Large-Scale Synthesis of Iron Ore@Biomass Derived ESBC to Degrade Tetracycline Hydrochloride for Heterogeneous Persulfate Activation

Abstract

Iron-based catalysts are widely used in water treatment and environmental remediation due to their abundant content in nature and their ability to activate persulfate at room temperature. Here, eggshell biochar-loaded natural iron slag (IO@ESBC) was successfully synthesized to remove tetracycline hydrochloride (TCH) by activated persulfate. The morphology, structure and chemical composition of IO@ESBC were systematically characterized. The IO@ESBC/PS process showed good performance for TCH removal. The decomposition rate constant (k) for IO@ESBC was 0.011 min⁻¹ and the degradation rate was 3690 mmol/g/h in this system. With the increase of PS concentration and IO@ESBC content, the removal rate of TCH both increased. The IO@ESBC/PS process can effectively remove TCH at pH 3–9. There are different effects on TCH removal for the reason that the addition of water matrix species (humic acid, Cl⁻, HCO₃⁻, NO₃⁻ and HPO₄²⁻). The IO@ESBC/PS system for degrading TCH was mainly controlled by both the free radical pathway (SO₄^•−, •OH and O₂^•−) and non-free radical pathway (¹O₂). The loading of ESBC slows down the agglomeration between iron particles, and more active sites are exposed. The removal rate of TCH was still above 75% after five cycles of IO@ESBC. This interesting investigation has provided a green route for synthesis of composite driving from waste resources, expanding its further application for environmental remediations.

1. Introduction

Tetracycline hydrochloride (TCH) belongs to the tetracycline class of antibiotics. It is widely used to treat infections in poultry, cattle, pigs, sheep and aquatic products [1]. For the large number of poultry on commercial farms, antibiotics are added directly to water, feed and aerosols [2]. However, part of the antibiotics is absorbed by animals, and the other part enters the environment through animal excrement. Different concentrations of antibiotics are released into municipal water, medical wastewater, and wastewater from livestock farms [3,4]. In particular, TCH is highly stable and cannot be removed by simple treatment, so it is of great significance to remove tetracycline hydrochloride from water [5]. There are some methods to remove pollutants from the aqueous solutions, such as adsorption [6] and coagulation [7], which could separate pollutants from water but could not remove pollutants completely [8]. Moreover, membrane technologies [9], filtration [10], and sedimentation [11] have high costs and low efficiency [12]. Therefore, it was necessary to explore effective treatment methods to remove these organic pollutants from wastewater.

Sulfate radical-based advanced oxidation processes (SR-AOPs) have been widely used because the highly reactive species produced could remove refractory pollutants present in water [13]. Reactive species include sulfate radical (SO₄^•−) [14], hydroxyl radicals (•OH) [2,7], and reactive oxygen (O₂^•−) [15], as well as by the generation of non-radical singlet oxygen (¹O₂) [16]. SR-AOPs has the advantages of low energy consumption, a high efficiency and simple operation [17]. Today, a host of catalysts have been used and developed in SR-AOPs and exhibited good potential. Activated persulfates by transition metal are commonly used, for example iron [18], cobalt [19], manganese [20] and copper [21], among which iron is the most popular because of easy access, low costly and no secondary pollution [22]. Researchers often prepared catalysts with specific features or functions to improve the performance of catalysts by introduced some macromolecular compounds or substances rich in functional groups in the process. Therefore, the preparation process is complicated, and the cost is high. Directly using waste materials to prepare catalysts not only has low cost and simple operation, but also does not cause secondary pollution to the environment. Herein, we have directly explored a natural iron ore residue (IO) from a mining company in the city of Pingdingshan, Henan Province, China, aiming to efficiently degrade organic pollutants by activating PS in the water. However, due to its own magnetic properties, iron-based catalysts often undergo agglomeration [23]. Therefore, in order to suppress agglomeration, the researchers combined the iron material with another substance.

In recent years, the composites which consist of iron-based catalysts and carbon have often been used in SR-AOPs. The rich pore structure in carbon materials enhance the mass transport of organic pollutant molecules in the pores and channels [24], the chance of contact between pollutants and metal active sites is increased, so the removal rate of pollutants is improved [25]. There have two ways to combine iron-based catalysts and carbon, one was to derive carbon from the metal organic framework using a one-step carbonization method; such as researches used MIL-88A-derived to a magnetic iron/carbon nanorod, which successfully decolorized Rhodamine B by activated PS [26]. Others include Fe-MIL-NM₂ derived Fe/Fe₃C@NC [27], MOF MIL-53(Fe)-driven Fe@porous carbon [28], Fe (Hbidc)-driven Fe@Cs [29]. Another way to do it is to combine the iron-based catalysts with pre-prepared carbon materials to form composites. For example, the researchers prepared composite activated persulfates by loading FeCo₂S₄ onto g-C₃N₄ to degrade sulfamethoxazole [30], sawdust biochar and FeS formed FeS@BC composites to remove TC by activated PS [31]. Table S1 show some statistical results of current research about Fe–Carbon composites for pollutants removal with SR-AOPs.

Therefore, iron ore (IO) enriched in iron active sites were found, but visible aggregation limited its application, and a pore structure was required to inhibit its aggregation. Biochar was a kind of multifunctional material with large specific surface area, abundant pore structure, abundant oxygen-containing functional groups and good acid and alkali resistance [32]. The majority of the carbon materials used have either been derived from metal–organic frameworks or materials such as graphene, sludge biochar and carbon nanofibers. There were no studies using eggshell biochar (ESBC). In theory, it was feasible to load IO onto ESBC to form composite materials activated PS to remove pollutants.

In this work, a composite material IO@ESBC was formed by loading natural iron slag onto eggshell biochar. The prepared materials were characterized by different characterization methods and applied to TCH removal. The reaction conditions were optimized and the effects of common substrates in water on the IO@ESBC/PS/TCH system were explored. The possible TCH removal pathways and existing mechanisms were explored by radical quenching experiments and EPR detection, too. This study may have important technical implications on TCH removal and material innovation.

2. Results and Discussion

2.1. Characterization of the Green Synthesized IO@ESBC

In order to explore the structural characteristics of IO@ESBC, the properties of IO, ESBC and IO@ESBC were analyzed by SEM, VSM, BET, XRD and Raman.

The aggregation between iron ore particles can be observed in Figure 1a. The apparent irregular pore structure on the ESBC surface could be observed in Figure 1b. The surface of IO@ESBC could be cleared to see IO particles interspersed in the pores of ESBC (Figure 1c), which increased the spacing between IO particles and inhibiting the agglomeration between particles. To prove that the addition of ESBC could inhibit IO aggregation, room temperature magnetization vs. field curves of IO and IO@ESBC detected by magnetic properties were exhibited in Supplementary Materials Figure S1. Notably, two samples showed typical S-type. It was also showed that the saturation magnetization (Ms), coercivity (Hc), remanence (Mr) value for two samples. When the IO loaded to the ESBC, the values of Ms and Mr became smaller and the value of Hc became larger. This was because the introduction of ESBC has increased the anisotropy of IO@ESBC [33]. The changes of these three values indicated that the addition of ESBC reduces the magnetism between IO particles and prevents the IO agglomeration.

Figure 1d shows the N₂ adsorption-desorption isotherms of the ESBC, IO and IO@ESBC. The samples belonged to IV isotherms. The hysteresis ring in N₂ adsorption desorption curve of ESBC belonged to H₃ type, indicated that the pores on ESBC surface were slit holes formed by the accumulation of lamellar particles, which could be confirmed by Figure 1b. The hysteresis ring in N₂ adsorption desorption curve of IO moved forward when ESBC were introduced into IO, which indicated that porosity decreased significantly. The decrease of BET specific surface and total pore volume could also explain this. It further confirmed that IO successfully loaded onto the ESBC surface.

The crystallographic structures of ESBC, IO and IO@ESBC were characterized by XRD (Figure 1e). Remarkably, the XRD pattern of samples exhibited obvious diffraction peaks, indicated that the crystalline structures were better. The peak from 20° indicated amorphous structure and incomplete graphitization in ESBC. The peaks were consistent with the crystal planes of Fe₃O₄ (JCPDS 88-0866), α-Fe₂O₃ (JCPDS 87-1165) and γ-Fe₂O₃ (JCPDS 39-1346) in IO. As was shown in Figure 1e, the characteristic peaks of IO and ESBC could be observed in the XRD patterns of IO@ESBC, which indicated that IO were successfully loaded onto ESBC.

Figure 1f shows the Raman spectrum of ESBC, IO and IO@ESBC. There were two peaks in the Raman spectrum, where the D band represents disordered carbon at 1380 cm⁻¹ and the G band corresponds to graphitic carbon at 1591 cm⁻¹ [34]. Compared with the IO@ESBC and ESBC, the intensity of D band and G band were smaller. It suggested that the ESBC was overwritten by IO and the IO were successfully loaded onto the ESBC. The I_D/I_G values of ESBC, IO and IO@ESBC were 0.82, 1.40 and 1.59. These values indicated that the degree of graphitization of ESBC was large, which could also be seen from the Figure 1b. However, the degree of defects of IO and IO@ESBC were larger, which could enhance the electron transfer between catalysts and PS [35,36], so the catalytic performance of the IO@ESBC was improved. The figure showed at 218.9, 292.2, 614.3 cm⁻¹ that were characteristic peaks of α-Fe₂O₃ [37]. At 408.3 and 660.2 cm⁻¹, which were attributable to Fe₃O₄, and at 507.3 cm⁻¹ that were assigned to γ-Fe₂O₃ [38]. The intensity of these peaks in IO@ESBC material were not significant, probably because of the IO loading in the pore structure of carbon.

The role of ESBC were analyzed in the following aspects. (1) The addition of ESBC inhibited IO agglomeration, increased the active site of catalyst, and improved the catalytic performance. (2) In the radical degradation process, ESBC as an electron transport agent caused PS decomposition to be accelerated. Since the iron active site rapidly transfers electrons to PS through special channels in ESBC, which resulted in the reactive radicals can be produced quickly. (3) In the ¹O₂ degradation process, the introduction of ESBC increased the defects on the catalyst surface, which accelerated the electron transport between IO@ESBC and TCH, leading to the decomposition of TCH.

2.2. Catalytic Performance of IO@ESBC

TCH was hardly removed when PS was added alone, while there was a significant removal of TCH after the continued addition of IO or IO@ESBC (Figure 2a). It indicates that PS alone cannot remove pollutants and requires the addition of catalysts. In IO/PS/TCH system, the removal rate of 58.94% for TCH was due to the presence of Fe(II) and Fe(III) on the surface of IO. Fe(II) and Fe(III) could activate PS to produce SO₄⁻ (Equations (1) and (2)), which could effectively remove pollutants [39]. Furthermore, the iron ions (0.5% of iron ore) would be leached from the solid phase catalysts to the solution. Thus, the experiment for contribution of released iron species on TCH removal in PS solution was conducted and shown in Figure 2a, confirming the contribution of leached iron on TCH degradation should be negligible. Nonetheless, the magnetic properties of IO made it easy to agglomerate, resulting in a reduction of active sites. The removal rate of 68.64% for TCH in ESBC/PS/TCH system, which because there were defective structures with abundant functional groups on the surface of ESBC [40]. However, ESBC were not good for recycling. The removal rate was 82.08% for TCH when IO@ESBC were added in PS/TCH system. Obviously, TCH removal rate increased by 23.14%. This indicated that IO@ESBC formed when IO were loaded onto ESBC had more active sites than IO, and the magnetic properties of IO were more conducive to recycling. In order to explore whether TCH removal was the result of adsorption of IO@ESBC or the activation of PS by IO@ESBC, the adsorption effect diagram of IO@ESBC on TCH was made (Figure 2b). The results showed that the residual amount of TCH remained at about 70.00% in 2 h as the dose of IO@ESBC increased. This indicated that the removal of TCH was mainly attributed to the activation of PS by the IO@ESBC. In addition, the XRD (Supplementary Materials Figure S2) and Raman (Supplementary Materials Figure S3) patterns of IO@ESBC after the reaction could be confirmed. In the XRD image of IO@ESBC after the reaction, it was obvious from the figure that the peak intensity of Fe₃O₄ weakened, while the peak intensity of Fe₂O₃ increased slightly. It indicated that there was a conversion of Fe₃O₄ to Fe₂O₃ during the reaction. In the Raman pattern of IO@ESBC after the reaction, the peak intensity decreases at positions 408.3 cm⁻¹ and 660.2 cm⁻¹ and becomes larger at other positions. It indicates that the Fe₃O₄ in the material was activated by PS and produced different levels of α-Fe₂O₃ and γ-Fe₂O₃.

$$Fe\left(\mathrm{II}\right)+S_2{O}_8^{2-}\to Fe\left(\mathrm{III}\right)+S{O}_4^{•-}+S{O}_4^{2-}$$

(1)

$$Fe\left(\mathrm{III}\right)+S_2{O}_8^{2-}\to Fe\left(\mathrm{II}\right)+S_2{O}_8^{•-}$$

(2)

2.3. Possible Mechanism of IO@ESBC System for TCH Removal

In order to investigate the pathway in the IO@ESBC/PS/TCH system, radical quenching experiment and EPR detection (Figure 3) were performed, respectively. TBA scavenging •OH (k_•OH,TBA = (3.8–7.6 × 10⁸) M⁻¹s⁻¹, k_SO4⁻,TBA = (4.0–9.1 × 10⁵) M⁻¹s⁻¹) [41], and MeOH scavenging SO₄⁻ and •OH (k_•OH,MeOH = 9.7 × 10⁸ M⁻¹s⁻¹, k_SO4⁻,MeOH = 3.2 × 10⁶ M⁻¹s⁻¹) [42] were added to the system respectively. It can be seen from Figure 3a that the addition of TBA with different concentrations in the IO@ESBC/PS/TCH system had no effect on the removal of TCH. When MeOH were added, there was an inhibition effect on the removal of TCH compared with the addition of TBA, which indicated the eliminate of SO₄⁻ causes the inhibition of TCH removal. The signals of DMPO-SO₄⁻ and DMPO-OH were also detected in EPR (Figure 3c). The appearance of DMPO-SO₄⁻ [43] was consistent with the results of quenching experiments. IO provided metal ions, involving Fe(II)-Fe(III)-Fe(II) redox cycle participate in the catalytic oxidation reaction and the PS could activated by iron ions to generate SO₄⁻ (Equations (1) and (2)) [44]. Quenching experiments showed that there was no •OH in IO@ESBC/PS/TCH system, but the DMPO-OH [45] were also detected in the EPR. This was because PS was decomposed to produce SO₄⁻ in the first stage, and SO₄⁻ combines with H₂O to produce •OH (Equation (10)) [46]. In addition to the presence of SO₄⁻ and •OH, it was also examined whether another free radical O₂⁻ plays a role in the reaction. PBQ were added to the IO@ESBC/PS/TCH system to quench O₂⁻ (k_O2⁻_,PBQ = 0.9–1.0 × 10⁹ M⁻¹s⁻¹) [47]. The presence of PBQ significantly inhibited the removal of TCH, and the higher the concentration of PBQ, the less TCH was removed (Figure 3a), which means that O₂⁻ played a dominant role in the system. The typical quadruple peaks of DMPO-O₂⁻ were showed in Figure 3c, which indicated the production of O₂⁻ in IO@ESBC/PS/TCH system. Since the existence of oxygen vacancies has been proved in the XPS, it could be inferred that the formation of O₂⁻ was mainly the electrons given by oxygen vacancies, which was consistent with the XPS O1s analysis of IO@ESBC.

O₂⁻ was unstable and it will continue to react to form ¹O₂ [48], so adding histidine to quench ¹O₂ (k_{¹O2,histidine} = 8.2 × 10⁹ M⁻¹s⁻¹) in the syetem [49]. The removal of TCH received a hindrance in IO@ESBC/PS/TCH/histidine system (Figure 3b). Furthermore, the EPR spectra showed the intense triplet peaks of TEMP-¹O₂ in Figure 3d. The ¹O₂ was generated in the system which because O₂⁻ readily continues to react to form ¹O₂ in an acidic environment (Equation (3)) [50]. To sum up, it could be concluded that there were two pathways that were found in this reaction by a combination of free radical quenching experiment and EPR detection. One was a free radical pathway where the major free radicals were SO₄^•−, •OH and O₂⁻; the other non-radical path was made up of ¹O₂.

As a result, the TCH removal process in IO@ESBC/PS/TCH system was depicted in Figure 4. Notably, it can be seen that the removal rate of TCH by iron released from the catalyst is only about 1% in Figure 2a, so the contribution of leached iron ions to TCH removal is negligible. Hence, the whole reaction takes place on the surface of the catalyst. Firstly, the TCH were enriched in the surface of IO@ESBC by specific recognition and the adsorption of ESBC. The PS reached the active sites (Fe) (Equations (1) and (2)) [51], then Fe(II) captured electrons from PS which resulted in O–O bond disconnect and SO₄^•− and Fe(III) generation (Equation (1)). The Fe(III) would react with PS to generate Fe(II) (Equations (2) and (8)). Thus, a cycle of iron between the Fe(II) and Fe(III) would be formed and continued, aiming to increase the quantity of SO₄^•− in an aqueous solution. Immediately, the SO₄^•− would also react with H₂O to generate •OH radicals in the reaction system. The oxygen vacancy on the surface of IO@ESBC reacted with PS to form O₂⁻ (Equation (7)), and then ¹O₂ produced by the recombination of O₂⁻ (Equation (3)). The formed reactive free radicals and ¹O₂ were efficiently contacted and oxidized TCH on the surface of IO@ESBC (Equation (4)).

The mineralization rate (TOC) for TCH in IO@ESBC/PS/TCH system within 120 min was determined as 44.3% (Supplementary Materials Figure S6), confirming that the TCH could not be completely mineralized. And some by-products were generated in IO@ESBC/PS/TCH system.

$$O_2^{•-}+2H^{+}\to {}^{1}O_2+H_2{O}_2$$

(3)

$$S{O}_4^{•-}+O_2^{•-}+{}^{1}O_2+TCH\to products$$

(4)

2.4. Effect of Different Conditions on TCH Removal

The actual effluent will change in pH due to environmental changes. Therefore, it was of great necessity to explore the feasibility of this reaction system in different pH environments (Figure 5a). It was clear that the removal rate of TCH in acidic environment was higher than that in alkaline and neutral environment. In general, the main free radicals were different in different acid-base environments [52]. Between pH 2 and 7, the major radicals were SO₄^•− [53]; between pH 7 and 10, the major radicals were SO₄^•− and •OH [54]. The reason for the higher removal of TCH in an acidic environment was the decomposition of PS and the rapid hydrolysis. Researchers thought that the acidic environment not only facilitated the decomposition of PS, but also enhanced the acid catalysis [55]. Furthermore, O₂^•− was produced in the presence of dissolved oxygen, but O₂^•− will continued to react to form ¹O₂ in the acid environment (Equations (5)–(9)) [56]. Hence the presence of SO₄^•−, O₂^•−and ¹O₂ allowed an increased removal of TCH in an acidic environment. In contrast, the removal rate decreases when SO₄^•− and •OH exist simultaneously in alkaline environment. This can be explained by the following reasons. On the one hand, •OH will be the dominant free radical because SO₄^•− could be quenched by •OH in an alkaline environment (Equations (11) and (12)) [57]. However, •OH reacted pollutants though hydrogen abstraction reaction, which it was slower than the reaction of SO₄^•− and pollutants by electron transfer [58]. On the other hand, the metal precipitates could be generated in an alkaline environment [59], which leads to a reduction in the active sites.

$$S_2{O}_8^{2-}+H^{+}\to H{S}_2{O}_8^{-}$$

(5)

$$H{S}_2{O}_8^{-}\to S{O}_4^{2-}+H^{+}+S{O}_4^{•-}$$

(6)

$$2O_V^{-}+S_2{O}_8^{2-}\to O_2^{•-}+S{O}_4^{•-}+S{O}_4^{2-}$$

(7)

$$Fe\left(\mathrm{III}\right)+O_2^{•-}\to Fe\left(\mathrm{II}\right)+O_2$$

(8)

$$S_2{O}_8^{2-}+H_{2}O\to 2HS{O}_4^{-}+\frac{1}{2}{O}_2$$

(9)

In addition, the pH after the reaction tended to decrease during the reaction process (Supplementary Materials Figure S4). Moreover, they were close to original pH that it was not adjusted with H₂SO₄ or NaOH in the IO@ESBC/PS/TCH system. It has already been reported that the H⁺ were released by the reaction of SO₄^•− with H₂O (Equation (10)) and OH⁻ were consumed by the reaction with SO₄^•− (Equation (11)). Furthermore, H⁺ had other sources. It was the reaction of SO₄^•− and •OH, S₂O₈²⁻ and H₂O all produced HSO₄⁻, which decomposed to produce H⁺ (Equations (12) and (13)) [60]. This decomposition will occur more easily in an alkaline environment, so the pH of the reaction solution decreases even at a high pH. Therefore, it was not necessary to adjust the pH for the IO@ESBC/PS/TCH system that could effectively in the practical application.

$$S{O}_4^{•-}+H_{2}O\to S{O}_4^{2-}+•OH+H^{+}$$

(10)

$$S{O}_4^{•-}+O{H}^{-}\to S{O}_4^{2-}+•OH$$

(11)

$$S{O}_4^{•-}+•OH\to HS{O}_4^{-}+\frac{1}{2}{O}_2$$

(12)

$$HS{O}_4^{-}\to S{O}_4^{2-}+H^{+}$$

(13)

Figure 5b showed the removal efficiency of TCH with the different IO@ESBC dosage. The residual TCH was 85.25% when the absence of IO@ESBC and the remaining amount of TCH was significantly reduced after adding IO@ESBC. The removal rate of TCH was increased (57.83%, 81.64%, 81.68%, 82.08%, 81.39%) with the increase of IO@ESBC content (0.50 g/L, 0.75 g/L, 1.00 g/L, 1.25 g/L, 1.50 g/L). This was because the amount of IO@ESBC determined the active site of Fe(II) in the reaction process [61], and the more active site of Fe(II) were more favorable to the removal of TCH to some extent. However, when the content of IO@ESBC exceeded 1.25 g/L, the removal rate of TCH was found to decrease. There were two reasons. One was that in this reaction system, not only the catalyst content will affect the removal of TCH, but also the amounts of PS [62]. Therefore, when the dosage of PS was consumed, the removal rate of TCH will not increase even if the catalyst dosage was increased. Secondly, the excess catalyst will react with SO₄^•− and •OH, which increased the consumption of additional PS and catalyst (Equations (14) and (15)) [18], thus inhibiting the further removal of TCH. Considering the utilization efficiency, 1.25 g/L was selected as the IO@ESBC dosage in the following experiment.

$$•OH+Fe\left(\mathrm{II}\right)\to O{H}^{-}+Fe\left(\mathrm{III}\right)$$

(14)

$$S{O}_4^{•-}+Fe\left(\mathrm{II}\right)\to S{O}_4^{2-}+Fe\left(\mathrm{III}\right)$$

(15)

The PS dosage was an important factor for the removal of TCH in the IO@ESBC/PS/TCH system because it was the source of SO₄^•−. In Figure 5c, when PS was not added, the removal rate of TCH was 30.61% within 2 h. Overall, the removal rate of TCH also increased with the increase of PS dosage, but when the PS dosage exceeded 6 mM, the removal rate of TCH basically remained the same (82.02%, 82.08%, 82.22%) with the increase of PS dose (8 mM, 10 mM, 12 mM). When the catalyst dosage was constant, PS was activated by the catalyst to produce reactive radicals for TCH removal, so the removal rate of TCH will increase with the increase of PS dosage. Nevertheless, excess PS will react with SO₄^•− (Equation (16)) [63], and the quenching reaction between SO₄^•− and SO₄^•− will occur (Equation (17)) [64], thus reducing the active site and further removal of TCH was inhibited. Thus, the PS dosage of 10 mM was considered as the most optimal dosage in the IO@ESBC/PS/TCH system.

$$S{O}_4^{•-}+S_2{O}_8^{2-}\to S{O}_4^{2-}+S_2{O}_8^{•-}$$

(16)

$$S{O}_4^{•-}+S{O}_4^{•-}\to 2S{O}_4^{2-}$$

(17)

The influence of original TCH concentration on TCH removal in IO@ESNC/PS/TCH system was investigated as described in Figure 5d. It could be seen from the graph that the removal rates of TCH were all around 80.00% at TCH concentrations ranging from 30 mg/L to 70 mg/L. This indicated that the system has good stability.

Common anions and natural organic matter in wastewater will compete for free radicals with pollutants [65]. They will affect the catalytic process and the pollutants removal process [66]. Hence, it was no wonder that explore the effect of various anions (Cl⁻, HCO₃⁻, NO₃⁻ and HPO₄²⁻) (Figure 6) and fulvic acid in the IO@ESBC/PS/TCH system (Supplementary Materials Figure S5).

In this work, it could be seen from Figure 6 that the presence of Cl⁻ and HCO₃⁻ did not show an inhibitory effect on the removal of TCH. Normally, the two anions mentioned above had an inhibitory effect on the removal of pollutants. This is because the form of reactive chlorine species (Cl•) by the reaction of Cl⁻ and SO₄^•− (Equation (18)) [67] and the form of ClOH^•− by the reaction of Cl⁻ and •OH (Equation (19)) [68], which could hardly remove pollutants [25]. On the- side, the HCO₃⁻ could competitively react with •OH and SO₄^•− to generated CO₃^•− (Equations (20)–(23)), which had a low redox potential (1.63 V) [69]. These were reactions between anions and •OH and SO₄^•− in the reaction system, resulting in inhibition of pollutant removal. However, in IO@ESBC/PS/TCH system, there had not only •OH and SO₄^•− but also had O₂^•− and non-free radical paths, so the existence of Cl⁻ and HCO₃⁻ had no influence on TCH removal. Compared with the Cl⁻ and HCO₃⁻, NO₃⁻ had a little boost to the removal of TCH. Although NO₃⁻ could also react with free radicals, their reaction rate was low (Equations (24) and (25)) [70], and a small amount of NO₃^• could be produced to further remove TCH [71]. Contrary to the above phenomenon, the higher the concentration of HPO₄⁻, the more difficult it was to remove TCH. Since the phosphate radicals were generated by the reaction •OH, SO₄^•− and HPO₄²⁻, which had a lower oxidation potential than •OH, SO₄^•− (Equations (26) and (27)) [72]. This result was consistent with previous reports that phenol was removed with active substances dominated by •OH, SO₄^•− and ¹O₂. The removal rate of phenol decreased by 55% even though the concentration of HPO₄²⁻ was very low [73].

As a representative of natural organic matter, fulvic acid was used to explore the influence of natural organic matter in the IO@ESBC/PS/TCH system (Supplementary Materials Figure S5). Surprisingly, the removal of TCH had no effect even though the concentration of fulvic acid was high, because of the presence of the active substance (•OH, SO₄^•−, O₂^•− and ¹O₂) in the reaction system. These active substances may counteract the competition between fulvic acid and TC for reactive substances [21].

$$C{l}^{-}+S{O}_4^{•-}\to C{l}^{•}+S{O}_4^{2-}\hspace{1em}\mathrm{k}=2.1 \times {10}^{10} {\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$

(18)

$$C{l}^{-}+•OH\to {\left[ClOH\right]}^{•-}\hspace{1em}\mathrm{k}=6.1 \times {10}^9 {\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$

(19)

$$C{O}_3^{2-}+S{O}_4^{•-}\to C{O}_3^{•-}+S{O}_4^{2-}\hspace{1em}\mathrm{k}=6.1 \times {10}^6 {\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$

(20)

$$•OH+C{O}_3^{2-}\to C{O}_3^{•-}+O{H}^{-}\hspace{1em}\mathrm{k}=3.9 \times {10}^8 {\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$

(21)

$$S{O}_4^{•-}+HC{O}_3^{-}\to S{O}_4^{2-}+C{O}_3^{•-}+H^{+}\hspace{1em}\mathrm{k}=2.8 \times {10}^6 {\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$

(22)

$$•OH+HC{O}_3^{-}\to C{O}_3^{•-}+H_{2}O\hspace{1em}\mathrm{k}=8.5 \times {10}^6 {\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$

(23)

$$N{O}_3^{-}+S{O}_4^{•-}\to S{O}_4^{2-}+N{O}_3^{•}\hspace{1em}\mathrm{k}=5.5 \times {10}^5 {\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$

(24)

$$N{O}_3^{-}+•OH\to O{H}^{-}+N{O}_3^{•}\hspace{1em}\mathrm{k} < 5.5 \times {10}^5 {\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$

(25)

$$HP{O}_4^{2-}+•OH\to O{H}^{-}+HP{O}_4^{•-}\hspace{1em}\mathrm{k}=1.5 \times {10}^5 {\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$

(26)

$$HP{O}_4^{2-}+S{O}_4^{•-}\to S{O}_4^{2-}+HP{O}_4^{•-}\hspace{1em}\mathrm{k}=1.2 \times {10}^6 {\mathrm{M}}^{-1}{\mathrm{s}}^{-1}$$

(27)

2.5. Reusability of IO@ESBC

For heterogeneous catalysts, both catalytic performance and reusability were important [74]. XPS analysis and repeated experiments were performed to evaluate the performance of IO@ESBC in removing TCH. The Fe2p, C1s and O1s peaks clearly appeared in XPS survey spectra of samples (Figure 7a). The Fe 2p XPS of samples were shown in Figure 7b. The Fe 2p3/2 peak for Fe(II) and Fe(III) emerged at 710.07 and 711.57 eV [75], respectively. Fe 2p1/2 peak for Fe(II) and Fe(III) emerged at 723.55 and 725.31 eV [76], respectively. In addition, the satellite peak positions were 718.95 and 732.65 eV [77]. The presence of α-Fe₂O₃, γ-Fe₂O₃ and Fe₃O₄ were confirmed by the presence of these peaks. The O 1s XPS of samples were fitted by two species (Figure 7c). It showed the spectra of O 1s that peak at 529.84 eV was assigned to Fe-O of Fe₂O₃ and Fe₃O₄ [78]. Moreover, the peak at 531.59 eV was typical of oxygen vacancies on the catalyst surface [79].

Figure 7d showed the spectra of C 1s that peaks at 284.86 and 289.13 eV were corresponded to the C–C bond from sp2-hybridized graphitic carbon and the carbon in C=O bond [80]. The C–C bond from sp2-hybridized graphitic carbon were the main groups for all biochar samples [81]. Obviously, the XPS diagram of IO@ESBC hardly changed before and after the reaction, indicating that the catalytic reaction did not cause the change of the catalyst structure. Further, IO@ESBC had good stability. As can be seen from Figure 7e, after IO@ESBC was reused five times, the system still maintained a high removal rate for TCH. With the increase of reaction times, the removal rate of TCH was 82.08%, 78.91%, 78.12%, 77.31%, 76.39%. The reactivity of IO@ESBC slightly decreased, which was because of the inevitable loss of catalysts during the reaction process and the increase of intermediates adsorbed on the catalysts surface during the reaction process. In order to explore the wider application of the reaction system, TCH was dissolved in river water, tap water and ultra-pure water respectively to simulate different water quality (Figure 7f). The results showed that different water quality had no effect on TCH removal, so this IO@ESBC/PS/TCH system had great potential in practical application.

3. Materials and Methods

3.1. Materials

The IO used in this study was taken from the ferromagnetic phase of a mining company in Pingdingshan City, Henan Province, China. Eggshells were taken from a bakery in Henan University of Urban Construction, Pingdingshan City, Henan Province, China. The TCH was purchased from Shanghai Aladdin Bio-Chem Technology Co., Ltd. (Shanghai, China). The reagents including PS (Na₂S₂O₈, 98%), hydrochloric acid (HCl,36%), methanol (MeOH, 99.5%), tert-butanol (TBA, 99.5%), P-benzoquinone (PBQ, 99%), 5, 5-dimethyl-1-pyrrolidine-N-oxide (DMPO, 97%), 2,2,6,6-tetramethylpiperidone (TEMP, 99%), histidine, sodium hydroxide (NaOH, 98%), fulvic acid, sodium chloride (NaCl, 99.5%), sodium bicarbonate (NaHCO₃, ≥99.5%), sodium nitrate (NaNO₃, 99%), dipotassium hydrogen phosphate (K₂HPO₄, 98%), and sulfuric acid (H₂SO₄, 98%) were all purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All the reagents used in this experiment were of analytical grade without further treatment.

3.2. IO@ESBC Synthesis

Iron ore sample (50 g) was passed through a 200-mesh sieve, aiming at the particles under the sieve. The obtained particles after sieving were added into 1000 mL ultrapure water, then stirred by JJ-1 precision booster electric stirrer at speed of 300 rpm. The ultrapure water in the stirring would be replaced many times until the supernatant was transparent. The iron ore sample after washing was filtered and dried at 105 °C for 12 h, then named IO.

The ESBC used in this study was obtained via slow pyrolysis of eggshells powder. Fistly, 100 g of eggshells are washed 5 times with tap water and ultra-pure water, then dried in a Desktop blast drying oven at 80 °C for 24 h. The washed eggshells are broken with a crusher, sifted through under a 200-mesh sieve. Next, the sample was calcined in a tube furnace with nitrogen flow at 150 mL/min. The pyrolysis temperature was 800 °C with 10 °C/min of the heating rate for 1 h. Then, the gained sample was dealt with HCl (2 M) until no bubbles were generated to entirely remove the calcium-containing component. The products were washed with ultra-pure water for five times and dried at 80 °C for 24 h and placed in a sealed bag.

A total of 10 g of ESBC was added to 750 mL of distilled water and stirred continuously using a JJ-1 precision booster electric stirrer (Electric mixer, MYP2011-100, Shanghai Mei Yingpu instrument and Meter Manufacturing Co., Ltd., Shanghai, China) to obtain a homogeneous suspension. Then, 50 g of IO was added and stirred for 1 h. The solid materials were filtered and rinsed with distilled water to remove the unloaded iron and other residues. Then, the solid catalyst was dried at 105 °C for 1 h and calcined at 600 °C for 1 h. The obtained sample was named IO@ESBC.

3.3. Characterization

The changes in morphologies of the samples were observed by the environmental scanning electron microscope (SEM, FEI QuANTA 200, USA). The crystal phases composition of the IO, IO@ESBC before and after reaction were measured by the D8 Advance X-ray diffractometer (XRD, Bruker, Karlsruhe, Germany) at 2θ ranging from 10° to 80°. The Raman spectroscopy (in Via confocal Raman, Renishaw Co., UK) were employed to examine the composition of the IO, IO@ESBC before and after reaction. The specific surface area and the pore size of the samples were determined by a Micromeritics TriStar II 3020 Version 3.02 analyze based on the Brunauer–Emmett–Teller (BET, Micromeritics Co., USA) method. The X-ray Photoelectron Spectroscopy (XPS, ESCALAB 250Xi, Thermo Fisher Scientific, Rockford, IL, USA) of the before and after reaction of IO@ESBC were measured by a Perkin Elmer PHI 6000C ECSA system to analyze their surface composition and elemental valence state. The free radicals (including •OH, SO₄⁻ and O₂^•−.) and no-radical (¹O₂) generated in the removal process were determined by the electron spin resonance (ESR, JES-FA200, JEOL Co., Japan) spectroscopy with a dose of 100 mM DMPO and 50 µM TEMP as spin-trapping agents (Supplementary Materials Text S1). The magnetic properties of the IO and IO@ESBC were analyzed by a physical property measurement system (Quantum Design, USA).

3.4. Catalytic Activity Tests of IO@ESBC

The catalytic performance of IO@ESBC was investigated in the PS containing aqueous solution by the removal rate of TCH. For the batch experiment, a certain amount of iron ore sample was added into 250 mL conical flask containing 150 mL of TCH (50 mg·L⁻¹) solution, followed by adding a certain dosage of PS into the mixing slurry. Then, the conical flasks containing mixing slurry were placed on a thermostatic shaker with a rotary speed of 240 rpm at a constant temperature of 25 °C. Every 10 min, 1 mL of reaction solution was quickly withdrawn from the slurry, then filtered by membrane (0.22 µm) to obtain the filtrate. Afterwards, the concentration of TCH in the filtrate was measured by UV-VIS (U7s, YOKE INSTRUMENT, Shanghai, China). The removal rate of TCH was calculated according to (C_t − C₀)/C₀. Nevertheless, the reaction pH was adjusted by 0.1M H₂SO₄ and 0.1 M NaOH solutions when exploring the effect of initial pH. Other experiments used the initial pH of the reaction solution. Different quenching agents (MeOH, TBA, PBQ and histidine) were added into the reaction system of IO@ESBC/PS/TCH in the experiment to explore the free radical species.

In order to investigate the effect of the organic and inorganic anions present in the water on the reaction system, different concentrations of humic acid, NaCl, NaHCO₃, NaNO₃, and K₂HPO₄ were added to the system. The IO@ESBC sample was recycled, washed repeatedly by the deionized water, and finally dried at 105 °C in a blast drying oven. The catalyst would be reused for five times to investigate its stability and recycling performance. In order to investigate the use of catalysts in a real environment, water samples were taken from laboratory tap water and Shahe, Pingdingshan City, Henan Province, respectively. All experiments were repeated three times.

4. Conclusions

Much of the current research requires the synthesis of catalysts using chemicals. In this study, a novel material was synthesized by combining IO with ESBC, which both came from real life and did not require the use of chemicals. It was used for the removal of TCH by activating PS. The addition of ESBC not only effectively alleviates the aggregation of IO and increases the number of active sites, but also effectively adsorbs TCH and increases the chance of contact between active free radical and TCH, and finally improving the removal rate of TCH. Free radical quenching and EPR detection showed that there were not only free radical paths (•OH, SO₄^•− and O₂^•−) but also non−free radical paths (¹O₂) in the reaction system. The removal rate of TCH was 82.08% in 2 h with the optimized conditions of 1.25 g/L IO@ESBC, 10 mM PS, 30 mg/L TCH, initial pH. In addition, fulvic acid, Cl⁻, HCO₃⁻ and NO₃⁻ often presented in water and water quality had no influence on the reaction system, but the presence of HPO₄²⁻ had a significant inhibitory effect on TCH removal. The reusability of catalyst was also investigated and the removal rate of TCH was not affected.