Environ Sci Pollut Res DOI 10.1007/s11356-015-5034-1

RESEARCH ARTICLE

Efficient degradation of trichloroethylene in water using persulfate activated by reduced graphene oxide-iron nanocomposite Ayyaz Ahmad 1 & Xiaogang Gu 1,2 & Li Li 1 & Shuguang Lv 2 & Yisheng Xu 1 & Xuhong Guo 1,3

Received: 23 May 2015 / Accepted: 3 July 2015 # Springer-Verlag Berlin Heidelberg 2015

Abstract Graphene oxide (GO) and nano-sized zero-valent iron-reduced graphene oxide (nZVI-rGO) composite were prepared. The GO and nZVI-rGO composite were characterized by transmission electron microscopy (TEM), Fourier transform infrared (FTIR), energy-dispersive spectroscopy (EDS), and Raman spectroscopy. The size of nZVI was about 6 nm as observed by TEM. The system of nZVI-rGO and persulfate (PS) was used for the degradation of trichloroethylene (TCE) in water, and showed 26.5 % more efficiency as compared to nZVI/PS system. The different parameters were studied to determine the efficiency of nZVI-rGO to activate the PS system for the TCE degradation. By increasing the PS amount, TCE removal was also improved while no obvious effect was observed by varying the catalyst loading. Degradation was decreased as the TCE initial concentration

was increased from 20 to 100 mg/L. Moreover, when initial solution pH was increased, efficiency deteriorated to 80 %. Bicarbonate showed more negative effect on TCE removal among the solution matrix. To better understand the effects of radical species in the system, the scavenger tests were performed. The •SO4− and •O2− were predominant species responsible for TCE removal. The nZVI-rGO-activated PS process shows potential applications in remediation of highly toxic organic contaminants such as TCE present in the groundwater. Keywords Reduced graphene oxide . Iron . Trichloroethylene degradation . Persulfate . Water treatment

Introduction Responsible editor: Santiago V. Luis Electronic supplementary material The online version of this article (doi:10.1007/s11356-015-5034-1) contains supplementary material, which is available to authorized users. * Yisheng Xu [email protected] * Xuhong Guo [email protected] 1

State-Key Laboratory of Chemical Engineering, East China University of Science and Technology, Shanghai 200237, China

2

State Environmental Protection Key Laboratory of Environmental Risk Assessment and Control on Chemical Process, East China University of Science and Technology, Shanghai 200237, China

3

Key Laboratory of Xinjiang Uygur Autonomous Region and Engineering Research Center of Xinjiang Bingtuan of Materials-Oriented Chemical Engineering, Shihezi University, Xinjiang 832000, China

Chlorinated volatile compounds are the most pervasive contaminants in groundwater. Trichloroethylene (TCE) is one of these chlorinated solvents which is frequently used as a metal degreaser, chemical intermediate extractant, and as a component of some consumer products (Kube et al. 2005). TCE, due to its carcinogenic effect to humans, poses a potential health hazard for liver, immune system, male reproductive system, kidney, central nervous system, and developing embryo/fetus (Chiu et al. 2013). Advanced oxidation processes are widely used for the treatment of these chlorinated compounds because these are considered to be fast and cost-effective compared with conventional treatment processes due to pumpand-treat and bioremediation (Andreozzi et al. 1999; Ko et al. 2012; Wang et al. 2012). Hydroxyl radicals (•OH) were widely employed to treat the water contaminants, but persulfate (PS) recently has gained considerable attention among these oxidation reduction processes (Gao et al. 2012; Liang et al. 2008a). PS has many advantages due to its strong

Environ Sci Pollut Res

oxidation potential (−2.01 V), non-selective reactivity and stability at room temperature (Oh et al. 2009). PS can be activated by heat, UV light, or transition metal (e.g., Fe2+) to generate the strong sulfate radicals (•SO4−) (Gao et al. 2012; Liang et al. 2004; Xu et al. 2014b). Recently, Danna et al. presented a new way to produce •SO4− for the treatment of acid orange 7 (Zhou et al. 2014). They used iron and sulfite to produce •SO4− radicals in the presence of light. •SO4− radical produced by co-mediated peroxymonosulfate (PMS) was used for the oxidation of 4-chloro-2-nitrophenol (Zhou et al. 2015). Bromate was obtained from the oxidation of free bromine through the •SO4− produced by Co-PMS system (Li et al. 2015b). Phosphorus-based buffer was used to activate the PMS for the generation of •SO4− to treat various contaminants (Lou et al. 2014). PS activation by heating requires a lot of energy and is not economically feasible while using UV light as PS activator; water should be pumped out from the underground water reservoir which is also a costly and complex process (Tan et al. 2012). Activation of PS with transition metals has shown some advantages over the UV or thermal processes. For example, transition metals can activate the PS at room temperature (Adewuyi and Sakyi 2013). The generation of radicals from oxidants using transition metals has been studied (Anipsitakis and Dionysiou 2004a; Thompson 1981). Co2+ was used to activate PMS for the oxidation of 2,4-dichlorophenol (Anipsitakis et al. 2006). Silver ions are considered as the most suitable metal ions for the activation of PS (Anipsitakis and Dionysiou 2004b). Liu et al. used Fe2+ and Cu2+ to activate PS for the oxidative reduction of propachlor (Liu et al. 2012). Among the transition metals, Fe2+ has received much attention because it is non-toxic, highly effective, and cheap. Many researchers have worked on the activation of PS by iron metal ions for the degradation of toxic contaminants (Liu et al. 2012; Usman et al. 2012; Zhen et al. 2012). As recognized, Fe2+ activate the PS in following way (Eq. 1) 2þ 3þ þ  SO−4 þ SO2− S2 O2− 8 þ Fe → Fe 4

ð1Þ

To prevent the rapid conversion of Fe2+ ions into Fe3+ ions and its scavenging effect, chelating agent like citric acid was used (Liang and Guo 2010; Zhang et al. 2014). However, it is still challenging to find a suitable activator for PS. However, zero-valent iron (ZVI) particles tend to aggregate, therefore hinder the particle reactivity due to its high surface energy and magnetic interaction (Phenrat et al. 2006). To resolve the issue, ZVI could be loaded on some suitable supports. Graphene as a two-dimensional structural support has drawn much attention of the researchers due to its excellent and unique properties (Wang et al. 2014a). Nanoparticles-graphene exhibited enhanced activity as compared to bare nanoparticles (Gupta et al. 2014). Zero-valent iron-supported graphene has also shown excellent

performance for removal of a variety of contaminants such as Cr(VI), Pb+2, U(VI), and methylene blue (Guo et al. 2012; Jabeen et al. 2011; Jabeen et al. 2013; Sun et al. 2014). To the best of our knowledge, the PS activation using nano-sized zero-valent iron-reduced graphene oxide (nZVIrGO) nanocomposite for the degradation of TCE was barely studied. In this work, we reported the synthesis of ZVI nanoparticles on the reduced graphene oxide and its application in the removal of TCE contaminants using PS as an oxidant. The graphene oxide and nZVI-rGO composite were characterized by transmission electron microscopy (TEM), Fourier transform infrared (FTIR), energy-dispersive spectroscopy (EDS), and Raman spectroscopy. The system of nZVI-rGO and PS was used for degradation of TCE at 20 °C, which is a highly toxic contaminant in water. The removal of TCE was investigated by using nZVI and nZVI-rGO in the absence and presence of PS. Furthermore, PS concentration, catalyst loading, initial TCE concentration, and initial solution pH and addition of various anions and humic acid were systematically studied to understand their effects on the degradation of TCE. In the end, scavenger tests were conducted to investigate the generation of the different radicals during TCE degradation in order to reveal the mechanism of the degradation process.

Experimental section Materials Natural graphite (Qingdao Nanshu Ruiying Graphite Co. Ltd., China); sodium persulfate (Na2S2O8, PS, 98.0 %); trichloroethylene (TCE, 99.0 %); 1,4-benzoquinone (C6H4O2, BQ, 97 %); and humic acid (HA, fulvic acid >90 %) were purchased from Aladdin Reagent Ltd. Co. (Shanghai, China). Ferric chloride (FeCl3•6H2O, 99.5 %); n-hexane (C6H14, 97.0 %); potassium iodide (KI, 99.5 %); sodium bicarbonate (NaHCO 3 , 99.5 %); potassium biphthalate (C 8 H 5 KO 4 , 99.0 %); 1,10-phenanthroline monohydrate (C12H8N2•H2O 98 %); hydroxylamine hydrochloride (NH 2 OH•HCl, 99.0 %); tert-butyl alcohol ((CH 3 ) 3 OH, tBA, 99.0 %); isopropanol (C3H8O, IPA, 99.5 %); sodium chloride (NaCl, 99.5 %); sodium nitrate; H2SO4; KMnO4; H2O2; and HCl were obtained from Shanghai Jingchun Reagent Ltd. Co. (Shanghai, China). The water used in the experiment was purified with a Millipore Milli-Q system. Synthesis of graphene oxide GO was synthesized from natural graphite flakes using the modified Hummers method (Cheng et al. 2012; Hummers and Offeman 1958). In the given process, graphite (2.0 g) and NaNO3 (1.0 g) were added to 50 mL of concentrated

Environ Sci Pollut Res

H2SO4 (98 %), and the mixture was mechanically stirred in an ice bath for 2 h. KMnO4 (7.3 g) was slowly added to the mixture, then 7 mL of H2O2 and 150 mL of de-ionized water were added, respectively. Thus, the color of the mixture changed from brown into bright yellow. The mixture was filtered and washed several times with 3 % HCl solution and de-ionized water. GO was vacuum-dried at 40 °C for 24 h and obtained as brown solid. The obtained product was dispersed in 500 mL water through ultra-sonication. In the end, solid form of GO was obtained by centrifugation and vacuum drying at 40 °C for 24 h. Synthesis of nZVI-rGO For the synthesis of nZVI-rGO composite, we followed the Jabeen’s method with little modification; we did not use nitrogen environment (Jabeen et al. 2011). In the typical reaction, 0.5 mg/mL GO dispersion was prepared through ultrasonication for half an hour. Then, we added 0.9 g of FeCl3•6H2O dissolved in 6 mL of water in GO dispersion and continued mixing for a couple of hours to ensure the exchange of Fe3+ on the GO. After mixing, 1.7 g of NaBH4 dissolved in 25 mL of water was added to the mixture dropwise at 80 °C for 2 h and 30 min. The color of mixture turned into black. The black solid was separated from the mixture through filtration by washing with water and ethanol, respectively. The obtained product was dried at the 50 °C under the vacuum for 24 h before the characterization or further use in an application. Characterization The high transmission electron microscopy (TEM) was performed using a JEOL-2100 electron microscope operating at 200 kV. To study the interaction between the nZVI and rGO, the Fourier transformation infrared spectra was performed using a 6700 Fourier transform spectrometer (Thermo Nicolet Corp.). X-ray diffraction (XRD) was carried out on a Bruker D8 Advance x-ray diffractometer with a scan rate of 6 °/min. Raman spectra of the samples were recorded by Raman spectroscopy. Catalytic studies All experiments were performed in a series of 24-mL screwcap borosilicate glass vials with TFE/silicone liners at constant temperature of 20±0.5 °C using a rotary shaker. A stock solution of TCE was prepared by mixing the pure non-aqueous phase liquid TCE with Milli-Q water under gentle stirring in the dark and then diluted to make the desired concentration. The initial concentration of TCE in all tests was 0.15 mM (20 mg/L), predetermined amount of nZVI-rGO was added and followed by adding the calculated amount of sodium

persulfate. 0.1 M NaOH or H2SO4 solution was used to adjust the pH of the solution. Samples were withdrawn at given time intervals and quenched immediately with n-hexane before analysis. In order to identify the radical involved in TCE degradation, the radical scavengers tests using TBA (•OH scavenger), IPA (both •OH and •SO4− scavenger), and BQ (•O2− scavenger) had been performed (Liang and Su 2009), Teel and Watts 2002). GC vials containing TCE were analyzed using a gas chromatograph (Agilent 7890A, Palo Alto, CA, USA) equipped with an electron capture detector (ECD), an autosampler (Agilent 7693), and a DB-VRX column (60-m length, 250 μm i.d., and 1.4-μm thickness). The temperatures of the injector and detector were 240 and 260 °C, respectively, and the oven temperature was kept at a constant temperature of 75 °C. One-microliter sample was injected into the GC at a split ratio of 20:1. The concentration of PS was measured via UV spectrophotometric method as described elsewhere (Liang et al. 2008b). The concentrations of ferrous ion (Fe2+) and total iron ions (Fe2+) and (Fe3+) were determined by using the 1,10-phenanthroline method (Tamura et al. 1974).

Results and discussion Characterizations of nZVI-rGO nanocomposite The morphology of GO and nZVI-rGO was observed using TEM as shown in Fig. 1a–c. GO exhibits thin layer structure while nZVI is decorated on the reduced graphene oxide sheet. It is observed that nZVI are widely dispersed and mostly spherical particles. In contrast to nZVI, rGO manifests a low contrast material due to its low electron density. The size of nZVI is around 6 nm, which is presumably to have higher activity than bare ZVI. EDS was also used to confirm the presence of ZVI, carbon, and oxygen in the system. Figure 1d clearly indicates that large amounts of C, O, and Fe were incorporated. Figure 2 presents the XRD patterns of bare nZVI and nZVI synthesized on the graphene sheets. The peaks at about 11 and 42.5 ° shown in Fig. S1 confirm the characterization peak of graphene oxide. Bare nZVI exhibits its peak at 2θ=45 and 65 ° while nZVI-rGO shows a peak only at 45 ° which is the typical band assigned to ZVI (Jabeen et al. 2013). FeOOH at 18 ° is also confirmed in the bare nZVI by XRD. In nZVI-rGO, 35 ° peak is attributed to M (Fe2O4/Fe2O3) (Liu et al. 2014). The peak of GO is not present in the composite which suggests that it has been reduced, while there is weak peak of rGO at around 25 °. According to the Raman spectra (Fig. 3a), the D band (sp3 carbon atoms of defects and disorders, at ~1370 cm−1) and G band (sp2 carbon atoms in graphitic sheets, at ~1586 cm−1) is

Environ Sci Pollut Res Fig. 1 TEM of GO (a), HRTEM of nZVI-rGO nanocomposite (b, c), and EDS of nZVI-rGO (d)

seen for GO, while nZVI/rGO composites exhibit peaks shift (D band at ~1343 cm−1 and G band at ~1595 cm−1). It is noted that the value of ID/IG ratio increases from 1.05 to 1.25 for nZVI/rGO nanocomposite, which is attributed to the changes in the GO structures. No noticeable peak is observed for nZVI. FTIR was used to characterize the functional groups present in the reduced graphene oxide with nanoparticles. Figure 3b shows the FTIR spectra of FeO and nZVI-

rGO. The spectrum of GO showed a broad absorption band at about 3400–3500 cm−1, which is related to the O–H stretching vibration. C–O shows a stretching peak at 1360 cm−1, which is the characteristic band of C–O groups in carbonyl and carboxyl moieties (Guo et al. 2012). In addition, epoxy C–O stretching peak at 1205 cm−1 is observed, which could be attributed to its characteristic peak. TCE degradation by nZVI-rGO-activated PS system Degradation of TCE was carried out by using nZVI and nZVIrGO in the presence and absence of PS as shown in Fig. 4. The control experiment shows negligible amount of degradation, which could be due to volatilization of TCE. The same result was obtained by using PS only because PS has to be activated by heat or transition metal as mentioned above. Zero-valent iron (ZVI) has been used as an alternative of Fe+2 according to Eqs. 2 and 3 (Furukawa et al. 2002: Liang and Lai 2008).

Fig. 2 XRD pattern of bare nZVI and nZVI-rGO

1 Fe þ O2 þ H2 O → Fe2þ þ 2OH− 2

ð2Þ

Fe þ 2H2 O → Fe2þ þ 2OH− þ H2

ð3Þ

Nineteen and 26 % degradation of TCE were achieved, respectively, when nZVI and nZVI-rGO were used without the addition of PS. nZVI-rGO apparently presents more efficiency in removal of TCE as compared to bare nZVI, which is

Environ Sci Pollut Res

Fig. 3 a Raman spectra of GO, nZVI-rGO, and nZVI. b FTIR spectra of nZVI and nZVI-rGO

probably due to the good stabilization of nZVI by graphene. nZVI and nZVI-rGO activated PS systems were also used to study the time-dependent removal of TCE. Seventy-two and 98 % of TCE were degraded within 2 min in the presence of PS. nZVI-rGO activated PS removes 26.5 % more TCE than nZVI/PS. In general, ZVI provides activation of PS in two ways. Fe2+ as well as ZVI decomposes PS while producing radicals (Eqs. 4 and 5) (Hussain et al. 2014). 2þ − þ SO2− Fe þ S2 O2− 8 → Fe 4 þ  SO4

ð4Þ

2þ 3þ − þ SO2− S2 O2− 8 þ Fe → Fe 4 þ  SO4

ð5Þ

It is likely because bare nZVI can aggregate or oxidize in the aqueous solution, hence reduces the efficiency of the catalyst. It was observed that the color immediately changed

Fig. 4 Degradation of TCE by nZVI and nZVI-rGO with and without activated PS (CTCE =0.15 mM, CPS =15 mM, Ccatalyst =1 g/L)

from black to yellowish after adding the PS to bare nZVI which might be due to the oxidation. A layer of FeOOH might form to decrease its efficiency as the reaction proceeded. On the other hand, nZVI-rGO shows better activity due to the smaller size of nZVI attached on the graphene sheet. The degradation of pollutants using ZVI-activated PS system has been widely demonstrated. For example, the dibutyl phthalate was degraded mainly by sulfate radicals produced during ZVI-PS system (Li et al. 2014). Xu et al. reported a high removal rate of ortho-nitrochlorobenzene while using ZVIPS (Xu et al. 2014a). Higher degradation of 2,4-dichlorophenol was achieved as compared to Fe2+ ions while •SO4− was generated in iron-activated PS environment (Li et al. 2015a). Further experiments were performed to investigate the activation effects of nZVI-rGO on PS for the TCE degradation. As known, Fe2+ and Fe3+ ions always play a vital role in the PS system for the removal of contaminant. In order to understand the role of nZVI-rGO in activation of PS, the oxidation of PS and concentrations of Fe2+ and Fe3+ ions were measured as a function of time as shown in Fig. 5. The concentration of Fe2+ ions was found to be much higher than Fe3+ in the beginning, but as the reaction proceeds, the Fe3+ concentration increases while Fe2+ concentration decreases rapidly. This conversion of Fe2+ to Fe3+ ions (Eq. 1) is supposed to enhance the reaction rate. Regeneration of Fe2+ ions from Fe3+ is another advantage of using ZVI. The iron sludge production can be reduced by using Fe°, which in turn reduces the cost (Wang et al. 2014b). During the conversion, PS degradation reaches around 90 % within 2 min, which might be due to leaching of these ferrous ions. After 2 min, concentrations of both ions are stabilized. We propose that the nZVI-rGO system showing this enhanced degradation of TCE is probably due to the rapid release of ferrous ions for PS activation to generate sulfate

Environ Sci Pollut Res

Effect of TCE initial concentration

Fig. 5 Decomposition of PS, Fe2+, and Fe3+ ion concentration during the TCE degradation

radical, which is one of the highest redox potential radicals. •O2− radicals might play an important role in PS decomposition which has been discussed briefly in scavenger test section.

Effect of different parameters on nZVI-rGO/PS system for TCE degradation Effect of persulfate concentration on TCE degradation The effect of PS concentration on TCE degradation was investigated using the same amount of nZVI-rGO (Fig. 6a). Degradation of TCE is about 27 % within 30 min by using 25:1 M ratio of PS/TCE. As the ratio of PS/TCE in the system is increased, TCE degradation increases accordingly. Nearly 51 and 85 % removal of TCE are achieved within 30 min by using 50:1 and 75:1 M ratio of PS. respectively. This is likely due to more radicals produced in the reaction. But when 100:1 M ratio of PS is used, nearly 99 % removal of TCE is achieved within 2 min. As we know that PS is activated in the presence of Fe2+ ions, nZVI-rGO provides these ions to activate the PS to produce •SO4− radicals. In addition, it is also being considered that through the chain reactions, other radicals like •O 2− and •OH could be involved in the TCE degradation.

Effect of nZVI-rGO loading Figure 6b shows the effects of catalyst loading on TCE removal. More than 98 % removal of TCE is achieved as the catalyst loading is 1 g/L or above, which is most probably contributed due to the excessive release of Fe2+. When the catalyst amount reduced to 0.4 and 0.8 g/L, 90 and about 92 % degradation of TCE were observed, respectively.

To investigate the effects of different initial TCE concentration on the degradation, we conducted experiments while keeping the other parameters constant. The degradation efficiency is not affected up to 20 mg/L of TCE concentration as shown in Fig. 6c. This may be due to the large amount of radicals present in the system. However, when the TCE initial concentration is increased to 50 or 100 mg/L, the degradation efficiency, reduces to 87 and 82 %, but still on a decent level as compared to other ZVI and PS systems used for TCE removal. The reduction of the efficiency might arise from the significant amount of SO42− generated in the system which forms a sulfate layer on ZVI (Al-Shamsi and Thomson 2013). Effect of initial solution pH H+ ions present in the system affect the removal of target contaminant. In the acidic environment, ZVI corrodes rapidly in contrast to basic environment (Lai et al. 2013). In addition to ZVI, PS can be oxidized in the acidic environment according to the following equations (Eqs. 6 and 7). þ − S2 O2− 8 þ H →HS2 O8

ð6Þ

− HS2 O−8 →Hþ þ SO2− 4 þ  SO4

ð7Þ

The ground water is around neutral pH, so the experiments at the initial pH of 5, 7, 9, and the control without adjusting the pH (pH 5.9) were conducted to study the degradation efficiency as shown in Fig. 6d. Apparently, the maximum degradation efficiency is achieved at acidic pH. As the pH increases, degradation slows down and maximum degradation is lowered. An oxide film on the active ZVI surface at high pH possibly limits its efficiency (Gomathi Devi et al. 2009). Effect of solution matrix on TCE degradation In ground water, a variety of anions or organic species can be presented, which affects the activity of PS and the degradation efficiency of TCE. Therefore, the degradation of TCE was examined in the presence of Cl−, HCO3−, and humic acid (HA) at different concentrations as depicted in Fig. 7. The effect of Cl− ions on TCE was studied in the range of 1 to 100 mM (Fig. 7). The chloride ion barely affects the TCE degradation at very low concentration (1 mM), while the severe inhibiting effect is observed at 10 and 100 mM. In case of 100 mM chloride concentration, the TCE removal efficiency decreases almost 30 %. When the chloride concentration increases, sulfate radicals are consumed by chloride anions

Environ Sci Pollut Res

Fig. 6 Effect of different parameters, PS concentrations (a), nZVI-rGO loading (b), TCE initial concentrations (c), and initial solution pH (d), on nZVIrGO-activated PS system for the degradation of TCE

causing a reduction in the degradation efficiency as described in Eqs. 8–10 (Liang et al. 2006). Weak radicals of •Cl2− might

be formed when chloride ion reacts with sulfate radicals, which could affect the TCE degradation (Fang et al. 2012).  SO−4 þ Cl− ↔SO2− 4 þ  Cl

ð8Þ

Cl þ Cl− ↔  Cl−2

ð9Þ



Fig. 7 Effect of anions and HA on degradation of TCE. a Chloride ions. b Bicarbonate ions. c Humic acid

Cl−2

þ 

Cl−2 →

−

2 Cl þ Cl2

ð10Þ

The bicarbonate ions have a scavenging effect on contaminant removal in contrast to chloride ions when the concentration of bicarbonate solution is used in the range of 1 to 100 mM. Little effect on removal efficiency is observed for bicarbonate at 1 mM. TCE degradation decreases dramatically from 98 to 82 and 55 %, respectively, when the bicarbonate concentration is increased to 10 and 100 mM as shown in Fig. 7. This reduction in efficiency can be attributed due to two reasons. First, bicarbonate ions consume sulfate radicals and secondly bicarbonate radicals are generated which have low oxidation potential to oxidatively degrade TCE (Zuo et al. 1999).

Environ Sci Pollut Res

Moreover, addition of bicarbonate ion causes an increase of pH accounting for the lower TCE degradation efficiency similar to the pH effect observed in Fig. 6d (Xu et al. 2014b). The natural organic materials (NOMs) comprising the various functional groups such as hydroxyls, amines, carboxyl acids, and phenolic could show interactions towards the surface of nZVIrGO, thus potentially affect its reactivity. Humic acid (HA) comprises of almost 50 % of NOMs (Kim et al. 2014; Roshani and Karpel Vel Leitner 2011). The effects of HA at 1, 5, and 10 mg/L concentrations on the degradation of TCE are shown in Fig. 7. The less inhibition effect of HA is observed in contrast to the anions. Addition of HA shows a ~2 % reduction in the removal efficiency regardless of concentration. The TCE degradation efficiency by ZVI-rGO is much better than the case of adding ferrous ions directly (Kim et al. 2014).

8 and 12 % by using 10 and 100 mM tBA. Therefore, it is concluded that sulfate radicals are the major radicals present in TCE degradation. 1,4-benzoquinone (BQ) was further used to detect the •O2− radicals participating in the degradation (Fig. 8). Thirty-five and 40 % of TCE degradation is scavenged by using 10 and 100 mM BQ showing a comparable effect as sulfate radicals. This is probably because •O2− radicals are produced in huge amounts at the start of reaction; they not only participate in the degradation but also help to generate sulfate radicals (Fang et al. 2013). In addition, the reactivity of •O2− could be enhanced due to the smaller size of nZVI attached on rGO, which is consistent with the previous studies (Furman et al. 2009: Vikesland et al. 2007).

Radical scavenger tests for nZVI-rGO and TCE system

Role of nZVI-rGO in nZVI-rGO/PS system

In order to better understand the reactive oxygen species or radicals present in our system, scavenger tests were conducted using tBA and BQ as a radical scavenger for •OH and •O2− radicals, and IPA was used to identify both the •SO4− and •OH radicals. It is well postulated that •SO4− radicals are produced when PS undergoes activation, while •OH can be formed from sulfate radicals under neutral or basic pH conditions (Liang et al. 2004; Liang et al. 2007). We have used different concentrations of these radical scavengers ranging from 1 to 100 mM to identify the radicals as shown in Fig. 8. It can be seen that the degradation of TCE decreases to 18 and 48 % in the presence of 10 and 100 mM IPA, respectively. Such tests confirm the presence of sulfate or hydroxyl radicals in the system. To figure out the dominant radical in the reaction, identification of •OH radical was further investigated by adding tBA, TCE degradation efficiency is affected only by

It is well known that PS can be activated by Fe+2 ions (Eq. 5) and nZVI is considered as a source of Fe2+ ions (Eqs. 3–4). nZVI itself can participate in the PS degradation as described above. It is seen from Fig. 6a that our system released the Fe2+ ions very quickly due to the small size of nZVI on reduced graphene oxide. It is found that the concentrations of Fe2+ and Fe3+ are 125 and 10 mg L−1, respectively, at the start of the reaction, and after 2 min, their concentration approached to 60 and 174 mgL−1 respectively. The concentration of Fe2+ is recovered slightly before reaching equilibrium. This data strongly indicates the conversion of nZVI into the dissolved iron with the effect of O2 (and H2O) that occurred initially (Eqs. 2 and 3), thus the PS is activated by the release of these ions accordingly. Another reason for the release of Fe2+ can be the acidic media. The XRD diagram of nZVI-rGO before and after the TCE degradation with PS is shown in Fig. 9. As compared to fresh

Fig. 8 Effect of scavenger tests for the degradation of TCE in nZVI-rGO activated PS system, a •SO4− and •OH radicals, b •OH radicals, and c •O2− radicals, respectively

Fig. 9 XRD data of nZVI-rGO (red) before use and (dark red) after use

Environ Sci Pollut Res

nZVI-rGO, iron peaked in the used sample decreases, suggesting that nZVI either reacts with PS itself or ions releases from the nZVI-rGO activated the PS. XRD data shows that nZVI is not fully converted to ions. The results of XRD are consistent with the experiment related to the measurements of ions concentration in the system. This result shows that nZVI attached on the rGO is the source of Fe2+, which mainly participated in the activation of PS in the nZVI-rGO/PS system. The radical chain reactions might be initiated by the sulfate radicals produced during the activation of PS. Therefore, nZVI decorated on the rGO sheet provides Fe2+ ions efficiently to activate the PS for the degradation of TCE.

Conclusion The nZVI-rGO and PS system had been shown to be an excellent oxidant system for the removal of TCE from the aqueous solution. In nZVI-rGO and PS system, the degradation of TCE was more rapid as compared to bare ZVI. This effective release of Fe2+ ions might be due to the nanocomposites of nZVI-rGO. The results showed that degradation was enhanced when the PS concentration was increased. Higher removal efficiency of TCE was achieved in the acidic medium although degradation in basic system was comparatively better than other system. One and 10 mM of anions had minor effects on TCE removal while 100 mM of bicarbonate had more scavenging effect as compared to 100 mM chloride solution. TCE degradation was negligibly affected due to humic acid. Scavenger tests had shown that •SO4− and •O2− radicals were mostly responsible for the degradation of TCE. •O2− radicals might have participated in the degradation of PS as was seen that PS was also degraded very quickly. The experimental results presented above had indicated that degradation was enhanced due to application of rGO. These results might provide us with a new strategy to degrade groundwater contaminants using nZVI-rGO composites and PS process.

Acknowledgments We acknowledge the financial support from the National Natural Science Foundation of China (nos. 51273063, 21476143, and 21306049), the Fundamental Research Funds for the Central Universities, the higher school specialized research fund for the doctoral program (222201313005 and 222201314029), 111 Project Grant (B08021), the Open Project of State Key Laboratory of Chemical Engineering (SKL-ChE-14C01).

References Adewuyi YG, Sakyi NY (2013) Removal of nitric oxide by aqueous sodium persulfate simultaneously activated by temperature and Fe2+ in a lab-scale bubble reactor. Ind Eng Chem Res 52:14687– 14697

Al-Shamsi MA, Thomson NR (2013) Treatment of organic compounds by activated persulfate using nanoscale zerovalent iron. Ind Eng Chem Res 52:13564–13571 Andreozzi R, Caprio V, Insola A, Marotta R (1999) Advanced oxidation processes (AOP) for water purification and recovery. Catal Today 53:51–59 Anipsitakis GP, Dionysiou DD (2004a) Radical generation by the interaction of transition metals with common oxidants. Environ Sci Technol 38:3705–3712 Anipsitakis GP, Dionysiou DD (2004b) Transition metal/UV-based advanced oxidation technologies for water decontamination. Appl Catal B Environ 54:155–163 Anipsitakis GP, Dionysiou DD, Gonzalez MA (2006) Cobalt-mediated activation of peroxymonosulfate and sulfate radical attack on phenolic compounds. Implications of chloride ions. Environ Sci Technol 40:1000–1007 Cheng P, Yang Z, Wang H, Cheng W, Chen M, Shangguan W, Ding G (2012) TiO2–graphene nanocomposites for photocatalytic hydrogen production from splitting water. Int J Hydrog Energy 37:2224–2230 Chiu WA, Jinot J, Scott CS, Makris SL, Cooper GS, Dzubow RC, Bale AS, Evans MV, Guyton KZ, Keshava N (2013) Human health effects of trichloroethylene: key findings and scientific issues. Environ Health Perspect 121:303–311 Fang GD, Dionysiou DD, Wang Y, Al-Abed SR, Zhou D-M (2012) Sulfate radical-based degradation of polychlorinated biphenyls: effects of chloride ion and reaction kinetics. J Hazard Mater 227:394– 401 Fang GD, Dionysiou DD, Al-Abed SR, Zhou D-M (2013) Superoxide radical driving the activation of persulfate by magnetite nanoparticles: implications for the degradation of PCBs. Appl Catal B Environ 129:325–332 Furman O, Laine DF, Blumenfeld A, Teel AL, Shimizu K, Cheng IF, Watts RJ (2009) Enhanced reactivity of superoxide in water–solid matrices. Environ Sci Technol 43:1528–1533 Furukawa Y, Kim JW, Watkins J, Wilkin RT (2002) Formation of ferrihydrite and associated iron corrosion products in permeable reactive barriers of zero-valent iron. Environ Sci Technol 36:5469–5475 Gao YQ, Gao NY, Deng Y, Yang YQ, Ma Y (2012) Ultraviolet (UV) light-activated persulfate oxidation of sulfamethazine in water. Chem Eng J 195–296:248–253 Gomathi Devi L, Girish Kumar S, Mohan Reddy K, Munikrishnappa C (2009) Photo degradation of methyl orange an azo dye by advanced fenton process using zero valent metallic iron: influence of various reaction parameters and its degradation mechanism. J Hazard Mater 164:459–467 Guo J, Wang R, Tjiu WW, Pan J, Liu T (2012) Synthesis of Fe nanoparticles@graphene composites for environmental applications. J Hazard Mater 225–226:63–73 Gupta VK, Atar N, Yola ML, Üstündağ Z, Uzun L (2014) A novel magnetic Fe@Au core–shell nanoparticles anchored graphene oxide recyclable nanocatalyst for the reduction of nitrophenol compounds. Water Res 48:210–217 Hummers WS, Offeman RE (1958) Preparation of graphitic oxide. J Am Chem Soc 80:1339 Hussain I, Zhang Y, Huang S (2014) Degradation of aniline with zerovalent iron as an activator of persulfate in aqueous solution. RSC Adv 4:3502–3511 Jabeen H, Chandra V, Jung S, Lee JW, Kim KS, Kim SB (2011) Enhanced Cr(vi) removal using iron nanoparticle decorated graphene. Nanoscale 3:3583–3585 Jabeen H, Kemp KC, Chandra V (2013) Synthesis of nano zerovalent iron nanoparticles—graphene composite for the treatment of lead contaminated water. J Environ Manag 130:429–435 Kim H-S, Ahn JY, Kim C, Lee S, Hwang I (2014) Effect of anions and humic acid on the performance of nanoscale zero-valent iron particles coated with polyacrylic acid. Chemosphere 113:93–100

Environ Sci Pollut Res Ko S, Crimi M, Marvin BK, Holmes V, Huling SG (2012) Comparative study on oxidative treatments of NAPL containing chlorinated ethanes and ethenes using hydrogen peroxide and persulfate in soils. J Environ Manag 108:42–48 Kube M, Beck A, Zinder SH, Kuhl H, Reinhardt R, Adrian L (2005) Genome sequence of the chlorinated compound-respiring bacterium Dehalococcoides species strain CBDB1. Nat Biotechnol 23:1269– 1273 Lai B, Chen Z, Zhou Y, Yang P, Wang J, Chen Z (2013) Removal of high concentration p-nitrophenol in aqueous solution by zero valent iron with ultrasonic irradiation (US–ZVI). J Hazard Mater 250–251:220– 228 Li H, Wan J, Ma Y, Huang M, Wang Y, Chen Y (2014) New insights into the role of zero-valent iron surface oxidation layers in persulfate oxidation of dibutyl phthalate solutions. Chem Eng J 250:137–147 Li R, Jin X, Megharaj M, Naidu R, Chen Z (2015a) Heterogeneous Fenton oxidation of 2,4-dichlorophenol using iron-based nanoparticles and persulfate system. Chem Eng J 264:587–594 Li Z, Chen Z, Xiang Y, Ling L, Fang J, Shang C, Dionysiou DD (2015b) Bromate formation in bromide-containing water through the cobaltmediated activation of peroxymonosulfate. Water Res. doi:10.1016/ j.watres.2015.06.019 Liang C, Guo YY (2010) Mass transfer and chemical oxidation of naphthalene particles with zerovalent iron activated persulfate. Environ Sci Technol 44:8203–8208 Liang C, Lai MC (2008) Trichloroethylene degradation by zero valent iron activated persulfate oxidation. Environ Eng Sci 25:1071–1077 Liang C, Su HW (2009) Identification of sulfate and hydroxyl radicals in thermally activated persulfate. Ind Eng Chem Res 48:5558–5562 Liang C, Bruell CJ, Marley MC, Sperry KL (2004) Persulfate oxidation for in situ remediation of TCE. II. Activated by chelated ferrous ion. Chemosphere 55:1225–1233 Liang C, Wang ZS, Mohanty N (2006) Influences of carbonate and chloride ions on persulfate oxidation of trichloroethylene at 20C. Sci Total Environ 370:271–277 Liang C, Wang ZS, Bruell CJ (2007) Influence of pH on persulfate oxidation of TCE at ambient temperatures. Chemosphere 66:106–113 Liang C, Huang CF, Chen YJ (2008a) Potential for activated persulfate degradation of BTEX contamination. Water Res 42:4091–4100 Liang C, Huang CF, Mohanty N, Kurakalva RM (2008b) A rapid spectrophotometric determination of persulfate anion in ISCO. Chemosphere 73:1540–1543 Liu CS, Shih K, Sun CX, Wang F (2012) Oxidative degradation of propachlor by ferrous and copper ion activated persulfate. Sci Total Environ 416:507–512 Liu F, Yang J, Zuo J, Ma D, Gan L, Xie B, Wang P, Yang B (2014) Graphene-supported nanoscale zero-valent iron: removal of phosphorus from aqueous solution and mechanistic study. J Environ Sci 26:1751–1762 Lou X, Wu L, Guo Y, Chen C, Wang Z, Xiao D, Fang C, Liu J, Zhao J, Lu S (2014) Peroxymonosulfate activation by phosphate anion for organics degradation in water. Chemosphere 117:582–585 Oh SY, Kim HW, Park JM, Park HS, Yoon C (2009) Oxidation of polyvinyl alcohol by persulfate activated with heat, Fe2+, and zero-valent iron. J Hazard Mater 168:346–351 Phenrat T, Saleh N, Sirk K, Tilton RD, Lowry GV (2006) Aggregation and sedimentation of aqueous nanoscale zerovalent iron dispersions. Environ Sci Technol 41:284–290

Roshani B, Karpel Vel Leitner N (2011) Effect of persulfate on the oxidation of benzotriazole and humic acid by e-beam irradiation. J Hazard Mater 190:403–408 Sun Y, Ding C, Cheng W, Wang X (2014) Simultaneous adsorption and reduction of U(VI) on reduced graphene oxide-supported nanoscale zerovalent iron. J Hazard Mater 280:399–408 Tamura H, Goto K, Yo tsuyanagi T, Nagayama M (1974) Spectrophotometric determination of iron(II) with 1,10phenanthroline in the presence of large amounts of iron(III). Talanta 21:314–318 Tan C, Gao N, Chu W, Li C, Templeton MR (2012) Degradation of diuron by persulfate activated with ferrous ion. Sep Purif Technol 95:44–48 Teel AL, Watts RJ (2002) Degradation of carbon tetrachloride by modified Fenton’s reagent. J Hazard Mater 94:179–189 Thompson RC (1981) Catalytic decomposition of peroxymonosulfate in aqueous perchloric acid by the dual catalysts silver(1+) and peroxydisulfate(2-) and by cobalt(2+). Inorg Chem 20:1005–1010 Usman M, Faure P, Ruby C, Hanna K (2012) Application of magnetiteactivated persulfate oxidation for the degradation of PAHs in contaminated soils. Chemosphere 87:234–240 Vikesland PJ, Heathcock AM, Rebodos RL, Makus KE (2007) Particle size and aggregation effects on magnetite reactivity toward carbon tetrachloride. Environ Sci Technol 41:5277–5283 Wang D, Bolton JR, Hofmann R (2012) Medium pressure UV combined with chlorine advanced oxidation for trichloroethylene destruction in a model water. Water Res 46:4677–4686 Wang H, Li S, Si Y, Sun Z, Li S, Lin Y (2014a) Recyclable enzyme mimic of cubic Fe3O4 nanoparticles loaded on graphene oxide-dispersed carbon nanotubes with enhanced peroxidase-like catalysis and electrocatalysis. J Mater Chem B 2:4442–4448 Wang X, Wang L, Li J, Qiu J, Cai C, Zhang H (2014b) Degradation of Acid Orange 7 by persulfate activated with zero valent iron in the presence of ultrasonic irradiation. Sep Purif Technol 122:41–46 Xu HB, Zhao DY, Li YJ, Liu PY, Dong CX (2014a) Enhanced degradation of ortho-nitrochlorobenzene by the combined system of zerovalent iron reduction and persulfate oxidation in soils. Environ Sci Pollut Res 21:5132–5140 Xu M, Du H, Gu X, Lu S, Qiu Z, Sui Q (2014b) Generation and intensity of active oxygen species in thermally activated persulfate systems for the degradation of trichloroethylene. RSC Adv 4:40511–40517 Zhang YQ, Xie XF, Huang SB, Liang HY (2014) Effect of chelating agent on oxidation rate of aniline in ferrous ion activated persulfate system at neutral pH. J Cent South Univ 21:1441–1447 Zhen G, Lu X, Zhao Y, Chai X, Niu D (2012) Enhanced dewaterability of sewage sludge in the presence of Fe(II)-activated persulfate oxidation. Bioresour Technol 116:259–265 Zhou D, Chen L, Zhang C, Yu Y, Zhang L, Wu F (2014) A novel photochemical system o ferrous sulfite complex: kinetics and mechanisms of rapid decolorization of Acid Orange 7 in aqueous solutions. Water Res 57:87–95 Zhou J, Xiao J, Xiao D, Guo Y, Fang C, Lou X, Wang Z, Liu J (2015) Transformations of chloro and nitro groups during the peroxymonosulfate-based oxidation of 4-chloro-2-nitrophenol. Chemosphere 446–451 Zuo Z, Cai Z, Katsumura Y, Chitose N, Muroya Y (1999) Reinvestigation of the acid–base equilibrium of the (bi)carbonate radical and pH dependence of its reactivity with inorganic reactants. Radiat Phys Chem 55:15–23