang, Electrochim. Acta, 2013, 94, 192–197.
4. J. Niu, Y. Bao, Y. Li and Z. Chai, Chemosphere, 2013, 92, 1571–1577.
5. J. Niu, H. Lin, J. Xu, H. Wu and Y. Li, Environ. Sci. Technol., 2013, 47, 4959.
6. E. Brillas and C. A. Martínez-Huitle, Appl. Catal., B, 2015, 166–167, 603–643.
7. M. Panizza and G. Cerisola, Chem. Rev., 2009, 109, 6541–6569.
8. C. Martínez-Huitle and S. Ferro, Chem. Soc. Rev., 2006, 35, 1324.
9. A. B. Velichenko, R. Amadelli, E. A. Baranova, D. V. Girenko and F. I. Danilov, J. Electroanal. Chem., 2002, 527, 56–64.
10. Y. Xia, Q. Dai and J. Chen, J. Electroanal. Chem., 2015, 744, 117–125.
11. G. Zhao, Y. Zhang, Y. Lei, B. Lv, J. Gao, Y. Zhang and D. Li, Environ. Sci. Technol., 2010, 44, 1754–1759.
12. Y. Shen, H. Xu, P. Xu, X. Wu, Y. Dong and L. Lu, Electrochim. Acta, 2014, 132, 37–41.
13. L. Du, J. Wu and C. Hu, Electrochim. Acta, 2012, 68, 69–73.
14. A. Fernandes, D. Santos, M. J. Pacheco, L. Ciríaco and A. Lopes, Appl. Catal., B, 2014, 148–149, 288–294.
15. L. Zhang, L. Xu, J. He and J. Zhang, Electrochim. Acta, 2014, 117, 192–201.
16. J. P. Carr and N. A. Hampson, Chem. Rev., 1972, 72, 679–703.
17. M. Panizza, A. Kapalka and C. Comninellis, Electrochim. Acta, 2008, 53, 2289–2295.
18. X. Hao, L. Jingjing, Y. Wei and W. Chu, Rare Met. Mater. Eng., 2013, 42, 885–890.
19. A. B. Velichenko and D. Devilliers, J. Fluorine Chem., 2007, 128, 269–276.
20. Y. Feng, Y. Cui, B. Logan and Z. Liu, Chemosphere, 2008, 70, 1629–1636.
21. Y. Feng, Y.-H. Cui, J. Liu and B. E. Logan, J. Hazard. Mater., 2010, 178, 29–34.
22. H. Xu, A. Li and X. Cheng, Int. J. Electrochem. Sci., 2011, 6, 5114–5124.
23. H. Lin, J. Niu, J. Xu, Y. Li and Y. Pan, Electrochim. Acta, 2013, 97, 167–174.
24. R. Berenguer, C. Quijada and E. Morallón, Electrochim. Acta, 2009, 54, 5230–5238.
25. A. Moncada, S. Piazza, C. Sunseri and R. Inguanta, J. Power Sources, 2015, 275, 181–188.
26. Y. Song, G. Wei and R. Xiong, Electrochim. Acta, 2007, 52, 7022–7027.
27. M. Manjaiah, S. Narendranath and S. Basavarajappa, Trans. Nonferrous Met. Soc. China, 2014, 24, 12–21.
28. W. Zhao, J. Xing, D. Chen, Z. Bai and Y. Xia, RSC Adv., 2015, 5, 26530–26539.
29. W. Zhang, H. Lin, H. Kong, H. Lu, Z. Yang and T. Liu, Int. J. Hydrogen Energy, 2014, 39, 17153–17161.
30. S. Iijima, Nature, 1991, 354, 56–58.
31. F. Hu, Z. Dong, X. Cui and W. Chen, Electrochim. Acta, 2011, 56, 1576–1580.
32. D. Yang, L. Zhu and X. Jiang, J. Electroanal. Chem., 2010, 640, 17–22.
33. F. Hu, X. Cui and W. Chen, Electrochem. Solid-State Lett., 2010, 13, F20.
34. X. Duan, F. Ma, Z. Yuan, L. Chang and X. Jin, J. Electroanal. Chem., 2012, 677–680, 90–100.
35. W. Zhao, J. Xing, D. Chen, Z. Bai and Y. Xia, RSC Adv., 2015, 5, 26530–26539.
36. H. Vogt, Electrochim. Acta, 1994, 39, 1981–1983.
37. S. Trasatti, Electrochim. Acta, 1991, 36, 1659–1667.
38. Y. Chen, L. Hong, H. Xue, W. Han, L. Wang, X. Sun and J. Li, J. Electroanal. Chem., 2010, 648, 119–127.
39. D. Pavlov, A. Kirchev, M. Stoycheva and B. Monahov, J. Power Sources, 2004, 137, 288–308.
40. J. Zhao, C. Zhu, J. Lu, C. Hu, S. Peng and T. Chen, Electrochim. Acta, 2014, 118, 169–175.
41. X. Zhu, J. Ni, H. Li, Y. Jiang, X. Xing and A. G. L. Borthwick, Electrochim. Acta, 2010, 55, 5569–5575.

---