Fast Degradation of Rhodamine B by In Situ H2O2 Fenton System with Co and N Co-Doped Carbon Nanotubes

In this study, an E-fenton oxidation system based on Co-N co-doped carbon nanotubes (Co-N-CNTs) was designed. The Co-N-CNTs system showed fast degradation efficiency and reusability for the degradation of rhodamine B (RhB). The XRD and SEM results showed that the Co-N co-doped carbon nanotubes with diameters ranging from 40 to 400 nm were successfully prepared. The E-Fenton degradation performance of Co-N-CNTs was investigated via CV, LSV and AC impedance spectroscopy. The yield of H2O2 could reach 80 mg/L/h within 60 min, and the optimal voltage and preparation temperature for H2O2 yield in this system was −0.7 V (vs. SCE) and 800 °C. For the target pollutant of RhB, the fast removal of RhB was obtained via the Co-N-CNTS/E-Fenton system (about 91% RhB degradation occurred during 60 min), and the •OH played a major role in the RhB degradation. When the Fe2+ concentrations increased from 0.3 to 0.4 mM, the RhB degradation efficiency decreased from 91% to about 87%. The valence state of Co in the Co-N-C catalyst drove a Co2+/Co3+ cycle, which ensured the catalyst had good E-Fenton degradation efficiency. This work provides new insight into the mechanism of an E-Fenton system with carbon-based catalysts for the efficient degradation of RhB.


Introduction
The electro-Fenton process based on hydroxyl radicals has attracted widespread attention due to its environmental friendliness, high efficiency and energy-saving properties [1,2], during which, H 2 O 2 is formed in situ by the reaction of the two-electron reduction of O 2 (2e-ORR) on the cathode [3,4]. Then, the in situ-produced H 2 O 2 reacts with the metal catalyst to produce radical •OH. The potential explosion hazard in the process of H 2 O 2 production and transportation could be avoided by the on-site electrosynthesis of H 2 O 2 during the electro-Fenton process [5]. Lower energy consumption for organic pollutants' elimination could be obtained by higher H 2 O 2 selectivity and regeneration rates of Fe 2+ , which are also the two key factors to promote the E-Fenton application [6,7].
Although excellent ORR catalytic activity and high H 2 O 2 selectivity have been achieved by using noble-metal-based catalysts [8], their practical application in wastewater treatment was hindered by their scarcity and high cost [9].
Alternatively, carbon materials, especially biochar catalysts, have attracted increasing attention owing to their abundance, environmental sustainability, lost-cost and ease of fabrication [10]. In addition, carbon-based catalysts have attracted more attention due to their environmental sustainability, ease of manufacturing, low cost and abundance [11,12]. Moreover, the surface chemistry and pore structure of carbon-based catalysts can be easily adjusted to obtain more accessible active sites and faster mass transfer according to the preparation conditions [13,14]. The doping of metal, nitrogen and oxygen functional groups has been proved to be an effective strategy to improve their catalytic activity and H 2 O 2

The Preparation of Electrode and Electrocatalytic Production of H 2 O 2
The preparation of the cathode for the electrocatalytic production of H 2 O 2 was carried out as follows: 5 mg Co-N-C powders were well dispersed into a mixed solution of isopropanol (0.95 mL) and polytetrafluoroethylene suspension (PTFE, 0.05 mL, 60 wt%) [24]. Slurry-like specimens were formed by heating in an 80 • C water bath, and then, the resulting black slurry was uniformly coated on the nickel foam collector. The working electrode was then dried at 60 • C for 6 h and pressed into tablets for use. The schematic diagram of the electro-Fenton experimental process is shown in Figure 1.
The concentration of H 2 O 2 in situ produced by the cathode under different conditions was tested using a spectrophotometer using the potassium titanium oxalate photometric method [25]. Then, 1.00 mL of a H 2 SO 4 solution (3 M) and 1.00 mL of a K 2 TiO(C 2 O 4 ) 2 solution (0.05 M) were added into 2 mL of the tested solution. The mass concentration of H 2 O 2 was obtained from the tested absorbance in the absorbance curve of the standard H 2 O 2 solution [26]. The concentration of H2O2 in situ produced by the cathode under different co was tested using a spectrophotometer using the potassium titanium oxalate pho method [25]. Then, 1.00 mL of a H2SO4 solution (3 M) and 1.00 mL of a K2TiO(C2O tion (0.05 M) were added into 2 mL of the tested solution. The mass concentration was obtained from the tested absorbance in the absorbance curve of the stand solution [26].

Degradation Experiments
An E-Fenton system, using Co-N-C as a cathode and carbon as an anode, structed for the degradation of RhB (20 mg/L). A UV-vis spectrophotometer wa measure the residual concentration of RhB in the solution. The initial pH of the R tion was about 6.5, and it was adjusted to 3 by H2SO4 (0.1 M) to promote E-Fen ciency and prevent the flocculation of iron ions. Typically, the RhB solution was p with oxygen for 20 min, then followed by the E-Fenton degradation process. A time interval, solution samples of 0.9 mL were collected and filtered with a 0.45 μ brane, and then quenched with MeOH (0.1 mL) at once [13].

Characterization of Co-N-C Composite Catalysts
As shown in Figure 2a, all three XRD patterns of the Co-N-CNTs synthesiz ferent temperatures included the diffraction peaks of the graphite (at 26.4°) and c 44.2°, 51.5° and 75.6°, respectively), which perfectly matched with graphite (hexa PDF41-1487) and cubic Co (Co 0 , PDF: 15-0806) and indicated the presence of grap Co 0 .
The FTIR results in Figure 2b clearly reveal the characteristic peaks of -O and aromatic C-H groups in all three Co-N-CNTs samples. The peaks at 2922, 876 cm −1 were related to the aromatic C-H bonds, while the two peaks at 1631 cm −1 were related to the stretching vibration of asymmetric νas (-COO) and symm COO). The FTIR spectra of Co-N-CNT showed that most of the organic carboxy removed after the pyrolysis process, which further verified the effective structu formation from cobalt salts to Co-N-CNTs. The wide band at 3432 cm −1 was attr the stretching vibration of the -OH group. When the temperature increased from

Degradation Experiments
An E-Fenton system, using Co-N-C as a cathode and carbon as an anode, was constructed for the degradation of RhB (20 mg/L). A UV-vis spectrophotometer was used to measure the residual concentration of RhB in the solution. The initial pH of the RhB solution was about 6.5, and it was adjusted to 3 by H 2 SO 4 (0.1 M) to promote E-Fenton efficiency and prevent the flocculation of iron ions. Typically, the RhB solution was pretreated with oxygen for 20 min, then followed by the E-Fenton degradation process. At the set time interval, solution samples of 0.9 mL were collected and filtered with a 0.45 µm membrane, and then quenched with MeOH (0.1 mL) at once [13].

Characterization of Co-N-C Composite Catalysts
As shown in Figure 2a, all three XRD patterns of the Co-N-CNTs synthesized at different temperatures included the diffraction peaks of the graphite (at 26.4 • ) and cobalt (at 44.2 • , 51.5 • and 75.6 • , respectively), which perfectly matched with graphite (hexagonal C, PDF41-1487) and cubic Co (Co 0 , PDF: 15-0806) and indicated the presence of graphite and Co 0 .
The FTIR results in Figure 2b clearly reveal the characteristic peaks of -OH, -COO and aromatic C-H groups in all three Co-N-CNTs samples. The peaks at 2922, 1087 and 876 cm −1 were related to the aromatic C-H bonds, while the two peaks at 1631 and 1424 cm −1 were related to the stretching vibration of asymmetric ν as (-COO) and symmetric ν s (-COO). The FTIR spectra of Co-N-CNT showed that most of the organic carboxylate was removed after the pyrolysis process, which further verified the effective structure transformation from cobalt salts to Co-N-CNTs. The wide band at 3432 cm −1 was attributed to the stretching vibration of the -OH group. When the temperature increased from 700 • C to 900 • C, the intensity of the -OH group slightly decreased, which indicated that the acidity of the surface of the catalyst slightly decreased. In this case, the decreased surface acidity of the catalyst meant a hindered reduction activity of O 2 to H 2 O 2 .
The pore distribution shown in Figure 2d proved the presence of a hybrid pore structure of a few micropores and the majority of mesopores with diameters centered at about 3 to 4 nm and BET specific surface areas of 142.89, 130.23 and 147.73 m 2 /g for Co-N-CNTs-700, Co-N-CNTs-800 and Co-N-CNTs-900, respectively. As shown in Figure 2c, the N 2 adsorption-desorption isotherm of the Co-N-C was a type IV adsorption isotherm with a hysteresis loop at P/P 0 from 0.4 to 0.5, which also indicated the irregular hybrid pore structure of micropores and mesopores in the Co-N-C. 700, Co-N-CNTs-800 and Co-N-CNTs-900, respectively. As shown in Figure 2c, the N2 a sorption-desorption isotherm of the Co-N-C was a type IV adsorption isotherm with hysteresis loop at P/P0 from 0.4 to 0.5, which also indicated the irregular hybrid pore str ture of micropores and mesopores in the Co-N-C. As shown in Figure 3a-c, numerous carbon nanotubes were formed after pyroly for all three specimens of Co-N-CNTs synthesized at different temperatures. The walls the nanotubes synthesized at 700 °C and 900 °C were rough and uneven, while the wa of those synthesized at 800 °C were smooth and uniform. The diameters of the nanotub ranged from 40 to 400 nm, while the diameters of the nanotubes synthesized at 800 were more even. As shown in Figure 3d, cobalt nanoparticles encapsulated at the enclos end of CNTs were also revealed by the images. The energy dispersive X-ray spectrosco (EDS) elemental mapping of a selected area on the Co-N-C composite (Figure 3e-i) vealed the existence of cobalt nanoparticles at the enclosed end of CNTs and uniform d tributions of the N, O and C elements across the tubular structure, which demonstrat the successful incorporation of Co, N and O into the carbon substrate. During the proc of pyrolysis, the entrapped cobalt acted as a catalyst to promote the formation of carb nanotubes [4,27]. As shown in Figure 3a-c, numerous carbon nanotubes were formed after pyrolysis for all three specimens of Co-N-CNTs synthesized at different temperatures. The walls of the nanotubes synthesized at 700 • C and 900 • C were rough and uneven, while the walls of those synthesized at 800 • C were smooth and uniform. The diameters of the nanotubes ranged from 40 to 400 nm, while the diameters of the nanotubes synthesized at 800 • C were more even. As shown in Figure 3d, cobalt nanoparticles encapsulated at the enclosed end of CNTs were also revealed by the images. The energy dispersive X-ray spectroscopy (EDS) elemental mapping of a selected area on the Co-N-C composite (Figure 3e-i) revealed the existence of cobalt nanoparticles at the enclosed end of CNTs and uniform distributions of the N, O and C elements across the tubular structure, which demonstrated the successful incorporation of Co, N and O into the carbon substrate. During the process of pyrolysis, the entrapped cobalt acted as a catalyst to promote the formation of carbon nanotubes [4,27].
As shown in Figure 4a, the X-ray photon spectroscopy (XPS) results also confirmed the existence of Co (0.80 at%), N (3.99 at%), O (3.40 at%) and C (91.81 at%) on the Co-N-C. The Co 2p XPS spectra of the freshly prepared Co-N-C also supported the existence of Co 0 (peak at 778.5 eV), as well as Co-N (peak at 780.5 eV, Co 2+ (peak at 782.8 eV) and Co 3+ (peak at 795.5 eV), which was consistent with the XRD result ( Figure 2a). The deconvolution results of C1s of the Co-N-C included four peaks attributed to C-C/C=C (at 284.8 eV), C-N (at 285.9 eV), C-O (at 286.8 eV) and N = C-N (at 287.5 eV). The N 1s of Co-N-C was deconvoluted into four peaks (402.52 ± 0.2 eV, 401.10 ± 0.2, 399.36 ± 0.2 and 398.64 ± 0.2, respectively, shown in Figure 4c), which corresponded to oxidized N, graphitic N, pyrrolic N and pyridinic N, respectively. Previous studies have reported that the introduction of doped nitrogen in CNTS could facilitate electron transport. The introduction of C-NHx into adsorbents could also boost the adsorptive removal of organics due to its outstanding capability for hydrogen bonding [26].  Figure 4a, the X-ray photon spectroscopy (XPS) results also confirm the existence of Co (0.80 at%), N (3.99 at%), O (3.40 at%) and C (91.81 at%) on the Co-C. The Co 2p XPS spectra of the freshly prepared Co-N-C also supported the existence Co 0 (peak at 778.5 eV), as well as Co-N (peak at 780.5 eV, Co 2+ (peak at 782.8 eV) and C (peak at 795.5 eV), which was consistent with the XRD result (Figure 2a). The deconvo tion results of C1s of the Co-N-C included four peaks attributed to C-C/C=C (at 284.8 e C-N (at 285.9 eV), C-O (at 286.8 eV) and N = C-N (at 287.5 eV). The N 1s of Co-N-C w deconvoluted into four peaks (402.52 ± 0.2 eV, 401.10 ± 0.2, 399.36 ± 0.2 and 398.64 ± 0 respectively, shown in Figure 4c), which corresponded to oxidized N, graphitic N, p rolic N and pyridinic N, respectively. Previous studies have reported that the introduct of doped nitrogen in CNTS could facilitate electron transport. The introduction of C-N into adsorbents could also boost the adsorptive removal of organics due to its outstandi capability for hydrogen bonding [26].  As shown in Figure 4a, the X-ray photon spectroscopy (XPS) results also confirmed the existence of Co (0.80 at%), N (3.99 at%), O (3.40 at%) and C (91.81 at%) on the Co-N-C. The Co 2p XPS spectra of the freshly prepared Co-N-C also supported the existence of Co 0 (peak at 778.5 eV), as well as Co-N (peak at 780.5 eV, Co 2+ (peak at 782.8 eV) and Co 3+ (peak at 795.5 eV), which was consistent with the XRD result (Figure 2a). The deconvolution results of C1s of the Co-N-C included four peaks attributed to C-C/C=C (at 284.8 eV), C-N (at 285.9 eV), C-O (at 286.8 eV) and N = C-N (at 287.5 eV). The N 1s of Co-N-C was deconvoluted into four peaks (402.52 ± 0.2 eV, 401.10 ± 0.2, 399.36 ± 0.2 and 398.64 ± 0.2, respectively, shown in Figure 4c), which corresponded to oxidized N, graphitic N, pyrrolic N and pyridinic N, respectively. Previous studies have reported that the introduction of doped nitrogen in CNTS could facilitate electron transport. The introduction of C-NHx into adsorbents could also boost the adsorptive removal of organics due to its outstanding capability for hydrogen bonding [26].   Figure 5 shows the CV, LSV and AC impedance spectroscopy curves of the Co-N-C electrode synthesized at 700 °C, 800 °C and 900 °C under oxygen saturation conditions. According to Figure 5, after oxygen permeation, all the three electrodes prepared by Co-N-C catalysts showed obvious O2 reductions at about −0.2 V to −0.3 V (vs. SCE), while there were no obvious O2 reductions under N2 saturation conditions. The LSV results were also consistent with the CV curves. From Figure 5c, we could see that semicircles appeared in the high-frequency region of the impedance diagram of all the three electrodes, indicating an electrochemically controlled step. In addition, the impedance arc radius of the Co-N-C electrode synthesized at 800 °C was the smallest, which indicated that this electrode had the smallest resistance (interface resistance of solution (Rs = 4.315 Ω) and contact resistance (RCT = 11.05 Ω)) and the best electron transfer efficiency and electro-catalytic performance of oxygen reduction compared to other electrodes in the electrochemical reaction process. These results implied the Co-N-C electrode synthesized at 800 °C would have had the best EF degradation efficiency for RhB.   Figure 5 shows the CV, LSV and AC impedance spectroscopy curves of the Co-N-C electrode synthesized at 700 • C, 800 • C and 900 • C under oxygen saturation conditions. According to Figure 5, after oxygen permeation, all the three electrodes prepared by Co-N-C catalysts showed obvious O 2 reductions at about −0.2 V to −0.3 V (vs. SCE), while there were no obvious O 2 reductions under N 2 saturation conditions. The LSV results were also consistent with the CV curves. From Figure 5c, we could see that semicircles appeared in the high-frequency region of the impedance diagram of all the three electrodes, indicating an electrochemically controlled step. In addition, the impedance arc radius of the Co-N-C electrode synthesized at 800 • C was the smallest, which indicated that this electrode had the smallest resistance (interface resistance of solution (R s = 4.315 Ω) and contact resistance (R CT = 11.05 Ω)) and the best electron transfer efficiency and electro-catalytic performance of oxygen reduction compared to other electrodes in the electrochemical reaction process. These results implied the Co-N-C electrode synthesized at 800 • C would have had the best EF degradation efficiency for RhB.

Electrochemical Performance Evaluation and Production Efficiency of H 2 O 2
The yield of H 2 O 2 was measured for the Co-N-C electrode synthesized at 800 • C under different applied voltages (as shown in Figure 6). There was little difference between the yield of H 2 O 2 under different applied voltages, and the yield of H 2 O 2 could reach 80 mg/L/h within 60 min, which was higher than that in many reported studies. From the literature, we know that a higher voltage can promote proton-coupled electron transfer in the formation of H 2 O 2 , while a side reaction of HER will also be promoted under higher voltages. Considering the influence of energy consumption and the HER side reaction, the optimal voltage for H 2 O 2 yield in this system was −0.7 V (vs. SCE).
Co-N-C electrode synthesized at 800 °C was the smallest, which indicated that this el trode had the smallest resistance (interface resistance of solution (Rs = 4.315 Ω) and cont resistance (RCT = 11.05 Ω)) and the best electron transfer efficiency and electro-cataly performance of oxygen reduction compared to other electrodes in the electrochemical action process. These results implied the Co-N-C electrode synthesized at 800 °C wou have had the best EF degradation efficiency for RhB.  The yield of H2O2 was measured for the Co-N-C electrode synthesized der different applied voltages (as shown in Figure 6). There was little differ the yield of H2O2 under different applied voltages, and the yield of H2O2 co mg/L/h within 60 min, which was higher than that in many reported stud literature, we know that a higher voltage can promote proton-coupled electr the formation of H2O2, while a side reaction of HER will also be promoted voltages. Considering the influence of energy consumption and the HER side optimal voltage for H2O2 yield in this system was −0.7 V (vs. SCE).

Efficient E-Fenton Degradation of RhB
The E-Fenton degradation efficiency of different systems for the remova explored. Figure 7a shows that all the Co-N-C displayed efficient E-Fenton rates of RhB, while that synthesized at 800 °C exhibited the highest RhB degr removal in 1 h). The reaction rate constants for the Co-N-C synthesized at 7 and 900 °C were 0.0353, 0.0366, and 0.0246, respectively. Figure 7c shows th N-C displayed efficient E-Fenton degradation rates of RhB, while that sy 800 °C exhibited the highest RhB degradation (91% removal in 1 h). The reac stants for the Co-N-C system with different Fe 2+ additions at 0.2 mM, 0.3 mM were 0.0228, 0.0366 and 0.0309, respectively. The highest RhB degradation tained with 0.3 mM Fe 2+ . During this process, the •OH could be generated f Co 2+ /Co 3+ cycle and the Fe 2+ /Fe 3+ cycle, which quickened the degradation of to the Co-Fe synergy effect. An increase in Fe 2+ concentration can promote t action to produce more ·OH, while an excess of Fe 2+ will compete electrons w reduce the contact of ·OH with organic compounds, which will lead to a de degradation efficiency of RhB.

Efficient E-Fenton Degradation of RhB
The E-Fenton degradation efficiency of different systems for the removal of RhB was explored. Figure 7a shows that all the Co-N-C displayed efficient E-Fenton degradation rates of RhB, while that synthesized at 800 • C exhibited the highest RhB degradation (91% removal in 1 h). The reaction rate constants for the Co-N-C synthesized at 700 • C, 800 • C and 900 • C were 0.0353, 0.0366, and 0.0246, respectively. Figure 7c shows that all the Co-N-C displayed efficient E-Fenton degradation rates of RhB, while that synthesized at 800 • C exhibited the highest RhB degradation (91% removal in 1 h). The reaction rate constants for the Co-N-C system with different Fe 2+ additions at 0.2 mM, 0.3 mM and 0.4 mM were 0.0228, 0.0366 and 0.0309, respectively. The highest RhB degradation rate was obtained with 0.3 mM Fe 2+ . During this process, the •OH could be generated from both the Co 2+ /Co 3+ cycle and the Fe 2+ /Fe 3+ cycle, which quickened the degradation of the RhB due to the Co-Fe synergy effect. An increase in Fe 2+ concentration can promote the Fenton reaction to produce more ·OH, while an excess of Fe 2+ will compete electrons with ·OH and reduce the contact of ·OH with organic compounds, which will lead to a decrease in the degradation efficiency of RhB.
were 0.0228, 0.0366 and 0.0309, respectively. The highest RhB degradation rate was ob tained with 0.3 mM Fe 2+ . During this process, the •OH could be generated from both th Co 2+ /Co 3+ cycle and the Fe 2+ /Fe 3+ cycle, which quickened the degradation of the RhB du to the Co-Fe synergy effect. An increase in Fe 2+ concentration can promote the Fenton re action to produce more ·OH, while an excess of Fe 2+ will compete electrons with ·OH an reduce the contact of ·OH with organic compounds, which will lead to a decrease in th degradation efficiency of RhB. A possible RhB degradation mechanism by the Co-N-C mediated E-Fenton system i illustrated in the following equations. In terms of the XPS results, Co-N-C contained Co 0 Co 2+ and Co 3+ . In addition to the Co ions, the metallic Co 0 also played a role in activating E-Fenton. H2O2 and •OH could be generated from a Co 2+ /Co 3+ redox cycle of the cobal element during the E-Fenton process (Equations (1)-(4)), which ensured a high remova rate of RhB in this system [4]. The EPR results in Figure 7e reveal the existence of •OH, in which the four-line signal with an intensity ratio of 1:2:2:1 was attributed to DMPO-•OH The active intermediate •OH could be generated from the self-formation process of E Fenton. As a result, RhB molecules were degraded into intermediates and inorganic smal molecules by these various ROSs involved in this system (Equations (5) and (6)) [28,29].
A possible RhB degradation mechanism by the Co-N-C mediated E-Fenton system is illustrated in the following equations. In terms of the XPS results, Co-N-C contained Co 0 , Co 2+ and Co 3+ . In addition to the Co ions, the metallic Co 0 also played a role in activating E-Fenton. H 2 O 2 and •OH could be generated from a Co 2+ /Co 3+ redox cycle of the cobalt element during the E-Fenton process (Equations (1)-(4)), which ensured a high removal rate of RhB in this system [4]. The EPR results in Figure 7e reveal the existence of •OH, in which the four-line signal with an intensity ratio of 1:2:2:1 was attributed to DMPO-•OH. The active intermediate •OH could be generated from the self-formation process of E-Fenton. As a result, RhB molecules were degraded into intermediates and inorganic small molecules by these various ROSs involved in this system (Equations (5) and (6)) [28,29].

Reusability of Co-N-CNTS
The reusability of Co-N-CNTs was studied by fulfilling cycling tests. As shown in Figure 8, after three cycling runs, the degradation efficiency of RhB decreased from about 89.4% to 85.8%, and the degradation rate of RhB only decreased by about 4%, which proved that the Co-N-CNTs had very good reusability. The above result indicates that during the process of RhB degradation, the Co nanoparticles were tightly embedded in the carbon nanotube and drove a Co 2+ /Co 3+ cycle to facilitate the E-Fenton degradation of RhB.
Materials 2023, 16, x FOR PEER REVIEW proved that the Co-N-CNTs had very good reusability. The above result indica during the process of RhB degradation, the Co nanoparticles were tightly embe the carbon nanotube and drove a Co 2+ /Co 3+ cycle to facilitate the E-Fenton degrad RhB.

Conclusions
An E-Fenton oxidation system based on Co-N-CNTs with a hybrid pore stru a few micropores and the majority of mesopores with diameters centered at abou nm and large BET specific surface areas was prepared, which enabled its good d tion efficiency for the degradation of RhB (91% at 60 min). The diameters of th CNTs with cobalt nanoparticles encapsulated at the enclosed end ranged from 4 nm, and the yield of H2O2 could reach 80 mg/L/h at −0.7 V (vs. SCE). The Co-N-CN had the smallest interface resistance of solution (Rs = 4.32 Ω) and contact resistanc 11.05 Ω), which meant it had the best electron transfer efficiency and electro-cataly formance of oxygen reduction. When the Fe 2+ was added, the •OH could be ge

Conclusions
An E-Fenton oxidation system based on Co-N-CNTs with a hybrid pore structure of a few micropores and the majority of mesopores with diameters centered at about 3 to 4 nm and large BET specific surface areas was prepared, which enabled its good degradation efficiency for the degradation of RhB (91% at 60 min). The diameters of the Co-N-CNTs with cobalt nanoparticles encapsulated at the enclosed end ranged from 40 to 400 nm, and the yield of H 2 O 2 could reach 80 mg/L/h at −0.7 V (vs. SCE). The Co-N-CNTs-800 had the smallest interface resistance of solution (R s = 4.32 Ω) and contact resistance (R CT = 11.05 Ω), which meant it had the best electron transfer efficiency and electro-catalytic performance of oxygen reduction. When the Fe 2+ was added, the •OH could be generated from both the Co 2+ /Co 3+ cycle and the Fe 2+ /Fe 3+ cycle, which quickened the degradation of the RhB due to the Co-Fe synergy effect. After three cycling runs, 96% of the degradation efficiency of RhB could be retained. The core-shell structure (Co nanoparticles encapsulated in carbon nanotubes) could prevent Co from leaching, which enabled its good reusability and convenient recycling for its magnetism.