Integrated system of electrocoagulation and electrooxidation for industrial indigo wastewater: Performance improvement and catalytic mechanism

2.1. Materials

Materials used in this study included: Al electrodes (4.0 cm × 4.0 cm × 0.1 cm, Zhoukou Shouheng Trading Co., Ltd.); Ru-Ir/Ti mesh electrodes (4.0 cm × 4.0 cm × 0.2 cm, Xingtai Bafang Metal Products Co., Ltd.); indigo dyeing wastewater (Table 1) showed the indicators for waste water, provided by Hebei Xindadong Textile Co., Ltd.); graphite felt (4.0 cm × 4.0 cm × 0.2 cm, Qinghe Xingye Metal Materials Co., Ltd.); tert-butyl alcohol (TBA), 1,4-benzoquinone (PBQ), Na₂SO₄, Na₂S₂O₄, NaOH (Tianjin Damao Chemical Reagent Co., Ltd.); acetone and sulfuric acid (Tianjin Kemiou Chemical Reagent Co., Ltd.); isopropanol (IPA), melamine (MEL); carbon powder, Nafion solution, hydrogen peroxide (H₂O₂), furfuryl alcohol (FA) (Shanghai Aladdin Biochemical Technology Co., Ltd.); potassium titanium oxalate (C₄H₂K₂O₁₀Ti, Shandong Xiya Chemical Co., Ltd.); and COD detection reagent kits (Henan Suijing Environmental Protection Technology Co., Ltd.).

2.2. Methods

2.3. Material characterization and performance testing

The surface morphology of the nitrogen-doped carbon cathodes was examined at 500× magnification using a MIRA LMS scanning electron microscope (SEM). The pore size distribution, pore volume, and specific surface area of the materials were analyzed using an ASAP 2460 fully automated surface area and porosity analyzer. Elemental composition and chemical states of surface species were characterized by X-ray photoelectron spectroscopy (XPS, K-Alpha), with a scanning range of 0–1200 eV. Elemental mapping and quantification of C, N, O, and F on the cathode surface were performed using an X-ray energy-dispersive spectrometer (EDS, Xplore 30). Raman spectra were collected with a 514 nm laser excitation source over a spectral range of 500–4000 cm⁻¹. Fourier-transform infrared spectroscopy (FTIR) was conducted using an ALPHA II spectrometer to analyze surface functional groups of the N-doped carbon cathode. The spectra were recorded in the 500–4000 cm⁻¹ range with a resolution of 4 cm⁻¹ and averaged over 64 scans. The morphology of flocs was observed at 40× magnification using an EX21 biological microscope. The Zeta potential of the flocs was measured using a JS94K micro-electrophoresis analyzer. Thermal characteristics of the flocs were analyzed by thermogravimetric-differential thermal analysis (TG-DTA) using a Rigaku TG/DTA 8122 instrument. Electrochemical performance of the anode was evaluated using a conventional three-electrode system, where the working electrode was an aluminum electrode, the counter electrode was a graphite electrode, and the reference electrode was a saturated calomel electrode (SCE). The electrolyte was a 0.5 mol/L Na₂SO_₄ solution. Electrochemical impedance spectroscopy (EIS) was conducted under open-circuit potential with a frequency range of 10⁻¹ to 10⁵ Hz and a sinusoidal amplitude of 10 mV. Tafel polarization curves were measured at a scan rate of 0.5 mV/s within a potential range of −1.6 to 0 V. ORR selectivity was evaluated using a rotating ring-disk electrode (RRDE) setup. The working, counter, and reference electrodes were a glassy carbon disk, a platinum sheet, and an Ag/AgCl electrode, respectively. The ring potential was held at 1.2 V (vs. RHE), with a rotation speed of 1600 rpm and a scan rate of 10 mV/s. The electron transfer number (n) and hydrogen peroxide selectivity (%H₂O₂) were calculated using Equations (1) and (2).

In the above equations, I_d represents the disk current, I_r is the ring current, and N (where N = 0.37) denotes the collection efficiency of the Pt ring electrode.

2.4. Wastewater treatment indicators

The removal efficiencies of COD and TOC were calculated using Equations (3) and (4), respectively:

In these equations, C₀ and C₁ refer to the COD values of the raw and treated wastewater, respectively; T₀ represents the TOC content of the raw wastewater, and T₁ corresponds to that of the treated effluent.

The decolorization efficiency of the indigo wastewater was determined using a UV–Vis spectrophotometer. The maximum absorption wavelength for indigo dye was identified at 670 nm, and the decolorization rate was calculated using Equation (5):

Here, B₀ and B₁ represent the absorbance of the raw and treated wastewater at 670 nm, respectively.

The specific energy consumption for COD removal and the operating cost of the system were calculated using Equations (6) and (7):

In these expressions, U is the applied voltage (V), I is the current (A), t is the reaction time (h), V is the treated wastewater volume (L), and ΔCOD denotes the change in COD concentration before and after treatment (g COD/L).

3.1. Characterization of Nx-GF materials

The surface morphology of the N-C cathode is shown in Figure 2a. A large number of particulate solids were observed adhered to the graphite felt surface, forming a porous structure that enhances the specific surface area and provides more active sites for the electrochemical reactions. As illustrated in Figure 2b-2e, the N0.1-GF sample contains C, N, O, and F elements, with nitrogen atoms uniformly distributed throughout the carbon matrix. The energy-dispersive spectroscopy (EDS) results in Figure 2(f) indicate that the contents of nitrogen and fluorine are 19.24% and 16.94%, respectively, where the presence of fluorine originates from the Nafion solution used in the electrode preparation. The N₂ adsorption–desorption isotherms of the Nx-GF materials are presented in Figure 2(g). According to the IUPAC classification, all samples exhibit Type IV isotherms with distinct H3 hysteresis loops and without apparent saturation plateaus, suggesting the presence of mesoporous structures [19]. Among them, the N_0.1-GF electrode shows the highest adsorption capacity and the most prominent hysteresis loop. Combined with the pore size distribution curve shown in Figure 2(h), these results suggest that nitrogen doping effectively promotes the development of the porous structure.

Close

Figure 2.

(a) SEM image of the nitrogen-doped carbon cathode at 500× magnification; (b–e) elemental mapping images of C, N, O, and F; (f) EDS spectrum; (g) N₂ adsorption–desorption isotherms; (h) pore size distribution curves.

Export to PPT

Figure 3 illustrates the structural changes in the carbon framework after nitrogen doping. The survey XPS spectrum of N_0.1-GF (Figure 3a) confirms the presence of carbon (C), nitrogen (N), and oxygen (O) on the cathode surface. The high-resolution C1s spectrum (Figure 3b) reveals four distinct carbon bonding environments: C–C/C=C (284.8 eV), C–O/C=N (285.4 eV), C–N (286.4 eV), and C=O (289.7 eV). The dominant peak at 284.8 eV indicates that sp² and sp³ hybridized carbon species are the major forms present in the structure [20]. The high-resolution N1s spectrum and nitrogen species distribution (Figures 3c-d) show that the N-doped carbon cathode surface contains pyridinic N (399.4 eV), pyrrolic N (400.3 eV), and graphitic N (401.3 eV) [21]. With increasing melamine dosage, the proportions of pyridinic and graphitic nitrogen increase, reaching their highest levels at a carbon-to-nitrogen mass ratio of 1:0.1. Both species are known to enhance the ORR. Figure 3(e) presents the FTIR spectra of Nx-GF. Compared with N0-GF, the N-doped cathodes exhibit enhanced intensities at 3420 cm^-1 (N–H stretching), 1540 cm^-1 (C=N stretching), and 1080 cm^-1 (C–N stretching) [22], confirming the incorporation of N–H functional groups into the carbon framework via C–N and C=N bonding configurations [23]. In the Raman spectra (Figure 3f), the peaks at 1355 cm⁻¹ and 1566 cm⁻¹ correspond to the D and G bands, respectively. The D band signifies the presence of structural defects in the carbon material, while the G band arises from the in-plane vibration of sp²-hybridized carbon atoms. Notably, the N_0.1-GF sample exhibits the highest I_D/I_G ratio of 1.6, indicating the highest defect density among all tested samples [24]

Close

Figure 3.

(a) XPS survey spectrum of the cathode; (b) high-resolution C1s spectrum; (c) high-resolution N1s spectrum; (d) distribution of nitrogen species; (e) FTIR spectra; (f) Raman spectra.

Export to PPT

3.2. Electrocatalytic performance of Nx-GF

To evaluate the catalytic performance of Nx-GF under different conditions, the concentration of H₂O₂ was measured using the potassium titanium oxalate spectrophotometric method at 400 nm [25]. As shown in Figure 4(a), the N_0.1-GF electrode produced the highest H₂O₂ concentration (15.67 mg L^-1) after 60 min of reaction. Since nitrogen atoms have higher electronegativity than carbon, they can act as electron donors and promote two-electron transfer. Moderate nitrogen doping enhances catalytic activity, whereas excessive doping impairs the material’s conductivity and hinders electron transfer [26], resulting in a decline in H₂O₂ production with further increases in nitrogen content. Figure 4(b) demonstrates that acidic conditions favor H₂O₂ generation, as the abundance of H⁺ facilitates the conversion of *OOH intermediates to H₂O₂ [27]. The effect of applied voltage is shown in Figure 4(c): the optimal H₂O₂ yield (15.88 mg L^-1) was achieved at 4 V. At voltages exceeding 4 V, side reactions may become more pronounced, or the catalytic activity of the cathode may decline, leading to reduced H₂O₂ output. As depicted in Figure 4(d), the initial use of the cathode yielded the highest H₂O₂ concentration (16.06 mg L^-1), which then stabilized in the range of 14.61–14.38 mg L^-1 over subsequent cycles. This trend may be attributed to the gradual deactivation of active catalytic sites on the cathode surface.

Close

Figure 4.

Effects of various factors on H₂O₂ generation by the cathode: (a) carbon-to-nitrogen ratio; (b) pH; (c) applied voltage; (d) reuse cycles.

Export to PPT

To further assess the two-electron oxygen reduction capability of the electrodes, rotating disk electrode (RDE) measurements were performed. As shown in Figure 5(a), the electrode with a carbon-to-nitrogen ratio of 1:0.1 exhibited a more positive onset potential and a higher disk current (60.29 μA at –0.6 V vs. Ag/AgCl). Based on the relationship between the disk and ring currents, the electron transfer number (n) and H₂O₂ selectivity were calculated, as shown in Figures 5(b and c). For N_0.1-GF, the n value was determined to be 3.41, indicating that both 2-electron and 4-electron oxygen reduction pathways occurred simultaneously during the ORR process [28]. After nitrogen doping, N_0.1-GF consistently exhibited the lowest n values among the tested electrodes, suggesting improved selectivity for the 2-electron pathway and enhanced H₂O₂ generation. The H₂O₂ selectivity of N0-GF was 29%, while that of N_0.2-GF and N_0.1-GF increased to 32% and 37%, respectively.

Close

Figure 5.

(a) Disk and ring currents obtained from RRDE measurements; (b) Electron transfer number (n); (c) H₂O₂ yield.

Export to PPT

3.3. Electrochemical treatment of indigo dyeing wastewater

During the cathodic generation of H₂O₂, a higher applied voltage facilitates greater electron transfer. As shown in Figure 6(a), when the voltage increased from 4 V to 12 V, the COD removal efficiency improved from 26.67% to 40.76%. The decolorization efficiency remained consistently above 99.5% and the floc mass continued to increase (Figure 6b). However, the improvement in COD removal from 10 V to 12 V was marginal, while energy consumption increased significantly. Therefore, a voltage of 10 V was selected for subsequent experiments. Under 10 V, the nitrogen-doped carbon cathode was reused for five treatment cycles. Additionally, the TOC removal rates of both the EC and EC–EO systems reached approximately 28% (Figure 6c). As shown in Figure 6d, the COD removal efficiency declined from 37.88% to 22.21%, whereas the decolorization efficiency remained above 99.5% throughout.

Close

Figure 6.

Effects of voltage on wastewater treatment performance: (a) COD removal efficiency, (b) decolorization efficiency and floc mass; (c) TOC removal efficiency; The x axis represents different treatment processes applied in the experiment. (d) cathode reuse performance.

Export to PPT

The wastewater treatment performance of EC and EC–EO systems under different electrolysis durations is shown in Figure 7. The highest COD removal efficiencies were observed at 50 min for both systems—47.4% for EC–EO and 28.85% for EC (Figure 7a). After this point, COD removal began to decline. Experiments showed that when wastewater was treated solely by EO, the COD removal rate reached approximately 10% after 50 min. Figure 7(b) shows that the EC–EO system achieved slightly higher decolorization efficiency than EC alone, reaching nearly 100% within 30 min. Energy consumption comparisons in Figures 7(c-d) reveal that the specific energy consumption for COD removal in the EC–EO system was as low as 0.09611 kWh/(g COD), which is 41.45% lower than that of the EC system. The total energy consumption was also reduced by 12.77%, indicating a significant improvement in energy efficiency.

Close

Figure 7.

Effect of electrolysis time on wastewater treatment performance: (a) COD removal efficiency; (b) decolorization efficiency; (c) specific energy consumption for COD removal; (d) total energy consumption.

Export to PPT

3.4. Mechanism of EC–EO synergy

To elucidate the synergistic mechanism of EC and EO in the EC–EO system, a comprehensive analysis was conducted on the floc adsorption and aggregation processes as well as the transformation of substances in the wastewater. As shown in Figure 8(a), the floc growth in the EC–EO system was more pronounced, indicating a higher capacity to adsorb suspended contaminants, including indigo dye, under the same electrolysis duration. Similarly, the Zeta potential measurements in Figure 8(b) show that with increasing electrolysis time, the Zeta potential values of both systems gradually increased. However, the absolute Zeta potential values in the EC–EO system remained consistently lower than those in the EC system. At 50 min, the Zeta potentials were-2.88 mV for EC–EO and -4.81 mV for EC, respectively. A lower absolute Zeta potential indicates weaker electrostatic repulsion between particles, which facilitates the formation of larger flocs and promotes more efficient sedimentation [29]. These findings suggest that the EC–EO system exhibits superior coagulation performance compared to EC alone.

Close

Figure 8.

Anodic dissolution behavior and floc characteristics: (a) floc morphology; (b) Zeta potential; (c) EIS spectra; (d) Tafel polarization curves; (e) floc mass; (f) electrode loss.

Export to PPT

To further investigate the anodic dissolution behavior and its contribution to floc formation in the EC–EO system, EIS and Tafel polarization measurements were conducted. As shown in Figure 8(c), the charge transfer resistance at the anode–electrolyte interface was significantly lower in the EC–EO system. Additionally, the corrosion potential shifted more negatively (Figure 8d), indicating enhanced anodic dissolution. This behavior is attributed to the nitrogen-doped cathode, which improves electron conductivity and facilitates electron flow toward the anode, thereby reducing the interfacial resistance and promoting anodic metal ion release [30]. These observations are consistent with the results in Figures 8(e and f), which show that the EC–EO system produces a higher floc mass and experiences slightly greater electrode loss, further supporting the conclusion that the introduction of electrocatalytic oxidation enhances anodic activity and floc formation efficiency.

As shown in Figure 9(a), during the electrochemical treatment process, the pH of the EC system remained consistently higher than that of the EC–EO system, indicating that the coupled system has a better capacity for stabilizing wastewater pH. This is attributed to the nitrogen-doped carbon cathode, which preferentially facilitates the ORR over the hydrogen evolution reaction (HER), thereby reducing the generation of OH⁻ and slowing the rise in pH [31]. Correspondingly, the conductivity of the EC–EO system was higher than that of the EC system (Figure 9b), and increased more rapidly, reaching nearly 1 mS/cm at 60 min. This can be attributed to the reduced interfacial resistance at the anode, which promotes the dissolution of Al³⁺ and elevates the ionic concentration in the solution. In addition, the cathodic oxidation of dithionite (S₂O₄^2-) to sulfate (SO₄^2-) further increases both the variety and quantity of ions in the system, collectively contributing to the observed enhancement in conductivity [32].

Close

Figure 9.

(a) pH variation; (b) conductivity change; (c) FTIR spectra of flocs; (d) FTIR spectra of dried supernatant; (e) TG curve of flocs in the EC system; (f) TG curve of flocs in the EC–EO system.

Export to PPT

The infrared (IR) spectra and thermogravimetric (TG) analysis of the flocs and the dried supernatant are presented in Figures 9(c-f). After treatment with the EC–EO system, characteristic functional groups of indigo dye—carbonyl (C=O) and imino (—NH—) groups—were still detectable, suggesting that ROS generated during the process had little direct reaction with the indigo dye. The TG results Figures 9(e and f) show that the flocs from both the EC and EC–EO systems exhibited similar weight loss trends, indicating a comparable composition. These flocs are likely composed of coagulated products formed by the adsorption of indigo dye. Differential thermogravimetric (DTG) analysis revealed clear endothermic peaks for both systems, with the maximum weight loss rates occurring at 141°C and 140°C, respectively. This weight loss in the 140–141°C range corresponds to the evaporation of physically adsorbed water from the electrocoagulated flocs [33]. These results confirm that the electrooxidation process did not directly participate in the degradation or removal of the indigo dye.

To eliminate the potential interference of indigo dye and organic particulates, EC–EO experiments were conducted using inorganic salt wastewater without dye. As shown in the FTIR spectra in Figures 10(a and b), the supernatant from the EC–EO system contained a higher concentration of sulfate ions (SO₄^2-), while the flocs had relatively low sulfate content. In contrast, only trace amounts of SO₄^2- were observed in the EC system supernatant. This confirms that ROS generated at the nitrogen-doped carbon cathode oxidized dithionite ions (S₂O₄^2-) to sulfate. High-resolution S2p spectra of the supernatant (Figures 10c and d) further support this conclusion: the EC–EO system showed a clear reduction in the S₂O₄^2- signal compared to the EC system, verifying the oxidative capacity of the EC–EO configuration. To identify the dominant reactive oxygen species responsible for S₂O₄^2- removal at 10 V, radical scavenging experiments were conducted. As shown in Figure 10(e), the addition of 1,4-benzoquinone (PBQ) led to a dramatic decrease in COD removal efficiency—from 37.88% to 14.26%—while other scavengers had relatively minor effects on S₂O₄^2- removal. This indicates that superoxide radicals •O₂ˉ are the primary active species responsible for oxidizing S₂O₄^2- Notably, the addition of other scavengers also partially suppressed the removal of S₂O₄^2-, suggesting that •O₂ˉ, hydroxyl radicals (·OH), and singlet oxygen (¹O₂) all contribute to the degradation process.

Close

Figure 10.

(a) FTIR spectrum of supernatant (EC system); (b) FTIR spectrum of supernatant (EC–EO system); (c) S2p spectrum of EC supernatant; (d) S2p spectrum of EC–EO supernatant; (e) Radical scavenging experiment results; (f) Proposed oxidation mechanism of the EC–EO system under near-neutral conditions.

Export to PPT

Based on these findings, the proposed oxidation mechanism of the EC–EO system is illustrated in Figure 10(f): under near-neutral pH conditions, O₂ is adsorbed on the surface of the nitrogen-doped carbon cathode and undergoes a 2-electron ORR to generate H₂O₂. Pyridinic and graphitic nitrogen sites facilitate the activation of O₂, leading to the formation of ROS such as ·OH, ¹O₂, and •O₂ˉ, which collectively oxidize S₂O₄^2- into SO₄²⁻. Meanwhile, the Al anode continuously releases Al³⁺, which undergoes hydrolysis and polymerization to form aluminum-based flocs. These growing flocs enmesh and sweep up suspended solids in the wastewater—including residual indigo dye and trace surfactants—through mechanisms such as enmeshment and charge neutralization, ultimately aggregating fine particles into larger flocs that can settle by gravity.