Heterogeneous activation of H2O2/Na2S2O8 with iron ore from water distribution networks for pollutant removal

In this study, we investigated to use the main composition of pipe deposits from water distribution networks as catalyst to activate dual-oxidant H2O2/Na2S2O8 system to produce radicals for perchloroethylene and chloramphenicol removal. According to the results, the degradation efficiency of perchloroethylene by H2O2/Na2S2O8 system was 92.05% within 8 h. Due to the slow conversion between ≡Fe3þ and ≡Fe2þ, the hydroxylamine was introduced to reduce reaction time. As the results, the degradation efficiency of chloramphenicol in the H2O2/Na2S2O8 system with hydroxylamine assistance was 73.31% within 100 min. Meanwhile, several key affecting factors and the kinetic models were investigated. The primary radicals were identified by electron paramagnetic resonance and radical scavenging tests. Eleven degradation products were confirmed by high-resolution liquid chromatography-mass spectrometry. The result of this study provided the theoretical basis for resource utilization of pipe deposits in water treatment in case of emerging contamination events.

including α-FeOOH, γ-FeOOH, γ-Fe 2 O 3 and Fe 3 O 4 , etc. (Oliveira et al. 2012;Cui et al. 2016). The pipe deposits could be used as inexpensive catalysts in heterogeneous Fenton-like oxidation, which attracted increasing attention (Oliveira et al. 2012). In our previous study, we explored the adsorption and degradation capacity of simulated pipe deposit(SGR, the mixture of the major constituents of actual pipe deposit (α-and γ-FeOOH)) for 5-bromosalicylic acid(5-BSA), 32.65% and 87.67% of 5-BSA could be removed by adsorption and SGR/H 2 O 2 oxidation (Huang et al. 2021). The findings of previous study would make an important contribution to pipe deposits recycling. However, the removal time was too long, and the mechanisms and kinetics were not discussed in detail.
In this study, we analyzed and prepared the main composition of pipe deposits, γ-FeOOH, which is used as a catalyst to activate HP/PS in order to produce radicals for PCE and CAP removal. The hydroxylamine (NH 2 OH, HA) was introduced to increase the degradation rate and shorten the degradation time. Furthermore, the underlying mechanism was investigated by radical scavenging test (RST), electron paramagnetic resonance (EPR) and X-ray photoelectron spectroscopy (XPS). The method may be expected to be used for in situ treatment of polluted water of the network emergency contamination.

Preparation of Lepidocrocite
Lepidocrocite (γ-FeOOH) was synthesized at 25°C. Firstly, 4 g FeCl 2 ·4H 2 O were added to 100 mL deionized water and continuously stirred, and maintained the solution pH at 6.8 + 0.1 by adding 1.0 mol L À1 NaOH. Then, air (36 À 72 L h À1 ) was sent into the solution and maintained pH at 6.8 + 0.1 until the color of suspension changed from dark green to orange. Finally, the prepared γ-FeOOH was centrifuged and washed with deionized water and dried (Antony et al. 2005).

PCE and CAP removal and detection
All batch removal experiments were conducted in a 100 mL capped bottle at 25°C. A certain number γ-FeOOH were added to 100 mL aqueous solution containing PCE or CAP. HA, HP and PS were added to the abovementioned solution at appropriate concentrations to perform degradation experiments. 0.1 mol L À1 HCl and 0.1 mol L À1 NaOH was used to adjust the pH of pollutant solution. The PCE was extracted by n-hexane and the residual concentration was detected by GC-ECD. The CAP residual concentration was detected by a UV-vis spectrophotometer at 277.6 nm. The detected peak did not shift which indicated that the addition of reagents did not influence CAP detection, as shown in Fig.S1. All experiments were conducted in triplicate.

Confirmed the rate constant of the pollutant with radicals
The rate constant of CAP with radicals (k ·OH,CAP ) was studied by competition kinetics method (Cao et al. 2020). pCBA (k ·OH, pCBA ¼ 5.00 Â 10 9 M À1 s À1 ) was used for the competitor. The adsorption of CAP and pCBA by γ-FeOOH and photodegradation were negligible. The rate constant of CAP with •OH could be estimated by the following equation: Rearranging Equation (1) got Equation (2) k •OH,pCBA was the reaction rate constant of •OH with pCBA. k •OH,CAP was the reaction rate constant of •OH with CAP.

Characterization
Scanning electron microscope (SEM, ZEISS Sigma 500, Instrument Parameter: EHT ¼ 15.00 kV, Signal A ¼ SE2, WD ¼ 8.4 mm, Mag ¼ 20.00 KX) was used to observe the morphology of γ-FeOOH. X-ray diffraction with 2θ range from 10 to 90°was used to analyzed the crystalline structures of γ-FeOOH(XRD, AXS D8 ADVANCE with CoKα radiation (λ ¼ 1.54060 Å)). X-ray photoelectron spectroscopy (XPS, AXIS ULTRA DLD using Al Ka radiation (1,486.6 eV)) were used to explore the surface valence state of γ-FeOOH before and after degradation. The chemical groups on γ-FeOOH surface were detected by Fourier transformation infrared spectroscopy (FTIR, Perkin Elmer Spectrum One). The main radicals were detected by Electron paramagnetic resonance (EPR, Bruker, Germany) with DMPO as the spin-trapping agent. The degradation products were determined by high-resolution LC-MS in negative ionization mode using an Agilent 1260 HPLC coupled with ABSciexQTrap 5500 MS with an electrospray ionization (ESI) source. The MS instrumental parameters were set as follows: source temperature, 500°C; Gas I, 50 arbitrary units; Gas II, 50 arbitrary units; negative ion spray voltage, 4,500 V; curtain gas, 50 arbitrary units; entrance potential (EP), 10 V; declustering potential (DP), 100 V; collision energy (CE), 30 V; collision cell exit potential (CXP), 20 V; PIS scan range, 100-1,000 amu.

RESULTS AND DISCUSSIONS
3.1. Morphology and structural characterization of γ-FeOOH γ-FeOOH was a rod-type structure by the SEM observation (Figure 1(a)). The XRD patterns (Figure 1(b)) showed that the diffraction peaks at 2θ ¼ 14.16°, 27.16°, 36.14°and 46.89°were attributed to (200), (210), (301) and (501)    . The stronger intensity and narrower half peak width suggested that γ-FeOOH was a consistent crystal structure. According to the FTIR spectrum ( Figure 1(c)), the peak appeared at 1,021 cm À1 was the in-plane of Fe-O-H bending vibrations, while the peak appeared at 753 cm À1 was for the out-plane (He et al. 2018). These results verified the successful preparation of γ-FeOOH. Figure 1(d) showed that the N 2 adsorptiondesorption isotherms of the prepared γ-FeOOH is type IV with H3 hysteresis loop, demonstrating its microporous/mesoporous structures Liu et al. 2019). Meanwhile, γ-FeOOH has a rather larger specific surface area (117.567 m 2 /g), average pore size (23.8672 nm) and the pore volume (0.6382 cm 3 /g). The larger specific surface area could expose more active sites, which was conducive to increasing reaction activity (Lyu et al. 2021).

The degradation PCE and CAP in different systems
The PCE degradation under different systems was explored, the result was showed in Figure 2(a). Due to the volatilization and adsorption, 15.33% PCE was removed. 92.05% PCE was removed in HP/PS system in 8 h. The PCE degradation efficiency in dual-oxidant system was higher than in single-oxidant system, HP/PS . HP . PS. The removal of CAP in γ-FeOOH, γ-FeOOH/HP/PS, γ-FeOOH/HA/HP, γ-FeOOH/HA/PS and γ-FeOOH/HA/HP/PS was shown in Figure 2(b). When γ-FeOOH was added alone, CAP was not removed indicating that γ-FeOOH had no adsorption. In addition, CAP could not be degraded effectively by γ-FeOOH/HP/PS system, indicating that γ-FeOOH cannot activate HP and PS to produce radicals. When the HA was introduced into system and the usage of oxidant was the same, the degradation efficiency of CAP in γ-FeOOH/HA/HP, γ-FeOOH/HA/PS and γ-FeOOH/HA/HP/PS system was 63.18%, 14.11% and 73.31% in 100 min, respectively. And the CAP degradation in HP/PS, HP and PS system could be well fitted by the first-order kinetics model as expressed by Equation (3) (Zhou et al. 2014), the detail fitting results of CAP were shown in Figure 2(c) and Table 1.
Water Science & Technology Vol 00 No 0, 4 Uncorrected Proof k is the kinetic rate constant, which can be calculated from the slope of the straight line.
The rate constant of γ-FeOOH/HA/HP/PS system was 0.0173 min À1 , which was higher than γ-FeOOH/HA/HP(0.0103 min À1 ) and γ-FeOOH/HA/PS (0.0023 min À1 ). The results indicated that the HP/PS system has the unique advantages in pollutant removal because of synergistic effect.

Effect of experimental conditions
AOPs are complicated systems, and many factors could influence their performance. In order to better clarify the mechanism of dual-oxidant system, the detailed effect of different experimental conditions on CAP degradation were studied.

Effect of the initial pH
The initial pH is a significant role in the AOPs. In the γ-FeOOH/HA/HP/PS system, the initial pH of solution was adjusted by using sodium hydroxide or hydrochloric acid and no buffering agents were introduced. As shown in Figure 3(a), the degradation efficiency was over 75% when the pH ranged from 3.0 to 9.0. However, the degradation was obviously inhibited at pH ¼ 11. The variation of solution H þ concentration was monitored (Figure 3(b)). The eventual pH all dropped to acidic, as shown in Table 2. The acidic conditions favored the reaction between γ-FeOOH and oxidant to produce radicals for target pollutant removal. According to the previous studies, if the eventual pH was higher than 4, the iron precipitation formed on the surface of γ-FeOOH that lead to a passivation phenomenon, which was harmful for the CAP degradation. The eventual pH was 6.35 when the initial pH ¼ 11, the CAP degradation was suppressed. Our results were consistent with previous studies (Rao et al. 2014). The increased H þ was because the reaction between SO †À 4 and H 2 O to form •OH via one-electron transfer process, as shown in Equation (4). These results demonstrated that the HP/PS dual-oxidant process can remove CAP in acidic and slightly alkaline.

Effect of the initial HP or PS concentration
HP and PS are the main source of radicals. It is important to optimize the dose of oxidants for the effective degradation of organic pollutants. Figure 3(c) and 3(d) described the effect of HP and PS concentration on CAP degradation. The increase of initial HP and PS concentration resulted in a noticeable enhancement in CAP degradation, which was because the enough radicals. According to the experimental results, 2 mmol L À1 PS and 4 mmol L À1 HP was the optimum dosage, the CAP degradation efficiency could reach to 78.43% in 100 min. However, the HP or PS concentration were further increased, the degradation efficiency decreased. The excessive HP and PS might result in the scavenge of •OH and SO †À 4 into the lower oxidizing ability species, which lead to the low degradation efficiency. The detail process was shown in Equations (5)-(8) (Cai et al. 2016;Gao et al. 2022)

Effect of the iron ore dosage
The influence of the iron ore dosage on CAP degradation was investigated using the catalyst from 1.4 mmol L À1 to 11.2 mmol L À1 , as shown in Figure 3(f). The increase of catalyst dosage provided more active sites for the reaction of HP/PS with antibiotics, resulting in high degradation rate. In contrast to our previous studies about the goethite activating HP/PS system ), the lepidocrocite system did not have an induction period because the lepidocrocite has a larger specific surface area than goethite, which causes the more ≡Fe 3þ to be reduced and to accumulate more radicals with the catalyst dosage at the same time.

The residual HP and PS concentration and the kinetics fitting curves
The Figure 4(a) exhibited the residual HP and PS concentration in γ-FeOOH/HA/HP /PS system. After 100 min, the consumption of HP and PS was 75.37% and 42.81%, and the decomposition of PS was two-stage reaction. The Figure 4(b) Water Science & Technology Vol 00 No 0, 6 Uncorrected Proof and 4(c) were the kinetics fitting curves of HP and PS decomposition in γ-FeOOH/HA/HP/PS system, respectively. The detail kinetic parameters were shown in Table 1. HP decomposition was fitted well by the first-order kinetics model and the rate constant was 0.0139 min À1 . PS decomposition was a two-phase reaction. In the initial phase(0-60 min), the rate constant was 0.0076 min À1 . In the second phase (60-100 min), the rate constant was 0.0024 min À1 . At the beginning of the reaction, PS was rapidly decomposed due to the accumulation of a large amount of ≡Fe 2þ by HA reduction. The HA was consumed which led to the slow conversion of ≡Fe 3þ to ≡Fe 2þ . Therefore, the PS decomposition entered a slow process.

The confirmation of active species and reaction rate constants
In order to confirm the primary radicals in CAP degradation by γ-FeOOH/HA/HP/PS system, DMPO was used as a trapping agent to conduct the EPR test. DMPO can react with radicals to generate adducts, such as DMPO-OH and DMPO-SO 4 . The  Uncorrected Proof strong EPR signal (shown in the Figure 5(a)) was DMPO-OH, simultaneously, the low peaks were most likely to be DMPO-SO 4 . PS may be first activated by γ-FeOOH to produce SO †À 4 and then react with the adsorbed water to lead to the production of abundant •OH via one-electron transfer process, which was consistent with the H þ increase. The similar phenomenon has reported in previous studies (Huang et al. 2014;Duan et al. 2016).
The competition kinetic method was used to test the reaction rate constants of CAP with radicals at original pH by using pCBA as the competitor (Cao et al. 2020), as shown in Figure 5(c). According to our previous research, the k ·OH,CAP was 6.10 Â 10 9 M À1 s À1 in γ-FeOOH/HA/HP system. The k ·OH,CAP was 4.24 Â 10 10 M À1 s À1 in the initial phase(0-60 min) and 8.15 Â 10 9 M À1 s À1 in the second phase(60-100 min) in γ-FeOOH/HA/HP/PS system, respectively. The rate was higher in double oxidant system than in single oxidant system. The result further confirmed the advantages of dual-oxidant system.

The role of hydroxylamine
The introduction of HA into HP/PS system could lead to a significant increase of the CAP removal efficiency. The degradation efficiency of CAP increased from 3.69% to 81.60% with HA concentration increasing from 0 mmol L À1 to 3 mmol L À1 , but the degradation efficiency decreased slightly when the HA concentration increase to 4 mmol L À1 , as shown in Figure 3(e). The HA accelerated the reduction of ≡Fe 3þ to produce ≡Fe 2þ for activating HP and PS. However, the competitive consumption of radicals with excess HA was the reason for the slight decrease of CAP degradation efficiency (Luo et al. 2021b). The dynamic fitting curves of CAP degradation under different HA dosage was shown in Figure 6(a). The whole reaction process was a two-stage reaction kinetics. In the fast reaction stage, plenty of ≡Fe 3þ was rapidly restored by HA, which caused the fast degradation of CAP. After HA was consumed, the generation of ≡Fe 2þ was slow, the formation of reactive species was hindered, leading to a follow-up slow oxidation stage .
The XPS spectra of γ-FeOOH before and after the catalytic reaction were favorable to evaluate the changes of Fe (Figure 6(b)) and O (Figure 6(c)) valence state in the HP/PS activation. The peaks at 711.61 eV and 725.21 eV belonged to the binding energy of 2p 3/2 and 2p 1/2 of ≡Fe 3þ of γ-FeOOH. After degradation, the new peaks at 710.65 eV and 724.25 eV appeared which were ≡Fe 2þ (Guo et al. 2019a;Wang et al. 2020). And ≡Fe 3þ and ≡Fe 2þ were 58.22% and 41.78% of the total iron surface atoms after reaction. This result suggested that the reduction of ≡Fe 3þ to ≡Fe 2þ occurred on the surface of γ-FeOOH. In addition, the O 1s peaks could be split into the lattice oxygen (O lat , Fe-O-Fe and Fe-OH) at 530.03 eV and 531.20 eV, the surface adsorbed oxygen (O surf ) at 532.04 eV and the adsorbed molecular water (O ads ) at 533.50 eV with atomic ratio of 69.84%: 23.78%: 6.38% before the degradation (Baltrusaitis et al. 2007;Kim et al. 2021). After the degradation, the atomic ratio of O lat , O surf and O ads was 45.71%: 37.35%: 16.95%, respectively. The ratio of O lat reduced, the ratio of O surf and O ads increased. The increase in the proportion of O ads might be attributed to the more adsorbed molecular water on the surface of γ-FeOOH. The slight decrease in O lat was attributed to the surface reduction from ≡Fe 3þ to Uncorrected Proof ≡Fe 2þ . If the leaching iron concentration was less than 3.92 mg L À1 , the process was heterogeneous catalysis (Chou et al. 2001). The leaching concentration of iron ion was 0.843 mg L À1 after 100 min in HP/PS system. All the results confirmed that the degradation process was occurs on the γ-FeOOH surface.
The degradation intermediates were detected through LC-MS, and eleven degradation intermediates were observed, and the details are exhibited in Figure 7 and Fig.S2. The oxidation and dehydration of CAP to produce P 1 . The center C connecting aromatic ring was attacked to produce P 2 and P 3 . Simultaneously, the ring was opened to form the intermediate P 4 (Yang et al. 2021). The radical break CAP by dechlorinate, dehydration and oxidation that result in the formation of P 5 -P 7 . The •OH attacked P 5 to generate the hydroxylated product P 8 (Guo et al. 2019b). A further attacking by radicals led to the the formation of P 9 -P 11 .
In brief, the CAP degradation process by γ-FeOOH could be depicted as follow. The HA restored the ≡Fe 3þ into ≡Fe 2þ . The HP and PS interacted with ≡Fe 2þ to produce highly oxidative radical species(•OH or SO †À 4 ). Then, these radical species attacked CAP and effectively transformed it into intermediates P 1 -P 11 .

CONCLUSIONS
In order to resource utilization of pipe deposits from water distribution networks, the primary composition of pipe deposits was employed as catalyst to active HP and PS to produce radicals for degradation pollutant. 92.05% PCE was oxidative  Uncorrected Proof degradation after 8 h. 73.31% CAP was removed in 100 min after the HA introduction. Furthermore, the reaction rate was HP/PS(0.0173 min À1 ) . HP(0.0103 min À1 ) . PS(0.0023 min À1 ) and the reaction rate constant of radicals with CAP in HP/ PS system was obvious higher. All the results implied that the dual-oxidant system has the unique advantages in pollutant removal. According to the result of XPS and ICP, the CAP degradation by HP/PS system was surface reaction. Eleven degradation intermediates were detected by LC-MS. The pipe deposits from water distribution networks has capability to remove PCE and CAP in water. Although the effects of water quality and hydraulic conditions were not investigated, but this study highlighted the advantage of HP/PS dual-oxidant system and the potential application of pipe deposits in water treatment in case of emerging contamination events.