Three-dimensional N-doped carbon electrodes activate peroxymonosulfate for tetracycline degradation

Abstract

Transition metal-based materials have been widely used as catalysts in both electrochemical oxidation (E) and peroxomonosulfate (PMS) processes to degrade pollutants in water. However, agglomeration of catalyst particles and leaching of metal ions hinder their practical application in pollutant decomposition. In this study, a novel N-doped carbon material (CPANI/CF) was prepared as a three-dimensional electrode by using low-cost synthesised polyaniline-loaded carbon felt as a precursor to improve the electrochemical oxidation capability and capacity to activate PMS in the electrochemical oxidation process. The constructed three-dimensional electrocatalytic system (E–PMS–CPANI/CF) achieved 90.59% TC degradation within 40 min, and the system still removed more than 80% of TC after repeating the experiment for five cycles. We investigated how different experimental conditions (such as the carbonation temperature of CPANI/CF electrode, voltage, PMS dosage and initial pH) affect the degradation of TC. Through quenching experiments, chemical probe experiments and electron spin resonance (ESR) analyses, we discovered that the 3D electrocatalytic system produced single-linear oxygen (¹O₂), hydroxyl radical (˙OH) and sulphate radical (SO₄˙⁻), which could effectively degrade the pollutants. The results of this study provide a useful strategy for efficient and cost-effective wastewater treatment.

---

1 Introduction

Over the past few years, antibiotics have been extensively utilized in both human and veterinary medicine to prevent and treat diseases through their ability to interfere with or eliminate microorganisms.^1,2 Tetracycline (TC) is the most commonly used antibiotic in animal husbandry. High concentrations of TC are frequently detected in aquatic environments due to its low uptake by animals.³ However, TC is toxic and persistent, and if not handled properly, it can seriously damage the ecological environment and endanger human health. Therefore, developing a highly efficient, inexpensive, simple and environmentally friendly method to remove TC from water is necessary.⁴

Conventional wastewater treatment methods (e.g., biological, physical, chemical, etc.) are inefficient in removing tetracycline from water. Researchers prefer advanced oxidation technologies (AOPs) because they are practical, energy-efficient, environmentally friendly and cost-effective. Electrochemical oxidation technology is a new type of oxidation technology that uses reactive oxygen species (ROS) during chemical reactions to oxidise and degrade pollutants. Although two-dimensional electrochemical oxidation technology (2D-E) for wastewater degradation is operationally simple and low-cost, its shortcomings, such as low mass transfer rate, high electrode consumption, and high energy consumption, limit the wide range of applications of 2D-E technology in industry.⁵ Three-dimensional electrochemical oxidation technology (3D-E) is revised on the basis of 2D-E technology by filling particle electrodes between the cathode and the anode, so that the particle electrodes form many microelectrodes under the action of an electric field. Under the influence of an electric field, the particle electrodes form many microelectrodes. This increases the reaction area and improves the mass transfer efficiency of the E system. 3D-E technology demonstrates superior electrochemical and electrocatalytic performances compared to traditional 2D-E technology.⁶

Peroxymonosulfate (PMS) oxidation is an advanced oxidation technology, which decomposes difficult-to-degrade organic pollutants into CO₂, H₂O and other low-toxicity intermediates by generating ROS (˙OH, SO₄˙⁻, ¹O₂, etc.).⁷ In general, PMS can be activated by electrocatalysis, photocatalysis and metal catalysis.^8–11 Using metal-based materials to activate PMS offers a low-cost and highly effective catalytic performance, but metal ions are inevitably precipitated during the activation of persulfate with metal-based materials, resulting in secondary fouling.¹² In contrast, carbon-based catalysts are advantageous due to their high chemical stability, low cost and absence of secondary pollution during PMS activation.^9,12–15 According to the report, doping nitrogen atoms into the carbon matrix during the preparation process significantly improves the catalytic performance of electrode materials.^16–18 The most commonly used method for the synthesis of N-doped carbon electrodes is to introduce an external N source (melamine, ammonium nitrate and urea, etc.) followed by high-temperature annealing, which can be an effective way to synthesize N-doped carbon electrode materials, but it is a cumbersome process, with a small amount of added N that is not sufficiently stable.¹⁹ Another method is to prepare N-doped carbon electrodes by carbonising the N-rich precursor. N-doped carbon electrode materials derived from polyaniline have been widely used in practice. Still, most of them are discretely distributed in the form of powder, which may result in catalyst build-up that is not easy to recycle if applied to electrocatalytic systems.^20–22 Accordingly, it's imperative to create a novel form of N-doped carbon electrode. Currently, N-doped carbon electrodes are usually prepared by coating the catalyst powder with an added binder on the carrier.^23,24 This method not only decreases the electrode's catalytic properties but can also cause environmental pollution if the material falls off in the process of use.²⁵ Polyaniline (PANI), which can be synthesised in situ on a carrier, can solve the above problems.

In summary, we propose the synergistic system of E–PMS–CPANI/CF. In this paper, the carrier material is carbon felt (CF). PANI was polymerized in situ on the CF surface without using a binder, and high-temperature carbonisation was carried out in an N₂ atmosphere to prepare CPANI/CF as a 3D electrode to improve the electrochemical oxidation capability and capacity to activate PMS in the 3D-E process. The proposed E–PMS–CPANI/CF synergistic system is an efficient, low-cost and sustainable wastewater treatment technology. The specific research procedure is as follows: (i) we used scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS), Brunauer–Emmett–Teller (BET), an X-ray Diffractometer (XRD) and Fourier transform infrared spectrometry (FTIR) to characterise the 3D electrodes. (ii) We researched the effects of voltage, carbonation temperature, pH and PMS addition on the electrode's electrocatalytic performance. The results of this study can provide insights into optimizing these parameters to enhance electrocatalytic performance. (iii) We used electron spin resonance (ESR) to test for the possible presence of ROS in different systems, and investigated the role of ROS in the system through radical quenching experiments and chemical probe experiments, and to investigate the mechanism of tetracycline degradation using the E–PMS–CPANI/CF process.

2 Materials and methods

2.1 Reagents and materials

CF was purchased from Tianjin Carbon Factory (Tianjin, China), ammonium persulfate (APS), aniline (ANI), concentrated hydrochloric acid (36–38% HCl), tetracycline (C₂₂H₂₄N₂O₈), potassium iodide (KI), peroxymonosulfate (PMS), anhydrous ethanol (EtOH), sodium sulphate (Na₂SO₄), sodium hydroxide (NaOH), sulfuric acid (H₂SO₄), 1,10-phenanthroline5,5dimethyl-1-pyrroline N-oxide (DMPO), sodium bicarbonate (NaHCO₃), ethanol (EtOH), phenol, coumarin, benzoic acid (BA), p-hydroxybenzoic acid (p-HBA), 1,3-diphenylisobenzofuran (DPBF), potassium titanium oxide oxalate dihydrate (C₄K₂O₉Ti·H₂O), furfuryl alcohol (C₅H₆O₂) and tert-butanol (TBA) were bought from Sinopharm Chemical Reagent Co. Ltd (S, China).

2.2 Synthesis of CPANI/CF

Firstly, start by cleaning a 2 cm × 4 cm × 0.2 cm CF block with 1 M HCl, ethanol, and deionized (DI) water to eliminate impurities and organic matter from its surface, then place the sample into a vacuum at 65 °C for 12 hours to dry.

The synthesis steps of CPANI/CF electrode materials are shown in Fig. 1. Solution A was prepared by dissolving 0.9 mL of aniline monomer in 30 mL of 1 M HCL and stirring the mixture. Similarly, Solution B was prepared by dissolving 0.4 g of APS in 30 mL of 1 M HCL and stirring it well. CF was impregnated through ultrasonic mixing in solution A for 0.5 h and then impregnated for 12 h. Then it was taken out of solution A and added to solution B. The mixture was mixed for 30 seconds, then it was polymerised at 4 °C for 9 h. After the sample was washed with 1 M HCl and DI water sequentially, it was dried in a vacuum oven at 65 °C for 24 hours to obtain PANI/CF. The N-doped carbon electrode (CPANI/CF) was made by carbonising the above obtained PANI/CF in a tube furnace kept at a temperature of 700 °C for 2 hours under a N₂ atmosphere.

Fig. 1 Schematic diagram of the CPANI/CF synthesis process.

2.3 Experimental procedure

During the experiment, various devices were used, including a cathode plate, an anode plate, a magnetic stirrer, an electrolysis tank and a direct current (DC) power supply (Fig. S1, ESI†). The electrolysis tank had a volume of 200 mL and was made of glass, the anode plate was constructed from graphite while the cathode plate was constructed from titanium. The cathode and anode had an effective working area of 12.5 cm² and were positioned parallel to each other within the reactor. Additionally, a three-dimensional electrode, CPANI/CF, was placed between the cathode and anode. Set the contaminant concentration in the system to 50 mg L⁻¹. The process was performed using an electrolyte of 0.05 mol L⁻¹ Na₂SO₄, pH = 3, plate spacing of 6 cm, voltage of 5 V, and magnetic stirrer speed of 2000 rpm for 40 min. The samples were taken at specific times, and the solution that needed testing was filtered through a membrane with a pore size of 0.45 μm. The UV spectrophotometer was used to measure the concentration of pollutants at 360 nm. All experiments used three parallel samples.

2.4 Characterisation

Scanning electron microscopy (SEM, Hitachi S4800) was used to observe the surface morphology of the electrodes. The composition of the electrodes was analyzed using X-ray photoelectron spectroscopy (XPS, NEXSA). The electrodes were tested for their purity and crystalline form using X-ray diffraction (XRD, D8 Discover). Functional groups on the 3D electrode surface were analysed using Fourier Transform Infrared Spectroscopy (FTIR, Nicolet iS50). 3D electrodes were subjected to analysis of their specific surface area and pore size by the fully automated physicochemical adsorption tester (BET, Autosorb-iQ-C). The active substances were identified using electron spin resonance (ESR, FA20). Pollutant concentrations were determined using a UV spectrophotometer.

2.5 Analytical methods

Eqn (1) was used to calculate the energy consumption required for removing pollutant.where V, U, I and T are the solution volume (L), applied voltage (V), average cell current (A) and reaction time (h), respectively.

The first order rate constant for TC degradation is determined according to equation eqn (2).²⁶

where k represents the first-order rate constant of the reaction (min⁻¹), C and C₀ represent the pollutant concentration at the specified time and the initial pollutant concentration, respectively.

The degradation contributions of ˙OH, SO₄˙⁻ and ¹O₂ could be determined by eqn (3)–(5).^27,28

where k_TBA is the rate constant when TBA is added, k_EtOH is the rate constant when EtOH is added, k_FFA is the rate constant when FFA is added, and k represents the first-order rate constant of the reaction (min⁻¹).

3. Results and discussion

3.1 Characterisation of the 3D electrode

In Fig. 2, the characterisation of the surface morphology of CF, PANI/CF and CPANI/CF electrodes using SEM is shown. The surface of CF before modification is smooth in Fig. 2(a). Whereas, in Fig. 2(b), the modified CF is relatively coarse. Fig. 2(c) shows that the polyaniline was uniformly covered on the surface of CF, and the polyaniline is a typical short rod structure. In Fig. 2(d)–(f), CPANI/CF was obtained after carbonisation, and the surface morphology changed from the original short and thick rod-like structure to a thin and long fibre-like one, forming an interconnected and cross-linked fibre porous network, and the macroscopic voids of CPANI/CF were widened. The calcination process likely caused the formation of micropores on the CPANI/CF due to the escape of large amounts of N₂ and O₂, which improved the electrode's surface area, adsorption capacity, and active sites for the reaction. When the carbonization temperature increased to 800°, the material structure collapsed, resulting in a partial buildup of material (Fig. 2(f)).

Fig. 2 SEM images of the 3D electrode: (a) CF, (b) CPANI/CF-700, (c) PANI/CF, (d) CPANI/CF-600, (e) CPANI/CF-700, and (f) CPANI/CF-800.

The elemental composition and chemical properties of CF, PANI/CF and CPANI/CF-t electrodes were analysed by XPS. As shown in Fig. 3(a), CF is primarily composed of O and C elements, PANI/CF and CPANI/CF-t consist mainly of C, N and O elements. This shows that the obvious N 1s signal of PANI/CF originated from polyaniline polymerised on the CF surface in Fig. 3(b), and the four typical peaks were located at 398.6 eV, 399.8 eV, 400.6 eV and 401.9 eV representing quinoid imine (–N=), benzoid amine (–NH–), doped amine (–NH⁺=) and doped imine species (–NH₂⁺–),^29,30 confirming the successful polymerisation of polyaniline on CF.¹⁹ An obvious N 1s signal appeared in CPANI/CF-t (Fig. 3(d)–(f)), and the three characteristic peaks were located at 398.6 eV, 400.4 eV, and 401.6 eV, corresponding to pyridinic N, pyrrolic N and graphitic N, respectively, indicating the successful doping of carbon fiber surfaces with N.^31,32 The relative atomic contents of N and O in the CPANI/CF electrodes gradually decreased with increasing carbonisation temperature, and the N content decreased from 7.79% (CPANI/CF-600) to 5.03% (CPANI/CF-800) (Table 1), which might be due to the C–N bond breaking when the carbonisation temperature is too high. The graphitic N content increased from 0.49% (CPANI/CF-600) to 1.41% (CPANI/CF-700) (Fig. 3(c)), and these changes could be attributed to the transformation of pyridine N into more heat-resistant graphitic N and nitric oxide during calcination.^15,33 Studies have demonstrated that the high energy levels of graphitic N and pyridine N have a significant impact on boosting the catalytic activity of electrode materials.³⁴

Fig. 3 (a) XPS surveys of CF, PANI/CF and CPANI/CF, (b) N 1s spectra of PANI/CF, (c) contents of three nitrogen species, (d) N 1s spectra of CPANI/CF-600, (e) N 1s spectra of CPANI/CF-700 and (f) N 1s spectra of CPANI/CF-800.

Table 1 Chemical compositions of the 3D electrodes

Samples	C%	N%	O%	N nitrogen species/Atomtotal (%)
				Pyridinic N	Pyrrolic N	Graphitic N
CF	84.34	—	15.65	—	—	—
PANI/CF	69.27	9.44	21.31	—	—	—
CPANI/CF-600	82.29	7.79	4.92	2.02	4.74	1.02
CPANI/CF-700	89.38	5.97	4.66	1.85	2.51	1.61
CPANI/CF-800	91.96	5.03	3.01	1.61	1.94	1.49

---

The N₂ adsorption–desorption isothermal curves were obtained through BET analysis, and the BJH method was employed to examine the distribution patterns of pore size. The pore size distribution curves demonstrated that CPANI/CF electrodes have a mesoporous structure, and the average pore size ranges from 3.408 to 3.412 nm in Fig. 4(a). The specific surface area increased from 35.194 m² g⁻¹ (CPANI/CF-600) to 39.753 m² g⁻¹ (CPANI/CF-700) with the increase of carbonation temperature. The specific surface area of the CPANI/CF electrode gradually decreased when the carbonisation temperature continued to increase, and the specific surface area of CPANI/CF-800 was only 27.506 m² g⁻¹ (Fig. 4(b)). The reason for this may be that the material architecture collapses when the carbonisation temperature is too high, leading to partial build-up of the material, and the specific surface area no longer increases. Similarly, the pore volume of CPANI/CF-600, CPANI/CF-700 and CPANI/CF-800 electrodes were 0.027 cm³ g⁻¹, 0.032 cm³ g⁻¹ and 0.017 cm³ g⁻¹ (Table 2). The larger specific surface area and pore volume of the CPANI/CF-700 electrode provided more active sites for the degradation of TC,³⁵ which improved the electrocatalytic performance of the 3D electrode.

Fig. 4 (a) Pore size distributions of CPANI/CF-t, (b) N₂ adsorption–desorption isotherms, (c) XRD patterns and (d) FTIR spectra of CF, PANI/CF and CPANI/CF.

Table 2 Structure properties of CPANI/CF-T electrodes

Samples	Surface area (m^2 g^−1)	Pore volume (cm^3 g^−1)	Pore diameter (nm)
CPANI/CF-600	35.194	0.027	3.412
CPANI/CF-700	39.753	0.032	3.409
CPANI/CF-800	27.506	0.017	3.408

---

The CF, PANI/CF and CPANI/CF electrode materials were characterised using XRD. In Fig. 4(c), PANI/CF has three sharp crystalline planes at 15°, 20.5° and 25.1°, which are attributed to the (011), (020) and (200) crystalline planes of polyaniline, indicating the successful synthesis of polyaniline. CF and CPANI/CF have three sharp crystalline planes at 25° and 44.1° only, which can be attributed to the reflections of graphite planes (002) and (100). In the CPANI/CF samples, the crystal planes (011), (020) and (200) of PANI disappeared. These results indicate the successful synthesis of CPANI/CF by in situ polymerisation and high temperature carbonation, and the results of XRD and XPS analyses show that CPANI/CF electrode materials have great purity and crystallinity.

The functional groups on the electrode surface were analysed using FTIR with IR spectra ranging from 400 cm⁻¹ to 4000 cm⁻¹ in Fig. 4(d). The characteristic peaks of PANI/CF appeared at 1579 cm⁻¹ (quinonoid (Q) ring stretching), 1481 cm⁻¹ (benzenoid (B) ring stretching), 1300 cm⁻¹ (CN– stretching), and 1232 cm⁻¹ (CN⁺– stretching), and the peaks at 1135 cm⁻¹ and 798 cm⁻¹ may be related to the aromatic ring C–H.³⁶ It is noteworthy that the characteristic peaks of polyaniline disappeared after carbonation of PANI/CF, which suggests that carbonation can break the chemical bond of polyaniline and form a new structure, and only two characteristic peaks are present in CPANI/CF, which appeared at 1537 cm⁻¹ (ring-stretching in conjugated heterocyclic aromatic) and 1256 cm⁻¹ (C–N stretching vibration in conjugated heterocyclic aromatic), which indicates the successful preparation of N-doped carbon in the CPANI/CF electrode.³⁷

3.2 Performance of different processes

3.2.1 Comparison of TC removal in various processes. As shown in Fig. 5(a), CF and CPANI/CF electrodes do not directly oxidatively degrade pollutants and have a poor adsorption capacity for pollutants, 3.11% and 8.21%, respectively. PMS is an oxidizing agent and reacts slowly with pollutants, degrading 26.19% of TC in 40 min at pH = 3, which may be due to the activation of a part of PMS by heat, light and other conditions.³⁸ According to the PMS–CF system analysis, it is known that the TC removal rate was 28.61%, which implies that CF alone is unable to activate PMS. In the PMS–CPANI/CF system, the TC degradation rate reaches 63.82%, which increased the TC removal rate by 2.43 and 7.77 folds compared with that of PMS alone and CPANI/CF alone, respectively, indicating that CPANI/CF could successfully activate PMS. There was a reciprocal interaction between CPANI/CF and PMS in the removal of TC.

Fig. 5 (a) The removal of TC, (b) TC removal under different systems, (c) influence of reaction kinetics and (d) the current and energy consumption of different systems. Reaction conditions: ([TC] = 50 mg L⁻¹, applied voltage = 5 V, initial pH = 6, [PMS] = 3 mmol L⁻¹ and [Na₂SO₄] = 0.05 mol L⁻¹).

In Fig. 5(b), the removal rate (C/C₀) of TC in various systems is illustrated. The removal of TC by E, E–CF and E–CPANI/CF systems was 41.91%, 47.97% and 45.45%, respectively, at pH = 3, voltage of 5 V, plate spacing of 6 cm, 0.05 M Na₂SO₄, and 3 mM PMS. This indicates that TC can be removed through direct and indirect oxidation in the E system. The introduction of CF and CPANI/CF as three-dimensional electrodes can shorten the cathode–anode distance, increase the current density, and improve the mass transfer effect. The removal rates of TC by the E–PMS, E–PMS–CF and E–PMS–CPANI/CF systems are 64.24%, 69.33% and 90.59% respectively, indicating that PMS can be electrically activated. Furthermore, the degradation rate of E–PMS–CPANI/CF for TC was increased by 48.64% and 26.35% respectively, compared with that of the traditional E and E–PMS. As shown in Fig. 5(c), the removal rates of TC by E, c, E–PMS, E–CPANI/CF, E–PMS–CF and E–PMS–CPANI/CF processes within the first 10 min conformed to first-order kinetics with k_obs of 0.0094 min⁻¹, 0.0146 min⁻¹, 0.0274 min⁻¹, 0.0279 min⁻¹, 0.0667 min⁻¹ and 0.1345 min⁻¹. The k_obs of the E–PMS–CPANI/CF system are 10.21 and 6.05 times higher than that of E and E–PMS respectively, indicating that the E–PMS–CPANI/CF system has significant advantages in pollutant degradation.

Combined with Fig. 5(d), it is evident that adding CF, CPANI/CF and PMS to the E system increased the current in the system to different degrees. This indicates that PMS not only plays the role of oxidising pollutants, but also acts as an electrolyte. The participation of CF and CPANI/CF electrodes shortened the cathode–anode distance and improved the electrocatalytic performance of the system.³⁹ The energy consumptions of the catalytic systems of E, E–CF, E–CPANI/CF, E–PMS, E–PMS–CF and E–PMS–CPANI/CF were 5.17, 4.79, 3.88, 3.16, 3.04 and 1.54 kW h m⁻³, respectively. By introducing the CPANI/CF electrode, the energy consumption of the system decreased and the catalytic effect improved, demonstrating the superiority of the E–PMS–CPANI/CF system for removing TC.

3.2.2 Influence of the operational parameters. We investigated the carbonation temperature of the CPANI/CF electrode in Fig. 6(a), and the removal rates of PANI/CF, CPANI/CF-600, CPANI/CF-700 and CPANI/CF-800 electrodes of the TC within 40 min were 73.45%, 77.34%, 90.59% and 84.75%, respectively. We can see that the removal of TC from the system increases significantly as the electrode calcination temperature increases from 0 °C to 700 °C accordingly. However, when the temperature is increased to 800 °C, this trend is reversed and the removal of TC by the system decreases by 5.84%. The CPANI/CF-700 electrode has the best ability to degrade TC, which may be due to the following two reasons: (1) the CPANI/CF-700 electrode has a high content of graphitic N. Graphitic N has a high adsorption and activation capacity for PMS, and it is also considered to be the most oxygen-reduction reaction (ORR) active N substance.⁴⁰ (2) The CPANI/CF-700 electrode has a higher specific surface area and pore volume with more active sites, which is favourable for TC adsorption and PMS activation.

Fig. 6 (a) The influence of carbonation temperature, (b) the influence of the initial pH, (c) the influence of voltage and (d) the effect of the amount of PMS added. Reaction conditions: ([TC] = 50 mg L⁻¹, applied voltage = 5 V, initial pH = 6, [PMS] = 3 mmol L⁻¹ and [Na₂SO₄] = 0.05 mol L⁻¹).

Typically, real wastewater is alkaline or acidic, therefore, it is important to develop an electrode material that is resistant to acidity or alkalinity.³⁹ As shown in Fig. 6(b), we investigated the effect of the E–PMS–CPANI/CF process on TC degradation at initial pH values of 3.0, 5.0, 7.0, 9.0, and 11.0, respectively. Satisfactorily, the removal rate of TC in the E–PMS–CPANI/CF system was 90.59%, 88.02% and 85.52% within 40 min of the system when the pH in the system was 3.0, 5.0 and 7.0, respectively. The results indicated that the E–PMS–CPANI/CF system was effective in removing TC from water under both acidic and neutral conditions. As the pH levels increased to 11.0, the removal rate of TC by the E–PMS–CPANI/CF system gradually decreased to 77.46%. Hydroxyl groups in alkaline solutions can break down PMS into SO₄²⁻, reducing ROS levels in the solution.⁴¹ An initial pH of 3.0 was selected for the subsequent batch experiments based on the results above.

When undergoing electrochemical oxidation, the strength of the applied voltage plays a significant role. In Fig. 6(c), it is demonstrated that after 40 minutes without a DC power supply, the E–PMS–CPANI/CF system achieved a 63.82% removal of TC. This result indicates that the CPANI/CF electrodes can activate PMS independently. Whereas when the voltage was 3 V, 5 V, 7 V and 9 V, the removal rates of TC were 77.93%, 90.59%, 83.22% and 80.22%. When the intensity of the applied voltage in the system is lower than 5 V, the bonding energy generated by the TC chemical bond breaking may not be reached.⁴² The removal rate of TC decreases significantly when the applied voltage is greater than 5 V. Higher current densities may lead to increased short-circuit and bypass currents, and side reactions such as the evolution reaction of the anode and hydrogen evolution reaction of the cathode may be intensified, inhibiting the removal of TC.⁴³ Given the above results, the applied voltage of 5 V was selected for the subsequent batch experiments.

The concentration of PMS is also significant in the electrochemical oxidation system. In Fig. 6(d), when the system did not have PMS added, the 3D-E system only removed 56.45% of TC. When 1 mmol L⁻¹, 2 mmol L⁻¹, 3 mmol L⁻¹, 4 mmol L⁻¹ and 5 mmol L⁻¹ of PMS were added to the E–PMS–CPANI system, the removal rates of TC by the process were 78.75%, 83.83%, 90.59%, 90.81% and 90.91%, respectively. The degradation rate of TC gradually increased with higher PMS concentrations, however, there was no significant change in the degradation effect of TC when the PMS concentration increased to 5 mmol L⁻¹. There are two reasons for this phenomenon: (1) with a moderate increase in PMS concentration, more ROS will be present in the 3D electrocatalytic system, which is beneficial to pollutant degradation.⁴⁴ (2) When the concentration of PMS is too high, it will consume ˙OH and SO₄˙⁻, resulting in a decrease in active substances.^45,46 For comprehensive consideration, the PMS dosage was selected to be 3 mmol L⁻¹.

3.2.3 The recyclability and stability of the CPANI/CF electrode. The stability and reusability of 3D electrodes are crucial in practical applications. Five consecutive cycle experiments were conducted using the same electrode (CPANI/CF) to evaluate its stability and reusability. From Fig. 6(e), it is shown that the removal of TC by the CPANI/CF electrode decreased from 90.59% (for the first use) to 86.78% (for the second use), 84.56% (for the third use), 82.19% (for the fourth use) and 80.25% (for the fifth use). As the usage increases, the removal of TC by the 3D electrode gradually decreases. This is probably because TC and intermediates are adsorbed onto the electrode's surface, taking up a small portion of its active sites. However, after five reaction cycles, the removal rate of TC by the 3D electrode could still reach 80.24%, which proved the stability and reusability of the CPANI/CF electrode.

3.2.4 Removal efficiency of other contaminants. To explore the versatility of the E–PMS–CPANI/CF system in industrial wastewater treatment, we investigated the removal of NOR (norfloxacin), PNP (p-nitrophenol), MO (methyl orange) and MB (methylene blue) by this system under the same conditions. In Fig. 6(f), it showed that the E–PMS–CPANI/CF process exhibited better removal of MB, MO, NOR and PNP of 96.57%, 97.36%, 90.88% and 85.89%, respectively, which proved that the E–PMS–CPANI/CF system could effectively treat a wide range of industrial wastewaters by electro-adsorption, radical and non-radical modes. It is applicable to the removal of different difficult-to-degrade pollutants.

3.3 Reaction mechanism

3.3.1 Identifying the reactive oxidizing species (ROS). In this section, ˙OH, SO₄˙⁻ and ¹O₂ were qualitatively and quantitatively analysed using electron spin resonance (ESR), free radical quenching experiments and chemical probe experiments.

Possible ROS in the E–PMS–CPANI/CF system were identified using an electron spin resonance (ESR) test. Using DMPO as ˙OH and SO₄˙⁻ radical trapping agents, in Fig. 7(a), the ESR spectrum of the E–PMS–CPANI/CF system showed a DMPO–˙OH four-peak signals (1 : 2 :2 : 1) and A DMPO–SO₄˙⁻ six-peak signals (1 : 1 : 1 : 1 : 1 : 1 : 1 : 1),⁴⁷ and the corresponding signals also appeared in the pairs of E–PMS–CF system and PMS–CPANI/CF system, but the intensities of the characteristic peaks were significantly lower than the E–PMS–CPANI/CF system. In contrast, no significant signal was observed in the PMS system. The results confirm the presence of ˙OH and SO₄˙⁻ in E–PMS–CPANI/CF, E–PMS and PMS–CPANI/CF systems, and the near absence of ˙OH and SO₄˙⁻ in the PMS system. Using TEMP as the ¹O₂ radical trapping agent, in Fig. 7(b), the characteristic three-peak signal (1 : 1 : 1) of TEMP–¹O₂ appeared in the ESR spectrum of the E–PMS–CPANI/CF system,⁴⁸ and the characteristic peak intensity signal was significantly higher in the system compared to the other systems, which indicated that a large amount of ¹O₂ existed in the E–PMS–CPANI/CF process.

Fig. 7 (a) ESR spectra of DMPO–˙OH in different processes and (b) ESR spectra of TEMP–¹O₂ in different processes. Reaction conditions: ([TC] = 50 mg L⁻¹, applied voltage = 5 V, initial pH = 6, [PMS] = 3 mmol L⁻¹ and [Na₂SO₄] = 0.05 mol L⁻¹).

In Fig. 8(a), the impact of ROS on the E–PMS–CPANI/CF system was further investigated through free radical quenching experiments. tert-Butanol (TBA, , k_˙OH = 6.0 × 10⁸ M⁻¹ s⁻¹), ethanol (EtOH, , k_˙OH = (1.6–7.7) × 10⁷ M⁻¹ s⁻¹) and furfuryl alcohol (FFA, k_¹O2 = 1.2 × 10⁸ M⁻¹ s⁻¹, k_˙OH = 1.5 × 10¹⁰ M⁻¹ s⁻¹) were added to the E–PMS–CPANI/CF system to determine the impact of the ROS (˙OH, SO₄˙⁻ and ¹O₂) in the E–PMS–CPANI/CF system on the TC removal rate, and phenol was added to determine the production of active substances on the electrode surface. In Fig. 8(a), the E–PMS–CPANI/CF process removed 90.59% of TC within 40 min without adding any quencher, which decreased to 76.99%, 71.95% and 25.75% with the addition of TBA, EtOH and FFA quenchers. The detected free radicals were further analysed by probe experiments. As in Fig. 8(b), the ˙OH in the system was indirectly detected using the production of hydroxycoumarin, the reaction product of coumarin with ˙OH, and SO₄˙⁻ in the system was indirectly detected using the production of p-benzoquinone (p-BQ), the main by-product of the reaction of p-hydroxybenzoic acid (p-HBA) with SO₄˙⁻. As shown in Fig. 8(c), the concentration of ¹O₂ was calculated from the degradation rate of DPBF using 1,3-diphenylisobenzofuran (DPBF) as a chemical probe. The contributions of ˙OH, SO₄˙⁻ and ¹O₂ to the degradation of TC by the E–PMS–CPANI/CF system were 33.14%, 12.90% and 46.04%, respectively, with ¹O₂ playing a dominant role in the degradation process. After adding phenol to the system, the removal of TC by the system decreased to 46.35%, indicating that most of the reaction occurred on the electrode surface.

Fig. 8 (a) TC removal efficiencies with the addition of different scavengers in the E–PMS–CPANI/CF system, (b) variation of the concentrations of ˙OH and SO₄˙⁻ during the reaction of the E–PMS–CPANI/CF system, and (c) the concentrations of ¹O₂ during the reaction of the E–PMS–CPANI/CF system.

3.3.2 Possible degradation mechanism. Based on the above ESR tests, probe experiments and free radical quenching experiments, we conclude that a large amount of ˙OH, SO₄˙⁻ and ¹O₂ exists in the E–PMS–CPANI/CF system. In this section, we systematically analyzed the pathways responsible for generating different types of ROS. In Fig. 9, for the generation of ROS in the E–PMS–CPANI/CF system, we summarised three pathways. First, the H₂O molecules on the anode surface are converted to ˙OH by the electric field (e.g., the conventional E system). Second, by combining Fig. 5(b) and the ESR spectra of the E–PMS system, we can conclude that PMS can be activated to generate ˙OH and SO₄˙⁻ under the action of the electric field. Third, CPANI/CF as a 3D electrode has a high content of graphitic N and pyridine N. It has been reported that the presence of graphitic N in the electrocatalytic system increases the production of H₂O₂, which subsequently decomposes to form ˙OH in the presence of pyridine N.^48–50 Graphitic nitrogen has a high electron density and a stronger affinity for PMS,^51,52 which allows it to extract electrons from neighbouring C, thus producing electron-deficient C, which is more susceptible to addition by nucleophilic PMS to produce large amounts of ¹O₂.⁵³ Pyrrole N as an adsorption site can shorten the reaction distance between the target pollutant and ¹O₂, thus effectively degrading TC.⁵³ In addition, a certain number of electrons are transferred to PMS via pyridine nitrogen and graphite nitrogen, leading to the breaking of the O–O bond in PMS and the production of ˙OH and SO₄˙⁻.⁵⁴ The ESR tests of the E–PMS–CPANI/CF process also provided evidence of the existence of such pathways. As a result, we can deduce that the E–PMS–CPANI/CF system degrades TC mainly through radical and non-radical pathways.

Fig. 9 Illustration of the oxidation mechanism involved in the E–PMS–CPANI/CF process.

4. Conclusion

In summarize, a nitrogen-doped three-dimensional electrode was synthesised using low-cost polyaniline-loaded carbon mats as precursors. The SEM showed that the polyaniline was successfully loaded on the surface of the carbon felt without the addition of binder, and the consistent results of XPS, XRD and FTIR proved the successful preparation of the nitrogen-doped carbon electrode, and the BET analysis showed that CPANI/CF-700 had a large specific surface area and pore volume. Compared with other systems, the E–PMS–CPANI/CF system has excellent removal efficiency and lower energy consumption for the pollutants, and the degradation mechanism was investigated and found that the ROS (˙OH, SO₄˙⁻ and ¹O₂) generated during the electrocatalytic process is a crucial factor in the removal of the pollutants, in which ¹O₂ plays a dominant role. The E–PMS–CPANI/CF system also showed excellent degradation of typical industrial wastewaters, such as NOR (norfloxacin), PNP (p-nitrophenol), MO (methyl orange) and MB (methylene blue), with certain stability and recyclability, which is of practical significance in wastewater treatment.

Conflicts of interest

There are no conflicts to declare.

References

1. S. Leong, D. Li, K. Hapgood, X. W. Zhang and H. T. Wang, Appl. Catal., B, 2016, 198, 224–233.
2. J. L. Xie, L. Chen, X. Luo, L. Huang, S. Y. Li and X. B. Gong, Sep. Purif. Technol., 2022, 281, 12.
3. R. Xu, J. J. Xiong, D. Liu, Y. R. Wang and Y. A. Ming, J. Colloid Interface Sci., 2022, 625, 397–404.
4. D. Y. Chen, Q. Bai, T. T. Ma, X. F. Jing, Y. Y. Tian, R. Zhao and G. S. Zhu, Chem. Eng. J., 2022, 435, 11.
5. B. Shen, X. H. Wen and X. Huang, Chem. Eng. J., 2017, 327, 597–607.
6. K. Zhao and Y. G. Zhang, Sep. Purif. Technol., 2023, 308, 11.
7. H. V. Lutze, S. Bircher, I. Rapp, N. Kerlin, R. Bakkour, M. Geisler, C. von Sonntag and T. C. Schmidt, Environ. Sci. Technol., 2015, 49, 1673–1680.
8. S. S. Zhu, X. C. Huang, F. Ma, L. Wang, X. G. Duan and S. B. Wang, Environ. Sci. Technol., 2018, 52, 8649–8658.
9. J. L. Wang and S. Z. Wang, Chem. Eng. J., 2018, 334, 1502–1517.
10. C. D. Qi, X. T. Liu, J. Ma, C. Y. Lin, X. W. Li and H. J. Zhang, Chemosphere, 2016, 151, 280–288.
11. X. Cheng, H. G. Guo, Y. L. Zhang, G. V. Korshin and B. Yang, Water Res., 2019, 157, 406–414.
12. M. X. Cheng, Y. C. Zhang, B. Lai, L. Z. Wang, S. J. Yang, K. L. Li, D. Q. Wang, Y. G. Wu, G. H. Chen and J. Qian, Chem. Eng. J., 2023, 455, 17.
13. D. K. Yoo, H. J. An, N. A. Khan, G. T. Hwang and S. H. Jhung, Chem. Eng. J., 2018, 352, 71–78.
14. R. Silva, D. Voiry, M. Chhowalla and T. Asefa, J. Am. Chem. Soc., 2013, 135, 7823–7826.
15. P. D. Hu, H. R. Su, Z. Y. Chen, C. Y. Yu, Q. L. Li, B. X. Zhou, P. J. J. Alvarez and M. C. Long, Environ. Sci. Technol., 2017, 51, 11288–11296.
16. X. G. Duan, Z. M. Ao, D. G. Li, H. Q. Sun, L. Zhou, A. Suvorova, M. Saunders, G. X. Wang and S. B. Wang, Carbon, 2016, 103, 404–411.
17. Y. W. Gao, T. Li, Y. Zhu, Z. H. Chen, J. Y. Liang, Q. Y. Zeng, L. Lyu and C. Hu, J. Hazard. Mater., 2020, 393, 10.
18. W. C. Peng, S. Z. Liu, H. Q. Sun, Y. J. Yao, L. J. Zhi and S. B. Wang, J. Mater. Chem. A, 2013, 1, 5854–5859.
19. T. G. Ma, M. J. Liu, T. T. Li, H. J. Ren and R. Zhou, Sep. Purif. Technol., 2022, 293, 11.
20. W. Zhou, L. Rajic, Y. W. Zhao, J. H. Gao, Y. K. Qin and A. N. Alshawabkeh, Electrochim. Acta, 2018, 277, 185–196.
21. D. T. Yue, X. F. Qian, M. Kan, M. Y. Fang, J. P. Jia, X. D. Yang and Y. X. Zhao, Appl. Catal., B, 2018, 229, 211–217.
22. N. Barhoumi, H. Olvera-Vargas, N. Oturan, D. Huguenot, A. Gadri, S. Ammar, E. Brillas and M. A. Oturan, Appl. Catal., B, 2017, 209, 637–647.
23. J. Y. Feng and Y. G. Zhang, New J. Chem., 2020, 44, 16584–16593.
24. Y. M. Zhang, Z. Chen, P. P. Wu, Y. X. Duan, L. C. Zhou, Y. X. Lai, F. Wang and S. Li, J. Hazard. Mater., 2020, 393, 10.
25. X. F. Huang and Y. G. Zhang, J. Iran. Chem. Soc., 2021, 18, 1913–1925.
26. H. Z. Wu, Z. Z. Hu, R. H. Liang, O. V. Nkwachukwu, O. A. Arotiba and M. H. Zhou, Appl. Catal., B, 2023, 321, 13.
27. J. J. Cai, M. H. Zhou, X. D. Du and X. Xu, Sep. Purif. Technol., 2021, 254, 11.
28. M. X. Yang, Z. X. Hou, X. Zhang, B. Y. Gao, Y. W. Li, Y. A. Shang, Q. Y. Yue, X. G. Duan and X. Xu, Environ. Sci. Technol., 2022, 11,  DOI:10.1021/acs.est.2c01261.
29. Z. Tong, Y. Yang, J. Wang, J. Zhao, B. L. Su and Y. Li, J. Mater. Chem. A, 2014, 2, 4642–4651.
30. H. M. Zhang, Q. Zhao, S. P. Zhou, N. J. Liu, X. H. Wang, J. Li and F. S. Wang, J. Power Sources, 2011, 196, 10484–10489.
31. X. G. Duan, H. Q. Sun, Y. X. Wang, J. Kang and S. B. Wang, ACS Catal., 2015, 5, 553–559.
32. G. L. Sun, L. Y. Ma, J. B. Ran, B. Li, X. Y. Shen and H. Tong, Electrochim. Acta, 2016, 194, 168–178.
33. X. L. Li, H. L. Wang, J. T. Robinson, H. Sanchez, G. Diankov and H. J. Dai, J. Am. Chem. Soc., 2009, 131, 15939–15944.
34. Y. Wang, Y. Lin, C. P. Yang, S. H. Wu, X. T. Fu and X. Li, Chem. Eng. J., 2023, 451 DOI:10.1016/j.cej.2022.138468.
35. S. Dutta, A. Bhaumik and K. C. W. Wu, Energy Environ. Sci., 2014, 7, 3574–3592.
36. J. B. Yin, X. A. Xia, L. Q. Xiang and X. P. Zhao, Carbon, 2011, 49, 739.
37. C. Z. Kuang, G. S. Zeng, Y. J. Zhou, Y. Y. Wu, D. X. Li, Y. F. Wang and C. H. Li, Water Res., 2023, 229, 11.
38. Z. H. Chen, L. B. Yang, X. W. Luo, D. L. Xie, C. Zhao, R. L. Qiu and Z. J. Huang, Sep. Purif. Technol., 2022, 296, 11.
39. K. Zhao and Y. G. Zhang, Sep. Purif. Technol., 2023, 308 DOI:10.1016/j.seppur.2022.122962.
40. W. Ren, L. L. Xiong, G. Nie, H. Zhang, X. G. Duan and S. B. Wang, Environ. Sci. Technol., 2020, 54, 1267–1275.
41. D. H. Ding, S. J. Yang, L. W. Chen and T. M. Cai, Chem. Eng. J., 2020, 392, 11.
42. C. Zhang, F. Li, R. B. Wen, H. K. Zhang, P. Elumalai, Q. Zheng, H. Y. Chen, Y. J. Yang, M. Z. Huang and G. G. Ying, Water Res., 2020, 182, 12.
43. K. Zhao and Y. G. Zhang, J. Chem. Technol. Biotechnol., 2023, 98, 940–948.
44. Y. H. Cui, X. D. Lv, J. X. Lei and Z. Q. Liu, Electrochim. Acta, 2017, 245, 193–202.
45. Y. J. Wang and G. Yu, J. Hazard. Mater., 2022, 436, 18.
46. L. Xu, B. R. Fu, Y. Sun, P. K. Jin, X. Bai, X. Jin, X. Shi, Y. Wang and S. T. Nie, Chem. Eng. J., 2020, 400, 12.
47. F. R. Guo, K. J. Wang, J. H. Lu, J. C. Chen, X. W. Dong, D. S. Xia, A. Q. Zhang and Q. Wang, Chemosphere, 2019, 218, 1071–1081.
48. Y. Wang, Y. H. Liu, K. Wang, S. Q. Song, P. Tsiakaras and H. Liu, Appl. Catal., B, 2015, 165, 360–368.
49. S. J. Jiang, Z. Li, H. Y. Wang, Y. Wang, L. N. Meng and S. Q. Song, Nanoscale, 2014, 6, 14262–14269.
50. P. Su, M. H. Zhou, X. Y. Lu, W. L. Yang, G. B. Ren and J. J. Cai, Appl. Catal., B, 2019, 245, 583–595.
51. P. Liang, C. Zhang, X. G. Duan, H. Q. Sun, S. M. Liu, M. O. Tade and S. B. Wang, Environ. Sci.: Nano, 2017, 4, 315–324.
52. W. J. Ma, N. Wang, Y. A. Fan, T. Z. Tong, X. J. Han and Y. C. Du, Chem. Eng. J., 2018, 336, 721–731.
53. S. H. Wu, C. P. Yang, Y. Lin and J. J. Cheng, J. Environ. Sci., 2022, 115, 330–340.
54. X. Chen, W. D. Oh, Z. T. Hu, Y. M. Sun, R. D. Webster, S. Z. Li and T. T. Lim, Appl. Catal., B, 2018, 225, 243–257.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3nj04311h

---