Efficient removal of trichloroethene in oxidative environment by anchoring nano FeS on reduced graphene oxide supported nZVI catalyst: the role of FeS on oxidant decomposition and iron leakage

Abstract

The performance of trichloroethene (TCE) removal was initially investigated in sodium persulfate (SPS) or potassium monopersulfate triple salt (PMS) oxidative environment by reduced graphene oxide (rGO) supported nZVI (nZVI-rGO) catalyst and further the role of sulphur by anchoring nano FeS on nZVI-rGO (FeS@nZVI-rGO) was evaluated. The high usage of oxidants and stability of FeS@nZVI-rGO catalyst were significantly improved due to the insoluble nature of this innovative catalyst by involvement of nano FeS which limited the rapid iron loss caused by the corrosion of active sites and mitigated rapid oxidants decomposition in FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems. The tests for target contaminant removal showed that over 95% TCE could be removed at 100 mg L⁻¹ FeS@nZVI-rGO and 1.2 mM SPS or 0.3 mM PMS dosages, in which over 85% TCE could be dechlorinated. The reactive oxygen radicals (ROSs) generation mechanisms and their contribution to TCE removal were investigated through radical scavenge tests in both systems, indicating that both HO^• and SO₄^−• were the major ROSs rather than O₂^−•. In conclusion, this study revealed the well function and fundamental mechanism of this innovative catalyst by anchoring nano FeS and worth of further demonstration of this technique in TCE contaminated groundwater remediation application.

1. Introduction

Due to the widespread prior use in various industrial application fields, chlorinated solvents remain a major contaminant in soil and groundwater, posing a serious potential risk to human beings. Trichloroethene (TCE), one of the primary chlorinated solvents, has been classified as a human carcinogen, and various techniques have been developed to remediate TCE contaminated groundwater. Advanced oxidation processes (AOPs) are effective technologies in removing TCE and other contaminants [1–3]. However, several shortcomings often limit AOPs application in-situ groundwater remediation applications to some extent. For example, the Fenton-based AOP employing hydrogen peroxide and dissolved iron is one standard approach, but it is often limited by pH restriction and the impact of iron precipitation. Thus, heterogeneous Fenton-like reaction using a solid iron catalyst has attracted attention recently for the remediation of polluted groundwater [4–6].

Nano zero-valent iron (nZVI) has been recognized as an efficient catalyst due to its nano size and high surface reactivity [6–8]. In order to take advantages of nZVI, the aggregation of nZVI caused by its stronger magnetism has to be solved before use. Therefore, a large number of solid supports with excellent physicochemical properties have been developed and applied to overcome the existing practical limitations of nZVI [9–13]. Jabeen and Chandra [14] have reported that the presence of graphene as nZVI support could successfully reduce agglomeration of nZVI and make them much water dispersible.

In recent years, sulfidation of nZVI has been reported to exhibit higher catalytic reactivity than pure nZVI [15–16]. The presence of an iron sulfide layer eliminates hydrolysis reactions at the nZVI surface, thereby improving the efficiency of electron transfer from nanoparticles to the target contaminant [17]. Although nZVI has been successfully applied for TCE removal due to the reduction reaction [7], to the best of our knowledge, the removal of TCE in sodium persulfate (SPS) or potassium monopersulfate triple salt (PMS) oxidative environments by anchoring nano FeS on reduced graphene oxide (rGO) supported nZVI catalyst (FeS@nZVI-rGO) has not been reported yet. Particularly, both the role of sulphur in promoting target contaminant removal in the process and the nano FeS function to stability of FeS@nZVI-rGO catalyst are still unclear. Further elucidation of fundamental mechanism of this innovative catalyst in generation of reactive oxygen radicals (ROSs) in system needs to be demonstrated. Therefore in this research, FeS@nZVI-rGO catalyst was synthesized by a chemical reduction method and activated using two SO₄^−• based oxic processes (SPS and PMS) to remediate TCE in aqueous solution.

The objectives of this research are：(1) to explore the physicochemical characteristics of FeS@nZVI-rGO via X-ray diffraction (XRD), scanning electron microscope (SEM), high-resolution transmission electron microscope (HR-TEM), energy-dispersive spectroscopy (EDS), and fourier-transform infrared spectrometer (FTIR) techniques; (2) to investigate the effect of the iron sulfide layer on the decomposition of oxidants and dynamic change of dissolved iron ion; (3) to assess the effect of catalyst and oxidant dosages in FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems; (4) to evaluate the effect of solution pH on TCE removal performance in both systems; (5) to elucidate the dominant reactive radicals for TCE removal in both systems through radical scavenge tests; and (6) to examine the dechlorination of TCE in FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems.

2. Materials and methods

2.1. Materials and chemicals

The following chemicals used for synthesis of FeS@nZVI-rGO were obtained from Aladdin Reagent Co. Ltd. (Shanghai, China): graphite power, ferrous sulfate heptahydrate (FeSO₄•7H₂O, 99.5%), sodium hydroxide (NaOH, 99.5%), sodium borohydride (NaBH₄, 98.0%), sodium dithionite (Na₂S₂O₄, 99.5%), trichloroethene (TCE, 99.0%), sodium persulfate (SPS, 98.0%) and potassium monopersulfate triple salt (PMS, ≥ 47.0% KHSO₅ basis, KHSO₅•0.5KHSO₄•0.5K₂SO₄). The following reagents were purchased from Shanghai Lingfeng Chemical Reagent Co. Ltd. (Shanghai, China): tertiary butyl alcohol ((CH₃)₃COH, TBA, 99%), isopropanol ((CH₃)₂CHOH, IPA, 99.7%), chloroform (CHCl₃, CF, 99%), sulfuric acid (H₂SO₄, 99%), hydrochloric acid (HCl, 37%), permanganate (KMnO₄) and n-hexane (C₆H₁₄, 97%). Ethanol (CH₃CH₂OH, 99.7%) was purchased from Shanghai Titan Scientific Co. Ltd. (Shanghai, China). The aqueous solution (ultrapure water) used in all experiments was prepared through Milli-Q water process (Classic DI, ELGA, Marlow, UK).

2.2. Synthesis of FeS@nZVI-rGO composites

Graphene oxide (GO) as the precursor of the support was prepared from graphite powder through the modified Hummers method [18]. The FeS@nZVI-rGO catalyst was synthesized according to the previously described method with some modification [19]. First, 0.5 g L⁻¹ GO was dispersed homogenously in a three-neck round-bottom flask with 2 h sonication, followed by the addition of 3 g FeSO₄•7H₂O with an additional 30 min sonication to ensure the exchange of Fe²⁺ on the GO support. Second, 3 g NaBH₄ was mixed with a specific amount of Na₂S₂O₄ in 100 mL ultrapure water to form a reducing solution and added drop-wise into the three-neck round-bottom flask with a magnetic stirrer at 400 rpm. The suspension was filtered with a 0.22 μm filter to recover the solid, which was rinsed with ultrapure water and ethanol several times. Finally, the FeS@nZVI-rGO catalyst was dried in a vacuum drying oven at 50°C for 12 h and stored for experiments.

2.3. Experimental procedures

The catalytic reactivity of FeS@nZVI-rGO was explored in batch experiments using TCE as the target contaminant. All experiments were conducted in a 250 mL cylindrical glass reactor (6.0 cm inner diameter and 9.0 cm height) with two openings at the top, one for dosing and the other for sampling. TCE solution was obtained by mixing the non-aqueous phase liquid TCE with ultrapure water under gentle stirring in dark until equilibration and then diluted to the desired concentration (initial TCE concentration = 0.15 mM). The temperature was controlled at 20 ± 0.5°C by a water bath (DC, Ningbo, China). A specific amount of FeS@nZVI-rGO with SPS or PMS was added into TCE solution and the solution was continuously mixed at 600 rpm with a magnetic stirrer. 1.0 mL aqueous samples were collected from the reactor at the desired intervals and extracted with n-hexane (1.0 mL). After 3 min vigorous stirring and 5 min quiescence to ensure separation, the samples were sealed for gas chromatography (GC) analysis. In the radical scavenge tests, the chemical scavengers along with TCE were added into the solution, and then the experiments started after the addition of FeS@nZVI-rGO and oxidant. For TCE dechlorination experiment, the aqueous samples (2.5 mL) were mixed with methanol (0.5 mL) to terminate the reaction and 1.0 mL samples were collected, left standing for 12 h, filtered and analyzed.

2.4. Analytical methods

XRD was conducted with a Rigaku D/max diffractometer (2550VB/PC) with Cu Kα radiation (λ= 0.154 nm) to investigate the crystal structure of FeS@nZVI-rGO at a scanning range from 5° to 75°. The SEM (S-3400N, Tokyo, Japan) with energy dispersive X-ray spectroscope and HR-TEM (JEM-1400, Japan) operated at 80 kV accelerating voltage were employed to identify the surface physicochemical properties of FeS@nZVI-rGO. The elemental analysis of FeS@nZVI-rGO was conducted by EDS and an EDAX advanced microanalysis AMETEK. FTIR spectroscopy (Thermo Nicolet Corp, 6700) was applied to characterize the different relationship of bond structures on the FeS@nZVI-rGO at a wave number range of 400–4000 cm⁻¹.

The GC method, particularly for TCE analysis, can be found in our previously published paper [20]. Chloride anions were analyzed by ion chromatography (Dionex ICS-1000, Sunnyvale, CA, USA). The solution pHs were recorded by a pH meter (Mettler-Toledo DELTA 320, Greifensee, Switzerland). The residual oxidant concentrations were measured by a spectrophotometric method (λ = 352 nm) [21]. The concentration of Fe²⁺ and total iron was determined by the method of o-phenanthroline spectrophotometry [22].

3. Results and discussion

3.1. Characterization of FeS@nZVI-rGO

Various catalyst characterization techniques (SEM, EDS, TEM, XRD, and FTIR) were used to explore the features of the FeS@nZVI-rGO catalyst. The SEM results (Fig. 1a) illustrate the very rough surface of the FeS@nZVI-rGO nanoparticles. According to the TEM results of the FeS@nZVI-rGO catalyst as shown in Fig. 1b and c, the size of FeS@nZVI-rGO catalyst was in a range of 100–300 nm. The results of TEM further displayed a core-shell structure of the catalyst that the black nZVI core was clearly observed and the surface of nZVI was covered with filamentous FeS which formed the FeS shell.

Fig. 1.

Catalyst surface morphologies of (a) SEM image of FeS@nZVI-rGO; (b) TEM image of FeS@nZVI-rGO distribution; (c) TEM image of single FeS@nZVI-rGO.

The X-ray diffraction (XRD) analysis was implemented to explore the crystal structures of FeS@nZVI-rGO with a wide 2θ range of 5–75°, indicating that all catalysts had a high-intensity diffraction peak at 2θ = 44.67°, corresponding to the α-Fe⁰ on the (1 1 0) diffraction plane [23] [Joint Committee on Powder Diffraction Standards (JCPDS) 06–0696] (Fig. 2). However, no obvious diffraction peaks of FeS were obtained on the XRD image of FeS@nZVI-rGO catalyst, which may be attributed to the low concentration or low crystalline degree [19]. According to the EDS analysis, the iron and sulfur element contents of FeS@nZVI-rGO account for 79.1 wt% and 3.6 wt%, respectively (Fig. 3). Thus, the iron type of FeS@nZVI-rGO was nZVI covered by less FeS, which was in consistent with our prepared stoichiometric composition. The peak of Mn presented in the EDS analysis of FeS@nZVI-rGO could be from the GO material during its preparation.

Fig. 2.

X-ray diffraction (XRD) images of nZVI, nZVI-rGO and FeS@nZVI-rGO.

Fig. 3.

Energy-dispersive spectroscopy (EDS) test of FeS@nZVI-rGO.

The functional groups and chemical bonds of the catalysts were explored by FTIR spectroscopy (Fig. 4). The wide band observed at 3423 cm⁻¹ indicated the presence of OH functional group. Other characteristic peaks of rGO support could be found at 1710, 1130, and 1085 cm⁻¹, corresponding to the C=O group, epoxy group, and deformed C-O stretching vibration, respectively. The peak at 1558 cm⁻¹ was assigned to carbon in the rGO support. A sharp peak at 578 cm⁻¹ indicated the Fe-O group on nZVI-rGO and FeS@nZVI-rGO catalysts. In comparison, the obvious new characteristic peak was observed at 478 cm⁻¹ and signified the presence of sulfur, illustrating that nano FeS particles were well dispersed on the surface of FeS@nZVI-rGO catalyst [24].

Fig. 4.

Fourier-transform infrared (FTIR) spectra of GO sheet, nZVI-rGO and FeS@nZVI-rGO.

3.2. TCE removal performance in FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems

Batch experiments were conducted with FeS@nZVI-rGO catalyst to explore the removal efficiency of TCE using SPS and PMS as oxidants, respectively. No significant TCE removal was observed in the absence of nZVI catalysts or oxidants. However, as shown in Fig. 5a, TCE was removed completely in the nZVI-rGO/SPS system, which was caused by the improved catalytic reactivity of nZVI-rGO, and this synergistic effect between rGO support and nZVI catalyst was also reported in our previous research [25].

Fig. 5.

TCE removal performance with different catalysts in (a) SPS and (b) PMS systems. [Catalyst]₀ = 100 mg L⁻¹, [SPS]₀ = 1.2 mM, [PMS]₀ = 0.3 mM, [TCE]₀ = 0.15 mM.

Further experiments were carried out to investigate the effect of FeS loading on FeS@nZVI-rGO catalyst. TCE removal efficiency reached 97.9% in FeS@nZVI-rGO/SPS system at Fe/S molar ratio of 2:1. However, a slight decrease of TCE removal efficiency (90.7%) was observed after changing the Fe/S molar ratio to 1:1. These results illustrate the double functions of FeS for FeS@nZVI-rGO catalyst. TCE removal efficiency could be improved at low sulfur content, which was caused by the following reasons. First, the electron transfer from the surface of FeS@nZVI-rGO to SPS was accelerated by the existence of doped FeS [26]. Second, Fe²⁺ was released by FeS on the surface of FeS@nZVI-rGO, and then catalyzed SPS to generate reactive oxygen radicals (ROSs) [27]. Third, because of the higher electronegativity of FeS (5.02 eV) than that of nZVI (4.04 eV), electrons generated by nZVI corrosion could spontaneously transfer to the FeS semiconductor that could promote TCE removal by electron transfer [28]. Conversely, previous research also reported that too much FeS could cover the nZVI core and decrease the number of active sites on the catalyst surface, which might delay the electron transfer between FeS@nZVI and SPS or TCE [29]. According to Fig. 5b, the reaction of PMS activated by iron catalyst was very quick and only took 10 min to complete TCE removal compared to SPS which needed 30 min. As the result of SPS asymmetrical structure, PMS was activated more easily by iron catalyst than SPS, thus, TCE removal was more susceptible to the features of catalysts. The result of Fig. 5b also shows that TCE removal efficiency within the initial 1 min reached 45.7% and 42.8% in the PMS system activated by FeS@nZVI-rGO at Fe/S molar ratio = 2:1 and 1:1, respectively. Therefore, in order to clearly explore the role of S in the FeS@nZVI-rGO catalyst, the Fe/S molar ratio of 2:1 was selected for the following experiments.

The effect of FeS@nZVI-rGO and SPS or PMS dosage on TCE removal performance was further studied in FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems. As shown in Fig. S1a and c, TCE removal efficiency increased from 25.5% to 97.9% in FeS@nZVI-rGO/SPS system with the SPS concentration increase from 0.15 mM to 1.2 mM, and this value was elevated from 40.7% to 99.3% with the PMS dosage increasing from 0.0375 mM to 0.3 mM in FeS@nZVI-rGO/PMS system. It could be observed that the removal of TCE was improved with the increase of oxidants dosage, strongly showing the effectiveness of SPS and PMS activated by FeS@nZVI-rGO in which the ROSs such as HO^• and SO₄^−• were generated. The dosage effect of FeS@nZVI-rGO in SPS and PMS systems was displayed in Fig. S1b and d. The increased TCE removal efficiency (from 73.5% to 97.9% for SPS; from 52.1% to 99.3% for PMS) along with the increasing dosage of FeS@nZVI-rGO catalyst illustrated the unsaturated catalyst dosage for the complete TCE removal up to 100 mg L⁻¹ in SPS and PMS systems [30]. Considering the results of this study, the best TCE removal was achieved at the FeS@nZVI-rGO and SPS dosages of 100 mg L⁻¹ and 1.2 mM in FeS@nZVI-rGO/SPS system or FeS@nZVI-rGO and PMS dosages of 100 mg L⁻¹ and 0.3 mM in FeS@nZVI-rGO/PMS system, respectively. Therefore the initial dosages of FeS@nZVI-rGO and SPS or PMS were set at 100 mg L⁻¹ and 1.2 mM or 0.3 mM as a control in the following experiments.

3.3. TCE removal mechanisms and dechlorination pattern

As mentioned in section 3.2, several ROSs may generated in the SPS and PMS oxidant systems that contribute to TCE removal. Hydroxyl radicals (HO^•) has been reported to co-exist with sulfate radicals (SO₄^−•) and superoxide radicals (O₂^−•) in SPS and PMS systems [31–32]. Therefore, in order to identify the contribution of the above radicals to TCE removal, the radical scavenge tests were implemented by using TBA as HO^• quenching agent, IPA as both HO^• and SO₄^−• quenching agent, and CF as O₂^−• quenching agent [33]. As shown in Fig. 6a, only 49.8% and 16.4% TCE were removed upon the addition of 100 mM TBA and 100 mM IPA in nZVI-rGO/SPS system, but over 94.5% TCE removal was still achieved in the presence of 50 mM CF. These results indicated that both HO^• and SO₄^−• played the important role in TCE removal, while O₂^−• did not contribute much to TCE removal. The removal of TCE in FeS@nZVI-rGO/SPS system was similar to nZVI-rGO/SPS system with the addition of scavengers, for instance, TCE also deceased to 62.3% and 24.8% in the presence of 100 mM TBA and 100 mM IPA, while over 85.5% TCE removal was reached in the presence of 50 mM CF, suggesting that both HO^• and SO₄^−• were the major ROSs except O₂^−•. The same results were obtained in nZVI-rGO/PMS and FeS@nZVI-rGO/PMS systems. TCE removal efficiency in both nZVI-rGO/PMS and FeS@nZVI-rGO/PMS systems suffered a significant inhibition after addition of excess TBA and IPA, indicating that HO^• and SO₄^−• also were the domain radicals responsible for TCE removal. However, a high TCE removal efficiency was maintained in both nZVI-rGO/PMS/CF and FeS@nZVI-rGO/PMS/CF systems, revealing that O₂^−• contributed less to TCE removal. The possible chemical reactions involved in SPS or PMS oxidative environments with nZVI-rGO and FeS@nZVI-rGO catalysts were listed in Table S1.

Fig. 6.

Effect of radical scavengers on TCE removal performance with different catalysts in (a) SPS and (b) PMS systems. [nZVI-rGO]₀ = [FeS@nZVI-rGO]₀ = 100 mg L⁻¹, [TBA]₀ = [IPA]₀ = 100 mM, [CF]₀ = 50 mM, [SPS]₀ = 1.2 mM, [PMS]₀ = 0.3 mM, [TCE]₀ = 0.15 mM.

Based on above discussion, a series of reaction mechanisms can be proposed for both FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems. (1) Large amounts of HO^• and SO₄^−• were generated on the surface of FeS@nZVI-rGO catalyst, and then reacted with adsorbed TCE; (2) Fe²⁺ released from FeS@nZVI-rGO catalyst would also diffuse into solution to activate oxidants, generating HO^• and SO₄^−• which participated in TCE removal. It has been pointed out that before transferring into the solution, part of Fe³⁺ on the catalyst surface could be reduced to Fe²⁺ by S²⁻ (Eq. (1)) [34–35]; (3) nZVI was determined to play a main role on the generation of ROSs along with the support by FeS, which was confirmed by 12.5 wt% and 3.5 wt% loss of iron and S element (Fig. S2).

$$2{\text{Fe}}^{3+}+{\text{HS}}^{-}\to 2{\text{Fe}}^{2+}{\text{S}}^0+{\text{H}}^{+}$$ (1)

Theoretically, the concentration of chloride ion (Cl⁻) in the solution increases at 3 times the rate of TCE removal when considering complete TCE degradation, which was analyzed and displayed in Fig. 7. When TCE loss (< 3%) caused by volatilization was taken into account, the final dechlorination of TCE reached 87.2% and 95.6% in FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems, respectively. These results suggested that the majority of TCE was completely dechlorinated in both systems and that FeS@nZVI-rGO/PMS system had a better TCE dechlorination extent than FeS@nZVI-rGO/SPS system. Therefore, FeS@nZVI-rGO/PMS system is more suitable to treat groundwater polluted by TCE.

Fig. 7.

Chloride ion liberation and its mass balance in (a) FeS@nZVI-rGO/SPS and (b) FeS@nZVI-rGO/PMS along with TCE removal. [nZVI@FeS-rGO]₀ = 100 mg L⁻¹, [SPS]₀ = 1.2 mM, [PMS]₀ = 0.3 mM, [TCE]₀ = 0.15 mM.

3.4. Identification of oxidant decomposition and Fe species

Although TCE could be removed completely with or without FeS coating on nZVI-rGO, in order to further distinguish the function of FeS@nZVI-rGO and nZVI-rGO, the difference of TCE removal performance in FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems was evaluated by measuring the residual concentration of oxidants. As shown in Fig. S3a, the residual SPS concentration was 0.39 mM in FeS@nZVI-rGO/SPS system and 0.25 mM in nZVI-rGO/SPS system after reaction, representing 32.5% and 24.6% of the initial SPS dosage, indicating that the usage of SPS decreased when FeS@nZVI-rGO was used. The residual PMS concentration was as less as 0.02 mM in both nZVI-rGO/PMS and FeS@nZVI-rGO/PMS systems, but the rate of PMS decomposition in FeS@nZVI-rGO/PMS system was slower than in nZVI-rGO/PMS system, which could be caused by less iron loss from the catalyst surface into the solution (Fig. S3b). Compared with nZVI-rGO, both FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems displayed relatively slow oxidant decomposition property, which may be caused by less iron loss from the catalyst surface to the solution. Doping FeS on the catalyst surface improved the insoluble feature of FeS@nZVI-rGO. Fan et al. [36] also confirmed that ZVI was unstable and easily corroded by dissolved oxygen and protons, especially under acidic condition, and excess ferrous iron consumed the generated reactive oxidants that further inhibited contaminant degradation. In contrast, FeS was more stable and insoluble, even under acidic condition, and the decomposition of SPS caused by FeS was relatively slow but consequently efficient compared with ZVI.

The changes of Fe species and total iron in the solution were explored to further confirm the insoluble features of FeS@nZVI-rGO catalyst. According to Fig. 8, the dissolved Fe²⁺ concentration was very low in SPS and PMS systems catalyzed by either nZVI-rGO or FeS@nZVI-rGO, indicating that surface Fe²⁺ had a more vital role than dissolved Fe²⁺ and that the majority of ROSs may be generated on the catalyst surface. The final concentration of dissolved total iron was 0.396 mM in nZVI-rGO/SPS system and 0.296 mM in FeS@nZVI-rGO/SPS system (Fig. 8a), showing about 25.3% less of dissolved Fe released in FeS@nZVI-rGO/SPS system when compared to nZVI-rGO/SPS system. In nZVI-rGO/PMS system (Fig. 8b), the final concentration of dissolved total iron (0.251 mM) was still higher than that in FeS@nZVI-rGO/PMS system (0.211 mM), also indicating about 15.9% less of dissolved Fe released in FeS@nZVI-rGO/PMS system compared to nZVI-rGO/PMS system. These results demonstrated a better corrosion resistance ability of FeS@nZVI-rGO in both SPS and PMS systems. Conversely, the rapid decomposition of SPS or PMS released large amounts of H⁺ and severely corroded the surface structure of nZVI-rGO without the protection of FeS. Higher concentrations of dissolved total iron were obtained in FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems within initial 3 min than those in nZVI-rGO/SPS and nZVI-rGO/PMS systems, which well corresponded to the higher TCE removal efficiency. The better electronegativity of FeS may accelerate the rapid electron transformation between nZVI and oxidants.

Fig. 8.

The concentration of dissolved Fe²⁺ and total iron in (a) FeS@nZVI-rGO/SPS and (b) FeS@nZVI-rGO/PMS systems. [FeS@nZVI-rGO]₀ = 100 mg L⁻¹, [SPS]₀ = 1.2 mM, [PMS]₀ = 0.3 mM, [TCE]₀ = 0.15 mM.

3.5. Effect of initial solution pH on TCE removal

It is widely known that pH plays a vital role on the generation of ROSs. Therefore, various initial solution pHs (3–11) were tested to explore the influence of pH on TCE removal performance in FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems. As shown in Fig. S4, TCE removal efficiency reached more than 90% for the initial solution pHs from 3 to 10 in both systems, suggesting that FeS@nZVI-rGO has a wider applicable pH range for efficient TCE removal. However, an obvious decline of TCE removal efficiency, merely achieving 14.7% and 23.2%, was observed in both FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems at initial solution pH = 11, which could be caused by the adverse effect of high final solution pH. For the initial solution pH of 3 to 10, the final solution pHs were all maintained at approximately 3, indicating that SPS or PMS was activated by FeS@nZVI-rGO successfully along with the release of large amounts of H⁺. However, the final solution pH was 10.8 when the initial solution pH was 11. In addition, Fig. S5 shows that the total iron concentration decreased from 0.68 mM to 0.001 mM with the initial solution pH increased from 3 to 11. These results indicated that high final solution pH may promote the generation of particulate amorphous and crystalline oxydroxides on FeS@nZVI-rGO surface, thus preventing the release of large amount of Fe²⁺ on catalyst surface.

4. Conclusion

The role of sulphur in FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems for the removal of TCE was investigated in this study. The catalyst, created by anchoring nano FeS particles on the surface of nZVI-rGO, was characterized and the results suggested that nano FeS particles could restrict the rapid iron loss caused by the corrosion of active sites, and therefore mitigate rapid oxidant decomposition in both FeS@nZVI-rGO/SPS and FeS@nZVI-rGO/PMS systems. Over 95% TCE removal could be achieved at the dosages of FeS@nZVI-rGO and SPS or PMS of 100 mg L⁻¹ and 1.2 mM or 0.3 mM, in which over 85% TCE could be dechlorinated. The generation of reactive oxygen radicals (ROSs) were confirmed to occur primarily on the surface of FeS@nZVI-rGO and radical scavenge tests identified SO₄^−• and HO^• as the dominant radicals in both systems. The effect of initial solution pH on TCE removal performance was tested further and the results indicated that TCE could be completely removed within a wider pH range of 3–10, thus extended the applicability of this technique. This study revealed the well function and fundamental mechanism of this innovative catalyst by anchoring nano FeS and demonstrated the great potential of this technique in the remediation of TCE contaminated groundwater.