Accepted Manuscript

0
Activating persulfate by Fe coupling with weak magnetic field: Performance and
mechanism

Xinmei Xiong , Bo Sun , Jing Zhang , Naiyun Gao , Jimin Shen , Jialing Li , Xiaohong
Guan
PII: S0043-1354(14)00414-X
DOI: 10.1016/[Link].2014.05.042
Reference: WR 10696

To appear in: Water Research

Received Date: 11 March 2014

Revised Date: 23 May 2014
Accepted Date: 25 May 2014

Please cite this article as: Xiong, X., Sun, B., Zhang, J., Gao, N., Shen, J., Li, J., Guan, X., Activating
0
persulfate by Fe coupling with weak magnetic field: Performance and mechanism, Water Research
(2014), doi: 10.1016/[Link].2014.05.042.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

1 Activating persulfate by Fe0 coupling with weak magnetic field: Performance and

2 mechanism

3 Xinmei Xiong1,2, Bo Sun3, Jing Zhang3, Naiyun Gao1, Jimin Shen3, Jialing Li1,

4 Xiaohong Guan1*

PT
1
5 State Key Laboratory of Pollution Control and Resources Reuse, College of

RI
6 Environmental Science and Engineering, Tongji University, Shanghai 200092, P. R.

7 China

SC
2
8 Department of Civil Engineering of Jiujiang College, Jiujiang 332005, Jiangxi, China

U
3
9 State Key Lab of Urban Water Resource and Environment, Harbin Institute of
AN
10 Technology, Harbin 150090, P. R. China

11
M

12
D

13 *Corresponding author contact information:

TE

14 Xiaohong Guan, Email: guanxh@[Link], Phone: +86-21-65980956

15
EP

16
C

17
AC

18

19

20

21

22

1
 ACCEPTED MANUSCRIPT

23 Abstract

24 Weak magnetic field (WMF) and Fe0 were proposed to activate PS synergistically

25 (WMF-Fe0/PS) to degrade dyes and aromatic contaminants. The removal rates of

26 orange G (OG) by WMF-Fe0/PS generally decreased with increasing initial pH

PT
27 (3.0-10.0) and increased with increasing Fe0 (0.5-3.0 mM) or PS dosages (0.5-3.0

mM). Compared to its counterpart without WMF, the WMF-Fe0/PS process could

RI
28

29 induce a 5.4-28.2 fold enhancement in the removal rate of OG under different

SC
30 conditions. Moreover, the application of WMF significantly enhanced the

U
31 decolorization rate and the mineralization of OG. The degradation rates of caffeine,
AN
32 4-nitrophenol, benzotriazole and diuron by Fe0/PS were improved by 2.1-11.1 fold

33 due to the superimposed WMF. Compared to many other sulfate radical-based

M

34 advanced oxidation technologies under similar reaction conditions, WMF-Fe0/PS

D

35 technology could degrade selected organic contaminants with much greater rates.
TE

36 Sulfate radical was identified to be the primary radical species responsible for the OG

37 degradation at pH 7.0 in WMF-Fe0/PS process. This study unraveled that the presence
EP

38 of WMF accelerated the corrosion rate of Fe0 and thus promoted the release of Fe2+,
C

39 which induced the increased production of sulfate radicals from PS and promoted the
AC

40 degradation of organic contaminants. Employing WMF to enhance oxidation capacity

41 of Fe0/PS is a novel, efficient, promising and environmental-friendly method since it

42 does not need extra energy and costly reagents.

43 Keywords: Weak magnetic field, Orange G, Emerging contaminants, Sulfate radicals,

44 Phenolic compounds

2
 ACCEPTED MANUSCRIPT

45 1. Introduction

46 As one of the advanced oxidation processes (AOPs), sulfate radical based AOPs has

47 become a hotspot and been widely studied for the decomposition of various kinds of

48 recalcitrant or hazardous organic compounds due to its high oxidizing ability. Sulfate

PT
49 radicals can be generated by activation of persulfate (PS) or peroxymonosulfate (PMS)

RI
50 with UV (Fang and Shang, 2012), heat (Johnson et al., 2008), sono (Chen and Su,

51 2012), base (Furman et al., 2010), granular activated carbon (Yang et al., 2011),

SC
52 quinones (Fang et al., 2013), phenols (Ahmad et al., 2013), transition metals

U
53 (Anipsitakis and Dionysiou, 2004), or magnetic spinel (Zhang et al., 2013; Ding et al.,
AN
54 2012).

55 Among these methods, activation of PS with transition metal ions is commonly

M

56 used and an efficient method to generate sulfate radicals. Among the transition metal
D

57 ions (Co2+, Cu2+, Ni2+, Fe2+, Ag+, Ru3+) effective for activating PS or PMS, Fe2+ is
TE

58 advantageous since it is cheap, non-toxic, naturally abundant and environmental

59 friendly. Fe2+ can activate PS to generate sulfate radicals via Eq. (1) (House, 1962).
EP

60 However, excess Fe2+ in Fe2+/PS process will consume the produced sulfate radicals
C

61 via Eq. (2) thereby lowering the concentration of sulfate radicals, the effective
AC

62 oxidants (Liang et al., 2004). In addition, iron hydroxide may be produced at neutral

63 pH, which leads to low oxidation efficiency.

64 S2 O 82− + Fe 2+ → SO •4− + Fe3+ + SO 24− (1)

65 SO •4− + Fe 2+ → Fe 3+ + SO 24− (2)

66 As an alternative, zero-valent iron (Fe0) can be used as a source of Fe2+ through

3
 ACCEPTED MANUSCRIPT

67 the corrosion of Fe0 under both aerobic and anaerobic conditions, especially under

68 acidic environments (Noubactep, 2009). The use of Fe0 as a source of Fe2+ to activate

69 PS would not only overcome the disadvantages of sulfate radical consumption by

70 excessive Fe2+ but also avoid the addition of other anions (Cl- or SO42-) to the treated

PT
71 system. Fe0 activation not only serves as a releasing source of dissolved Fe2+, but also

RI
72 provides another way that does involve direct electron transfer from ZVI to PS to

73 produce sulfate radicals (Eq. (3)) (Oh et al., 2010). Meanwhile, Fe2+ could be

SC
74 regenerated at the Fe0 surface (Bremner et al., 2006), as shown in Eq. (4), and thus the

U
75 precipitation of ferric hydroxides during the reaction could be lessened.
AN
76 2S 2 O 82 − + Fe 0 → 2SO •4− + Fe 2 + + 2SO 24 − (3)

77 2Fe3+ + Fe0 → 3Fe 2+ (4)

M

78 Kusic et al. (2011) reported that the Fe0/PS process yielded a higher rate of azo
D

79 dye mineralization than the Fe2+/PS process under the optimal conditions. As a result,
TE

80 employing Fe0/PS process in water and wastewater treatment to decompose organic

81 contaminants has attracted extensive attention. The Fe0/PS process has been employed
EP

82 to degrade various organic pollutants (Oh et al., 2010; Liang and Guo, 2010; Zhao et
C

83 al., 2010; Le et al., 2011). These studies showed that Fe0 is an alternative source of
AC

84 Fe2+ to activate PS and the enhanced release of Fe2+ can accelerate the removal rate of

85 organics in Fe0/PS process (Hussain et al., 2012).

86 In our previous study, we found that the presence of weak magnetic field (WMF),

87 significantly accelerated Se(IV) removal by Fe0 since WMF promoted the corrosion

88 of Fe0 (Liang et al., 2014). The accelerated Fe0 corrosion and Fe2+ release in the

4
 ACCEPTED MANUSCRIPT

89 presence of WMF may be advantageous to the Fe0/PS process. However, this

90 speculation has never been verified. Therefore, a novel method of activating PS by

91 Fe0 coupling with WMF (WMF-Fe0/PS) was proposed in this study to degrade

92 organic contaminants. Ten selected organic contaminants, including orange G (OG),

PT
93 acid organic 7 (AO7), reactive blue 19 (RB19), 4-nitrophenol (4-NP), caffeine,

RI
94 benzotriazole, diuron, 2,4-dichlorophenol (2,4-DCP), 4-chlorophenol (4-CP), and

95 phenol, were employed as the target compounds. The objectives of this study were to

SC
96 1) explore the feasibility and the mechanisms of applying WMF to increase the

U
97 degradation of OG by Fe0/PS process; 2) investigate the influence of WMF on OG
AN
98 removal and mineralization by Fe0/PS under different operation conditions; 3)

99 determine the wide suitability of this method by comparing the half-lives (t1/2) of
M

100 removing ten target compounds by WMF-Fe0/PS process with those reported in the
D

101 literature under similar reaction conditions.

TE

102 2. Materials and methods

103 2.1 Chemicals

EP

104 All chemicals were analytical grade and were used as received without further
C

105 purification. All solutions were prepared in deionized water and stored at 4 oC in the
AC

106 refrigerator. OG was purchased from Aladdin chemistry Co., Ltd. (Shanghai, China)

107 and its physiochemical characteristics are shown in Table S1. Fe0 powder ( 98%

108 pure, and BET surface area 0.87 m2 g-1) were obtained from Shanghai Jinshan smelter

109 (Shanghai, China). The Fe0 particles were agglomerated with median particle size of

110 ∼24.9 µm, as shown in Figure S1. Benzotriazole, caffeine, 4-NP, and diuron were

5
 ACCEPTED MANUSCRIPT

111 purchased from Sigma-Aldrich. The reagents (acetonitrile and methanol) for

112 chromatographic analyses were of HPLC grade and purchased from J&K Chemical

113 Co.. All the other chemicals including sodium persulfate (Na2S2O8, 98% purity),

114 2,4-DCP, 4-CP, phenol, AO7, RB19, Na2S2O3, KI, NaOH, H2SO4, methanol (MeOH)

PT
115 and tert-butyl alcohol (TBA) were purchased from the Sinopharm Group Chemical

RI
116 Reagent Co., Ltd. (Shanghai, China).

117 2.2 Experimental procedures

SC
118 Batch experiments were conducted to explore the influences of initial pH, PS dosage,

U
119 and Fe0 dosage on the performance of OG degradation in presence and absence of
AN
120 WMF. The experiments were performed open to the air and in a series of 500-mL

121 borosilicate glass jars that were placed in a water bath to keep temperature constant at
M

o
122 25 C. The WMF was generated by positioning two thin cylindrical
D

123 neodymium-iron-boron permanent magnets under the water bath. The experimental
TE

124 setup is schematically illustrated in Fig. S2. The magnetic field intensity was

125 determined with a Teslameter (HT201, Shanghai Hengtong Magnetic & Electric
EP

126 Technology Co., Ltd) to be 10-40 mT at the bottom of the reactor. Control
C

127 experiments without applying WMF were also performed to offer reference. The
AC

128 solution was mixed at 400 rpm using a mechanical stirrer (D2004W, Shanghai Sile

129 Instrument Co., Ltd). With this stirring intensity, Fe0 could be evenly distributed in the

130 solution and no aggregation of Fe0 was observed at the bottom of the reactor.

131 Preliminary study showed that OG degradation by PS was negligible and thus

132 batch experiments were initiated by adding Fe0 (0.5-3.0 mM) to a 0.5 L solution

6
 ACCEPTED MANUSCRIPT

133 containing Na2S2O8 (0.5-3.0 mM) and OG (0.2 mM). Before dosing Fe0, pH of the

134 solution containing PS and OG was adjusted to the pre-determined value with NaOH

135 or H2SO4 (0.1 M). To explore the consumption of PS and the generation of Fe2+/Fe3+

136 with time, Fe0 (0.5-3.0 mM) was supplied to a 0.5 L solution containing 2.0 mM

PT
137 Na2S2O8 but no OG at pH 7.0. Quenching experiments were conducted with the

RI
138 addition of MeOH and TBA for identifying the primary radical species formed in the

139 absence or presence of WMF. During these experiments, pH was not controlled. The

SC
140 experiments with other organic contaminants followed the same procedure but with

U
141 different reagent concentrations.
AN
142 At the given time intervals, the samples aliquots were collected with plastic

143 syringes and mixed immediately with appropriate amounts of methanol to quench any
M

144 further oxidation reactions (Anipsitakis and Dionysiou, 2004). The samples were
D

145 filtered through a 0.22 µm membrane filter before measurement. For TOC
TE

146 measurement, sodium thiosulfate, an inorganic salt, was chosen as the quencher to

147 minimize any interference of quenching agent in TOC analysis. All experiments were
EP

148 run in duplicates or triplicates, and all points in the figures are the mean of the results
C

149 and error bars represent standard deviation of the means.

AC

150 2.3 Analytical methods

151 The concentration of dyes was analyzed with a Purkinje TU-1902 automatic scanning

152 UV-Vis spectrophotometers with a spectrometric quartz cell (1 cm). The maximum

153 absorbance wavelengths of OG, RB19 and AO7 were observed at 478, 595 and 484

154 nm, respectively. Concentrations of other contaminants were directly determined by

7
 ACCEPTED MANUSCRIPT

155 UPLC (Waters) with a Symmetry C18 column (2.1*100 mm, 1.7 µm) and UV-visible

156 detector. The details for UPLC-UV analyses are listed in Table S2. The concentrations

157 of Fe2+ and Fe3+ ion (after reduction to Fe2+ with hydroxylamine hydrochloride) were

158 determined by spectrophotometric method at 510 nm after complexing with

PT
159 1,10-phenanthroline using an UV-Vis spectrophotometer. The concentration of PS was

RI
160 monitored using a rapid spectrophotometric determination method, as described by

161 Liang et al. (2008). It should be specified that no quenching reagent was added to the

SC
162 samples to determine the concentrations of Fe3+ and PS and the analyses were

U
163 immediately performed once sampled. Total organic carbon (TOC) was monitored
AN
164 using a Shimadzu TOC analyzer (model TOC-V CPN, Japan) to identify the

165 mineralization of OG.

M

166 3. Results and discussion

D

167 3.1 Influence of WMF on OG removal by Fe0/PS at various Fe0 dosages

TE

168 Without WMF, negligible OG was degraded by PS alone (2.0 mM) and less than 2.5%

169 of the OG was removed by Fe0 alone (1.0 mM) over 180 min, as illustrated in Fig.
EP

170 S3(a). The application of WMF had no influence on OG removal by PS alone or Fe0
C

171 alone, as shown in Fig. S3(b). However, addition of 1.0 mM of Fe0 to PS solution
AC

172 resulted in very rapid and pronounced degradation of OG, with ∼60.1% removal in

173 180 min without WMF, indicating that Fe0 can activate PS to initiate OG oxidation.

174 Surprisingly, 95.2% of OG removal was achieved in 20 min under the same condition

175 but with WMF. Obviously, the application of WMF could remarkably enhance the

176 removal rate of OG by Fe0/PS.

8
 ACCEPTED MANUSCRIPT

177 The influence of WMF on OG removal by Fe0/PS was further investigated at Fe0

178 dosage varying from 0.5 mM to 3.0 mM while keeping the initial concentration of PS

179 fixed, as demonstrated in Fig. 1. The rate of OG oxidation was greater with increasing

180 Fe0 dosage, in both Fe0/PS and WMF-Fe0/PS systems. Moreover, the removal of OG

PT
181 by Fe0/PS was significantly accelerated in the presence of WMF. It was observed that

the more rapid removal of OG by Fe0/PS was accompanied with a faster decrease in

RI
182

183 pH. The final pH was in the range of 3.0-3.4 for all the cases except when Fe0 was

SC
184 dosed at 0.5 mM in the absence of WMF.

U
185 As shown in Fig. 1, the removal efficiency of OG increased almost linearly with
AN
186 elapsed time, especially at the beginning of the reaction. The zero-order rate law,

187 shown in Eq. (5), could be employed to determine the initial removal rate of OG by
M

188 Fe0/PS or WMF-Fe0/PS.

D

189 C = C0 − kOG
'
t (5)
TE

'
190 where kOG is the zero-order rate constant of OG removal (mM min-1).

191 Since the initial concentration of OG was kept constant (0.2 mM) in this study,
EP

192 Eq. (5) was modified to Eq. (6):

C

'
C kOG
193 =1− t = 1 − kOGt (6)
AC

C0 C0

194 where kOG is the observed reaction rate constant (min-1).

195 Furthermore, the ratio of the observed reaction rate coefficients was used in this

196 study to roughly characterize the extent of promotion induced by WMF and could be

197 expressed by Eq. (7).

9
 ACCEPTED MANUSCRIPT

kOG2
198 f = (7)
kOG1

199 where f is the promotion factor and kOG1 and kOG2 are the observed reaction rate

200 constants of OG removal by Fe0/PS and WMF-Fe0/PS, respectively.

It was found that zero-order rate law could simulate the initial kinetics of OG

PT
201

202 removal reasonably well with regression coefficient (R2) above 0.91, as shown in Fig.

RI
203 S4 and summarized in Table S3. Increasing the dosage of Fe0 from 0.5 to 3.0 mM

SC
204 resulted in an increase of kOG1 from 0.0018 to 0.0078 min-1 without WMF and kOG2

205 from 0.0329 to 0.1324 min-1 with WMF. The value of promotion factor, f, was as high

206
U
as 15.6-18.0, indicating that WMF accelerated the decolorization rate constants of OG
AN
207 by over ten times at various Fe0 dosages. In addition, the removal rate constants of
M

208 OG with and without WMF were linearly correlated to the Fe0 dosage in the range of

209 0.5-3.0 mM, as illustrated in Fig. S5, implying that there was ample amount of PS to
D

be activated by Fe2+ generated from Fe0. Since the initial concentration of PS was
TE

210

211 fixed, the linear improvement in the rate constants of OG removal with increasing Fe0
EP

212 dosage should be associated with the increased amount of available activating agent

213 (Fe2+). Oh et al. (2010) reported that PS may be activated through electron transfer at
C

the Fe0 surface or by Fe2+ in the aqueous phase. Both mechanisms were consistent
AC

214

215 with the data in Fig. S5 because the reactive Fe0 surface area and dissolved Fe2+

216 concentration would both increase with increasing Fe0 dose, resulting in higher rates

217 of PS activation and OG oxidation. However, the mechanism of activating PS by Fe0

218 surface area could not explain the acceleration of OG removal by applying WMF

219 since WMF did not change the Fe0 surface area under vigorous mixing conditions.
10
 ACCEPTED MANUSCRIPT

220 Thus, the activation of PS by Fe0 was attributable to the release of Fe2+ from Fe0. The

221 application of WMF enhanced release rate of Fe2+ and thus the removal rate of OG by

222 Fe0/PS, which would be further verified in the following part.

223 3.2 Role of WMF in the WMF-Fe0/PS process

PT
224 3.2.1 Influence of WMF on Fe2+ release in Fe0 corrosion process

In the Fe0/PS system, Fe0 may react with O2, H2O, and H+ to generate Fe2+, which will

RI
225

226 then activate PS to produce sulfate radicals and Fe3+, following Eq. (1). The generated

SC
227 Fe3+ may be reduced by Fe0 and generate Fe2+, following Eq. (3). The produced Fe3+

U
228 will also hydrolyze following Eq. (8). Both the hydrolysis of Fe3+ and the conversion
AN
229 of SO4•− to HO• (Dogliotti and Hayon, 1967), shown in Eq. (9), will generate H+ and

230 lead to a drop in pH.

M

231 Fe 3+ + xH 2 O → Fe(OH ) 3x− x + xH + (8)

D

232 SO •4− + H 2 O → HO • + H + + SO 24 − (9)

TE

233 Our experiments confirmed that WMF had no influence on the kinetics of OG

234 removal by Fe2+ activated PS (data not shown), thus, the WMF-induced acceleration
EP

235 of OG removal by Fe0/PS should be ascribed to the promoted Fe0 corrosion and Fe2+
C

236 generation. To verify this hypothesis, the influence of WMF on Fe2+ release from Fe0
AC

237 at pH 3.0 (without PS and OG) was explored and presented in Fig. S6. Obviously, the

238 superimposed WMF accelerated Fe2+ generation at various initial Fe0 dosages and the

239 trends of Fe2+ release generally followed zero-order rate law, especially at the initial

240 stage. In the field of electrochemistry, it had been also reported that an applied

241 magnetic field could enhance the active dissolution of iron, which was mainly

11
 ACCEPTED MANUSCRIPT

242 associated with the ferromagnetic nature of Fe0 and the paramagnetic nature of Fe2+

243 (Ragsdale et al., 1998; Sueptitz et al., 2009; Rakoczy, 2010). In an applied WMF, the

244 Fe0 particles are magnetized and an induced inhomogeneous magnetic field is

245 generated around the Fe0 particles. Magnetic field tends to cause an additional

PT
246 convective transfer of paramagnetic Fe2+ ion due to the Lorentz force, FL, and Fe0

RI
247 corrosion is promoted consequently. Moreover, the field gradient force, FB, tends to

248 move Fe2+ ions along the field gradient. The uneven distribution of Fe2+ on Fe0

SC
249 surface will result in localized corrosion and thus corrosion is accelerated in the

U
250 presence of WMF (Liang et al., 2014).
AN
251 3.2.2 Influence of WMF on Fe3+ generation and PS consumption in the Fe0/PS process

252 Since the presence of OG interfered in the analysis of Fe2+ and Fe3+, the influence
M

253 of WMF on the Fe2+/Fe3+ generation and PS consumption in the reaction of Fe0 with
D

254 PS (without OG) at different initial Fe0 dosages was investigated and illustrated in Fig.
TE

255 S7. No Fe2+ was detected (detection limit: 0.03 mg L−1) in all experiments in both

256 Fe0/PS and WMF-Fe0/PS systems, indicating that activation of PS by the generated
EP

257 Fe2+ was very rapid and release of Fe2+ from Fe0 was the limiting step in the reaction
C

258 of Fe0 with PS. Fig. S7 showed that a larger Fe3+ generation rate was accompanied
AC

259 with a greater pH drop with time and a higher PS consumption rate.

260 Although the Fe3+ generation and PS consumption showed complex kinetic

261 behavior, the kinetics of Fe3+ generation and PS consumption at the beginning of

262 reaction followed zero-order rate law, as shown in Eqs. 10 and 11, respectively.

263 C = k Fe t
3+ (10)

12
 ACCEPTED MANUSCRIPT

C
264 1− = k PS t (11)
C0

265 The initial rate constants of Fe3+ generation and PS consumption were determined,

266 as demonstrated in Fig. 2. The Fe3+ generation rate constants increased linearly with

increasing Fe0 dosages in the absence or presence of WMF, as shown in Fig. S8, and

PT
267

268 the Fe3+ generation rates in the presence of WMF were dozens of times of those in the

RI
269 absence of WMF. This was because that the application of WMF induced a

SC
270 considerable increase in Fe2+ release, which was immediately oxidized by PS or

271 sulfate radicals and thus more Fe3+ was generated. It could be inferred from the above

272
U
discussion that Fe2+ release from Fe0 was the limiting step of OG oxidation by Fe0/PS
AN
273 or WMF-Fe0/PS process. Moreover, it was found that the PS consumption rates were
M

274 well correlated with the Fe3+ generation rates in the absence or presence of WMF, as

275 illustrated in Fig. 2, suggesting that PS was mainly activated by Fe2+ released from
D

Fe0. Therefore, the observed zero order degradation behavior of OG should be mainly
TE

276

277 ascribed to the fact that the Fe2+ release from Fe0 followed zero order reaction kinetics,
EP

278 as illustrated in Fig. S6. On a closer inspection, the OG removal rate constants

279 obtained at various Fe0 dosages were also well correlated with the Fe3+ generation
C

rates, as demonstrated in Fig. S9, confirming that OG removal by Fe0/PS was limited
AC

280

281 by the Fe2+ release step.

282 3.2.3 Identification of radical species in the Fe0/PS and WMF-Fe0/PS processes

283 TBA is effective quenching agent for hydroxyl radicals (k = 3.8-7.6×108 M-1 s-1),

284 but it cannot efficiently quench sulfate radicals since it reacts with sulfate radicals

285 much slower (k = 4.0-9.1×105 M-1 s-1) than with hydroxyl radicals (Anipsitakis and
13
 ACCEPTED MANUSCRIPT

286 Dionysiou, 2004). The rate constants for reactions of MeOH with hydroxyl radicals

287 and sulfate radicals are 8.0-10×108 M-1 s-1 and 0.9-1.3×107 M-1 s-1, respectively, and

288 thus MeOH can effectively quench both sulfate and hydroxyl radicals (Buxton et al.,

289 1988; Eibenberger et al., 1978). The dosing of TBA resulted in a slight drop in OG

PT
290 degradation efficiency with and without WMF at the end of reaction although there

RI
291 was a slight enhancement at the initial stage of reaction, as shown in Fig. S10.

292 However, the OG removal efficiencies decreased remarkably from 60.1% to 11.8%

SC
293 and from 95.2% to 34.1%, respectively, in Fe0/PS and WMF-Fe0/PS systems due to

U
294 the addition of MeOH. Therefore, sulfate radicals were the dominant active species
AN
295 responsible for the oxidation of OG in the Fe0/PS system, regardless of the presence

296 or absence of WMF.

M

297 3.2.4 Discussion on the role of WMF in the WMF-Fe0/PS process

D

298 These results confirmed that Fe0 was the source of Fe2+, which activated PS in the
TE

299 Fe0/PS system to produce sulfate radicals. Once PS was activated, sulfate radicals

300 formed would initiate the degradation of OG. Meanwhile, Fe3+ was generated and
EP

301 hydrolyzed to produce H+, resulting in a drop in pH. The application of WMF did not
C

302 change the radical species primarily responsible for OG degradation but it
AC

303 significantly improved the removal of OG by accelerating the Fe2+ release from Fe0 in

304 the Fe0/PS process. The enhanced Fe2+ generation would induce the increased

305 production of sulfate radicals and Fe3+, and thus the improved OG removal and faster

306 pH drop. The sharp drop in pH would further increase the Fe0 corrosion and the

307 influence of WMF was enlarged. Increasing the dosage of Fe0 from 0.5 to 3.0 mM

14
 ACCEPTED MANUSCRIPT

308 only led to a 3.3-fold increase in the OG removal rate constant. However, the rate

309 constants of OG removal were raised by 14.6-17.0 times due to the application of a

310 WMF at the same Fe0 dosage.

311 3.3 Influence of WMF on OG removal by Fe0/PS at different initial pH levels

PT
312 Since pH is a critical factor influencing Fe0 corrosion and then the efficiency of

Fe0/PS, the influence of WMF on OG removal by Fe0/PS was investigated at initial

RI
313

314 pH ranging from 3.0 to 10.0 and the results are presented in Fig. 3. The removal rate

SC
315 of OG by Fe0/PS was significantly promoted under WMF over the pH range of

U
316 3.0-10.0 (Fig. 3a-e). Generally, the OG removals at pH 3.0-9.0 followed zero-order
AN
317 rate law in both Fe0/PS and WMF-Fe0/PS systems. The kinetics of OG removal at pH

318 10.0 exhibited self-acceleration characteristics and involved a lag period before the
M

319 initiation of a rapid removal period, which may be ascribed to the removal of the iron
D

320 oxide layer (Alowitz and Scherer, 2002; Bokare and Choi, 2009). The application of
TE

321 WMF considerably shortened the lag period at pH 10.0 by promoting Fe0 corrosion.

322 Lower pH facilitated the abrasion of passive iron oxide layer and thus no lag period
EP

323 was observed at lower initial pH, even in the absence of WMF.
C

324 The observed reaction rate constants of OG (kOG) at pH 3.0-9.0 were determined,
AC

325 as shown in Fig. 3f. The values of kOG experienced a decrease with increasing initial

326 pH from 3.0 to 7.0 in the absence or presence of WMF, which was associated with the

327 larger Fe0 corrosion rate at lower pH and the consequent greater SO4•− generation rate.

328 Interestingly, the kOG at pH 9.0 is greater than that at initial pH 7.0, which may be

329 ascribed to the generation of HO• (E0 = 2.7 V) following Eq. (12) with a rate constant

15
 ACCEPTED MANUSCRIPT

330 of 6.5×107 M-1 s-1 under alkaline conditions (Buxton et al., 1988; Hayon et al., 1972).

331 Sun et al. (2006) reported that the azo linkage (–N=N–) of OG was particularly

332 susceptible to electrophilic attack by HO• radical. As pH increased (e.g., >8.5), the

333 conversion of SO4•− to HO• became increasingly important, but the precondition was

PT
334 the generation of enough SO4•− (Dogliotti and Hayon, 1967; Norman et al., 1970). The

removal of OG at pH 10.0 was mainly limited by the slow Fe0 corrosion rate.

RI
335

336 SO •4− + OH − → HO • + SO 24 − (12)

SC
337 The removal rates of OG at pH 3.0-9.0 with WMF were 5.4-28.2 fold greater

338 than those without WMF. At pH 10.0, the application of a WMF shortened the

339
U
reaction time from 180 min to 20 min to achieve 80% of OG removal. At all
AN
340 examined pH values, the pH dropped as the reactions proceeded and the final pH was
M

341 in the range of 2.8-4.0 at the end of the decolorization reaction with or without WMF.

342 The variation of pH with time was closely correlated to the removal rate of OG.
D

343 3.4 Influence of WMF on OG removal by Fe0/PS at different PS dosages

TE

344 Figure 4 shows that the kinetics of OG removal at different PS dosages at pH 7.0 with

345 and without WMF followed zero order reaction law. The observed rate constants of
EP

346 OG removal increased progressively from 0.0010 to 0.0036 min-1 and from 0.0118 to

0.0629 min-1, respectively, in the Fe0/PS and WMF-Fe0/PS systems, by increasing the
C

347
AC

348 PS dosage from 0.5 to 2.0 mM. However, the rate constants of OG removal almost

349 leveled off with further increase in PS dosage, which may be due to the lack of Fe0.

350 There was only 1.0 mM Fe0, which was not enough to activate PS dosed at over 2.0

351 mM. Anyway, the application of WMF induced a more than 10-fold increase in the

352 OG removal rate by Fe0/PS at pH 7.0, owing to the improved Fe0 corrosion in the

353 presence of WMF.

16
 ACCEPTED MANUSCRIPT

354 3.5 Influence of WMF on decolorization and mineralization of OG by Fe0/PS

355 The UV-vis spectra of the OG decolorization evolution by Fe0/PS and WMF-Fe0/PS

356 are presented in Fig. S11. The absorption spectra of OG were scanned in the range of

357 250-600 nm. The spectrum of light absorption by the OG solution before reaction

PT
358 consists of three main peaks at 259, 328, and 478 nm, plus a shoulder peak at 421 nm,

RI
359 which was consistent with the observations of Xu and Li (2010). The peak at 478 nm

360 is attributed to the absorption of the π → π∗ transition related to the –N=N– group,

SC
361 while additional bands at 259 and 328 nm are assigned to its aromatic ring in the OG

U
362 molecule. As the reaction proceeded, the two characteristic absorption peaks at 328
AN
363 and 478 nm decreased dramatically and almost disappeared after 15 min in the

364 presence of WMF, showing that the chromophore and conjugated system were
M

365 completely destroyed but the aromatic rings were still present. The intensity of the
D

366 peaks at 328 and 478 nm decreased with a much smaller rate without WMF than that
TE

367 with WMF, confirming that the application of WMF accelerated the decolorization of

368 OG.
EP

369 The influence of WMF on the mineralization of OG by Fe0/PS was determined by

C

370 varying the dosage of PS from 2.0 to 6.0 mM with a fixed PS/Fe0 molar ratio of 2.0
AC

371 and the results are shown in Fig. S12. Without WMF, no mineralization was observed

372 when PS was dosed at 2.0 mM and 2.1% of OG was mineralized when PS was dosed

373 at 4.0-6.0 mM. The OG mineralization by Fe0/PS was remarkably improved by

374 applying a WMF. The removal efficiency of TOC with WMF was increased from 7.6%

375 to 36.4% in 3 h by increasing the PS dosage from 2.0 mM to 6.0 mM. Therefore, the

17
 ACCEPTED MANUSCRIPT

376 application of WMF significantly increased not only the removal rate but also the

377 mineralization of OG.

378 3.6 The wide suitability of WMF-Fe0/PS technology

379 As summarized in Table S3, the removal rates of OG by WMF-Fe0/PS were 5.4-28.2

PT
380 times greater than those by Fe0/PS. The WMF induced improvement in OG

degradation by Fe0/PS was so surprising that the influence of WMF on the removal of

RI
381

382 four other refractory contaminants, including caffeine, 4-NP, benzotriazole and diuron,

SC
383 by Fe0/PS was investigated, as shown in Fig. 5. Similar to the case of OG, the

U
384 superimposed WMF had remarkable influence on these four contaminants
AN
385 degradation by Fe0/PS. Zero-order rate law could well simulate the kinetics of

386 degradation of these four contaminants by Fe0/PS and WMF-Fe0/PS at the initial stage
M

387 of reaction and the observed reaction rate constants were determined and listed in
D

388 Table S3. The removal rates of these four contaminants were increased by 2.1-11.1
TE

389 times due to the application of WMF.

390 To further confirm the advantage of WMF-Fe0/PS technology for removing

EP

391 organic contaminants, the half-lives (t1/2) of three dyes including OG, AO7, and RB19,
C

392 and three phenolic compounds, including 2,4-DCP, 4-CP and phenol degradation by
AC

393 WMF-Fe0/PS and those obtained from the literature under similar reaction conditions

394 were compared, as summarized in Table 1. The WMF-Fe0/PS technology could

395 degrade all the six organic contaminants with much greater rates than many other

396 sulfate radical-based advanced oxidation technologies. Anipsitakis and Dionysiou

397 (2004) reported that Ag(I) was the most efficient transition metal for activating PS

18
 ACCEPTED MANUSCRIPT

398 and the t1/2 of 2,4-DCP degradation was 110 min under their conditions. Replacing

399 Ag(I) with same dosage of Fe0 in the presence of WMF, the t1/2 of 2,4-DCP

400 degradation was dramatically lowered to 7.1 min. Le et al. (2011) found that the t1/2 of

401 Reactive blue 19 removal by Fe0/PS was 16.7 min and it was only 11.8 min although

PT
402 the Fe0 dosage employed in this study was only 1/7 of that employed in their study

and the temperature in this study was 5 oC lower than that in their study.

RI
403

404 4. Conclusions

SC
405 A novel method of activating PS by Fe0 coupling with WMF was proposed to degrade

U
406 organic contaminants. Fe0 was the source of Fe2+, which was verified to be the
AN
407 efficient activating agent for PS to produce sulfate radicals. Once PS was activated,

408 the generated sulfate radicals would initiate the degradation of OG. Meanwhile, Fe3+
M

409 was generated and hydrolyzed to produce H+, resulting in a drop in pH. The
D

410 application of WMF did not change the radical species primarily responsible for OG
TE

411 degradation but it significantly improved the removal and mineralization of OG by

412 accelerating the Fe2+ release from Fe0 in the Fe0/PS process. The enhanced Fe2+
EP

413 generation would induce the increased production of sulfate radicals and Fe3+, and
C

414 thus the improved OG removal and greater pH drop. The sharp drop in pH would
AC

415 further increase the Fe0 corrosion and the influence of WMF was enlarged. Compared

416 to its counterpart without WMF, the WMF-Fe0/PS process could induce a 5.4-28.2

417 fold increase in the removal rate of OG at different Fe0 dosages, initial pH levels and

418 different PS dosages. Applying a WMF could also induce a 2.1-11.1 fold increase in

419 the degradation rates of caffeine, 4-nitrophenol, benzotriazole and diuron by Fe0/PS.

19
 ACCEPTED MANUSCRIPT

420 Moreover, WMF-Fe0/PS technology could degrade dyes and phenolic compounds,

421 with much greater rates than many other sulfate radical-based advanced oxidation

422 technologies under similar reaction conditions. A much shorter hydraulic retention

423 time or much smaller chemical dosages was necessary to achieve similar removal

PT
424 efficiency of organic pollutants by replacing the traditional activation methods with

the combination of Fe0 and WMF. Employing WMF to enhance organic contaminants

RI
425

426 destruction by Fe0/PS technology is a promising and environmental-friendly method

SC
427 since it does not need extra energy source and costly reagents. Although applying a

U
428 magentic field around a treatment unit in real practice is unlikely, pretreatment of Fe0
AN
429 powder, which is ferromagnetic, in a magnetic field and then taking advantage of its

430 residual magnetism may be a feasible method. Of course, further studies and
M

431 optimization are necessary.

D

432 Acknowledgements
TE

433 This work was supported by the Shanghai Rising-Star Program (12QA1403500), the

434 National Natural Science Foundation (21277095), and Major Science and Technology
EP

435 Program for Water Pollution Control and Treatment (2012ZX07403-001).

C

436 REFERENCES
AC

437 Ahmad, M., Teel, A.L., Watts, R.J., 2013. Mechanism of persulfate activation by

438 phenols. Environmental Science & Technology 47 (11), 5864-5871.

439 Alowitz, M.J., Scherer, M.M., 2002. Kinetics of nitrate, nitrite, and Cr (VI) reduction

440 by iron metal. Environmental Science & Technology 36 (3), 299-306.

441 Anipsitakis, G.P., Dionysiou, D.D., 2004. Radical generation by the interaction of

20
 ACCEPTED MANUSCRIPT

442 transition metals with common oxidants. Environmental Science & Technology 38

443 (13), 3705-3712.

444 Bokare, A.D., Choi, W., 2009. Zero-valent aluminum for oxidative degradation of

445 aqueous organic pollutants. Environmental Science & Technology 43 (18),

PT
446 7130-7135.

RI
447 Bremner, D.H., Burgess, A.E., Houllemare, D., Namkung, K.C., 2006. Phenol

448 degradation using hydroxyl radicals generated from zero-valent iron and hydrogen

SC
449 peroxide. Applied Catalysis B: Environmental 63 (1), 15-19.

U
450 Buxton, G.V., Greenstock, C.L., Helman, W.P., Ross, A.B., 1988. Critical review of
AN
451 rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl

452 radicals. Journal of Physical and Chemical Reference Data 17 (2), 513-886.
M

453 Chen, W.S., Su, Y.C., 2012. Removal of dinitrotoluenes in wastewater by

D

454 sono-activated persulfate. Ultrasonics Sonochemistry 19 (4), 921-927.

TE

455 Ding, Y., Zhu, L., Wang, N., Tang, H., 2012. Sulfate radicals induced degradation of

456 tetrabromobisphenol A with nanoscaled magnetic CuFe2O4 as a heterogeneous

EP

457 catalyst of peroxymonosulfate. Applied Catalysis B: Environmental 129, 153-162.

C

458 Dogliotti, L., Hayon, E., 1967. Flash photolysis of persulfate ions in aqueous
AC

459 solutions. Study of the sulfate and ozonide radical anions. The Journal of Physical

460 Chemistry 71 (8), 2511-2516.

461 Eibenberger, H., Steenken, S., O'Neill, P., Schulte-Frohlinde, D., 1978. Pulse

462 radiolysis and electron spin resonance studies concerning the reaction of SO4•− with

463 alcohols and ethers in aqueous solution. The Journal of Physical Chemistry 82 (6),

21
 ACCEPTED MANUSCRIPT

464 749-750.

465 Fang, G., Gao, J., Dionysiou, D.D., Liu, C., Zhou, D., 2013. Activation of persulfate

466 by quinones: Free radical reactions and implication for the degradation of PCBs.

467 Environmental Science & Technology 47 (9), 4605-4611.

PT
468 Fang, J.Y., Shang, C., 2012. Bromate formation from bromide oxidation by the

RI
469 UV/persulfate process. Environmental Science & Technology 46 (16), 8976-8983.

470 Furman, O.S., Teel, A.L., Watts, R.J., 2010. Mechanism of base activation of

SC
471 persulfate. Environmental Science & Technology 44 (16), 6423-6428.

U
472 Hayon, E., Treinin, A., Wilf, J., 1972. Electronic spectra, photochemistry, and
AN
473 autoxidation mechanism of the sulfite-bisulfite-pyrosulfite systems. SO2-, SO3-,

474 SO4-, and SO5- radicals. Journal of the American Chemical Society 94 (1), 47-57.
M

475 House, D.A., 1962. Kinetics and mechanism of oxidations by peroxydisulfate.

D

476 Chemical Reviews 62 (3), 185-203.

TE

477 Hussain, I., Zhang, Y., Huang, S., Du, X., 2012. Degradation of p-chloroaniline by

478 persulfate activated with zero-valent iron. Chemical Engineering Journal 203,
EP

479 269-276.
C

480 Johnson, R.L., Tratnyek, P.G., Johnson, R.O.B., 2008. Persulfate persistence under
AC

481 thermal activation conditions. Environmental Science & Technology 42 (24),

482 9350-9356.

483 Kusic, H., Peternel, I., Koprivanac, N., Bozic, A.L., 2011. Iron-activated persulfate

484 oxidation of an azo dye in model wastewater: Influence of iron activator type on

485 process optimization. Journal of Environmental Engineering 137(6), 454-463.

22
 ACCEPTED MANUSCRIPT

486 Le, C., Wu, J.H., Li, P., Wang, X., Zhu, N.W., Wu, P.X., Yang, B., 2011.

487 Decolorization of anthraquinone dye reactive blue 19 by the combination of

488 persulfate and zero-valent iron. Water Science & Technology 64 (3), 754-759.

489 Liang, C., Bruell, C.J., Marley, M.C., Sperry, K.L., 2004. Persulfate oxidation for in

PT
490 situ remediation of TCE. I. Activated by ferrous ion with and without a persulfate–

RI
491 thiosulfate redox couple. Chemosphere 55 (9), 1213-1223.

492 Liang, C.J., Guo, Y.Y., 2010. Mass transfer and chemical oxidation of naphthalene

SC
493 particles with zerovalent iron activated persulfate. Environmental Science &

U
494 Technology 44 (21), 8203-8208.
AN
495 Liang, C.J., Huang, C.F., Mohanty, N., Kurakalva, R.M., 2008. A rapid

496 spectrophotometric determination of persulfate anion in ISCO. Chemosphere 73 (9),

M

497 1540-1543.
D

498 Liang, L.P., Sun, W., Guan, X.H., Huang, Y.Y., Choi, W., Bao, H.L., Li, L.N., Jiang,
TE

499 Z., 2014. Weak magnetic field significantly enhances selenite removal kinetics by

500 zero valent iron. Water Research 49, 371-380.

EP

501 Lin, Y.T., Liang, C.J., Chen, J.H., 2011. Feasibility study of ultraviolet activated
C

502 persulfate oxidation of phenol. Chemosphere 82 (8), 1168-1172.

AC

503 Niu, C.G., Wang, Y., Zhang, X.G., Zeng, G.M., Huang, D.W., Ruan, M., Li, X.W.,

504 2012. Decolorization of an azo dye orange G in microbial fuel cells using Fe

505 (II)-EDTA catalyzed persulfate. Bioresource Technology 126, 101-106.

506 Norman, R., Storey, P., West, P., 1970. Electron spin resonance studies. Part XXV.

507 Reactions of the sulphate radical anion with organic compounds. Journal of the

23
 ACCEPTED MANUSCRIPT

508 Chemical Society B: Physical Organic, 1087-1095.

509 Noubactep, C., 2009. Characterizing the discoloration of methylene blue in Fe0/H2O

510 systems. Journal of Hazardous Materials 166 (1), 79-87.

511 Oh, S.Y., Kang, S.G., Chiu, P.C., 2010. Degradation of 2,4-dinitrotoluene by

PT
512 persulfate activated with zero-valent iron. Science of the Total Environment 408

RI
513 (16), 3464-3468.

514 Ragsdale, S.R., Grant, K.M., White, H.S., 1998. Electrochemically generated

SC
515 magnetic forces. Enhanced transport of a paramagnetic redox species in large,

U
516 nonuniform magnetic fields. Journal of the American Chemical Society 120 (51),
AN
517 13461-13468.

518 Rakoczy, R., 2010. Enhancement of solid dissolution process under the influence of
M

519 rotating magnetic field. Chemical Engineering and Processing: Process

D

520 Intensification 49 (1), 42-50.

TE

521 Rastogi, A., Al-Abed, S.R., Dionysiou, D.D., 2009. Effect of inorganic, synthetic and

522 naturally occurring chelating agents on Fe (II) mediated advanced oxidation of

EP

523 chlorophenols. Water Research 43 (3), 684-694.

C

524 Sueptitz, R., Koza, J., Uhlemann, M., Gebert, A., Schultz, L., 2009. Magnetic field
AC

525 effect on the anodic behaviour of a ferromagnetic electrode in acidic solutions.

526 Electrochimica Acta 54 (8), 2229-2233.

527 Sun, J.H., Wang, X.L., Sun, J.Y., Sun, R.X., Sun, S.P., Qiao, L.P., 2006. Photocatalytic

528 degradation and kinetics of orange G using nano-sized Sn (IV)/TiO2/AC

529 photocatalyst. Journal of Molecular Catalysis A: Chemical 260 (1), 241-246.

24
 ACCEPTED MANUSCRIPT

530 Xu, X.R., Li, X.Z., 2010. Degradation of azo dye orange G in aqueous solutions by

531 persulfate with ferrous ion. Separation and purification technology 72 (1), 105-111.

532 Yang, S., Yang, X., Shao, X., Niu, R., Wang, L., 2011. Activated carbon catalyzed

533 persulfate oxidation of azo dye acid orange 7 at ambient temperature. Journal of

PT
534 Hazardous Materials 186 (1), 659-666.

RI
535 Zhang, T., Zhu, H., Croué, J.P., 2013. Production of sulfate radical from

536 peroxymonosulfate induced by a magnetically separable CuFe2O4 spinel in water:

SC
537 Efficiency, stability, and mechanism. Environmental Science & Technology 47 (6),

U
538 2784-2791.
AN
539 Zhao, J., Zhang, Y., Quan, X., Chen, S., 2010. Enhanced oxidation of 4-chlorophenol

540 using sulfate radicals generated from zero-valent iron and peroxydisulfate at
M

541 ambient temperature. Separation and Purification Technology 71 (3), 302-307.

D

542
TE
C EP
AC

25
 ACCEPTED MANUSCRIPT

543 Figure Captions

544 Fig. 1 - Influence of WMF on OG removal by Fe0 activated PS at different initial Fe0

545 dosages (a, [Fe0]0 = 0.5 mM; b, [Fe0]0 = 1.0 mM; c, [Fe0]0 = 2.0 mM; d, [Fe0]0 = 3.0

546 mM) and the corresponding pH variations. Reaction conditions: [OG]0 = 0.2 mM,

PT
547 [PS]0 = 2.0 mM, pH0 = 7.0.

RI
548

549 Fig. 2 - The correlation of Fe3+ generation rates versus the consumption rates of PS at

SC
550 various initial Fe0 dosages in the (a) Fe0/PS system; (b) WMF-Fe0/PS system. The

U
551 insets show PS consumption and Fe3+ generation as a function of time. Reaction
AN
552 conditions: [PS]0 = 2.0 mM, [Fe0]0 = 0.5-3.0 mM, pH0 = 7.0.

553
M

554 Fig. 3 - Influence of WMF on OG removal by Fe0 activated PS at different initial pH

D

555 levels and the corresponding pH variations (a, pH0 3.0; b, pH0 5.0; c, pH0 7.0; d, pH0
TE

556 9.0; e, pH0 10.0) as well as the change of zeroth-order decolorization rate constants

557 with pH0 (f). Reaction conditions: [OG]0 = 0.2 mM, [PS]0 = 2.0 mM, [Fe0]0 = 2.0
EP

558 mM.
C

559
AC

560 Fig. 4 - The correlation of decolorization rate constants of OG by Fe0 activated PS

561 versus PS dosages in the (a) Fe0/PS system; (b) WMF-Fe0/PS system. The insets show

562 the decolorization of OG as a function of time. Reaction conditions: [OG]0 = 0.2 mM,

563 [Fe0]0 = 1.0 mM, [PS]0 = 0.5-3.0 mM, pH0 = 7.0.

564

26
 ACCEPTED MANUSCRIPT

565 Fig. 5 - Influence of WMF on caffeine, 4-nitrophenol, benzotriazole and diuron

566 removal by Fe0/PS process at pH 7.0. The dots are experimental data and the solid

567 lines are fits to zero-order kinetics following Eq. 8. Reaction conditions:

568 [Contaminants]0 = 0.02 mM, [PS]0 = 0.5 mM, [Fe0]0 = 0.5 mM.

PT
569

570

RI
U SC
AN
M
D
TE
C EP
AC

27
 ACCEPTED MANUSCRIPT

Table 1 - Comparison of degradation rates of organic pollutants by WMF-Fe0/PS with those by PS/PMS activated by other methods.
t1/2
Activation method Pollutant Reaction conditions Reference

PT
(min)
WMF-Fe0/PS Caffeine 29 [pH]0=7.0, [Caffeine]0=0.02 mM; [Na2S2O8]0=0.5 mM, [Fe0]0=0.5 mM,

RI
Fe0/PS 238 T=25 oC
WMF-Fe0/PS 4-NP 27 [pH]0=7.0, [4-Nitrophenol]0=0.02 mM; [Na2S2O8]0=0.5 mM, [Fe0]0=0.5 mM,

SC
Fe0/PS 161 T=25 oC
This study
WMF-Fe0/PS Benzo- 40 [pH]0=7.0, [Benzotriazole]0=0.02 mM; [Na2S2O8]0=0.5 mM, [Fe0]0=0.5 mM,
Fe0/PS triazole 122 T=25 oC

U
WMF-Fe0/PS Diuron 17 [pH]0=7.0, [Diuron]0=0.02 mM; [Na2S2O8]0=0.5 mM, [Fe0]0=0.5 mM,

AN
Fe0/PS 208 T=25 oC
WMF-Fe0/PS OG 2.9 [pH]0=3.0, [OG]0=0.2 mM, [Na2S2O8]0=2.0 mM, [Fe0]0=2.0 mM, T=25 oC This study

M
o
Fe(II)/PS 49.5 [pH]0=3.5, [OG]0=0.1 mM, [Na2S2O8]0=1.0 mM, [Fe(II)]0=1.0 mM, T=20 C Xu and Li (2010)
Fe(II)-EDTA/PS in 60 [pH]0=3.0, [OG]0=0.1 mM, [K2S2O8]0=2.0 mM, [Fe(II)]0=1.0 mM, Niu et al. (2012)

D
microbial fuel cells [EDTA]0=1.0 mM, T=25 oC
WMF-Fe0/PS AO7 1.5 [pH]0=5.1, [AO7]0=0.057 mM, [Na2S2O8]0=3.0 mM, [Fe0]0=3.0 mM, T=25 oC This study

TE
GAC/PS 105 [pH]0=5.1, [AO7]0=0.057 mM, [Na2S2O8]0=5.7 mM, [GAC]0=1.0 g/L, Yang et al. (2011)
o
T=25 C
EP
WMF-Fe /PS0
RB19 11.8 [pH]0=5.0, [RB19]0=0.2 mM, [Na2S2O8]0=2.0 mM, [Fe0]0=2.0 mM, T=25 oC This study
0 0 o
Fe /PS 16.7 [pH]0=5.0, [RB19]0=0.3 mM, [Na2S2O8]0=10 mM, [Fe ]0=14.29 mM, T=30 C Le et al. (2011)
0
C

0
WMF-Fe /PS 2,4-DCP 7.1 [pH]0=3.0, [2,4-DCP]0=0.311 mM, [Na2S2O8]0=1.244 mM, [Fe ]0=1.244 mM, This study
o
T=25 C
AC

Ag(I)/K2S2O8 110 [pH]0=3.0, [2,4-DCP]0=0.311 mM, [K2S2O8]0=1.244 mM, [Ag2SO4]0=0.622 Anipsitakis and
mM, room temperature Dionysiou (2004)
WMF-/Fe0/PS 4-CP 14.6 [pH]0=7.0, [4-CP]0=0.156 mM; [Na2S2O8]0=0.78 mM, [Fe0]0=1 mM, T=25 oC This study
 ACCEPTED MANUSCRIPT

Fe0/PS 33 [pH]0=6.0, [4-CP]0=0.156 mM; [Na2S2O8]0=0.78 mM, [Fe0]0=3.57 mM, room Zhao et al. (2010)
temperature

PT
Fe(II)/ 75 [pH]0=7.0, [4-CP]0=0.396 mM; [PMS]0=3.96 mM, [Fe(II)]0=0.99 mM, Rastogi et al.
Pyrophosphate/ [Pyrophosphate]0=0.99 mM (2009)

RI
PMS
WMF-Fe0/PS Phenol 1.1 [pH]0=4.2, [Phenol]0=0.5 mM; [Na2S2O8]0=8.4 mM, [Fe0]0=4.2 mM, T=25 oC This study

SC
UV/PS 9.4 [pH]0=4.2, [Phenol]0=0.5 mM; [Na2S2O8]0=8.4 mM, 15W mercury lamp Lin et al. (2011)
(UV-C, 254 nm, Sparkie GLQ-D287), T=25 oC

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
100 a 7 100 b 7
OG removal (%)

OG removal (%)
80 6 80 6
0
60 OG removal, Fe /PS 60

pH

pH
OG removal, WMF-Fe0/PS
5 5
0
40 pH variation, Fe /PS 40
pH variation, WMF-Fe0/PS 4 4
20 20
3 3
0 0
0 30 60 90 120 150 180 0 30 60 90 120 150 180

PT
Time (min) Time (min)
100 c 7 100 d 7

RI
OG removal (%)

OG removal (%)
80 6 80 6

60 60

pH

pH
5 5

SC
40 40
4 4
20 20
3 3
0 0

U
0 30 60 90 120 150 180 0 20 40 60 80 100 120
Time (min)
AN
Time (min)

Fig. 1
M
D
TE
C EP
AC

1
 ACCEPTED MANUSCRIPT

9 1.0
a 1.0 b
120

PS (C/C0)
PS (C/C0)
0.8
0.9
8 0.6
0.8 100 0.4
min )

min )
-1

-1
0.7 0.2
7 0 10 20 30 [Fe0]0=0.5 mM 80 0 3 6 9 12 15 [Fe0]0=0.5 mM
T (min) T (min)
-3

[Fe0]0=1.0 mM

-3
[Fe0]0=1.0 mM
kPS (x10

kPS (x10
(x10 mM)
6

(x10 mM)
80 0
[Fe ]0=2.0 mM
y=2.707x+2.978 60 y=0.9315x+12.97 600
0
[Fe ]0=2.0 mM
60 [Fe0]0=3.0 mM 2 [Fe0]0=3.0 mM

-3
R2=0.99 R =0.99

-3
400
5 40
40 200
20

3+

3+
Fe

Fe
0 0
4

PT
0 10 20 30 20 0 5 10 15 20
T (min) T (min)
0.5 1.0 1.5 2.0 20 40 60 80 100

kFe3+ (x10-3 mM min-1) kFe3+ (x10-3 mM min-1)

RI
Fig. 2

U SC
AN
M
D
TE
C EP
AC

2
 ACCEPTED MANUSCRIPT
100 a 3.1 100 b 5.0 100 c 7
3.0

OG removal (%)

OG removal (%)
OG removal (%) 80 80 80
4.5 6
2.9
60 60 60
4.0

pH

pH

pH
2.8 5
40 OG removal, Fe0/PS 40 40
2.7 3.5
OG removal, WMF-Fe0/PS 4
20 20 20
pH variation, Fe0/PS 2.6 3.0
pH variation,WMF-Fe0/PS 3
0 0 0
2.5
0 10 20 30 40 50 60 0 30 60 90 120 150 180 0 30 60 90 120 150 180

Time (min) Time (min) Time (min)

100 d 9 100 e 10 f
160
8
OG removal (%)

OG removal (%)

(x10 min )
80

-1
80

PT
8
7 120
60 60 Fe0/PS

-3
pH

pH
6 WMF-Fe0/PS
40 6 80
40 5

kOG
20 4 20 40
4

RI
0 3 0 0
0 30 60 90 120 150 180 0 30 60 90 120 150 180 3 5 7 9
Time (min) Time (min) pH0

SC
Fig. 3

U
AN
M
D
TE
C EP
AC

3
 ACCEPTED MANUSCRIPT
4 a b
60
y=1.646x+0.33 y=33.71x-3.99
3
R2=1.00 R2=1.00
kOG (x10-3 min-1)

kOG (x10-3 min-1)

1.0 [PS]0=0.5 mM
1.2
40 [PS]0=1.0 mM
2 0.8 1.0
[PS]0=2.0 mM

OG (C/C0)

OG (C/C0)
0.8 [PS]0=3.0 mM
0.6
[PS]0=0.5 mM 0.6
1 0.4 20
[PS]0=1.0 mM 0.4
0.2 [PS]0=2.0 mM
0.2
[PS]0=3.0 mM
0 0.0 0.0
0

PT
0 30 60 90 120 150 180 0 10 20 30 40 50 60
Time (min) Time (min)

0.5 1.0 1.5 2.0 2.5 3.0 3.5 0.5 1.0 1.5 2.0 2.5 3.0 3.5

[PS]0 (mM) [PS]0 (mM)

RI
Fig. 4

U SC
AN
M
D
TE
C EP
AC

4
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M

Fig. 5
D
TE
C EP
AC

5
 ACCEPTED MANUSCRIPT

Highlights

1. A novel method of activating PS by Fe0/WMF was proposed.

2. Applying WMF induced a 5-28 fold increase in the removal rate of OG by Fe0/PS.

PT
3. The presence of WMF improved the Fe0 corrosion and thus the production of

RI
SO4•−.

4. The WMF-Fe0/PS was superior to many activation methods for removing

SC
organics.

U
5. This method did not cost extra energy or chemicals and is environmentally
AN
friendly.
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

1 Supporting Information

2 Activating persulfate by Fe0 coupling with weak magnetic field: Performance and

3 mechanism

4 Xinmei Xiong1,2, Bo Sun3, Jing Zhang3, Naiyun Gao1, Jimin Shen3, Jialing Li1,

PT
5 Xiaohong Guan1*

RI
6 State Key Laboratory of Pollution Control and Resources Reuse, College of

7 Environmental Science and Engineering, Tongji University, Shanghai 200092, P. R.

SC
8 China

U
2
9 Department of Civil Engineering of Jiujiang College, Jiujiang 332005, Jiangxi, China
AN
3
10 State Key Lab of Urban Water Resource and Environment, Harbin Institute of

11 Technology, Harbin 150090, P. R. China

M

12
D

13 *Corresponding author contact information:

TE

14 Xiaohong Guan, Email: guanxh@[Link], Phone: +86-21-65980956

15
EP

16 16 Pages
C

17 12 Figures
AC

18 3 Tables

19

20

1
 ACCEPTED MANUSCRIPT

a b 3.0

2.5

Volume (%)
2.0

1.5

1.0

.5

0.0
0 20 40 60 80 100 120

PT
21 Size (µm)

22 Fig. S1 - (a) SEM image and (b) Particle size distribution of Fe0 employed in this

RI
23 study.

SC
24

25

26
U
AN
27
M

28

29
D
TE

30

31
C EP
AC

2
 ACCEPTED MANUSCRIPT

PT
RI
SC
32

33
U
Fig. S2 - Diagram of the experimental setup.
AN
34
M
D
TE
C EP
AC

3
 ACCEPTED MANUSCRIPT
100 100
a b
PS alone
80 80
Fe0 alone
OG removal (%)

OG removal (%)
60 60

40 40

20 20

PT
0 0
0 30 60 90 120 150 180 0 30 60 90 120 150 180

35 Time (min) Time (min)

RI
36 Fig. S3 - Removal of OG by PS alone or Fe0 alone in the (a) Fe0/PS system; (b)

SC
37 WMF-Fe0/PS system. Reaction conditions: [OG]0 = 0.2 mM, [PS]0 = 2.0 mM, [Fe0]0

38 = 1.0 mM, pH0 = 7.0.

U
39
AN
M
D
TE
C EP
AC

4
 ACCEPTED MANUSCRIPT

1.0 [Fe0]0 = 0.5 mM 1.0

[Fe0]0 = 1.0 mM
0.8 [Fe0]0 = 2.0 mM 0.8
[Fe0]0 = 3.0 mM
OG (C/C0)

OG (C/C0)
0.6 0.6

0.4 0.4

0.2 0.2

PT
0.0 a 0.0 b
0 30 60 90 120 150 180 0 5 10 15 20 25
Time (min) Time (min)
40

RI
41 Fig. S4 - Simulation of the kinetics of OG removal by Fe0/PS at various Fe0 dosages

SC
42 in the (a) Fe0/PS system; (b) WMF-Fe0/PS system. Reaction conditions: [OG]0 = 0.2

43 mM, [PS]0 = 2.0 mM, pH0 = 7.0.

44
U
AN
45
M

46

47
D
TE
C EP
AC

5
 ACCEPTED MANUSCRIPT

140

120
Fe0/PS
WMF-Fe0/PS
100
kOG (x10 min )
-1 y=36.87x+17.24
R2=0.96
80
-3

60

PT
40

y=2.26x+0.91
20 2

RI
R =0.99

SC
0 1 2 3
0
[Fe ]0 (mM)
48

U
49 Fig. S5 - The correlation of removal rate constants of OG by Fe0 activated PS versus
AN
50 Fe0 dosages. Reaction conditions: [OG]0 = 0.2 mM, [PS]0 = 2.0 mM, [Fe0]0 = 0.5-3.0

51 mM, pH0 = 7.0.

M

52
53
D

54
55
TE

56
57
58
EP

59
60
61
C

62
63
AC

64
65
66
67
68

6
 ACCEPTED MANUSCRIPT

0.3
Fe0 alone 0.4
WMF-Fe0
0.3

(mM)
(mM)

0.2

0.2
2+

2+
Fe

Fe
0.1
0.1

PT
0.0 a 0.0 b
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Time (min) Time (min)

RI
0.5 0.5

0.4 0.4

SC
(mM)

(mM)
0.3 0.3
2+

2+
0.2 0.2
Fe

Fe

U
0.1 0.1
AN
0.0 c 0.0 d
0 5 10 15 20 25 30 0 5 10 15 20 25 30

69 Time (min) Time (min)

M

70 Fig. S6 - Influence of WMF on releasing of Fe2+ at different initial Fe0 dosages (a,

71 [Fe0]0 = 0.5 mM; b, [Fe0]0 = 1.0 mM; c, [Fe0]0 = 2.0 mM; d, [Fe0]0 = 3.0 mM).
D

Reaction conditions: pH0 = 3.0, T = 25 oC.

TE

72

73
C EP
AC

7
 ACCEPTED MANUSCRIPT
0.2 a 1.0 b 1.0
0.4
0.8 0.8
0.3
(mM)

(mM)
PS (C/C0)

PS (C/C0)
7
0.6 Fe0/PS 0.6
Fe3+, Fe0/PS
0.1 WMF-Fe0/PS
PS, Fe0/PS

pH
3+

3+
0.2 5
Fe

Fe
3+ 0
Fe , WMF-Fe /PS 0.4 0.4
PS, WMF-Fe0/PS 3
0.1
0.2 0 50 100 150 0.2

0.0 0.0 0.0 0.0

PT
0 30 60 90 120 150 180 0 30 60 90 120 150 180
Time (min) Time (min)
1.0
c 1.0 d 1.0

0.8 1.2
0.8 0.8

RI
(mM)

(mM)
PS (C/C0)

PS (C/C0)
0.6 0.9
0.6 7 0.6
Fe0/PS
3+

3+
0.4 0.6 0
WMF-Fe /PS
Fe

pH
Fe

SC
0.4 5 0.4

0.2 0.3 3
0.2 0.2
0 50 100 150

0.0 0.0 0.0 0.0

U
0 30 60 90 120 150 180 0 20 40 60 80 100 120

74 Time (min) Time (min)

AN
75 Fig. S7 - Influence of WMF on Fe3+ generation and PS consumption in the reaction of

76 Fe0 with PS at different initial Fe0 dosages (a, [Fe0]0 = 0.5 mM; b, [Fe0]0 = 1.0 mM; c,
M

77 [Fe0]0 = 2.0 mM; d, [Fe0]0 = 3.0 mM). The insets show the influence of WMF on pH
D

78 variation with time. Reaction conditions: [OG]0 = 0.0 mM, [PS]0 = 2.0 mM, [Fe0]0 =
TE

79 0.5-3.0 mM, pH0 = 7.0.

80
EP

81
C
AC

8
 ACCEPTED MANUSCRIPT

120

100 Fe0/PS
WMF-Fe0/PS
y=43.06x-13.64
kFe3+ (x10 mM min ) R2=1.00
-1 80

60
-3

PT
40

20 y=0.7022x+0.04932
R2=0.88

RI
0

SC
0 1 2 3
0
[Fe ]0 (mM)
82

U
83 Fig. S8- Correlation of the Fe3+ generation rate constants and Fe0 dosages. Reaction
AN
84 conditions: [OG]0 = 0.0 mM, [PS]0 = 2.0 mM, [Fe0]0 = 0.5-3.0 mM, pH0 = 7.0.
M
D
TE
C EP
AC

9
 ACCEPTED MANUSCRIPT
10
a 140
b
8 120
min )

min )
y=0.8463x+29.49
-1

-1
100
y=2.767x+1.278 R2=0.94
6
R2=0.83 80
-3

-3
kOG (x10

kOG (x10
4 60

40
2
20

PT
0 0
0.0 0.5 1.0 1.5 2.0 0 20 40 60 80 100 120
-3 -1 -3 -1
kFe3+ (x10 mM min ) kFe3+ (x10 mM min )
85

RI
86 Fig. S9 - Correlation of the observed decolorization rate constants of OG by Fe0

SC
87 activated PS at various initial Fe0 dosages in the (a) Fe0/PS system; (b) WMF-Fe0/PS

88 system. Reaction conditions: [OG]0 = 0.2 mM, [PS]0 = 2.0 mM, [Fe0]0 = 0.5-3.0 mM,

89
U
pH0 = 7.0) versus the Fe3+ generation rate obtained in the control experiments ([PS]0 =
AN
90 2.0 mM, [Fe0]0 = 0.5-3.0 mM, pH0 = 7.0.
M

91
D
TE
C EP
AC

10
 ACCEPTED MANUSCRIPT
100 100
Fe0/PS, 1 M MeOH WMF-Fe0/PS, 1 M MeOH
80 Fe0/PS, 1 M TBA 80 WMF-Fe0/PS, 1 M TBA

OG removal (%)
OG removal (%)
0 0
Fe /PS WMF-Fe /PS

60 60

40 40

20 20

PT
0 a 0 b
0 30 60 90 120 150 180 0 5 10 15 20

92 Time (min) Time (min)

RI
93 Fig. S10 - Effect of different radical quenching agents on OG removal by Fe0

SC
94 activated PS in the (a) Fe0/PS system; (b) WMF-Fe0/PS system. Reaction conditions:

95 [OG]0 = 0.2 mM, [PS]0 = 2.0 mM, [Fe0]0 = 1.0 mM, pH0 = 7.0.

96
U
AN
97
M

98

99
D
TE
C EP
AC

11
 ACCEPTED MANUSCRIPT

3 3
0 min 0 min
10 min 3 min
20 min 5 min
30 min 7 min
60 min 10 min
2 120 min 2 15 min
180 min 20 min
Abs

Abs
1 1

PT
a b
0 0
300 400 500 600 300 400 500 600

RI
λ (nm) λ (nm)
100

101 Fig. S11 Variation of the UV-Vis spectra of OG with time in the (a) Fe0/PS system; (b)

SC
102 WMF-Fe0/PS system. Reaction conditions: [OG]0 = 0.2 mM, [Fe0]0 = 1.0 mM, [PS]0

U
103 = 2.0 mM, pH0 = 7.0.
AN
104

105
M
D
TE
C EP
AC

12
 ACCEPTED MANUSCRIPT

40
Fe0/PS
WMF-Fe0/PS

30
TOC removal (%)

20

PT
10

RI
0
0 0 0
=1:1 =1:2 =1:3

SC
] :[PS] 0 G] :[PS] 0 ] :[ PS] 0
[OG 0 [O 0 [OG 0

106 Added oxidants

U
107 Fig. S12 - Influence of WMF on mineralization of OG by Fe0/PS at different PS
AN
108 dosages with fixed PS/Fe0 molar ratio of 2.0. Reaction conditions: [OG]0 = 0.2 mM,

109 [PS]0 = 2.0-6.0 mM, pH0 = 7.0, Reaction time = 180 min.
M

110
D

111
TE

112

113
EP

114
C

115
AC

116

117

118

119

120

13
 ACCEPTED MANUSCRIPT

121 Table S1 - The characteristics of OG.

Chemical structure

PT
Chemical class Monoazo

RI
Formula C16H10N2Na2O7S2

λmax (nm) 478

SC
MW (g mol−1) 452.37

U
CAS No. 1936-15-8
AN
122

123
M
D
TE
C EP
AC

14
 ACCEPTED MANUSCRIPT

124 Table S2 - Details for UPLC-UV analyses.

Compounds Mobile phase UV Flow rate

Water 0.1% formic Methanol Acetonitrile (nm) (mL min-1)

(%) acid-Water (%) (%)

(%)

PT
Phenol 45 55 273 0.4
4-CP 70 30 284 0.4

RI
2,4-DCP 70 30 290 0.4

SC
Caffeine 70 30 273 0.4
4-NP 60 40 318 0.4
Benzotriazole 80 20 275 0.4
Diuron 50
U 50 254 0.4
AN
125
M

126

127
D

128
TE

129
EP

130

131
C
AC

132

133

134

135

136

15
 ACCEPTED MANUSCRIPT

137 Table S3 - The observed zeroth-order reaction rate constants for OG and other

138 contaminants removal by Fe0/PS with and without WMF under different conditions.

[Fe0]0 [PS]0 Fe0/PS WMF-Fe0/PS

Contaminants pH0 f
kOG1 kOG2
(mM) (mM) R12 R22
(min-1) (min-1)

PT
OG 7.0 0.5 2.0 0.0018 0.91 0.0329 0.99 18.0
OG 7.0 1.0 2.0 0.0036 0.98 0.0629 0.97 17.7

RI
OG 7.0 2.0 2.0 0.0052 1.00 0.0805 0.91 15.6
OG 7.0 3.0 2.0 0.0078 0.99 0.1324 0.94 17.1

SC
OG 3.0 2.0 2.0 0.0271 0.96 0.1732 0.91 6.4
OG 5.0 2.0 2.0 0.0050 0.98 0.1452 0.98 29.2
OG 7.0 2.0 2.0
U0.0052 1.00 0.0805 0.91 15.6
AN
OG 9.0 2.0 2.0 0.0077 0.99 0.1129 0.95 14.7
OG 7.0 1.0 0.5 0.0010 0.98 0.0118 0.95 11.6
M

OG 7.0 1.0 1.0 0.0022 0.92 0.0314 0.97 14.3

OG 7.0 1.0 2.0 0.0036 0.98 0.0629 0.97 17.7
D

OG 7.0 1.0 3.0 0.0039 0.93 0.0629 0.94 16.2

TE

Caffeine 7.0 0.5 0.5 0.0021 0.94 0.0175 0.94 8.3

4-NP 7.0 0.5 0.5 0.0031 0.95 0.0189 0.95 6.1
EP

Benzotriazole 7.0 0.5 0.5 0.0041 0.96 0.0126 0.98 3.1

Diuron 7.0 0.5 0.5 0.0024 0.90 0.0291 0.95 12.1
C

139 Note: All experiments were run in duplicates or triplicates, and the data were averaged.
AC

140 The standard deviations were usually within 5%.

16