Enhanced PMS Activation by Highly Dispersed Mn-Ce Bimetallic Oxide on Carbon Nanotubes for Degradation of Phenol

Abstract

Peroxymonosulfate (PMS) activation is an intriguing technology for refractory organic pollutant removal in wastewater treatment. Herein, a highly dispersed Mn-Ce bimetallic oxide on carbon nanotubes (MCC) was synthesized and applied to catalyze PMS for the degradation of phenol. The material was well characterized using a transmission electron microscope (TEM), N₂ adsorption–desorption isotherms, X-ray powder diffraction (XRD), and X-ray photoelectron spectroscopy (XPS). The synthesized MCC showed superior activity for PMS activation. The k value of phenol removal with MCC is 0.135 min⁻¹, which is greatly superior to that of CNT (6.17 × 10⁻⁵ min⁻¹) and Mn-Ce bimetallic oxide (3.18 × 10⁻⁴ min⁻¹). Electron paramagnetic resonance (EPR), along with radical quenching experiments, revealed that the activation of PMS by MCC for phenol degradation involves both radical and non-radical reaction pathways. Moreover, a synergic effect between Mn-Ce bimetallic oxide and CNT was identified to be responsible for the outstanding catalytic activity.

1. Introduction

Recently, advanced oxidation technologies (AOTs) founded on sulfate radicals (SO₄^•−) have been capturing rising attention owing to the high oxidative capacity (2.5–3.1 V), long-term survivability (30–40 μs), and broad pH flexibility (2–8) [1,2,3]. Typically, SO₄^•− is produced from catalyzing persulfate (PS) by heating [4], UV irradiation [5], and catalysts [6,7,8,9,10]. The generation of SO₄^•− from activating PS is accompanied by the O–O bond break. Compared with peroxydisulfate (PDS), PMS is recognized more easily to be dissociated to produce SO₄^•− because of its weaker O–O bond [11]. Up to the present, various metal-based catalysts have been employed for PMS activation, and Co²⁺ was proven to be the most effective homogeneous catalyst [12]. However, Co²⁺ is difficult to recycle and reuse, restricting its application. Heterogenous catalysts based on cobalt oxides such as Co₃O₄ [13], CoMn₂O₄ [14], and ZnCo₂O₄ [15] also displayed good catalytic performance for PMS activation. However, the relatively high toxicity and potential carcinogenicity of Co limits its practical application for environmental remediation. Hence, the development of environmentally friendly and high-efficiency catalysts becomes a priority for PMS activation.

Manganese-based catalysts have been supposed to be superior substitutes for PMS activation because they are abundant, low-cost, and eco-friendly, but their catalytic activities are generally low [16]. Even Mn₂O₃, which is considered the most efficient Mn-based catalyst, cannot completely remove 2,4-DCP during PMS activation [17]. Several Mn-based catalysts achieved complete pollutant degradation, yet at the expense of a high dose of catalysts or oxidants [18]. Thus, effective strategies should be adopted to construct high-efficiency Mn-based catalysts for PMS activation. Previous works have demonstrated that the construction of bimetallic oxide structures is available for enhancing the catalytic performance owing to a synergism between the different metal species [19]. Recently, various bimetallic oxides such as MnFe₂O₄ [20], Mn-Fe-mixed Oxide [21], Ce-doped Mn₂O₃ [22], and Co₂TiO₄ [23] showed their outstanding catalytic performance for PMS activation. Zhang et al. [24] presented a method to improve the activity of Co₃O₄ towards PMS activation by regulating the electronic structure through single-atom Zr doping. Zhang and co-workers [25] employed PdO/CeO₂ for catalytic ozonation. The Pd-Ce bimetallic oxide showed higher activity than that of a monometallic oxide, as the synergism between PdO and CeO₂ allows the produced active species to react preferentially with the adsorbed oxalic acid. Huang and co-workers [26] found the performance of the Mn-Fe bimetallic oxide was notably higher than that of Mn₃O₄ and Fe₃O₄ owing to the synergistic effect between Mn (active site) and Fe (adsorption site). Tian and co-workers [22] prepared Ce-doped Mn₂O₃ for PMS activation and found the presence of Ce could enhance the electron transfer, leading to increased catalytic activity. Wu et al. [24] studied the catalytic performance of Co, Cu, and Ce-doped manganese oxide towards PMS activation. The results indicated that Co and Ce doping could accelerate the redox cycle of Mn with a different valent, which is beneficial for the conversion of PMS into active species. Additionally, our previous study also revealed that Mn-Ce bimetallic oxide illustrated a greater catalytic activity than Mn/Ce monometallic oxide towards catalytic ozonation [27]. Consequently, it was speculated that Mn-X bimetallic oxides could give rise to generating synergistic effects and enhanced catalytic performance towards PMS activation.

Herein, in this study, highly dispersed Mn-Ce bimetallic oxide on CNT was synthesized and used for PMS activation. CNTs possess a large outer surface for anchoring Mn-Ce bimetallic oxides. The highly dispersed metal oxide on CNT could bring more exposed active sites for PMS activation, as well as reduce the mass transfer resistance of a reactant, which thereafter could result in its superior catalytic activity. Moreover, the abundant free electrons in CNT can provide an impetus for the electron transfer in the reaction, which is conducive to the conversion of high-valent Mn and Ce species into low-valent species. The activity was evaluated in a search for phenol, a typical refractory pollutant. The influences of reaction parameters, including MCC dosage, PMS concentration, solution pH, and reaction temperature, on catalytic activity were investigated. To acquire more information on the variable effects involved in the catalytic process, response surface methodology (RSM) using a Box–Behnken design (BBD) was carried out. The possible generated reactive species were identified using EPR measurements and quenching experiments. Moreover, the catalytic mechanism was delved and proposed.

2. Materials and Methods

2.1. Chemicals

All the reagents were used in their original status without further purification. Multi-wall carbon nanotubes (carboxylated, length < 15 μm, outer diameter: 30–50 nm, purity: >98%), cerium (Ⅲ) nitrate hexahydrate (Ce(NO₃)₃·6H₂O), potassium permanganate (KMnO₄), sodium hydrogen carbonate (NaHCO₃), sodium hydroxide (NaOH), tert-butyl alcohol (TBA), methanol, ethanol, potassium monopersulfate triple salt (KHSO₅·0.5KHSO₄·0.5K₂SO₄, PMS), and phenol were supplied by Aladdin Bio-Chem Technology Co., Ltd. (Shanghai, China). The utilization of ultrapure water (resistivity of 18.25 MΩ·cm) generated by an ultra-pure water system (HOKEE, Hefei, China) penetrated all experiments.

2.2. Synthesis of MCC

The highly dispersed Mn-Ce bimetallic oxide on carbon nanotubes was synthesized by an oxidation-reduction reaction of Mn (VII) and Ce (III). In brief, 0.4 g of carboxylated-CNT was added to 90 mL of H₂O and treated by ultrasonic dispersion for 30 min. Then, 0.0923 g of Ce(NO₃)₃·6H₂O was mixed well with the above suspension. The KMnO₄ (0.08 g) was dissolved in 10 mL of H₂O and added gradually to the above suspension by a syringe under continuous stirring. Afterward, the pH of the mixture was altered to 6.0, and the reaction was maintained at 60 °C. After 2 h of reaction, the sediment was collected via vacuum filtration and purified by water till the filter liquor approached pH 7.0. Finally, the obtained product was dried at 60 °C for 2 h and calcinated at 300 °C for 3 h in air. The highly dispersed Mn-Ce bimetallic oxide on CNT was achieved and designated as MCC. For comparison, Mn-Ce bimetallic oxide designated as MnCeO was also synthesized through the same process except for the absence of CNT.

2.3. Characterization

The morphologies and structures of the as-obtained MCC were characterized utilizing a Talos F200S TEM (FEI, Alhambra, CA, USA) and a SmartLab SE XRD (Rigaku, Tokyo, Japan). The surface chemical composition and elemental valence state of MCC before and after the reaction was analyzed through XPS (Thermo Scientific K-Alpha, Waltham, MA, USA) analysis. The surface area and average pore size of materials were investigated by a fully automatic surface area analyzer (Micromeritics ASAP2020, Norcross, GA, USA). The chemical composition of MCC was studied by ICP-MS (PerkinElmer 8300, Waltham, MA, USA).

2.4. Phenol Removal Tests

The catalytic activity of synthesized MCC towards PMS activation was studied by the degradation of phenol. All experiments were performed in a beaker holding 100 mL of 20 mg L⁻¹ phenol. Typically, 0.02 g of MCC was first dispersed in the solution by sonication for 10 min. Then, 0.02g of PMS was mixed with the above-mentioned suspension under continuous agitating to trigger the reaction. Next, at a predetermined time, samples were withdrawn and quenched immediately with methanol (volume ratio of 10:1). Finally, the filtrate was filtrated for HPLC detection.

2.5. Analysis

The concentrations of phenol during the reaction were determined by an LC16 HPLC-UV (Shimadzu, Kyoto, Japan) coupled with a C18 column. The mobile phase was made of methanol and ultra-pure water (volume ratio of 7:3), and the velocity of flow was 0.8 mL min⁻¹. The radicals generated during the reaction were measured by an EMXplus-6/1 EPR spectrometer (Bruker, Mannheim, Germany) with DMPO employed as a spin-trapping agent.

3. Results and Discussion

3.1. Physicochemical Characteristics of Catalysts

The morphology of the prepared MCC was characterized by TEM. As shown in Figure 1a, no obvious nanoparticles belonging to Mn-Ce bimetallic oxide can be observed on the surface of CNT. It indicates that the synthesized Mn-Ce bimetallic oxide is highly dispersed on CNT, which is conducive to generating more exposed active sites. By further increasing the magnification of the TEM image, particles with a particle size of approximately 5.4 nm can be observed (the inset of Figure 1a). To further investigate the dispersion of Mn and Ce on the CNT surface, EDX analysis was carried out and shown in Figure 1b–f. As can be seen, Mn and Ce species are uniformly embedded in the carbon matrix. The Mn-Ce bimetallic oxide supported on the outer surfaces of CNT are well accessible, enabling the prepared material excellent performance in PMS activation.

Figure 2a illustrates the XRD patterns of pristine CNT, MCC, and MnCeO. The broad diffraction peak located at around 26° and the weak peak at around 43° correspond to the (002) reflections and the diffraction plane (100) (PDF# 41-1487), respectively [28]. Moreover, no obvious diffraction peaks belonging to Mn-Ce bimetallic oxide can be observed. This phenomenon could be owed to the low load contents of metal species on the CNT surface. The XRD data was analyzed using Jade 6.5 software. The lattice constant was calculated to be 0.242 nm, and the density of the material was 2.16 g cm⁻³.

Figure 2b shows the N₂ adsorption–desorption isotherms of different materials. The BET-specific surface areas (SSAs) and pore size of pristine CNT, MCC, and MnCeO were determined and shown in

Table 1. Structural characteristics of different samples.

Materials	SSAs (m2 g−1)	Average Pore Diameter (nm)	Pore Volume (cm3 g−1)
pristine CNT	127.7	11.2	0.36
MCC	177.1	7.5	0.33
MnCeO	158.0	6.6	0.26

. The calculated SSAs of CNT, MCC, and MnCeO were 127.7, 177.1, and 158.0 m² g⁻¹, respectively. The average pore diameter of the prepared materials ranges from 6.6 to 11.2 nm. The surface chemical composition of MCC was gained from XPS measurements. The survey XPS spectrum (Figure 2d) further confirms the existence of Mn and Ce species on CNT. The atomic ratio of Mn and Ce is calculated to be 3.36% and 1.30%, respectively. Thus, the stoichiometric ratio of Mn, Ce, and O for MCC is calculated to be 0.72:0.28:1.6 (Mn_0.72Ce_0.28O_1.6).

3.2. Catalytic Performance of Catalysts

The activity of MCC for PMS activation was studied using phenol degradation. The reaction was controlled at room temperature and pH ~ 7.0. As depicted in Figure 3a, PMS used alone shows little phenol removal. Similarly, with the simultaneous presence of the PMS and CNT or MnCeO, less than 5% phenol removal was achieved in 40 min. In contrast, when the PMS and MCC were employed together, complete phenol removal was reached in 30 min of reaction. The pseudo-first-order kinetic constants (k) of different reaction systems were fitted and displayed in Figure 3b. The k values of phenol degradation during MCC/PMS, CNT/PMS, and MCC/PMS systems are 0.135 min⁻¹, 6.17 × 10⁻⁵ min⁻¹, and 3.18 × 10⁻⁴ min⁻¹, respectively. The above results indicate that the catalytic performance of the prepared MCC for PMS activation was significantly better than those of CNT or MnCeO. Additionally, the desorption capacity of phenol on MCC was determined and showed less than 3% of phenol removal, revealing the superior catalytic activity of MCC.

The influences of various experimental factors, including catalyst dose, oxidant concentration, solution pH, and reaction temperature, on phenol removal were also tested. Figure 4a exhibits that a higher phenol removal efficiency was achieved at an increased catalyst dosage. With the dosage increasing from 0.1 to 0.5 g L⁻¹, the phenol degradation efficiency is enhanced from 58% to 100% (after 30 min of reaction). Additionally, the corresponding k value rises from 0.028 min⁻¹ at 0.1 g L⁻¹ to 0.23 min⁻¹ at 0.50 g L⁻¹ (Figure 4d). This phenomenon could be driven by the fact that the increased MCC could bring more active sites to get involved in the PMS activation. Similarly, as shown in Figure 4e, the k value rises from 0.028 min⁻¹ ([PMS] = 0.5 mmol L⁻¹) to 0.23 min⁻¹ ([PMS] = 0.8 mmol L⁻¹) as the concentration of PMS increased. Nevertheless, when the concentration of PMS was further increased to 1.0 mmol L⁻¹, no obvious enhancement in phenol removal was detected. Notably, the removal of phenol even illustrates a slight decrease when further increasing the PMS concentration to 1.5 mmol L⁻¹. The phenomenon also occurred during the activation of PMS using nitrogen-doped porous carbon [1]. It might be attributed to the self-quenching reaction of SO₄^•− by HSO₅⁻ (Equation (1)). The surplus PMS could react with the generated SO₄^•− and convert them into SO₅^•− with low oxidation potential, which thereby reduces the removal efficiency of phenol [29]. Accordingly, 0.8 mmol L⁻¹ of PMS was employed in the follow-up reactions.

$${\mathrm{HSO}}_5^{-}+{\mathrm{SO}}_4^{•-}\to {\mathrm{SO}}_5^{•-}+{ \mathrm{SO}}_4^{2-}+{\mathrm{H}}^{+}$$

(1)

The influence of solution pH on phenol removal was studied and is shown in Figure 4c. The initial solution was adjusted to a wide range of around 4–10, covering the pH of most actual water (acidic, neutral, or alkaline). As can be seen, the phenol degradation efficiencies exhibit negligible changes under initial solution pH of 4.1, 5.5, 7.0, 8.8, and 10.4. This result indicates that the effect of solution pH on phenol removal is not significant during MCC/PMS system, implying the great potential in actual water remediation.

As is well known, reaction temperature also plays a critical role during PMS activation. As displayed in Figure 5, the removal of phenol shows a positive dependence on the reaction temperature. The k values of the catalytic reaction were 0.062, 0.135, 0.178, and 0.198 min⁻¹ at 15, 25, 35, and 45 °C, respectively. To define the temperature sensitivity of the catalytic process, activation energy (Ea) was computed using Equation (2). On account of the correlation between ln(k) and 1/T, the Ea for phenol removal during MCC/PMS system was calculated to be 28.9 kJ mol⁻¹, which is lower than those of Co₃O₄@CNT (35.8 kJ mol⁻¹) [30], Fe-Co hydroxide (59.71 kJ mol⁻¹) [31], N-doped Co (51 kJ mol⁻¹) [32], α-MnO₂ (38.7 kJ mol⁻¹) [18], and α-Mn₂O₃ (44.9 kJ mol⁻¹) [18]. The result implies that the prepared MCC possesses a more constructive performance for PMS activation.

$$\ln k=\mathrm{lnA}-\frac{{\mathrm{E}}_{\mathrm{a}}}{\mathrm{R}}\left(\frac{1}{\mathrm{T}}\right)$$

(2)

To study the significance of the parameters and the interactions of variable effects, including MCC dosage (A), PMS concentration (B), initial solution pH (C), and reaction temperature (D), on the removal efficiencies of phenol during the catalytic processes, RSM using BDD was carried out. The experimental results were analyzed by Design-Expert 13, in which 29 sets of experiments were conducted. The ANOVA method for the quadratic model was performed to evaluate the significance of variable effects. The values of R², adjusted R², and predicted R² are 0.9905, 0.9810, and 0.9542, respectively. The high value of R² and the little difference between adjusted R² and predicted R² (less than 0.2) indicate that the correlation of the variable effects is well described by the model.

The model F-value of 23.18 indicates the model is significant. In addition, the F-values of A, B, C, and D are 568.5, 322.55, 6.80, and 45.17, respectively. This suggests that the predicted significance of different factors on phenol removal efficiency is as follows: A > B > D > C. Thus, in this case, MCC dosage, PMS concentration, and reaction temperature are significant model terms, while initial solution pH with a p-value of 0.0206 has little effect on phenol removal efficiency. These results are consistent with the above experimental results. Furthermore, the interactions of variable factors on phenol removal efficiencies were analyzed using three-dimensional (3D) response surface plots and depicted in Figure 6. The dosage of MCC and PMS significantly affects the removal efficiency of phenol because they play a critical role in the generation of active species. As displayed in Figure 6a–e, the phenol removal efficiency is altered remarkably owing to the interaction between MCC and PMS. Increasing the amount of MCC or PMS used can lead to an increase in the removal efficiency of phenol because more active sites and oxidants are available for the reaction. Nevertheless, a further increase in the amount of MCC or PMS becomes ineffective or even detrimental to the catalytic process. The reaction temperature also illustrates a positive effect on the efficiency of phenol removal. Additionally, little effects on phenol removal can be observed under a wide initial solution pH range, further suggesting the potential application of the prepared MCC for actual water treatment.

3.3. Active Species Generated during the Catalytic System

Given the fact that •OH and SO₄^•− could be generated in PMS activation by metal-based materials [3,16], EPR measurements were performed to study the radicals involved in the catalytic system. DMPO was utilized as the spin-trapping agent for both •OH and SO₄^•−. As shown in Figure 7a, with the presence of PMS alone, no characteristic peaks assigned to •OH or SO₄^•− were identified. This indicates PMS alone does not generate any detectable •OH or SO₄^•−. In stark contrast, a seven-fold peak with a relative intensity of 1:2:1:2:1:2:1 assigning to DMPOX could be observed when PMS and MCC were added simultaneously. DMPOX is an oxidation product of DMPO reacting with strong oxidants. Thus, the characteristic peaks of DMPOX might have contributed to the presence of •OH or SO₄^•− during MCC/PMS system. Remarkably, Figure 7b shows that the intensity of the seven-fold peak becomes much stronger with time elevated from 5 min to 10 min, indicating the generation and accumulation of more reactive species in the system.

To further penetrate the possible radicals in the MCC/PMS system, quenching experiments were also performed. TBA was chosen as the quencher of •OH because its affinity for •OH (k(•OH) = (3.8–7.6) × 10⁸ M⁻¹ s⁻¹) is much higher than that for SO₄^•− (k(SO₄^•−) = (4.0–9.1) × 10⁵ M⁻¹ s⁻¹). Methanol reacts quite fast with the radicals and was used for quenching both •OH and SO₄^•− (k(•OH) = (1.2–2.8) × 10⁹ M⁻¹ s⁻¹, k(SO₄^•−) = (1.6–7.7) × 10⁷ M⁻¹ s⁻¹) during MCC/PMS system. As exhibited in Figure 8, the phenol removal efficiency was inhibited remarkably in the attendance of TBA or methanol. The k values reduced from 0.135 to 0.06 and 0.04 min⁻¹, respectively, when 0.5 mol L⁻¹ TBA or methanol was added. This result makes further efforts to identify that •OH and SO₄^•− participated in the degradation of phenol during the MCC/PMS system. Nevertheless, the k values could still reach 0.05 and 0.02 min⁻¹, respectively, when 1 mol L⁻¹ TBA or methanol is added. These phenomena indicate that the activation of PMS by MCC for phenol degradation involves both radical and non-radical pathways, which is in line with the previous studies [22].

3.4. The Synergistic Effect between Bimetallic Oxide and CNT

To look into the potential synergic effect of bimetallic oxide and CNT during the PMS activation process, the performance of mechanical mixed MnCeO and CNT for phenol removal was also studied. As displayed in Figure 9, the mixture performed 4.2% removal of phenol in 40 min. The catalytic efficiency shows little enhancement compared to CNT- or MnCeO-catalyzed PMS. Furthermore, the removal of phenol is much lower than that of MCC. The results demonstrate that loading Mn-Ce bimetallic oxide on CNT is beneficial to catalyzing PMS for highly efficient phenol degradation, and a strong interaction between them is vital for bringing about the synergistic effect.

To further disclose the catalytic mechanism of MCC for PMS activation, the changes in the physical–chemical properties of the MCC before and after the reaction were examined. As illustrated in Figure 10, the XRD patterns of MCC depict little change after the reaction. As reported in the previous literature, the activation of PMS by metal-based catalysts usually involves the oxidation process of reductive metal species (e.g., Fe, Co, Mn, Ce) [22,33,34]. Considering the pivotal role of low-valent metal species in PMS activation, the changes in valence state before and after the reaction were detected by XPS measurements.

The high-resolution XPS spectra of Ce 3d of fresh-MCC, fresh-MnCeO, and used-MCC are displayed in Figure 11a–c, respectively. The spectrogram was well fitted with a ten-fold peak, including v⁰, v, v′, v″, v′″, u⁰, u, u′, u″, and u′″. The atomic ratio of Ce(Ⅲ)/Ce(Ⅳ) was obtained by calculating the ratio of the peak area of v⁰, v′, v′″, and u′ to the peak area of v, v″, u⁰, u, u″, and u′″. The ratio of Ce(III)/Ce(IV) for fresh-MCC is 0.84, which is higher than that for fresh-MnCeO (0.79). The result could be ascribed to the fact that the introduction of CNT with reducing ability during the preparation process favors the generation of low-valent Ce, which thereafter enhances the catalytic performance. Additionally, it was mentioned above that the activation of PMS by Ce-based catalysts is accompanied by the oxidation process of Ce(III); thus, the ratio of Ce(III)/Ce(IV) increases to 1.12 for the used-MCC.

The high-resolution XPS spectra of Mn 2p of fresh-MCC, fresh-MnCeO, and used-MCC are depicted in Figure 12a–c, respectively. The characteristic peaks assigned to Mn 2p_1/2 and Mn 2p_3/2 can be observed, while there have no peaks assigned to Mn(II). This phenomenon implies the absence of Mn(II) in fresh-MCC, fresh-MnCeO, and used-MCC. To clarify the valence state of Mn, Mn3s spectra were studied and displayed in Figure 12e,f. The average oxidation state (AOS) of Mn was gained from Equation (3) [35].

AOS = 8.95 − 1.13ΔE

(3)

where ΔE is the peak splitting amplitude of the Mn 3s spectrum.

The ΔE values of fresh-MCC and fresh-MnCeO are 5.15 and 4.9, respectively. Hence, the AOS of Mn on fresh-MCC is “+3.13”, which is slightly lower than that of fresh-MnCeO (AOS = 3.41). The result also could be attributed to the fact that the CNT favors the generation of low-valent Mn, which is beneficial for improving catalytic activity. It is notable that the ΔE for the used-MCC is still 5.15, indicating the AOS did not change after the reaction. As mentioned above, the activation of PMS into ROS was catalyzed by Mn-based materials along with the oxidation of Mn(III) to Mn(IV), and the conversion rate of Mn(IV)/Mn(III) cycling is considered a critical step. Tian et al. [22] confirmed that the substitution of Mn by Ce can give rise to an enhanced charge transfer capability and reductive capability. This could be instrumental in the reduction of Mn(IV) into Mn(III). Apart from the motive of the Ce dopant, CNT also plays an essential role in the catalytic process. Attributing to the abundant free electrons in CNT, the reaction exhibits an accelerated electron transfer rate, which is conducive to the conversion of high-valent Mn species into low-valent species. Together, these give rise to the excellent catalytic performance of MCC for PMS activation.

4. Conclusions

In summary, a highly dispersed Mn-Ce bimetallic oxide on CNT was successfully synthesized by a redox method. It displayed superior activity towards PMS activation for phenol removal. The MCC depicted a synergic effect between Mn-Ce bimetallic oxide and CNT for PMS activation. The activity of MCC was significantly greater than that of CNT, MnCeO, and the mixture of CNT and MnCeO. The excellent performance of MCC could be attributed to the abundant exposed active sites of highly dispersed Mn-Ce bimetallic oxides, the improved mass transfer by the large outer surfaces of CNT, and the facilitated redox Mn(IV)/Mn(III) cycling by Ce dopant and the abundant free electrons in CNT.