Introduction

Over the past few years, water contamination by endocrine disruptors — mainly pharmaceutical compounds, has become a matter of great concern in our societies, and this has drawn considerable attention among researchers and other stakeholders worldwide (Kasonga et al. 2021; Pironti et al. 2021). The increasingly growing demand for pharmaceutical products (including antibiotics) intended for the treatment and prevention of diseases, ailments and pains has given rise to a rampant use of these compounds and their haphazard disposal in water bodies; this has led to the frequent detection of these pharmaceutical substances in various aqueous matrices (Kovalakova et al. 2020). Norfloxacin (NOR) is a popular antibiotic which is widely used because of its high antimicrobial content; this compound has been detected in wastewater and other water bodies (Médice et al. 2021) and has been classified as a persistent contaminant due to its low biodegradability and toxicity to microorganisms. NOR has been found to pose serious risks to human health and to the ecosystem as it promotes the proliferation of resistant bacteria when present in aqueous systems (Michael et al. 2013). The presence of this kind of contaminants in water bodies has become a matter of great concern since conventional methods of water treatment have been found to be incapable of effectively removing these substances in water, thus leading to their accumulation in the environment (Li et al. 2021; Van Doorslaer et al. 2014).

Taking the above considerations into account, it is clear that there is an urgent need for the development of efficient treatment techniques which are capable of effectively degrading or removing antibiotics and other harmful contaminants from water/wastewater. In this context, electrochemical advanced oxidation processes (EAOPs) have emerged as highly efficient alternative methods for the treatment of contaminants due to their clean, environmentally friendly nature as well as their proven efficiency in treating water contaminated by recalcitrant compounds, mainly through the generation of hydroxyl radicals (•OH) (Ganiyu et al. 2021; Khan et al. 2020; Taoufik et al. 2021). Electro-Fenton (EF) process is regarded as one of the most economically attractive EAOPs (Brillas et al. 2009; Sirés and Brillas 2021). In this process, hydrogen peroxide (H2O2) is electrochemically generated and •OH species are formed via the electrochemically assisted Fenton reaction (involves the combination of H2O2 and ferrous ions) [H2O2 + Fe2+ → •OH+ HO− + Fe3+] (Brillas et al. 2009; Petrucci et al. 2016).

More recently, the use of heterogeneous electro-Fenton (HEF) process for the treatment of water has gained considerable traction among researchers as it allows the treatment to be performed at varying pH levels and provides one with the possibility of reutilizing the heterogeneous catalyst for the conduct of consecutive analyses (Alizadeh and Rezaee 2022; Ghanbari et al. 2021; Jonoush et al. 2021; Wang et al. 2021). The successful application of the HEF process requires the use of highly efficient electrocatalysts with suitable properties (Jonoush et al. 2020, 2022). Bearing that in mind, the use of new cathodic materials with outstanding properties such as iron supported on carbonaceous materials in HEF processes has sparked the interest of researchers due to the high surface area of these materials (Han et al. 2022; Li et al. 2022). Carbon felt (CF) is a suitable material known to be characterized by good electronic conductivity and a sizeable number of active sites (Le et al. 2017). The modification of CF with carbon black and poly(tetrafluoroethylene) (PTFE) leads to the production of a highly efficient electrocatalyst with an enhanced ability to generate a huge amount of H2O2 compared to an unmodified CF (Zhang et al. 2020).

In general, a good iron-based catalyst supported on carbonaceous materials should possess the following desirable properties: elevated electrocatalytic activity, stability at different pH levels, reusability, and environmental compatibility (Luo et al. 2021). The main challenges encountered when it comes to the development of an efficient electrocatalyst for application in HEF processes are as follows: (i) the stability of the catalyst when the pH becomes acidic (leaching of iron species); and (ii) the complexity involving the preparation of the catalyst (Ganiyu et al. 2018). Under the HEF mechanism, the activation of H2O2 to generate •OH species may occur homogeneously through the Fe3+/Fe2+ redox pair (under acidic pH conditions), or through the process catalyzed superficially by the metallic species ≡FeIII−OH to ≡FeII−OH, or through both, depending on the operating pH level (Wan and Wang 2017). Thus, further studies need to be conducted with a view to developing novel, highly efficient iron-based cathodes which are endowed with a considerable number of sites for H2O2 electrolysis (oxygen reduction reaction via the 2e– mechanism); this is because the efficiency of the HEF process is primarily dependent on the rapid activation of the electrogenerated H2O2 by the iron species supported on the electrode surface (Wang et al. 2021).

With that in mind, the present study reports the development and application of novel iron-based cathodes in HEF process using two different approaches: (i) iron oxide is incorporated into a carbon matrix on CF surface; and (ii) iron oxide is deposited directly on CF previously modified with a carbonaceous matrix. Based on the application of the two approaches, we expect to obtain an efficient and reusable cathode capable of operating under a near-neutral pH range. The performance of the cathodes will be investigated by monitoring H2O2 generation and activation under different synthesis conditions. To evaluate the efficiency of the HEF process, the degradation of NOR and its by-products will be monitored by high-performance liquid chromatography (HPLC), ion chromatography (IC), and total organic carbon (TOC). Finally, the mechanisms involving the oxidation of NOR based on the application of the proposed materials will be elucidated.

Experimental section

Materials

The following materials were employed in the experiments: Norfloxacin (99%), Iron (III) chloride hexahydrate (FeCl3·6H2O) (97%), ethylene glycol (99.8%), citric acid (99.5%), n-Butanol (99.5%), ammonium molybdate, and short-chain carboxylic acid standards (a.r.; including oxalic and formic acids) — all the materials were purchased from Sigma Aldrich. Carbon felt (CF), Printex® L6 carbon (PL6C) powder, and poly(tetrafluoroethylene) (PTFE) (60% — aqueous dispersion), used in the experiments, were acquired from LMTerm, Evonik Ltd., and Dupont®, respectively. Other reagents used in the experiments were as follows: potassium sulphate (>99%; Vetec), nitric acid (65%; Synth), sulfuric acid (98% Neon), sodium bicarbonate (>99%; J.T. Baker), sodium carbonate (>99%; J.T. Baker) and acetonitrile (HPLC grade, J.T. Baker). All chemicals were used as received. Ultrapure water (Millipore® Milli-Q system, ρ≥ 18.1 MΩ cm) was used to prepare the solutions.

Preparation of the cathodes

CF with dimension of 4 cm × 5 cm × 1 cm (length × width × height) was used to prepare the modified carbon felts (MCF). PLC6/PTFE dispersion in the proportion of 25:75 was prepared by sonicating a mixture of 0.3 g of PL6C, 1.5 g of PTFE dispersion, 30 mL of ultrapure water and n-Butanol (3% V/V) in an ultrasonic bath. The dispersion was then loaded onto the CF which was immediately calcined in an oven at 360 °C for 1 h at a heating rate of 10 °C min−1 to prepare the MCF (Yu et al. 2015).

To optimize the mass loading of PL6C/PTFE, the MCF cathodes were prepared by the deposition of one and two layers of the dispersion. The deposition with one and two layers (cycles) of dispersion yielded 60 mg (~3 mg cm−3) and 120 mg (~6 mg cm3) electrodes, respectively; the electrodes were then subjected to calcination. To determine the final mass deposited, the initial mass of the CF substrate was subtracted from the mass obtained after calcination (mdeposited = minitial – mafter calcination). The value obtained was then divided by the volumetric area of CF (20 cm3). These electrodes will be denoted hereinafter as MCF-(3) and MCF-(6).

To obtain the iron-based MCF electrodes, two different approaches were employed: (1) the iron oxide was incorporated on the felt fibers by soaking the CF in a previously prepared dispersion containing PL6C, PTFE and FeCl3·6H2O (one-step approach); and (2) the iron oxide was deposited on CF previously modified with PL6C/PTFE through thermal decomposition of a polymeric iron precursor (prepared based on the procedure reported in (Santos et al. 2020)) (two-step approach). The cathodes will be denoted hereinafter as MCF/Fe-incorporated and MCF/Fe-deposited for the one-step and two-step approaches, respectively.

Physical characterization

The surface morphology of the cathodes was characterized by scanning electron microscopy using scanning electron microscope (SEM; JEOL model JSM 6510 V) equipped with an X-ray detector for EDS spectroscopy. X-ray diffraction (XRD; Rigaku diffractometer RINT 2000/PC) analysis was carried out using CuKα radiation (wavelength = 1.5406 Å) in a scanning interval of 2ϴ range between 20 and 60°, in continuous scan mode, with steps of 0.02° min–1. The crystalline phases were indexed using the X’Pert High Score Plus software (version 2.2.2) according to database of Joint Committee on Powder Diffraction Standards (JCPDS) database. To verify the hydrophobicity of the materials, contact angle values were recorded using a tensiometer (Teclis Tracker, IT Concept). A 3 μL volume of water was dropped on the surface of the materials and the contact angle between them was estimated.

Electrochemical characterization

The electrochemical characterization (linear sweep voltammetry (LSV), cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS)) of the materials were performed using potentiostat/galvanostat (MetrohmAutolabPGSTAT-128N). The measurements were performed using a one-compartment three-electrode cell equipped with the prepared cathodes (used as working electrodes), a dimensional stable anode (DSA®) composed of RuO2-TiO2 (used as counter electrode), and Ag/AgCl (3 mol L−1 KCl) — used as reference electrode. In all the analyses, 0.05 mol L−1 K2SO4 solution at pH 7 (without adjustment) was used as supporting electrolyte. The oxygen reduction reaction (ORR) was evaluated by LSV using scan rate of 10 mV s−1 and potential interval of −1.3–0 V vs Ag/AgCl (3 mol L−1 KCl). EIS analysis was conducted based on the application of different voltages in the frequency range of 1000 to 0.01 Hz, with AC amplitude of 5 mV.

H2O2 electrogeneration and NOR degradation

The experiments involving H2O2 electrogeneration and the degradation of NOR were carried out in a one-compartment cell (batch operation mode) using potentiostat/galvanostat (MetrohmAutolab PGSTAT302). DSA® (4 cm2) and Ag/AgCl (3 mol L−1 KCl) were used as counter (anode) and reference electrodes, respectively. Unmodified (CF) and modified (MCF-(3), MCF-(6), MCF/Fe-incorporated and MCF/Fe-deposited) carbon felts of 20 cm2 (geometric area) prepared according to the procedure described above were used as cathodes. All the experiments were performed using 200 mL of 0.05mol L−1 K2SO4 solution (pH ~7) under continuous agitation with the aid of a magnetic stirrer (500 rpm), under applied temperature of 25 °C. A continuous flow of O2 was fed into the electrolyte at a flow rate of 1 L min−1 during the experiments.

The experiments involving the electrogeneration of H2O2 were performed using the MCF-(3) and MCF-(6) cathodes, with the application of different electric currents of 0.1, 0.15 and 0.3 A (i.e., j = 5, 7.5 and 15 mA cm−2, respectively) relative to the geometric area of the cathode (the values of j in terms of volumetric area are equivalent). After the optimization of the applied current, 0.15 A (7.5 mA cm−2) was chosen for the conduct of subsequent studies on H2O2 electrogeneration using the iron oxide–based cathodes. In this step, an analysis was conducted in order to evaluate the influence of pH on the amount of electrogenerated oxidant using MCF-(3), MCF/Fe-deposited and MCF/Fe-incorporated. Finally, the degradation of NOR (63 μmol L−1) was evaluated under the optimal conditions obtained.

NOR removal and mineralization analyses

The concentration of electrogenerated H2O2 was measured after adding 0.5 mL of the sample to 4.0 mL of 2.4×10−3mol L−1 ammonium molybdate solution — which turns into a yellowish complex in the presence of H2O2 (with absorbance at 350 nm) (Forti et al. 2007). The electrogenerated H2O2 was then quantified spectrophotometrically using UV–vis spectrophotometer (Shimadzu UV-1900).

The degradation of NOR was monitored by HPLC/UV-DAD detector using HPLC equipment, Shimadzu model LC-20AT, coupled to an SPD-20A detector. A Phenomenex reverse-phase C18 column (250 mm × 4.6 mm, 5 μm particle size) was used as the stationary phase and a mixture of 50:50 acetonitrile/acidified water (5% formic acid) was used as mobile phase in an isocratic mode. The conditions applied for this analysis were as follows: flow rate of 0.8 mL min−1, injection volume of 20 μL, detection wavelength of 275 nm, and oven temperature of 40 °C. The degree of mineralization was evaluated using Shimadzu TOC analyzer (model TOC-VCPN). The carboxylic acids were identified by ion chromatography using Metrosep Organic Acids 250/7.8 column, with the application of 0.5 mmol L−1 H2SO4 (flow rate of 0.5 mL min−11) as eluent, injection volume of 20 μL, and column temperature of 60 °C. The retention time of formic acid and acetic acid was compared to that of the standards which have been previously analyzed. Isopropanol was employed as quenching agent to investigate the role of hydroxyl radicals in the degradation mechanism of norfloxacin where the second-order rate constants with the hydroxyl radical is k = 2.8×109 M−1 s−1. (Liang and Su 2009). The iron leached was quantified using the orthophenantroline complexometric method (Murti et al. 1970).

The percentage of current efficiency (CE) was calculated based on Eq. (1), where 2 corresponds to the number of electrons required for oxygen reduction, F is the Faraday constant (96,487 mol−1), \({\mathrm C}_{{\mathrm H}_2{\mathrm O}_2}\) is the H2O2 concentration in mol L−1, Vs is the volume of solution in L, I is the applied current in A, and t is the time of electrolysis (s) (Fortunato et al. 2020b).

$${\mathrm{CE}}_{{\mathrm{H}}_2{\mathrm{O}}_2}\ \left(\%\right)=\frac{\ 2\ F{C}_{{\mathrm{H}}_2{\mathrm{O}}_2}{V}_S}{It}\times 100\%$$
(1)

The energy consumption (ECH2O2) was derived from Eq. (2), where Ecell is the average of cell potential (V), I is the applied current (A), t is the time of electrolysis (hour), \({\mathrm C}_{{\mathrm H}_2{\mathrm O}_2}\) is the concentration of H2O2 (mg L−1), and Vs is the volume of solution (L) (Barros et al. 2015b).

$$\mathrm{EC}\ \left(\mathrm{kWh}\ {\mathrm{kg}}^{-1}\right)=\frac{\ 1000\ {E}_{cell} It}{C_{{\mathrm{H}}_2{\mathrm{O}}_2}{V}_S}$$
(2)

The energy consumption (ECTOC) per TOC removal was calculated based on Eq. (3), where Ecell is the cell potential (V), I is the applied current (A), t is the electrolysis time (s), Vs is the volume of solution (L), and ∆[TOC] is the change in TOC concentration (mg L–1) (Fortunato et al. 2020a).

$${\mathrm{EC}}_{\mathrm{TOC}}\ \left(\mathrm{kWh}\ {\mathrm{g}}^{-1}\right)=\frac{E_{cell} It}{V_S\varDelta \left[ TOC\right]}$$
(3)

The figure-of-merit, specific electrical energy (EEO), corresponds to the electrical energy (kW h−1) required to decrease the concentration of the contaminant in one order of magnitude. The EEO was used to compare the data obtained from the application of other AOPs in terms of cost-efficiency based on Eq. (4), where Ecell (V) is the average cell potential, I is the current applied (A), Vs is the volume of solution (L), and k1 is the pseudo-first-order constant (min−1) (Bolton et al. 1996; Lanzarini-Lopes et al. 2017).

$${\mathrm{E}}_{\mathrm{E}\mathrm{O}}\ \left(\mathrm{kWh}\ {\mathrm{m}}^{-3}{\mathrm{order}}^{-1}\right)=\frac{38.4\times {10}^{-4}{E}_{cell}I}{V_S{k}_1}$$
(4)

Results and discussion

Optimization of the amount of PL6C used in H2O2 electrogeneration

Physical and electrochemical characterization

SEM analyses of the unmodified CF and modified MCF-(3) and MCF-(6) cathodes were performed in order to investigate the effect of PL6C/PTFE loading on the morphology of the materials. Figure 1a–c shows that the PL6C/PTFE mixture is homogenously deposited on the carbon fibers with some agglomerated regions. The EDS analysis of the regions of each material showed the presence of carbon on pure CF; as expected, after modification with PL6C/PTFE, the modified materials exhibited the presence of fluorine element from PTFE (which has (C2F4)n composition) (Karatas et al. 2022; Rofaiel et al. 2012).

Fig. 1
figure 1

SEM images and EDS spectra of: (a) pure CF, (b) MCF-(3), (c) MCF-(6) and (d) cyclic voltammetries and (e) linear sweep voltametries of CF, MCF-(3), MCF-(6) recorded at 20 mV s−1 in 0.05 mol L−1 K2SO4

Full size image

CV and LSV techniques were applied for the electrochemical characterization of the materials (Fig. 1d–e); these analyses were performed in order to evaluate the effect of the catalyst loading on the electrochemical properties of the cathodes. The electrochemically active surface area (ECSA) of MCF-(3) and MCF-(6) was determined as the relative measure of the voltammetric charge by integrating the CV curves in the range of −1.4–0 V vs Ag/AgCl at 20 mV s−1 (Fig. 1d). As can be observed in Fig. 1, the ECSA values obtained were quite close to each other; this means that an increase in PL6C/PTFE loading does not lead to a significant increase in the active sites for electrochemical reaction

To gain a better understanding of the ORR on the CF, MCF-(3) and MCF-(6) cathodes in the system studied at neutral pH, LSV measurements were performed in the negative region (Fig. 1e). The results obtained from the LSV measurements showed that both MCF cathodes presented relatively higher currents for ORR compared to the unmodified CF. This outcome is in line with the results obtained in previous studies reported in the literature which showed that the modification of graphite felt with carbon black/PTFE led to improvements in the electrocatalytic activity for O2 reduction and conductivity of the material (Huang et al. 2021; Yu et al. 2015). The electrode containing higher PL6C/PTFE loading (i.e., MCF-(6)) exhibited relatively lower current density compared to the MCF-(3); this points to a reduction in the active sites for O2 reduction, which is likely to have led to the electrogeneration of lower amounts of H2O2.

Effect of applied current on H2O2 electrogeneration

The high number of sites for the electrogeneration of H2O2 under the 2e– ORR mechanism is found to be of great importance in the HEF process; this is because the efficiency of the process in terms of the degradation of organic pollutants is deeply related to the rapid activation of H2O2 by the heterogeneous catalyst (Zhou et al. 2019). Fig. 2 shows the concentration of H2O2 electrogenerated in 60 min of electrolysis based on the application of the MCF-(3) and MCF-(6) electrodes at different current densities using 0.05 mol L−1 K2SO4 solution as electrolyte solution. Primarily, the formation of H2O2 occurs via the electrochemical reduction of oxygen dissolved in the solution based on chemical Equation (5) (Perry et al. 2019). Looking at Fig. 2a, it is clear that H2O2 accumulation follows a pseudo-zero order kinetics, which is reflected by the linear increase in concentration observed in the first 30 min of electrolysis (the same profile can be observed in Fig. 2b). After 30 min of electrolysis, H2O2 electrogeneration can be found to have stabilized; this pattern of behavior has been widely documented in the literature, and it is attributed to different side/parallel reactions that may affect the H2O2 electrogeneration process. These parallel reactions include the following: self-decomposition (Eqs. (6)–(8)), cathodic reduction to water (Eq. (9)) or oxidation to oxygen at the anode, to name a few (see Eq. (10)) (Brillas et al. 1995; Carneiro et al. 2015; Samanta 2008).

Fig. 2
figure 2

Effect of PL6C/PTFE loading on H2O2 electrogeneration (a, b) and current efficiency and energy consumption (c, d) at different current densities. Electrolyte solution: 0.05 mol L−1 K2SO4 at pH 7

Full size image
$${\mathrm{O}}_{2\left(\mathrm{g}\right)}+2{\mathrm{H}}^{+}+2{\mathrm{e}}^{-}\to {\mathrm{H}}_2{\mathrm{O}}_2$$
(5)
$$2{\mathrm{H}}_2{\mathrm{O}}_2\to 2{\mathrm{H}}_2\mathrm{O}+{\mathrm{O}}_2$$
(6)
$${\mathrm{H}}_2{\mathrm{O}}_2\leftrightarrow \mathrm{H}{\mathrm{O}}_{2^{\hbox{--} }}+{\mathrm{H}}^{+}$$
(7)
$${\mathrm{H}}_2{\mathrm{O}}_2+\mathrm{H}{\mathrm{O}}_{2^{\hbox{--} }}\to {\mathrm{H}}_2\mathrm{O}+{\mathrm{O}}_2+\mathrm{O}{\mathrm{H}}^{\hbox{--} }$$
(8)
$${\mathrm{H}}_2{\mathrm{O}}_2+2{\mathrm{H}}^{+}+2{\mathrm{e}}^{-}\to 2{\mathrm{H}}_2\mathrm{O}$$
(9)
$${\mathrm{H}}_2{\mathrm{O}}_2-2{\mathrm{e}}^{\hbox{--}}\to {\mathrm{O}}_2+2{\mathrm{H}}^{+}$$
(10)

The kinetics of H2O2 electrogeneration was analyzed for the MCF-(3) and MCF-(6) electrodes taking into account the first 30 min of electrolysis. The results obtained showed that MCF-(3) recorded the following kinetics values: 4.5, 10.5 and 13.2 mg L−1 min−1, at the applied current densities of 0.1, 0.15 and 0.3 A, respectively. Interestingly, by doubling the PL6C/PTFE loading MCF-(6), the kinetics values obtained were 5.3, 10.7 and 12.5 mg L−1 at the applied current densities of 0.1, 0.15 and 0.3 A, respectively; this shows that both cathodes recorded similar amounts of electrogenerated H2O2.

The current efficiency (CE) and energy consumption (EC) of the MCF-(3) and MCF-(6) electrodes were also calculated (see Fig. 2c–d). As can be observed, the highest current efficiencies were obtained by the application of current density of 0.15 A, with MCF-(3) and MCF-(6) recording 90.1% and 75.5% of current efficiency, respectively. With regard to energy consumption, the lowest amounts of energy consumed by MCF-(3) and MCF-(6) were 8.4 and 10.0 kWh kg−1, respectively. These results point to the following: (i) the applied current of 0.15 A is the most suitable current for H2O2 electrogeneration; and (ii) the MCF-(3) electrode exhibited better performance in terms of CE and EC. In view of that, the MCF-(3) electrode was chosen as the optimal mass loading as it is able to provide sufficient electrochemical active area and diffuse the amount of O2 required for H2O2 electrogeneration. These data are in agreement with those discussed above related to the physical and electrochemical characterization of the materials (Fig. 1).

Effect of iron supported on PL6C/PTFE in H2O2 electrogeneration

Physical and electrochemical characterization

Over the past few years, there has been a considerable interest among researchers in the development of highly effective, low-cost iron oxide-based cathodes for use in HEF systems under neutral operating environment (Ganiyu et al. 2018; Perry et al. 2019). In the present study, after the analysis of the optimal PL6C/PTFE loading, the iron oxide-based cathodes were prepared by modifying the MCF-(3) cathode. The images obtained from the SEM analysis pointed to the successful deposition (MCF/Fe-deposited) and incorporation (MCF/Fe-incorporated) of iron oxide onto the CF fibers — see Fig. 3a–b. In a recent study conducted by Huang et al. (2021), with the aid of SEM images, the authors showed that the modification of graphite felt with a mixture of iron-manganese binary oxide, activated carbon, carbon black and PTFE led to a good distribution of the particles formed over the fibers of the graphite felt electrode (Huang et al. 2021). Under the two approaches employed in our present study, the presence of iron and oxygen was detected through EDS mapping images; this is clearly indicative of the successful deposition or incorporation of the iron on the PL6C/PTFE material. The XRD patterns obtained for the unmodified CF and the three MCF cathodes (Fig. 3e) exhibited diffraction peaks related to carbon (Huang et al. 2021; Su et al. 2019), and the two iron oxide–based cathodes prepared were composed of Fe2O3 (JCPDS Card No 96-9770) (Su et al. 2019); however, only the Fe-incorporated cathode presented a peak related to zero valent iron (Fe0) (JCPDS Card No 06-0696). The aforementioned results are in agreement with the findings reported in previous studies related to the development and application of iron-based cathodes (Huang et al. 2021).

Fig. 3
figure 3

SEM images and EDS mapping images of: (a) MCF/Fe-deposited and (b) MCF/Fe-incorporated (b); EDS spectra (elemental analysis) of (c) MCF/Fe-deposited and of (d) MCF/Fe-incorporated; and XRD patterns obtained for the unmodified CF substrate and CF modified with and without iron oxide (prepared by deposition or incorporation)

Full size image

The analysis of contact angle (CA) was carried out in order to investigate the wettability of the cathodes (Fig. SM1). In this technique, liquid drop (usually water) is placed on a solid surface, forming an angle based on the gas-liquid-solid interface — this is referred to as the contact angle (ϴ). Materials with CA values (ϴ) > 90° have low degree of wettability (i.e., hydrophobic surface), and the materials with ϴ values > 150° are found to be superhydrophobic in nature, while those with ϴ values < 90° are typically hydrophilic in nature (Kim et al. 2014). The results obtained from the analysis of the contact angle of the materials reflected the following tendency: ϴCF < ϴMCF/Fe-incorporated < ϴMCF-(3) < ϴMCF/Fe-deposited. These results show that the wettability of the materials changes according to the modification of MCF; in other words, the CA of the CF is found to increase after modification — as expected by the presence of PTFE. The MCF/Fe-incorporated exhibited the lowest degree of wettability among the modified cathodes; this may be attributed to the formation of Fe2O3 nanoparticles on the felt fibers. In contrast, the MCF/Fe-deposited exhibited the highest CA values; this may be attributed to the formation of Fe2O3 as deposits between the felt fibers.

Mohamed et al. developed an efficient modified anode via the deposition of Fe/Fe2O3 nanoparticles for energy production and wastewater treatment; the authors found that the wettability of carbonaceous materials, including carbon felt, was enhanced after modification with iron/iron oxide (Mohamed et al. 2018).

The electrochemical behavior of the cathodes was evaluated using CV and EIS. CV profiles obtained for MCF, MCF/Fe-deposited and MCF/Fe-incorporated are shown in Fig 4a. Looking at the CV profile of the iron oxide–modified cathodes, one can observe the presence of the Fe(II)/Fe(III) redox couple with the anodic peaks at around 0.6–0.8 V vs Ag/AgCl and the cathodic peaks at 0.1–0.2 V vs Ag/AgCl (Kabtamu et al. 2018). Among the iron oxide-modified cathodes, the MCF/Fe-incorporated exhibited the largest voltammetric area with extremely high currents for oxidation and reduction reactions. As pointed out in the literature, the electrochemically active surface area (ECSA) is a useful tool for measuring the intrinsic electrochemical activity of the materials investigated (Ren et al. 2020). The ECSA of MCF, MCF-(3) and MCF-(6) was calculated by the double layer capacitance method (Fig. 4b–c) — this is a common technique used to study the electrochemical activity of carbon-based materials (Cordeiro-Junior et al. 2020; McCrory et al. 2013). The results obtained from this analysis showed that the MCF/Fe-incorporated electrode exhibited the greatest surface area, followed by the MCF/Fe-deposited electrode; the iron oxide-free MCF exhibited the lowest surface area. These results show that the modification of the MCF with iron oxide leads to a significant improvement in the surface area of the material.

Fig. 4
figure 4

CV curves obtained for the modified CF at scan rate of 50 mV s−1 (a), with capacitive current density average at −0.35 V vs scan rate (b); ECSA for the different cathodes (c) and charge transfer resistance obtained from the application of different polarization voltages (d). Conditions: 0.05 mol L−1 K2SO4 at pH 7 employed as electrolyte

Full size image

EIS measurements were performed using different MCF electrodes applied in 0.05 mol L−1 K2SO4 (electrolyte solution) at different potentials (Fig. SM2a–c). The values related to the ohmic resistance of the electrolyte solution, electrodes and contacts (RΩ) as well as the charge transfer resistance (Rct) were obtained from the Nyquist diagrams; these values help evaluate the conductivity of the materials investigated. These parameters related to the adjustment of the EIS data are shown in Table 1.

Table 1 EIS data obtained for the modified carbon felt cathodes at the following potentials: −0.3, −0.35, −0.4, and −0.45 V vs Ag/AgCl
Full size table

The polarization voltages applied for the EIS measurements were chosen based on negative LSV curves and the corresponding Nyquist plots obtained can be found in Fig. SM2a–c. The voltages applied varied from the ORR onset potential (−0.3 V) to more negative values (−0.35, −0.40, and −0.45 V). Both the MCF/Fe-deposited and MCF/Fe-incorporated (in particular) exhibited lower Rct values (Fig. 4d); this behavior may be attributed to the ion redox reaction of iron species that occur on the electrode surface. It has been well documented in the literature that the main functional groups typically found in carbonaceous matrix — commonly observed by X-ray photoelectron spectroscopy or Fourier transform infrared spectroscopy techniques, are ether (–C–O), carboxyl –COOH, (–C–OH), carbonyl (–C=O) and hydroxyl (–OH) groups (Abaalkhail et al. 2022; Cordeiro-Junior et al. 2020; Paz et al. 2018; Wang et al. 2014; Ye et al. 2016). Thus, we assume that the iron ions gradually connect with the active sites on the fiber surface (i.e.,–OH, –COOH) to form C–O–Fe bonds; these bonds contribute to the electron transfer reactions, leading to lower Rct values. The iron oxide-deposited cathode also exhibited a similar pattern of behavior; in other words, the possible formation of C–O–Fe bonds in the cathode led to relatively smaller Rct compared to the unmodified MCF cathode although the Rtc values of the iron oxide deposited cathode were higher than that of the iron-incorporated cathode.

Effect of modifying CF cathode with iron on H2O2 electrogeneration at different pH values

An analysis was conducted in order to evaluate the effect of pH (3, 7 and 11) on the concentration of H2O2 electrogenerated in 90 min of electrolysis (Fig. 5). Interestingly, the application of neutral pH contributed toward the electrogeneration of higher amounts of H2O2, which is favorable in terms of efficiency when it comes to the removal of contaminants and the application of the HEF process under suitable pH conditions that are compatible with environmental systems. On the other hand, an increase in the pH value to 11 led to the production of the lowest amount of H2O2. Under basic condition, H2O2 coexists with hydroperoxide ion (HO2−); this compound has been reported to be involved in the auto-decomposition of H2O2 in strongly basic solutions (pH > 9) (HO2– + H2O2 → H2O + O2 + OH–) (Barros et al. 2015a). A previous study conducted by Zhou et al. based on the application of hydrazine-modified CF also showed that an increase in pH from 3 to 8.1 led to a decrease in the amount of electrogenerated H2O2 from 193.9 to 162.2 mg L−1, respectively (Zhou et al. 2008); this finding is consistent with the results obtained in the present work — where the application of pH 11 led to a shift in the equilibrium of the reaction to H2O2 decomposition. It should be noted however that, other studies published in the literature have reported the production of higher amounts of H2O2 in basic medium, especially when gas diffusion electrodes are applied (GDE). Similarly, Soltani et al. showed that an increase in the pH value from 6 to 9 resulted in an increase in the amount of H2O2 electrogenerated from 66.52 to 122.6 μM (Soltani et al. 2013). Barros et al. also investigated H2O2 electrogeneration in alkaline medium based on the application of GDE with PL6C where they reported to have obtained good results in terms of H2O2 production with the consumption of relatively low amount of energy (Barros et al. 2015a). It is worth pointing out that in our present work, CF is employed as an immersed cathode; this is totally different from gas diffusion cathodes where O2 is continuously injected through the cathode. The results obtained in this study show that the electrode performance at different pH values varies according to the nature of the cathode material and the modifications carried out in this material.

Fig. 5
figure 5

Effect of the pH value on H2O2 electrogeneration based on the application of three different MCF cathodes. Conditions applied: 0.05 mol L−1 K2SO4 employed as electrolyte solution; pH values evaluated: 3, 7, and 11.

Full size image

Based on the results obtained, it can be concluded that the modified carbon felts recorded their best performance at pH close to neutral, followed by acidic pH environment. These findings are in agreement with those reported by Yu et al. who showed that modifying the cathode with carbon black contributed to a slight increase in H2O2 production at pH 7 (Yu et al. 2015).

NOR degradation

The performance of the MCF cathodes was investigated through degradation experiments using 63 μmol L−1 NOR in 0.05 mol L−1 K2SO4 at pH 7. As can be seen in Fig. 6a, the NOR removal percentages recorded for MCF and MCF-deposited in 90 min of treatment were 50.8% and 81.1%, respectively, while the MCF/Fe-incorporated cathode exhibited 100% of NOR removal in 30 min of experiment. These results help to confirm that the presence of iron oxide in the cathode enhances the efficiency of the HEF system applied under the H2O2 electrogeneration mechanism. In addition, the NOR removal process follows a pseudo-first-order kinetics model (inset in Fig. 6b). The degradation rate constant obtained for the MCF/Fe-incorporated cathode (k = 0.057 min−1) was 5.34 and 3.54–fold higher than the value recorded for the iron oxide-free cathode (MCF) and the MCF/Fe-deposited cathode, respectively. A comparison of the iron oxide–deposited cathode with the iron oxide–incorporated cathode shows that the latter exhibited a 1.5-fold increase in NOR degradation rate constant relative to the former.

Fig. 6
figure 6

Results obtained for (a) NOR decay, (b) degradation rate constants, and (c) short-chain carboxylic acids evolution during NOR removal under the HEF treatment mechanism based on the application of the MCF/Fe-incorporated cathode. Conditions applied — electrolyte solution: 0.05 mol L−1 K2SO4 at pH 7

Full size image

The application of the MCF/Fe-incorporated cathode resulted in TOC removal efficiency of 52.3% after 90 min of treatment under the HEF system; the application of the other two cathodes did not yield any change in TOC concentration (Table 2). Regarded as an important figure-of-merit, the electrical energy per order (EEO) was used to compare the efficiency of different advanced electrochemical oxidation techniques (Bolton et al. 2001). As can be seen in Table 2, the application of the MCF/Fe-incorporated cathode under the HEF treatment process resulted in the lowest EEO and the highest TOC removal efficiency. Two short-chain carboxylic acids were detected during the degradation process using the MCF/Fe-incorporated cathode (Fig. 6c); these acids, which consisted of formic and acetic acids, were found to be the terminal oxidation intermediates. The presence of formic and acetic acids has been previously reported in the literature during the degradation of other fluoroquinolone antibiotics, including levofloxacin, enoxacin, norfloxacin, and ciprofloxacin, under the application of the electro-Fenton process (Annabi et al. 2016; Barhoumi et al. 2015; Carneiro et al. 2020).

These results show that the application of the MCF/Fe-incorporated cathode at neutral pH generates the optimal conditions for the effective degradation/mineralization of NOR with high electrical energy consumption efficiency. As can be seen in Table 3 the degradation efficiency obtained from the application of the MCF/Fe-incorporated cathode under the HEF treatment process reported in the present work was found to be better than that obtained by previous studies reported in the literature which involved the use of other advanced oxidation processes (AOPs).

Table 2 Percentages of total organic carbon (TOC) removed after 90 min of electrolysis and values related to electrical energy per order (EEO) obtained under the application of different treatment processes
Full size table
Table 3 Comparative analysis of NOR removal based on the application of different AOPs reported in the literature
Full size table

Mechanistic study

In HEF systems, the kinetics of the process is usually complex, depending on the nature of the catalyst (solubility) and the working pH; the most critical step in the process is the catalytic surface production of •OH species (Munoz et al. 2015). The mechanism that is widely accepted in the literature for the oxidation process under the HEF technique is similar to that of the classic homogeneous Fenton; the surface mechanisms involving this process can be found in Equations (11)–(16) below (Ganiyu et al. 2018; Wang et al. 2013). The mechanism of activation is initiated by the H2O2 electrogenerated (Eq. (1)) and by the reduction of ≡FeIII─OH (Eq. (11)) which occur simultaneously on the cathode. The surface complex ≡FeIII─OH(H2O2) can be formed on the inner or external surface of the cathode (Eq. (12)). Also, the ≡FeII─OH(HO2∙) can be formed via electron transfer (Eq. (13)) and be deactivated to form FeII─OH (Eq. (14)). After that, H2O2 undergoes catalytic decomposition to form FeIII─OH and •OH (Eq. (15)), which oxidize the organic pollutants, leading to the formation of intermediaries or complete mineralization to CO2, water and inorganic ions (Eq. (16)).

$$\equiv {\mathrm{Fe}}^{\mathrm{II}\mathrm{I}}\hbox{--} \mathrm{OH}+{\mathrm{e}}^{-}\to {\mathrm{Fe}}^{\mathrm{II}}\hbox{--} \mathrm{OH}$$
(11)
$$\equiv {\mathrm{Fe}}^{\mathrm{III}}\hbox{--} \mathrm{OH}+{\mathrm{H}}_2{\mathrm{O}}_2\leftrightarrow \equiv {\mathrm{Fe}}^{\mathrm{III}}\hbox{--} \mathrm{OH}{\left({\mathrm{H}}_2{\mathrm{O}}_2\right)}_{\left(\mathrm{s}\right)}$$
(12)
$$\equiv {\mathrm{Fe}}^{\mathrm{II}\mathrm{I}}\hbox{--} \mathrm{OH}{\left({\mathrm{H}}_2{\mathrm{O}}_2\right)}_{\left(\mathrm{s}\right)}\to \equiv {\mathrm{Fe}}^{\mathrm{II}}\hbox{--} \mathrm{OH}\ {\left(\mathrm{H}{{\mathrm{O}}_2}^{\bullet}\right)}_{\left(\mathrm{s}\right)}+{\mathrm{H}}^{+}$$
(13)
$$\equiv {\mathrm{Fe}}^{\mathrm{II}}\hbox{--} \mathrm{OH}\ \left(\mathrm{H}{\mathrm{O}}_2\bullet \right)\left(\mathrm{s}\right)\to \equiv {\mathrm{Fe}}^{\mathrm{II}}\hbox{--} \mathrm{OH}+\mathrm{H}{{\mathrm{O}}_2}^{\bullet }+{\mathrm{H}}^{+}$$
(14)
$$\equiv {\mathrm{Fe}}^{\mathrm{II}}\hbox{--} \mathrm{OH}+{\mathrm{H}}_2{\mathrm{O}}_2\to \equiv {\mathrm{Fe}}^{\mathrm{II}\mathrm{I}}\hbox{--} \mathrm{OH}+\bullet \mathrm{OH}+{\mathrm{O}\mathrm{H}}^{-}$$
(15)
$$\bullet \mathrm{OH}+\mathrm{pollutant}\to \mathrm{intermediaries}\to \mathrm{C}{\mathrm{O}}_2+{\mathrm{H}}_2\mathrm{O}$$
(16)

Quenching experiments were performed using isopropanol (•OH scavenger) in order to study the role of active species in the heterogeneous EF process under neutral pH (Fig. 7). The results obtained in these experiments showed that in the presence of isopropanol, NOR removal decreased from 100 to 58% in 30 min of reaction when the MCF/Fe-incorporated cathode was applied, while the MCF/Fe-deposited cathode recorded a decrease in NOR removal from 34.64 to 17.2%. The system equipped with the MCF cathode also recorded a decline in NOR removal from 23 to 11.1% (Santos et al. 2021). In the case of the MCF cathode, the reduction observed in the degradation rate can be attributed to the blockage of •OH species on the DSA® anode surface — which produces the higher state superoxide that is responsible for the partial oxidation of organic compounds (Martínez-Huitle and Andrade 2011; Martínez-Huitle et al. 2015).

Fig. 7
figure 7

Effect of •OH scavenger on the degradation of NOR during electrolysis using MCF (anodic oxidation) (a), MCF/Fe-deposited (b) and MCF/Fe-incorporated (c) electrodes

Full size image

It can be clearly observed that •OH species played a major role in the system equipped with the MCF/Fe-incorporated cathode; this points to the occurrence of a quick reaction between the deposited iron oxide (FeIII/FeII) and the H2O2 electrogenerated in situ to produce highly reactive radicals on the cathode surface. On the other hand, the iron-deposited electrode exhibited a less pronounced reduction in degradation rate; this outcome can be related to the lower surface area of this electrode, as can be seen in the electrochemical characterization images/data. That is, iron oxide partially blocks the felt fibers (as can be seen in the SEM images in Fig. 3a), and this leads to a decrease in the ECSA value. Finally, in the case of the MCF-(3) electrode, the inhibition of NOR removal was found to be less pronounced; this means that anodic oxidation via direct electron transfer and HO∙ species on the anode surface is the main mechanism responsible for NOR degradation.

Overall, the possible mechanism involving NOR degradation via the application of the two different iron oxide–based cathodes prepared in this study appears to be the same; the surface reactions are found to be the main mechanism involved in NOR degradation under the HEF treatment process. The iron deposited or incorporated into the cathode yields ≡FeII; this product in turn reacts to form H2O2 and O2 which generate •OH and O2•−. The •OH and O2•− generated then act to promote NOR degradation on the cathode surface, thus enhancing the degradation efficiency.

Reusability and leaching of the MCF/Fe-incorporated cathode

Reusability of the cathode is crucial to evaluate the cost-effectiveness and feasibility of the proposed process. In this view, cycling tests were performed using the MCF/Fe-incorporated cathode (Fig. SM4) and as can be observed, degradation efficiency does not decrease after four successive cycles (keeping 100% removal within 90 min treatment). Although in neutral medium, FeIII is expected to be insoluble; the contribution of dissolved iron must also be accounted in HEF systems even at neutral pH. Thus, a further investigation of iron leaches from the electrode surface was conducted (Fig. SM.5). As a result, the total iron measured at the end of the experiment was 2.36 ± 0.3 mg L−1 which shows a controlled leaching of iron in each cycle in the HEF system with MCF/Fe-incorporated electrode. For comparison, the value found for the MCF/Fe-deposited (0.17± 0.005 mg L−1) is much lower. Clearly, the higher concentration of leached iron promotes the enhanced efficiency of HEF with MCF/Fe-incorporated for NOR degradation, as it would lead to a faster activation of H2O2 to produce radicals compared to MCF/Fe-deposited.

Conclusions

The present study reported the development of a novel synthesis technique, which involved the modification of CF with PL6C/PTFE and iron oxide (deposited or incorporated) and the application of the modified CF electrode in HEF system for the treatment of water contaminated by recalcitrant and hazardous compounds, including antibiotics. The main conclusions drawn from this study are as follows:

  • Increasing PL6C/PTFE loading does not cause a significant effect on H2O2 electrogeneration. The modification of CF with only 3 mg cm−3 catalyst loading led to the generation of sufficient active sites for H2O2 electrogeneration with higher current efficiency.

  • Both iron-based cathodes investigated in this study were synthesized successfully; the iron oxides were homogenously dispersed on the carbon felt fibers. The one-step synthesis approach (i.e., MCF/Fe-incorporated) was found to be the best route for the construction of highly efficient iron-modified CF in terms of morphology with homogenous deposition of iron oxide without losing the 3D character and electrochemical properties (highest ECSA and greater conductivity).

  • After the modification of CF with iron, a decrease was observed in H2O2 production due to the quick reaction of H2O2 with the iron to form •OH species. For the degradation experiments performed under neutral pH, the MCF/Fe-incorporated cathode exhibited the best efficiency in terms of NOR removal, mineralization and energy consumption.

  • From mechanistic study, the coexistence of heterogeneous (surface reactions) and homogeneous reactions (activation of H2O2 in solution by the iron leached) helps to explain the outstanding degradation results exhibited by the MCF/Fe-incorporated cathode.

  • Recyclability of MCF-Fe incorporated with controlled iron leaching points out that heterogeneous supported iron can be used in several consecutive cycles proving that cathode easy-to-prepare is a suitable alternative material for the degradation of contaminants from aqueous solutions.

Finally, the application of the iron-modified carbon-based materials, prepared by a simple, facile and low-cost methodology, under the HEF treatment process is a highly efficient alternative technique for the degradation and mineralization of organic pollutants in aqueous medium under pH conditions close to neutral; this technique helps expand the application range of the treatment process.