Environmental problems concerning water pollution has drawn much attention of scholars in recent years. One of the prominent sewage types is effluent containing synthetic dyes discharged from pharmaceutical, printing, food, paper, and textile industries. The synthetic dyes in water pose serious threat to animal lives and human health^[1]. Therefore, it is of great significance to find an effective method to remove synthetic dyes, e.g. methyl orange (MO), from water. MO is a typical anionic dye that belongs to azo group. Owing to its sturdy molecular structure, MO is difficult to be degraded by conventional biochemical methods^[2]. In recent decades, advanced oxidation processes (AOPs), such as Fenton oxidation, photocatalytic oxidation, ozone oxidation and peroxymonosulfate (PMS) oxidation, have gained widespread attention as an efficient approach to sewage treatment^[3⇓⇓-6]. Owing to their strong oxidation and high removal efficiency, AOPs have been employed to treat industrial sewage and sludge with high concentration pollutant and refractory to biochemical degradation^[6]. Various organics, including antibiotic, phenolic compound, surfactant and synthetic dyes, could be converted to CO₂, H₂O and other harmless inorganic products by applying these methods^[7-8].

Among them, PMS-based advanced oxidation process has high reactivity, strong oxidation capacity and produces little secondary pollution^[9]. Therefore, it has great potential for industrial application. The strong oxidation comes from the reactive oxygen species (ROS) generated during the catalytic reaction, including hydroxyl radical (•OH), superoxide radical (O₂^•-), sulfate radical (SO₄^•-) and singlet oxygen (¹O₂)^[6]. To improve the oxidation performance of PMS, various catalysts, including metals, metal oxides and heterogeneous metal-based catalysts, have been studied^[10-11]. Spinel ferrite (MeFe₂O₄, Me=Co, Ni, Mn, Cu and Zn, etc.), a heterogeneous metal-based catalyst, has attracted much attention recently, due to its excellent catalytic activity^[7]. Among them, CoFe₂O₄ has good catalytic activity owing to the full utilization of the catalytic activity of both Fe and Co, and the synergistic interaction between them^[7,11]. CoFe₂O₄/PMS system was used to oxidize various organic pollutants in aqueous solution, such as antibiotics, pesticides, synthetic dyes and surfactant, with a degradation rate more than 90%^[7,12⇓-14].

Since CoFe₂O₄ is ferromagnetic, it can be separated from water efficiently by magnetic separation, which not only improves the solid-liquid separation, but also facilitates the recovery of the catalyst. However, strong magnetism makes CoFe₂O₄ particles conductive to agglomerate, especially for the nanoscale, which greatly reduces their specific surface area so as to weaken the catalytic performance^[15]. To address the problem, CoFe₂O₄ nanoparticles were supported on carriers. In order to increase the effective supporting amount of catalyst and enhance the structure stability of their composite, materials with high surface area were selected as carriers. The carrier material is also required to have high acid-base stability and thermal stability to adapt to the harsh chemical environment in the synthesis process and using process. Various materials, including silica, titanium dioxide, zirconia and graphene, were employed as catalyst carriers^[16]. Zeolite, as a cheap and stable inorganic porous material, well meets the above requirements, so it could be one of the best candidates for catalyst carrier.

Herein, magnetic CoFe₂O₄@zeolite (CFZ) was prepared by a co-precipitation hydrothermal method. Contents of the study were included as follows: (I) CFZ was synthesized and characterized. (II) The degradation of MO in different systems was compared. (III) Influence factors on the degradation were investigated. (IV) Stability and reusability of the catalyst were tested. (V) The presence of different ROS was determined.

Coal gangue obtained from Huainan, China, was chosen as the raw material to prepare zeolite. The typical silica alumina ratio of Na-A zeolite is 1.0-2.0. To meet the requirement, Na₂SiO₃·9H₂O was used to adjust the ratio. The coal gangue was successively pulverized and sieved. Then an aluminosilicate precursor was obtained by calcining the coal gangue powder for 2 h at 750 ℃^[17]. 5.5 g precursor and 1 g Na₂SiO₃·9H₂O were dispersed in 30 mL deionized water, and stirred for 30 min, and then solution A was obtained. Next, 6.0 mmol FeCl₃·6H₂O and 3.0 mmol CoCl₂·6H₂O were dissolved into 25.0 mL deionized water. The mixture was slowly added into 25 mL NaOH solution (3 mol/L) heating in a water bath at 90 ℃, and continued heating for 5 min. Then solution B was obtained. Both solution A and B were transferred into a hydrothermal reaction bottle and reacted at 90 ℃ for 10 h. The obtained solid products were washed with deionized water and ethanol, then dried at 60 ℃ in vacuum for 6 h. After grinding, magnetic CFZ catalyst were obtained.

Degradation experiments were carried out in 250 mL beakers at room temperature. Before the reaction, reagents were added into 100 mL MO solution. Part of the supernatant was extracted every few minutes to measure its absorbance. The removal rate of MO can be calculated as follows:

Where, R, C₁ and C₀ are the removal rate, current concentration and initial concentration of MO, respectively.

In the study, the effects of dosage of CFZ and PMS, solution pH, MO concentration, as well as common coexisting anions on the degradation were investigated. In addition, the stability and reusability of CFZ was tested. In the end, the contributed ROS in the CFZ/PMS system was determined.

Experimental details are described in the supporting materials.

X-ray diffraction (XRD) patterns of the zeolite, CoFe₂O₄ and CFZ are exhibited in Fig. 1(a). It can be seen that the diffraction peaks of CFZ all correspond to zeolite and CoFe₂O₄. The peaks at 2θ = 12.4°, 16.1°, 20.3°, 23.9°, 27.1°, 33.3°, 40.1°, 56.3° and 68.5° are attributed to the (111), (210), (220), (311), (321), (421), (521), (722) and (900) crystal plane diffraction peaks of Na-A zeolite (PDF#38-0214), separately. While the characteristic diffraction peaks located at 2θ = 18.3°, 30.1°, 35.4°, 37.1°, 43.1°, 53.4°, 57.0°, 62.6° and 74.0° correspond to the (111), (220), (311), (222), (400), (422), (511), (440) and (533) plane diffraction of CoFe₂O₄ (PDF#22-1086), respectively. There are no miscellaneous peaks in the diffraction patterns, indicating that CFZ is a composite material containing CoFe₂O₄ and Na-A zeolite. Since CoFe₂O₄ is ferromagnetic, CFZ has magnetism. As illustrated in the hysteresis loops (Fig. 1(b)), the saturation magnetization of CFZ is 29.0 A·m²·kg^-1. Although this value is lower than pure CoFe₂O₄ (51.8 A·m²·kg^-1)^[18], it is adequate for effective magnetic separation. As shown in the inset of Fig. 1(b), more than 99.5% (mass ratio) of CFZ could be collected by a hand magnet.

The morphology of CFZ was characterized by scanning electron microscope (SEM). As shown in Fig. 2(a), the prepared zeolite with a cuboidal shape is 2-3 μm in diameter, which is the typical shape of Na-A zeolite. The surface of the prepared zeolite is smooth, but its morphology is changed obviously after the modification of CoFe₂O₄. As displayed in Fig. 2(b), the zeolite is totally covered and the surface becomes bumpy, but the cube prototype is faintly visible, which suggests that the shell layer is not thick. As shown in Fig. 1(c), size distribution is calculated by ImageJ software, indicating that the shell is composed of 10-130 nm particles, and the mean is around 46.6 nm.

In order to learn about the distribution of elements in CFZ, EDS mapping analyses (Fig. 2(c-f)) were carried out. It can be seen that the content of O, Si and Al elements in CFZ is higher than Fe and Co, and all of them distribute evenly in the catalyst. Transmission electron microscope (TEM) (Fig. 3(a)) and elemental line scanning were selected to investigate its overall structure. Clear difference between the inside and outside of the catalyst is shown in the elemental line scanning spectrum (Fig. 3(b)). O, Si and Al elements mainly exist in the interior, while Fe and Co are in the exterior, clearly showing the core-shell structure of CFZ.

To further figure out the chemical component of CFZ, the surface chemical property was characterized by X-ray photoelectron spectroscope (XPS). The survey spectrum (Fig. 3(c)) clearly shows that Fe, Co, O, Si and Al elements exist in CFZ. As shown in the Fe2p spectrum (Fig. 3(d)), the peaks located at 724.5 and 710.8 eV are ascribed to Fe2p_1/2 and Fe2p_3/2, respectively, evidencing the existence of Fe³⁺ and absence of Fe²⁺ on the surface. Fig. 3(e) shows that the Co2p spectrum has two main peaks at around 780.4 and 796.4 eV, associated with the Co2p_1/2 and Co2p_3/2, separately. Two corresponding satellite peaks are present at 786.1 and 802.3 eV. It is a solid proof for the presence of Co²⁺ instead of Co^3+[19]. The O1s peak is located at 530.6 eV, as illustrated in Fig. 3(f). With Avantage software fitting, it could be divided into two peaks at 530.6 and 532.8 eV, corresponding to the binding energy of lattice oxygen and the oxygen in H₂O, respectively^[11,20]. The lattice oxygen in CFZ may belong to CoFe₂O₄, Al₂O₃ and SiO₂. The Fe/Co atomic ratio of CFZ calculated from XPS and element line scanning are 1.93 and 2.07, respectively, which are close to the stoichiometric ratio in CoFe₂O₄ molecule. The XPS results clearly verify the formation of CoFe₂O₄ on the surface, which gives further proof for the fabrication of magnetic CFZ composite.

N₂ adsorption/desorption isotherms of CFZ and zeolite are shown in Fig. 4(a). It is found that both the zeolite and CFZ have a H3 hysteresis adsorption isotherm, indicating their porous structure^[21]. The specific surface area (S_BET) of the zeolite is 30.13 m²/g. Interestingly, the S_BET of CFZ is 107.06 m²/g, more than three times that of zeolite (Table 1). The increase of S_BET could be related to the accumulation of nanoparticles in the CoFe₂O₄ shell. As displayed in Fig. 4(b), the pore volumes of micropores, mesopores and macropores all increase obviously, which could come from the shell layer composed of tightly packed nanoparticles. The accumulation of nanoparticles can produce various pores. Owing to the highly porous structure of shell layer, the pore structure of the zeolite is still able to be fully utilized. The increase of S_BET would be beneficial to surface adsorption and catalytic degradation of organic pollutant.

The removal of MO in sole-PMS, sole-CFZ and PMS/CFZ systems was determined and compared. As shown in Fig. 4(c), the removal rates of MO in the sole-PMS and sole-CFZ system were 23.9% and 3.5%, respectively, after 15 min reaction. While in the CFZ/PMS system, the removal rate increased significantly. The removal rate reached 91.3%, which was more than 3.8 times of that using PMS alone. The removal by sole-CFZ could be attributed to adsorption by the porous CoFe₂O₄ shell. Obviously, compared with the total removal, the adsorption is only a small amount. The low adsorption might be owing to the small dosage and short adsorption time. The removal of MO in sole-PMS system could be in a non-radical way as the degradation of sulfachloro-pyridazine, moxifloxacin and sulfonamides^[15,22-23]. The degradation rate and utilization of oxidant were relatively low. The significant enhancement of removal efficiency in the CFZ/PMS system suggests that the degradation mechanism change a lot in the presence of CFZ. As the previous studies on the catalytic process of PMS oxidation, ROS could play an important role^[11]. The production of ROS is determined by the structure and property of the catalyst. Therefore, the high removal rate of the CFZ/PMS system could be related to the synergistic action between CFZ and PMS, which generated a tremendous amount of ROS^{[23⇓⇓-26]}.

The effect of catalyst dose in the range of 0.02-0.25 g/L on the degradation was studied in the CFZ/PMS system. As shown in Fig. 4(d), even if only 0.02 g/L CFZ was added, the degradation rate could be twice that of sole-PMS. The MO removal increased from 55.9% to 97.1% with the increase of the CFZ dosage from 0.02 to 0.20 g/L. The enhancement of the degradation might be ascribed to an increase in active catalytic sites, which accelerated the conversion reaction between PMS and ROS. Hence, more ROS were generated, such as O₂^•- and ¹O₂^[27]. However, when the dosage exceeded 0.20 g/L, the removal of MO did not increase continuously, suggesting there is an optimal ratio between CFZ and PMS. When the CFZ dosage exceeded this value, the available PMS was not adequate so as not to produce more ROS^[28]. Considering the balance between catalytic performance and the cost, 0.20 g/L was chosen as the optimal dosage.

In order to detect the influence of initial solution pH on the catalytic activity of CFZ, the degradation at initial solution pH 2-9 was investigated. As shown in Fig. 4(e), in the range of pH 5-9, the MO removal all reached 97% after around 15 min. Since the CFZ catalyst has good catalytic activity in a wide pH range, it is quite potential in practical application for industrial sewage treatment. However, the catalytic activity decreased fast with the increase of acidity as initial solution pH<5. The removal efficiency reduced to 37.6%, when pH decreased to 2. The decrease of catalytic activity could be owing to the enhanced stability of PMS molecules under low initial solution pH condition, which hindered the production of ROS and lowered its catalytic activity^[7]. Besides, the high concentration of H⁺ could capture the generated ROS, e.g. SO₄^•- and •OH (Eq. (2, 3))^[29-30]. It was also reported that the formation of H-bond between H⁺ and the O-O groups of HSO₅^- in acidic pH could hinder the interaction between HSO₅^- and positively charged catalyst surface^[31]. Interestingly, the MO degradation also declined under strong alkaline condition. The removal efficiency decreased to 65.1% at pH 10. The reason is probably due to that the CFZ catalyst is negatively charged and repulsive to both HSO₅^- and SO₅^2-[28].

PMS dosage also has significant effect on the MO degradation. As exhibited in Fig. 4(f), when the PMS dosage increased from 0.3 to 1.0 mmol/L, the removal efficiency raised from 60.9% to 97.1%, respectively, suggesting that a certain concentration of PMS is necessary to produce high concentration of ROS in the presence of CFZ catalyst. The more PMS, the faster ROS would be produced, unless all the full catalytic activity of CFZ was excited. Further adding PMS did not improve the removal efficiency. Furthermore, excessive PMS dosage in reaction system would reduce the amount of ROS by the ego depletion as well^[29,32-33]. As shown in Fig. 4(f), the removal rate at 1.5 mmol/L PMS addition was not higher than that at 1.0 mmol/L addition. Therefore, the optimal dosage of PMS was about 1.0 mmol/L.

Actual industrial wastewater usually contains various inorganic ions, which may significantly impact the degradation reaction. Herein, SO₄^2-, Cl^-, H₂PO₄^- and CO₃^2- with a concentration of 50 mg/L were selected to explore their influence on the catalytic activity of CFZ. As shown in Fig. 5(a), the removal efficiency decreased to 95.6%, 95.3%, 95.1% and 90.5% in the presence of SO₄^2-, Cl^-, H₂PO₄^- and CO₃^2-, respectively. The inhibition was ranked as CO₃^2- > H₂PO₄^- > Cl^- > SO₄^2-. Only CO₃^2- caused a slightly larger effect on the catalytic performance. The effect of CO₃^2- could be owing to its reaction with ROS, consuming part of •OH, SO₄^•- and HSO₅^- to generate some free radicals with low redox potential (Eq. (4, 5))^[13,28,34]. As shown in Fig. S2, when the concentration of coexisting anions increased to 1000 mg/L, the removal efficiency still achieved 90.7%. It turned out that the influence of most coexisting anions on the catalytic degradation was quite limited, thus the CFZ catalyst could be applied in high salinity industrial wastewater.

To verify the continuous catalytic activity of CFZ, it was recycled for 5 times to estimate its recyclability. As revealed in Fig. 5(b), the removal efficiency of MO still remained 94.7% after 5 cycles. The slight decline (2.4%) of the removal efficiency could be attributed to the deactivation of reactive sites resulted from minor amount of leached Co and Fe from CFZ. Serious measurement found that 0.09 mg/L cobalt ions was released from CFZ in the examined reaction time (15 min), which could meet the emission standard (<1 mg/L). To further prove its stability, XRD and SEM were used to characterize the catalyst cycled for 5 times. As shown in Fig. S3, compared with the fresh catalyst, the XRD pattern of the used sample had little change and no new diffraction peak presented. SEM images (Fig. S4) indicate that most of the CFZ particles preserved the original morphology and structure. The cyclic test and discussion above suggest that CFZ has excellent reusability and stability.

It was reported that SO₄^•-, •OH, O₂^•- and ¹O₂ play an important role in the process of PMS activation. To figure out the existence of ROS and their contribution to the degradation, methanol (MeOH), tert-butanol (TBA), p-benzoquinone (p-BQ) and furfuryl alcohol (FFA) were used as scavengers to rapidly quench SO₄^•-, •OH, O₂^•- and ¹O₂, respectively. As illustrated in Fig. 5(c), the removal efficiency declined from 97.1% to 94.4%, 94.3%, 67.2% and 74.8% with the addition of MeOH, TBA, p-BQ and FFA, separately. It is suggested that p-BQ and FFA inhibit the degradation significantly, while MeOH and TBA only have a slight impact. The result implies that O₂^•- and ¹O₂ could play a predominant role in the catalytic oxidation.

According to the discussion above, it could be derived that CFZ could activate PMS to selectively generate O₂^•- and ¹O₂ (Fig. 6). The generation of ¹O₂ from PMS was reported previously and applied in degrading p-chlorophenol and sulfamethoxazole^[35-36]. ¹O₂ could be produced by several pathways, as described in Eq. (6-14). O₂^•- might be produced by the hydrolysis of PMS (Eq. (9)), and there is no extra side reaction. As shown in Eq. (6-8, 11), O₂^•- could also be the precursor to produce ¹O₂^[11,37]. In this reaction, •OH would also be produced at the same time. Since little •OH was tested by the scavenger experiment, the reaction of Eq. (7, 8, 11) could be ruled out. As a result, the high concentration of ¹O₂ could be generated from the reaction among HSO₅^-, SO₅^2-, SO₅^•-and H₂O (Eq. (10, 12-14))^{[11,37⇓⇓-40]}. To sum up, the excellent catalytic performance of CFZ should be closely related to its promotion in the reaction that produces O₂^•- and ¹O₂. In the CFZ/PMS system, part of these reactions were catalyzed and enhanced by the synergistic interaction between Fe, Co and PMS^[7]. As a result, tremendous amount of ¹O₂ and O₂^•-were employed to react with MO. Since the oxidation of ¹O₂ and O₂^•-is quite strong, MO was decomposed efficiently.

In this study, magnetic CFZ catalyst with high catalytic activity was synthesized via the co-precipitation hydrothermal method, and exploited as a PMS activator to degrade MO. Careful investigations indicated that highly porous layer composed of nano-sized CoFe₂O₄ particles was covered on the prepared Na-A zeolite. Compared to sole-PMS and sole-CFZ, the CFZ/PMS system exhibited much better catalytic performance. The MO degradation efficiency achieved 97.1% under the optimum reaction condition ([MO]=50 mg/L, [PMS]=1 mmol/L, 0.2 g/L CFZ, pH 8 and T=25 ℃). Except CO₃^2-, coexisting anions (H₂PO₄^-, Cl^- and SO₄^2-) have slight impact on the degradation, suggesting that the CFZ/PMS system has the potential to be used in high salinity wastewater treatment. With efficient magnetic separation technology, the used CFZ is easy to be recycled. After 5 cycles, the MO degradation rate is still achieved 94.7%. The ROS quenching tests suggest that ¹O₂ and O₂^•- play a main role in the degradation.

Supporting materials related to this article can be found at https://doi.org/10.15541/jim20220591.

Enhanced Degradation of Methyl Orange with CoFe₂O₄@Zeolite Catalyst as Peroxymonosulfate Activator: Performance and Mechanism

WANG Lei¹, LI Jianjun^1,2, NING Jun³, HU Tianyu^1,2, WANG Hongyang¹, ZHANG Zhanqun¹, WU Linxin¹

(1. School of Materials Science and Engineering, Anhui University of Science and Technology, Huainan 232001, China; 2. Institute of Environment-friendly Materials and Occupational Health, Anhui University of Science and Technology, Wuhu 241000, China; 3. Anhui Shuiyun Environmental Protection Co., Ltd., Wuhu 241000, China)

Reagents: PMS (KHSO₅·0.5KHSO₄·0.5K₂SO₄, KHSO₅), MO, cobalt chloride (CoCl₂·6H₂O), ferric chloride (FeCl₃·6H₂O), sodium metasilicate nonahydrate (Na₂SiO₃·9H₂O), methanol (MeOH), tert-butyl alcohol (TBA), p-Benzoquinone (p-BQ), furfuryl alcohol (FFA), sodium carbonate (Na₂CO₃), sodium sulfate (Na₂SO₄), sodium chloride (NaCl) and potassium phosphate monobasic (KH₂PO₄) were purchased from Aladdin Reagent Co., Ltd (Shanghai, China). Coal gangue were collected from Huainan, China. All chemicals above are analytical grade, and deionized water was used throughout all experiments.

Relationship between the concentration and absorbance: Before the degradation experiment, MO solutions with concentration of 0, 2, 4, 6, 8 and 10 mg/L were prepared. The absorbance was measured by ultraviolet-visible spectrophotometer (UV-5100, China) at 464 nm. After linear fitting of MO concentration and absorbance, a functional relationship can be obtained:

Where, X and Y are the concentration and absorbance of MO, respectively. It can be seen from the value of R² that there is a good linear relationship between the concentration and absorbance of MO.

Characterization methods: The fresh CFZ were investigated by XRD, Smartlab SE, Japan). Brunauer-Emmett-Teller (BET, V-sorb2800p, China) was used to measure the N₂ adsorption-desorption isotherms, and the specific surface area were calculated based on the BET model. The surface morphology, elements distribution and overall structure of CFZ were characterized by Scanning electron microscope (SEM, ZEISS-G500, Germany) equipped with an energy dispersive spectrometer (EDS) and transmission electron microscope (TEM, Tecnai G2 F20, Netherlands). The chemical states of Fe, Co, O were studied by X-ray photoelectron spectroscope (XPS, ESCALAB 250Xi, USA). The saturation magnetization was measured by vibrating sample magnetometer (VSM, HH-20, China) at room temperature. The leached cobalt ions concentration was determined by an inductively coupled plasma mass spectrometry (ICP-MS, Agilent 7700x, USA).

The degradation performance of the CFZ/PMS system at various MO concentrations was also investigated. As shown in Fig. S1, with the increase of MO concentration from 20 to 200 mg/L, the removal efficiency decreased from 98.1% to 92.0%, respectively. It turned out that the CFZ/PMS system has excellent degradation for both low and high concentration dyes, which is of great significance for its practical application.

Treatment of the SO₄ product during degradation: It was reported that ettringite precipitation is an effective method for handling high-concentration SO₄^2-. In the approach, lime and aluminum salts are added into the sewage in turn to react with SO₄^2- to form insoluble ettringite (Ca₆Al₂(OH)₁₂(SO₄)₃·26H₂O) that can be used as a cement grouting material^[S1]. Owing to its high removal efficiency and cost-effectiveness, relative researches have employed this method to deal with industrial sewage, such as textile industries and mine water^[S2-S3].

Degradation of methylene blue (MB)

References:

[S1] FANG P, TANG Z J, CHEN X B, et al. Removal of high-concentration sulfate ions from the sodium alkali FGD wastewater using ettringite precipitation method: factor assessment, feasibility, and prospect. Journal of Chemistry, 2018, 2018: 1265168.

[S2] DOU W, ZHOU Z, JIANG L M, et al. Sulfate removal from wastewater using ettringite precipitation: magnesium ion inhibition and process optimization. Journal of Environmental Management, 2017, 196: 518.

[S3] TOLONEN E T, HU T, RAMO J, et al. The removal of sulphate from mine water by precipitation as ettringite and the utilisation of the precipitate as a sorbent for arsenate removal. J. Environ. Manage., 2016, 181: 856.