English

• Recently, antibiotics have been widely used to control infectious diseases and their presence in the environment has aroused dramatic attention, since surface water, groundwater, and even drinking water are contaminated with antibiotics residues [1]. As a general consensus, the existence of antibiotics leads to the development of antibiotic-resistant pathogens that pose a potential threat to ecosystems function and human health [2]. As a broad-spectrum bacteriostatic agent, tetracycline (TC) is widely used in the livestock and aquaculture industry [3]. However, due to the strong bacterial resistance, TC could not be efficiently degraded by traditional methods of wastewater treatment [4]. Therefore, it is necessary to exploit a new process that can effectively degrade it.

  Advanced oxidation processes (AOPs) have been exploited to degrade the existing refractory organics in wastewater due to their remarkable treatment effects [5]. AOPs are based on the formation of hydroxyl radicals (^•OH) in situ, which have high redox potential (2.7 eV) and non-selectivity. In addition, they can oxidize most organic pollutants. Recently, several AOPs, such as O₃ [6], ultraviolet [7], ultrasound [8], Fenton [9], and electro-Fenton processes [10], have been applied to degrade organic contaminants from wastewater. However, these approaches have various drawbacks, such as high cost and energy consumption (O₃, ultraviolet, ultrasound, and electro-Fenton), the need for pH adjustment (Fenton and electro-Fenton), and sludge generation (Fenton). Recently, heterogeneous Fenton-like process becomes a promising choice to overcome the disadvantages of traditional AOPs (massive iron sludge and incapable catalyst recovery) [11].

  At present, carbon-based materials have been widely used as catalysts for environmental remediation [12-14]. Specifically, activated carbons (ACs) [15], graphene [16], carbon aerogels [17, 18], g-C₃N₄-based materials [19-21], and biochars (BCs) [22], have attracted more and more attentions as H₂O₂ activators. BCs, which are produced from agricultural and industrial waste through the direct thermal decomposition of biomass under anaerobic conditions, could be used as substitutes for expensive carbon-based materials. Although BCs have large specific surface areas, multiple pores, abundant active functional groups, and a stable structure, the catalytic performance of original BCs is unsatisfactory. For instance, only 33.3% sulfamethoxazole could be removed by BC/H₂O₂ system, which was much lower than that of by Schwertmannite@BC/H₂O₂ system (100%) under the same conditions [23]. Additionally, the original BCs are difficult to separate. Metal loadings (Fe and Co) not only enhance the catalytic performance of BCs, but they also effectively solve the problem of catalyst recyclability. As Fe spices are non-toxic, relatively inexpensive, and magnetic, they were chosen as metal loads on BCs. Recently, Fe-BC, as an activator for H₂O₂, has aroused widespread attention in removing organic contaminants [24]. The Fe phases loaded on BC has an important effect on the structure and reactivity of Fe-BC. Meanwhile, the contribution of each component in Fe-BC to contaminant degradation was also different. However, only few studies have focused on the performance of Fe additions and the Fe-BC catalytic mechanisms for removing pollutants, especially the generation of by-products and their toxicities in Fe-BC/H₂O₂ process. Therefore, a thorough investigation of Fe-BC as the catalyst to activate H₂O₂ for the removal of TC from water is needed, which can guide the application of Fe-BC/H₂O₂ process in wastewater treatment.

  In this study, hydrothermal-pyrolysis, which can uniformly load iron on BC [25], was used to prepare a new Fe-BC composite using wheat stalk and FeSO₄·7H₂O as precursors, and a Fe-BC/H₂O₂ system was proposed to treat tetracycline (TC). The effects of catalyst and H₂O₂ dosage, initial pH, pollutant concentration, and the water matrix on the wastewater treatment were studied. Meanwhile, the reusability of Fe-BC was also studied. Additionally, the catalytic mechanisms of the Fe-BC/H₂O₂ process were studied through the characterization analysis of materials, electron paramagnetic resonance (EPR) detection, chemical quenching experiments, and H₂O₂ consumption detection. Moreover, the degradation products of TC were determined and the possible degradation pathways were speculated. Additionally, the ecotoxicities of TC and its intermediates were evaluated. The characterization details of Fe-BC and the experimental procedures are provided in Supporting information.

  The morphology of BC-based materials were detected by scanning electron microscopy (SEM) (Fig. S1 in Supporting information). The BCs retained the irregular flaky structure of the wheat straw with slight surface protrusions. By comparing Figs. S1a and b, it can be seen that there is no obvious difference in the surface morphology of BC before and after the reaction, indicating that BC alone has no or little effect on the TC pollutants. A large number of tiny particles could be clearly seen attached to the Fe-BC surface. These particles are known to be loaded with iron or iron oxide based on the element distribution. These structures provided more specific surface areas and active sites which can in favor of enhancing catalytic activity, compared to original BCs (Table S3 and Figs. S2a in Supporting information). Compared with Figs. S1c and d, it was noticed that the particles attached to the Fe-BC surface before and after the reaction had significantly changed and the particles became coarser and tended to aggregate after the reaction, indicating that the Fe loaded on the BC played an important role in the reaction process. The microstructure and compositional distribution of the BC and Fe-BC were further investigated using transmission electronic microscopy (TEM), energy dispersive X-ray detector (EDX), and EDX-mapping. The TEM image (Fig. S3a in Supporting information) showed that Fe was uniformly and densely doped on the surface of Fe-BC, which is in agreement with the SEM results. The distribution of the C, O, and Fe components on the surface of the materials was determined using EDX-mapping. As presented in Figs. S3b and c and Fig. S4 (Supporting information), compared with the pristine BC sample, the Fe content of the Fe-BC sample was significantly increased.

  In Fig. 1a, the XRD patterns of the prepared BC and Fe-BC before and after the reaction clearly demonstrated the structures and changes of these materials. The broadened peaks cantered at 2θ = 23° and 42.2° were found in BC before and after the reaction, corresponding to the (002) and (100) planes of crystalline carbon, respectively [26]. The apparent peaks at 2θ = 33.24°, 43.28° and 53.73° (JCPDS No. 19-0629) in the fresh and used Fe-BC were assigned to the (220), (400) and (422) crystal planes of magnetite [25]. Moreover, the 2θ = 30.1° and 44.8° might have been corresponded to γ-Fe₃O₄ and α-Fe, respectively [27]. The XRD patterns demonstrated that Fe was successfully loaded on the BC surface, which is in agreement with the SEM results. The positions of the diffraction peaks were almost unchanged before and after the reaction, illustrating that the crystal structures of the material were relatively stable. The diffraction peak intensity of the iron species changed, showing that Fe played a key role in the degradation of TC.

  Figure 1

  

  Figure 1.  XRD patterns of BC and Fe-BC before and after reaction (a), Raman spectra (b) and XPS survey of BC, Fe-BC and used Fe-BC (c), and magnetic hysteresis loops of Fe-BC (d).

  DownLoad: Full-Size Img PowerPoint

  Raman spectroscopy was applied to further uncover the crystallization and the defect degree on the catalysts surface, as described in Fig. 1b. The D band (~1350 cm⁻¹) was ascribed to the disordered graphitic carbon or amorphous carbon, and the G band (~1580 cm⁻¹) was attributed to the sp²-hybridised graphitic carbon [28]. The carbon-based material defects could be expressed using the relative intensity ratio of I_D/I_G. It could be inferred from the Raman spectra that the I_D/I_G of BC was 0.99 and that of Fe-BC was 1.01, indicating that Fe-BC formed more defects in the generation process [29].

  X-ray photoelectron spectroscopy (XPS) measurements were conducted to probe the surface element composition and valence state data. As illustrated in Fig. 1c, the survey spectra of XPS described the presence of C and O for both BC and Fe-BC samples. The peak at 708.8 eV (Fe 2p) was observed for Fe-BC, which is in agreement with the results of EDX-mapping and XRD, suggesting that Fe was successfully loaded onto the BC surface in the form of Fe⁰, Fe²⁺, and Fe³⁺. In addition, we found that the O content of used Fe-BC was higher than that of fresh Fe-BC, indicating oxygen-containing functional groups play an important role in removal of TC, which also discussed in the following results.

  The magnetic hysteresis loops in Fig. 1d reflect the magnetization properties of Fe-BC. The small coercivity indicates that Fe-BC is a soft magnetic material, and the small remanence indicate that Fe-BC is a typical superparamagnetic material. Moreover, Fe-BC had a high saturation magnetization value of about 6.2 emu/g, indicating that it was easily attracted to external magnets and that it could be easily separated from the system after the reaction.

  The catalytic performance of materials was estimated and the results were shown in Figs. 2a and b. Less than 1% TC was eliminated only in the H₂O₂ system, indicating that H₂O₂ can hardly degrade TC effectively in the absence of external catalysis. The removal efficiency of TC was 5.4% by BC within 60 min, illustrating that BC can remove a part of TC due to its porous structure. The removal efficiency of TC was 15.3% in the BC/H₂O₂ process, confirming that BC can promote the decomposition of H₂O₂ [30]. In addition, 56% TC was removed by Fe-BC alone in 60 min. The TC removal may have been caused by adsorption (N₂ adsorption-desorption isotherm results in Fig. S2 in Supporting information) and the micro-electrolysis because of the existence of Fe species (Fe⁰) on Fe-BC surface (XRD and XPS results). In the Fe-BC/H₂O₂ process, ~92% of TC was removed, so it can be inferred that Fe plays a key role in TC removal. The k_obs of the Fe-BC/H₂O₂ process was 0.155 min⁻¹, which is 23.85 times higher than that of the BC/H₂O₂ process (0.0065 min⁻¹), indicating that Fe doping endowed the material with higher adsorption and reactivity.

  Figure 2

  

  Figure 2.  Influence of different system (a and b), Fe-BC dosage (c), H₂O₂ concentration (d), initial pH values (e), and initial TC concentration (f) on the TC removal using the Fe-BC/H₂O₂ system. Reaction conditions: [TC] = 20 mg/L, [H₂O₂] = 1 mmol/L, [Fe-BC] = 0.2 g/L, and initial pH 3.

  DownLoad: Full-Size Img PowerPoint

  Fig. 2c shows the influence of the Fe-BC dosage on the TC removal. The removal efficiency of TC increased with the increase of Fe-BC dosage. 89% of the TC was removed within 60 min under the condition of 0.1 g/L Fe-BC. When the Fe-BC dosage was 0.2 g/L, 94.2% of TC could be removed. The dosage of Fe-BC was further increased to 0.6 g/L and 0.8 g/L, a similar removal efficiency was obtained within only 10 min and 5 min, respectively. The k_obs of the Fe-BC/H₂O₂ process increased with the increase of the Fe-BC dosage from 0.105 min⁻¹ of 0.1 g/L to 0.355 min⁻¹ of 0.8 g/L. These results indicate that increasing the dose of Fe-BC can provide more active sites.

  The H₂O₂ concentration effect on the TC removal is shown in Fig. 2d. 87% of TC was removed within 60 min with 0.5 mmol/L H₂O₂. When the H₂O₂ increased to 1.0 mmol/L, the removal efficiency of TC reached 94.2% within 60 min. However, with a further increase in H₂O₂ (1.5 and 2.0 mmol/L), the removal of TC was not further enhanced, corresponding to the trend of k_obs (0.088, 0.155, 0.165 and 0.145 min⁻¹). One possible for the results was that the excess H₂O₂ in the process consumed ^•OH according to Eq. 1 [5].

  (1)

  pH is usually considered as a critical factor that affects heterogeneous Fenton reactions. In addition, TC is an amphoteric molecule with multiple pK_a and appeared in different chemical state at different pH values [31]. Therefore, the TC removal using the Fe-BC/H₂O₂ process at different initial pH values (3.0-9.0) was evaluated. As exhibited in Fig. 2e, TC could be effectively removed at pH of 3.0-4.0. while the removal efficiency of TC slightly decreased from 91.37% to 88.37% with increasing pH from 4.0 to 9.0 under experimental conditions, which is consist with previous report [32]. Similarly, the values of k_obs vary with pH, which are 0.155, 0.159, 0.149, 0.145, 0.151, and 0.144 min⁻¹ at pH 3.0-9.0. The reasons why TC can be effectively removed under neutral or alkaline conditions are explained as follows: TC present different forms under different pH value, i.e., cationic (pH < 3.3), zwitterionic (3.3 < pH < 7.7), and anionic species (pH > 7.7). When pH > 3.3, TC mainly exists in the form of zwitterionic (TCH₂⁰) and anions (TC⁻), which are vulnerable to attack by reactive oxygen species (ROS) due to the higher electronic density in the ring system. In addition, previous study found that H₂O₂ could directly attack TC molecules to form hydroxyl or to remove certain groups to achieve TC removal under alkaline conditions [33].

  The TC concentration is also an important factor for pollutants removal. As shown in Fig. 2f, at a concentration of 10 mg/L, 90% TC could be removed in 15 min and 98% in 60 min. As the TC concentration increased to 20, 30, 40 and 50 mg/L, the degradation rates correspondingly decreased to 94%, 91%, 90%, and 87%, respectively. The quantity of the produced free radicals and active sites in the Fe-BC/H₂O₂ process was limited and could not respond to the excessive amount of TC. Therefore, the k_obs decreased with the increase in the TC concentration.

  Based on previous experience, it was predicted that the free radical and non-free radical pathways are the main causes of TC removal using the Fe-BC/H₂O₂ process [34]. Therefore, quenching experiments were executed to confirm the reaction mechanism. The results of the quenching experiments are shown in Fig. 3a. First, the role of the free radicals was discussed, and tert-butyl alcohol (TBA) was applied to quench ^•OH with rate constants of 4.8-7.6 × 10⁸ L mol⁻¹ s⁻¹ [35]. The removal efficiency of TC decreased from 94% to 41% within 60 min after adding TBA, thus confirming the formation of ^•OH in the Fe-BC/H₂O₂ process. The production of ^•OH in the Fenton reaction is generally considered to be the reaction of Fe²⁺ with H₂O₂. Additionally, p-benzo-quinone (p-BQ) was chosen to scavenge O₂^−• (k = 0.9-1.0 × 10⁹ L mol⁻¹ s⁻¹) [36]. The TC degradation effects were significantly affected, from 94% to 18% in 60 min, and the k_obs decreased from 0.155 min⁻¹ to 0.008 min⁻¹ after the addition of p-BQ. This indicated that O₂^−• played a pivotal role in the TC removal in the Fe-BC/H₂O₂ process. The generation of O₂^−• came from the reaction of Fe²⁺/Fe³⁺ with O₂ Eqs. 2 and 3.

  (2)

  (3)

  Figure 3

  

  Figure 3.  Influence of different scavengers on the removal efficiency of TC (a), the EPR spectra of the Fe-BC/H₂O₂ process with DMPO and TEMP (b), and the possible activate mechanism of the TC removal using the Fe-BC/H₂O₂ process (c). Reaction conditions: [TC] = 20 mg/L, [H₂O₂] = 1.0 mmol/L, [Fe-BC] = 0.2 g/L, [TBA]₀ = 100 mmol/L, [BQ]₀ = [FFA]₀ = [KI]₀ = 20 mmol/L, [DMPO]₀ = [TEMP]₀ = 10 mmol/L, and initial pH 3.

  DownLoad: Full-Size Img PowerPoint

  As for the non-free radical pathway, furfuryl alcohol (FFA) was considered to be the quenching agent for ¹O₂, which was used in this study to confirm the existence of ¹O₂ (k = 1.2 × 10⁸L mol⁻¹ s⁻¹) [37]. The removal efficiency of TC ranged from 94% to 36% with the presence of FFA, with the k_obs ranging from 0.155 min⁻¹ to 0.029 min⁻¹, suggesting that the TC removal was significantly inhibited. In the Fe-BC/H₂O₂ process, ¹O₂ was generated, as shown in Eqs. 4-6 [38]. Potassium iodide (KI) quenched ROS on the material surfaces [39], and the TC removal efficiency decreased from 94% to 54% after adding KI. It was proven that the Fe-BC surface bound some ROS, which was crucial in the reaction.

  (4)

  (5)

  (6)

  EPR was also measured and used to validate the TC degradation pathway, and the results are demonstrated in Fig. 3b. In the tests, 5, 5-dimethyl-1-pyrroline N-oxide (DMPO) and 2, 2, 6, 6-tetramethyl-4-piperidinol (TEMP) were applied to capture radicals. When captured with DMPO, typical DMPO-OH signals were clearly identified, indicating the generation of ^•OH in Fe-BC/H₂O₂ process. Also, DMPO was a spin trapping agent for O₂^−•. With the assistance of DMPO, representative quartet characteristic EPR signals appeared, corresponding to the production of O₂^−•. Additionally, ¹O₂ was trapped by TEMP, through which the triplet EPR spectrum (1:1:1) was detected and attributed to signals for TEMP-¹O₂, thus implying the generation of ¹O₂. These results clearly confirmed that the removal of contaminants in the Fe-BC/H₂O₂ process involved free radical and non-free radical pathways, which is in agreement with the quenching results. To further clarify the role of iron species and H₂O₂ in Fe-BC/H₂O₂ system, the ROS formed in Fe-BC and BC/H₂O₂ process were also detected by EPR and the results showed that the presence of iron species and H₂O₂ facilitated the production of O₂^−• and ¹O₂ (the details were shown in Fig. S5 in Supporting Information). Moreover, the H₂O₂ consumption and Fe²⁺ dissolution in the Fe-BC/H₂O₂ process further demonstrates that non-free radical pathways exist (Fig. S6 in Supporting information).

  In order to understand components transformation on the surface of Fe-BC, the used Fe-BC were characterized by EDX, Raman, XPS, and FTIR and the results were displayed in Fig. S7 (Supporting information). The EDX spectra of used Fe-BC were showed in Fig. S7a and Table S2 (Supporting information). Compared with the fresh Fe-BC, the Fe contents on the surface of the used Fe-BC decreased from 27.96% to 18.65% while the O content increased from 14.15% to 23.56%, illustrating that a part of Fe was dissolved and oxidized during the reaction and that Fe plays a key role in the TC removal. Raman spectra were also obtained to further understand the changes of crystallization and defects of the catalysts after reaction. Fig. S7b showed that The Fe-BC defects changed after the reaction, and the I_D/I_G of the used Fe-BC decreased from 1.01 to 0.98, which might have been caused by the damage of the defective edges, the change of the functional groups, and the interactions between the adsorbed pollutants and sp² hybrid carbon system [40]. Thus, the defective sites of Fe-BC took part in the degradation of TC. The high-resolution C 1s O 1s, and Fe 2p XPS spectra were presented in the Figs. S7c-e. The XPS spectra of C 1s were divided to four peaks (Fig. S7c) at around 288.5, 286.7, 285.6, and 284.7 eV, corresponding to C=O/C=N, C−O−C, C−OH/C−N and C−C/C=C, respectively [41]. The spectra of O 1s were divided into three peaks (Fig. S7d) at 533, 530.7 and 529.9 eV, corresponding to C=O, C−OH and Fe−O, respectively [42]. Interestingly, it was found that C=O increased and that C−OH decreased in both the C 1s and O 1s XPS spectra after the reaction, suggesting that C=O might play a catalytic role in the Fe-BC/H₂O₂ process [43]. The Fe 2p spectra were resolved to a variety of states of iron species (Fe³⁺, Fe²⁺ and Fe⁰) (Fig. S7e). Notably, the amounts of Fe⁰ of Fe-BC decreased, while Fe²⁺ and Fe³⁺ increased after the reaction, demonstrating that Fe⁰ was oxidized. FTIR technology was used to distinguish the functional groups of BC-based materials. Fig. S7f shows the FTIR spectra of Fe-BC before and after the reaction. The broad absorption peak appearing at 3426 cm⁻¹ is ascribed to the stretching vibrations of the −OH groups: 2930, 1588, 1103 and 807 cm⁻¹ to −CH₂, C=C, C−O and C−H, respectively [44]. The absorption peak at 550 cm⁻¹ is probably associated with the Fe-O bond stretching vibrations of Fe₃O₄ nanoparticles [45], or possibly the binding of iron adhering to the surface of Fe-BC with the oxygen-containing functional groups [46]. It should be noted that the band of C=C and Fe−O slightly diminished, confirming that C=C and Fe−O provided active sites for the TC removal. These results are consistent with the XPS results.

  On the basis of above results, the possible reaction mechanisms for the pollutants removal by the Fe-BC/H₂O₂ process were put forward, as shown in Fig. 3c. The TC removal was primarily related to the adsorption and oxidative degradation (the species distribution of TC on Fe-BC were shown in Fig. S8 in Supporting Information). First, the TC was adsorbed on the surface of Fe-BC, then H₂O₂ is activated by Fe-BC to produce ROS, and finally the adsorbed TC was in situ attacked by ROS. The oxidative degradation mainly included two pathways (radical pathway and non-radical pathway). The iron species on the Fe-BC surface could activate O₂ and H₂O₂ to generate O₂^−• and ^•OH. Meanwhile, H₂O₂ reacted with C−OH to form ^•OH and HO₂^•. These free radicals could rapidly oxidize and decompose pollutants. It was found that persistent free radicals (PFRs) were produced during the BC pyrolysis, and the PFRs existence was confirmed using EPR spectroscopy (Fig. S9 in Supporting information). The PFRs in the BC could directly activate H₂O₂ to generate ^•OH, and they could also transfer electrons to the oxygen molecules to form O₂^−•, as described in a previous work [22]. Additionally, H₂O₂ was decomposed to produce ¹O₂ (Eq. 4), which could remove pollutants using electron transfer. Furthermore, Fe-BC provided electrons primarily via a graphite electron donor-transfer complex in the BC and directly transferred electrons to the H₂O₂ as an electronic shuttle. The combination of radical and non-radical processes could degrade the organic contaminants.

  The degradation intermediates of TC were identified using LC-MS. Base on the transformation intermediates identified in the present work and previous research [47], three possible degradation pathways of TC in the Fe-BC/H₂O₂ process were illustrated in Fig. S10 (Supporting information). Pathway Ⅰ: product P1 (m/z 417) was generated by N-demethylation of TC attacked by ROS, the amino group of P1 was destroyed to form P2 (m/z 376), and the benzene ring and amino group of P2 were attacked by ROS to form P3 (m/z 339), which then removed −OH and methylates to form P4 (m/z 309), Then, P4 was further broken into P5 (m/z 287) under the attack of ROS. The similar mechanisms have been reported in previous work [47]. Pathway Ⅱ: Due to the active substance presented, the deamidation reaction of TC could occur, resulting in the loss of dimethylamino group to form P6 (m/z 362) [48]. Subsequently, P6 underwent the opening of ring to form P7 (m/z 274). Pathway Ⅲ: TC could be converted into P8 (m/z 461) by hydroxylation [49], and then the C=C on P8 was oxidized by ROS to generate hydroxyl and ketone groups, forming P9 (m/z 477). As the reaction proceeds, these intermediates formed by the above three pathways were further degraded into small molecular weight organics, such as P10 (m/z 242) and P12 (m/z 218), then they were further decomposed into the smaller molecules P11 (m/z 114), P13 (m/z 118), P14 (m/z 60), P15 (m/z 90)) under the action of ROS. Finally, the intermediates were eventually oxidized to CO₂, H₂O, and NH₄⁺. In addition, the acute toxicity (represented by oral rat LD₅₀) and mutagenicity of intermediates were predicted by the quantitative structure-activity relationship (QSAR) method using the Toxicity Estimation Software Tool (T.E.S.T.) (Fig. S11 in Supporting information)

  Inorganic anions, e.g., Cl⁻, NO₃⁻, and SO₄²⁻, and natural organic matter (humic acid, HA) are usual water ingredients in the actual water. To estimate the influence of inorganic anions on the Fe-BC/H₂O₂ process, 20 mmol/L of Cl⁻, NO₃⁻, and SO₄²⁻ were added into the process, and compared with control experiment. As depicted in Fig. 4a, negligible effects on the 60 min removal were obtained in the presence of these matrices. However, when NO₃⁻, Cl⁻ and SO₄²⁻ were added, the obtain k_obs values (0.144, 0.147 and 0.145 min⁻¹) were slightly decreased compared to the control, presumably due to the clearing of a small amount of reactive radicals as exhibited in Eqs. 7-9 [50-52]. Moreover, HA can promote the degradation of TC to some extent, the k_obs value was 0.156 min⁻¹, which might be ascribed to the fact that HA could expedite circulation of iron and enhance the pollutants elimination [53]. In addition, HA can act as a catalyst to activate H₂O₂ to form ROS therefore promoting the degradation performance [54]. In conclusion, these results suggested that Fe-BC in this study could achieve excellent results for TC removal in practical water treatment.

  (7)

  (8)

  (9)

  Figure 4

  

  Figure 4.  Influence of inorganic ions and HA on TC removal by the Fe-BC/H₂O₂ process (a), TC removal in deionized water, tap water, and river water (b), the removal of different contaminants in the Fe-BC/H₂O₂ system (c), and the reusability of Fe-BC on removal of TC (d). Reaction conditions: [TC]= 20 mg/L, [H₂O₂] = 1.0 mmol/L, [Fe-BC] = 0.2 g/L, [NO₃⁻]₀ = [Cl⁻]₀ = [SO₄²⁻]₀ = 20 mmol/L, [HA]₀ = 10 mg/L, and initial pH 3.

  DownLoad: Full-Size Img PowerPoint

  To further investigate the practicability of the prepared Fe-BC catalyst in removing TC pollutants from real aquatic systems, the catalytic degradation reactions were carried out in different water processes. As demonstrated in Fig. 4b, the removal efficiencies of TC were 90.3% and 86.7% within 60 min in tap water and river water, respectively, and the corresponding values of k_obs were 0.136 and 0.037 min⁻¹. Compared to deionized water (94.2% and 0.155 min⁻¹), the removal efficiency and rate constant of TC were inhibited in river water. The reason might be that the coexisting ions and organic matters in the river water would consume ROS [41]. As we all know, there are some organic pollutants in the nature water bodies, such as azo dyes and endocrine disruptors. To prove the feasibility of Fe-BC for other organic pollutants, the degradations of methyl orange (MO), orange Ⅱ (OGII), and bisphenol A (BPA) were studied. Fig. 4c showed that the removal efficiencies of these organic pollutants reached more than 90%. These results indicated that Fe-BC had a great prospect in actual wastewater treatment.

  Reusability of catalysts plays a significant role in practical application. To test the reusability performance of Fe-BC in this study, the catalysts were recovered by magnetic separation, and washed several times with ultrapure water, then directly put into the next cycle. As illustrated in Fig. 4d, the removal efficiency of Fe-BC gradually decreased as the number of cycles increased, probably because the pore structure of the catalyst was blocked by contaminants or the active sites were occupied, which was consist with the results of N₂ adsorption-desorption isotherms of the used Fe-BC. Even so, the degradation effect of the fourth time still reached 80.5% at 60 min, which indicated that Fe-BC is promising for practical wastewater treatment. In addition, the magnetic separation property of Fe-BC is shown in Fig. S12 (Supporting information). As depicted, Fe-BC showed a great magnetic response when a permanent magnet was nearby, indicating the Fe-BC catalyst was recyclable, which helped to avoid secondary pollution.

  Fe-BC catalysts were successfully fabricated using a simple and effective method in this work. The obtained Fe-BC displayed outstanding activation performance for the H₂O₂ in TC removal, which was improved with a moderate increase in Fe-BC and in the H₂O₂ dosage (favored at acidic condition). The EPR and quenching experiments verified that the TC removal mechanisms by the Fe-BC/H₂O₂ process depended on both radical (^•OH and O₂^−•) and non-radical pathways (¹O₂ and electron transfer). Moreover, the defective sites and PFRs on the catalysts facilitate the removal of TC. The degradation mechanisms of TC by Fe-BC/H₂O₂ process were proposed based on above results. Based on the identified intermediate products, the possible reaction pathways of TC were proposed. Furthermore, the ecotoxicity of TC and degradation intermediates were estimated and the acute toxicity of most intermediates was much lower than that of TC. Finally, the Fe-BC showed remarkable reusability and the adaptability for removal of TC in real samples (river water and tap water).

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  This work was supported by National Natural Science Foundation of China (No. 21906045), The 2020 Scientific Research Project for Postgraduates of Henan Normal University (No. YL202029).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi: https://doi.org/10.1016/j.cclet.2021.08.054.

  
  
• 1. [1]

     Q. Wang, Q. Xue, T. Chen, et al., Chin. Chem. Lett.32 (2021) 609-619. doi: 10.1016/j.cclet.2020.10.025

  2. [2]

     J. Cao, Z. Xiong, B. Lai, Chem. Eng. J. 343 (2018) 492-499. doi: 10.1016/j.cej.2018.03.036

  3. [3]

     D. Zhang, Q. He, X. Hu, et al., Colloid. Surf. A: Physicochem. Eng. Asp. (2021) 615. doi: 10.1007/s11042-020-09420-5

  4. [4]

     Z. He, X. Wang, Y. Luo, et al., Chemosphere 277 (2021) 130327. doi: 10.1016/j.chemosphere.2021.130327

  5. [5]

     X. Li, M. Zhou, Y. Pan, Chemosphere 212 (2018) 853-862. doi: 10.1016/j.chemosphere.2018.08.144

  6. [6]

     P. Ormad, S. Cortes, A. Puig, J.L. Ovelleiro, Water Res. 31 (1997) 2387-2391. doi: 10.1016/S0043-1354(97)00066-3

  7. [7]

     A. Zuorro, M. Fidaleo, M. Fidaleo, R. Lavecchia, J. Environ. Manage. 133 (2014) 302-308. doi: 10.1016/j.jenvman.2013.12.012

  8. [8]

     T. Huang, G.M. Zhang, S. Chong, et al., Ultrason. Sonochem. 37 (2017) 676-685. doi: 10.1016/j.ultsonch.2017.02.032

  9. [9]

     X.L. Hao, L.Y. Zou, G.S. Zhang, Y.B. Zhang, Chin. Chem. Lett. 20 (2009) 99-101. doi: 10.1016/j.cclet.2008.09.058

  10. [10]

      J. Li, Y. Li, Z. Xiong, et al., Chin. Chem. Lett. 30 (2019) 2139-2146. doi: 10.1016/j.cclet.2019.04.057

  11. [11]

      X. Li, M. Zhou, Y. Pan, J. Hazard. Mater. 353 (2018) 454-465. doi: 10.1016/j.jhazmat.2018.04.035

  12. [12]

      L. Fu, R. Wang, C. Zhao, et al., Chem. Eng. J. 414 (2021) 128857. doi: 10.1016/j.cej.2021.128857

  13. [13]

      H. Yang, C. He, L. Fu, et al., Chin. Chem. Lett. 32 (2021) 3202–3206. doi: 10.1016/j.cclet.2021.03.038

  14. [14]

      R. Wang, C. He, W. Chen, et al., Chin. Chem. Lett. 32 (2021) 3821–3824. doi: 10.1016/j.cclet.2021.05.024

  15. [15]

      V.P. Santos, M.F.R. Pereira, P.C.C. Faria, J.J.M. Órfão, J. Hazard. Mater. 162 (2009) 736-742. doi: 10.1016/j.jhazmat.2008.05.090

  16. [16]

      Q. Wang, H. Li, J. -H. Yang, et al., Appl. Catal. B: Environ. 192 (2016) 182-192. doi: 10.1016/j.apcatb.2016.03.047

  17. [17]

      M. Liu, Z. Feng, X. Luan, et al., Environ. Sci. Technol. 55 (2021) 6042-6051. doi: 10.1021/acs.est.0c08018

  18. [18]

      F. Xiao, Z. Wang, J. Fan, et al., Angew. Chem. Int. Ed. 60 (2021) 10375-10383. doi: 10.1002/anie.202101804

  19. [19]

      B. Li, L. Nengzi, R. Guo, et al., Chin. Chem. Lett. 31 (2020) 2705-2711. doi: 10.1016/j.cclet.2020.04.026

  20. [20]

      M. Ding, J. Zhou, H. Yang, et al., Chin. Chem. Lett. 31 (2020) 71-76. doi: 10.1016/j.cclet.2019.05.029

  21. [21]

      Q. Yang, Z. Feng, M. Liu, et al., Chin. Chem. Lett. 32 (2021) 2787–2791. doi: 10.1016/j.cclet.2021.01.012

  22. [22]

      X. Zhang, P. Sun, K. Wei, et al., Chem. Eng. J. 385 (2020) 123921. doi: 10.1016/j.cej.2019.123921

  23. [23]

      Z. Yang, P. Zhu, C. Yan, et al., Chemosphere 266 (2021) 129175. doi: 10.1016/j.chemosphere.2020.129175

  24. [24]

      X. Li, Y. Qin, Y. Jia, et al., Chemosphere 274 (2021) 129766. doi: 10.1016/j.chemosphere.2021.129766

  25. [25]

      X. Li, Y. Jia, M. Zhou, et al., J. Hazard. Mater. 397 (2020) 122764. doi: 10.1016/j.jhazmat.2020.122764

  26. [26]

      A.L. Cazetta, T. Zhang, T.L. Silva, et al., Appl. Catal. B: Environ. 225 (2018) 30-39. doi: 10.1016/j.apcatb.2017.11.050

  27. [27]

      Y. Long, Y. Huang, H. Wu, et al., Chem. Eng. J. 369 (2019) 542-552. doi: 10.1016/j.cej.2019.03.097

  28. [28]

      A. Shahzad, J. Ali, J. Ifthikar, et al., J. Hazard. Mater. 392 (2020) 122316. doi: 10.1016/j.jhazmat.2020.122316

  29. [29]

      S. Zhu, X. Huang, F. Ma, et al., Environ. Sci. Technol. 52 (2018) 8649-8658. doi: 10.1021/acs.est.8b01817

  30. [30]

      P. Sun, Y. Li, T. Meng, et al., Water Res. 147 (2018) 91-100. doi: 10.1016/j.watres.2018.09.051

  31. [31]

      S. Ma, J. Jing, P. Liu, et al., J. Hazard. Mater. 392 (2020) 122437. doi: 10.1016/j.jhazmat.2020.122437

  32. [32]

      S. Xin, G. Liu, X. Ma, et al., Appl. Catal. B: Environ. 280 (2021) 119386. doi: 10.1016/j.apcatb.2020.119386

  33. [33]

      Y. Chen, Y. Ma, J. Yang, et al., Chem. Eng. J. 307 (2017) 15-23. doi: 10.1016/j.cej.2016.08.046

  34. [34]

      Q. Jin, Z. Chen, Q. Chen, et al., J. Hazard. Mater. 404 (2021) 124142. doi: 10.1016/j.jhazmat.2020.124142

  35. [35]

      Y. -d. Chen, Y. -C. Lin, S. -H. Ho, et al., Bioresour. Technol. 259 (2018) 104-110. doi: 10.1016/j.biortech.2018.02.094

  36. [36]

      C. Chen, T. Ma, Y. Shang, et al., Appl. Catal. B: Environ. 250 (2019) 382-395. doi: 10.1016/j.apcatb.2019.03.048

  37. [37]

      B.M. Peterson, A.M. McNally, R.M. Cory, et al., Environ. Sci. Technol. 46 (2012) 7222-7229. doi: 10.1021/es301105e

  38. [38]

      T. Jedsukontorn, V. Meeyoo, N. Saito, M. Hunsom, Chin. J. Catal. 37 (2016) 1975-1981. doi: 10.1016/S1872-2067(16)62519-6

  39. [39]

      Z. Huang, H. Bao, Y. Yao, et al., Appl. Catal. B: Environ. 154-155 (2014) 36-43. doi: 10.1016/j.apcatb.2014.02.005

  40. [40]

      Z. Pi, X. Li, D. Wang, et al., J. Clean. Prod. 235 (2019) 1103-1115. doi: 10.1016/j.jclepro.2019.07.037

  41. [41]

      C. Liu, L. Chen, D. Ding, T. Cai, Appl. Catal. B: Environ. 254 (2019) 312-320. doi: 10.1016/j.apcatb.2019.05.014

  42. [42]

      F. Xiao, W. Li, L. Fang, D. Wang, J. Hazard. Mater. 308 (2016) 11-20. doi: 10.1016/j.jhazmat.2016.01.011

  43. [43]

      C. Liu, Z. Diao, W. Huo, et al., Environ. Pollut. 239 (2018) 698-705. doi: 10.3390/app8050698

  44. [44]

      Z. Zhao, Y. Cao, S. Li, Y. Zhang, Bioresour. Technol. 320 (2021) 124295. doi: 10.1016/j.biortech.2020.124295

  45. [45]

      C.M. Hung, C.P. Huang, C.W. Chen, et al., Environ. Pollut. 265 (2020) 114914. doi: 10.1016/j.envpol.2020.114914

  46. [46]

      Z. Li, Y. Sun, Y. Yang, et al., J. Hazard. Mater. 383 (2020) 121240. doi: 10.1016/j.jhazmat.2019.121240

  47. [47]

      S. Yang, Y. Feng, D. Gao, et al., J. Hazard. Mater. 407 (2021) 124361. doi: 10.1016/j.jhazmat.2020.124361

  48. [48]

      Y. Liu, J. Li, L. Wu, et al., Sci. Total Environ. 761 (2021) 143956. doi: 10.1016/j.scitotenv.2020.143956

  49. [49]

      Y. Cao, K. Cui, Y. Chen, et al., Solid State Sci. 113 (2021) 106548. doi: 10.1016/j.solidstatesciences.2021.106548

  50. [50]

      Z. Zhao, J. Zhao, C. Yang, Chem. Eng. J., 327 (2017) 481-489. doi: 10.1016/j.cej.2017.06.064

  51. [51]

      S.P. Mezyk, T.D. Cullen, K.A. Rickman, B.J. Mincher, Int. J. Chem. Kinet. 49 (2017) 635-642. doi: 10.1002/kin.21103

  52. [52]

      J. Wang, L. Tang, G. Zeng, et al., Appl. Catal. B: Environ. 209 (2017) 285-294. doi: 10.1504/IJCSE.2017.10008136

  53. [53]

      X. Huang, X. Hou, F. Jia, et al., ACS Appl. Mater. Inter. 9 (2017) 8751-8758. doi: 10.1021/acsami.6b16600

  54. [54]

      B. Yang, X. Cheng, Y. Zhang, et al., J. Hazard. Mater. 407 (2021) 124853. doi: 10.1016/j.jhazmat.2020.124853

• 

  Figure 1  XRD patterns of BC and Fe-BC before and after reaction (a), Raman spectra (b) and XPS survey of BC, Fe-BC and used Fe-BC (c), and magnetic hysteresis loops of Fe-BC (d).

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Influence of different system (a and b), Fe-BC dosage (c), H₂O₂ concentration (d), initial pH values (e), and initial TC concentration (f) on the TC removal using the Fe-BC/H₂O₂ system. Reaction conditions: [TC] = 20 mg/L, [H₂O₂] = 1 mmol/L, [Fe-BC] = 0.2 g/L, and initial pH 3.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Influence of different scavengers on the removal efficiency of TC (a), the EPR spectra of the Fe-BC/H₂O₂ process with DMPO and TEMP (b), and the possible activate mechanism of the TC removal using the Fe-BC/H₂O₂ process (c). Reaction conditions: [TC] = 20 mg/L, [H₂O₂] = 1.0 mmol/L, [Fe-BC] = 0.2 g/L, [TBA]₀ = 100 mmol/L, [BQ]₀ = [FFA]₀ = [KI]₀ = 20 mmol/L, [DMPO]₀ = [TEMP]₀ = 10 mmol/L, and initial pH 3.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Influence of inorganic ions and HA on TC removal by the Fe-BC/H₂O₂ process (a), TC removal in deionized water, tap water, and river water (b), the removal of different contaminants in the Fe-BC/H₂O₂ system (c), and the reusability of Fe-BC on removal of TC (d). Reaction conditions: [TC]= 20 mg/L, [H₂O₂] = 1.0 mmol/L, [Fe-BC] = 0.2 g/L, [NO₃⁻]₀ = [Cl⁻]₀ = [SO₄²⁻]₀ = 20 mmol/L, [HA]₀ = 10 mg/L, and initial pH 3.

  下载: 全尺寸图片 幻灯片
•