Modeling the Kinetics of Contaminants Oxidation and the Generation

Subscriber access provided by GAZI UNIV

Article

Modeling the Kinetics of Contaminants Oxidation and the Generation of Manganese(III) in the Permanganate/Bisulfite Process Bo Sun, Hongyu Dong, Di He, Dandan Rao, and Xiaohong Guan Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.5b05207 • Publication Date (Web): 28 Dec 2015 Downloaded from http://pubs.acs.org on December 30, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36

Environmental Science & Technology

Modeling the Kinetics of Contaminants Oxidation and the Generation of Manganese(III) in the Permanganate/Bisulfite Process Bo Sun1,2, Hongyu Dong1, Di He3, Dandan Rao1, Xiaohong Guan1,4* 1

State Key Laboratory of Pollution Control and Resources Reuse, Tongji University, Shanghai 20092, P. R. China

2

State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin 150090, P. R. China

3

School of Civil and Environmental Engineering, University of New South Wales, Sydney, NSW 2052, Australia 4

Key Laboratory of Yangtze River Water Environment, Ministry of Education,

College of Environment Science and Engineering, Tongji University, Shanghai 200092, China

*Contact/Corresponding author contact information: Email: [email protected] (X.H. Guan); Phone: +86-21-65980956.

1

ACS Paragon Plus Environment


1

ABSTRACT:

2

Permanganate can be activated by bisulfite to generate soluble Mn(III)

3

(non-complexed with ligands other than H2O and OH−) which oxidizes organic

4

contaminants at extraordinarily high rates. However, the generation of Mn(III) in the

5

permanganate/bisulfite (PM/BS) process and the reactivity of Mn(III) toward

6

emerging contaminants have never been quantified. In this work, Mn(III) generated in

7

the PM/BS process was shown to absorb at 230-290 nm for the first time and

8

disproportionated more easily at higher pH and thus the utilization rate of Mn(III) for

9

decomposing organic contaminant was low under alkaline conditions. A Mn(III)

10

generation and utilization model was developed to get the second-order reaction rate

11

parameters of benzene oxidation by soluble Mn(III) and then benzene was chosen as

12

the reference probe to build a competition kinetics method, which was employed to

13

obtain the second-order rate constants of organic contaminants oxidation by soluble

14

Mn(III). The results revealed that the second-order rate constants of aniline and

15

bisphenol A oxidation by soluble Mn(III) were in the range of 105-106 M-1s-1. With the

16

presence of soluble Mn(III) at micromolar concentration, contaminants could be

17

oxidized with observed rate several orders of magnitude higher than those by common

18

oxidation processes, implying the great potential application of the PM/BS process in

19

water and wastewater treatment.

2


Page 2 of 36

Page 3 of 36


20

INTRODUCTION

21

In recent years, there has been growing concern on the occurrence of emerging

22

contaminants in the aquatic environment.1 Many emerging contaminants, such as

23

pharmaceuticals, personal care products and hormones, persist at least partially during

24

conventional wastewater treatment and were detected in secondary effluents and

25

receiving surface waters worldwide.2-4 Therefore, oxidation processes are widely

26

applied in water treatment for the purpose of ensuring the safety of drinking water and

27

in wastewater treatment to minimize the discharge of emerging contaminants to the

28

receiving waters.5-8 Chemical oxidants, such as chlorine,9,

29

ozone,6, 13, 14 permanganate,15-19 and hydroxyl radicals (HO•),20,21 have been widely

30

applied for the transformation/elimination of undesired emerging contaminants from

31

wastewater and drinking water in pilot scale while ferrate,22,23 and sulfate radicals

32

(SO4•-)24-27 have been mainly investigated in lab scale for eliminating organic

33

contaminants.28 The transformation efficiency mainly depends on the reactivity of the

34

oxidant to target contaminants and to the matrix components of water such as

35

dissolved organic matter (DOM).28

10

chlorine dioxide,11, 12

36

Over the past two decades, numerous studies have been carried out to evaluate

37

the reactivity of the above-mentioned oxidants toward various emerging contaminants

38

by employing rate laws and rate constants.28,

39

(non-complexed with ligands other than H2O and OH−) is known to be a strong

40

one-electron oxidizing agent,30 very few studies have employed soluble Mn(III) as

29

Although soluble Mn(III)

3



41

active oxidant to achieve rapid transformation of emerging contaminants. Due to its

42

tetragonally distorted electron configuration, Mn(III) is labile and susceptible to

43

disproportionate to Mn(II) and Mn(IV).31 As such, up to now, most of the studies on

44

transformation or sequestration of contaminants by Mn(III) were performed with

45

long-lived Mn(III) species, including Mn(III)-rich MnO2,32, 33 an abundant naturally

46

occurring Mn(III/IV) oxide, and soluble Mn(III) complexes.34, 35 Until fairly recently,

47

our group became the first to report that soluble aquo and/or hydroxo Mn(III) (i.e.

48

non-complexed with ligands other than H2O and OH−) could be generated in the

49

process of permanganate reduction by bisulfite, serving as an strong oxidant for

50

extremely rapid degradation of organic contaminants.36 However, the generation of

51

Mn(III) in permanganate/bisulfite (PM/BS) process and the reactivity of Mn(III)

52

toward emerging contaminants have never been quantified. Moreover, it is difficult to

53

quantify the generated Mn(III) experimentally due to the instability of hydrolyzed

54

Mn(III).

55

Competition kinetics method has been well developed for quantification of the

56

unknown rate constants.37 By comparing to the available rate constant of the reference

57

probe toward the oxidant applied, the rate constant of the target compound can be

58

calculated. Song et al. proposed a stopped-flow spectrophotometric competition

59

kinetics method to determine the oxidation rate constants of selected contaminants

60

based on that of a reference probe by detecting the suppression of selected

61

contaminants on the generation of product from reference probe.37 To employ this 4


Page 4 of 36

Page 5 of 36


62

method to obtain the rate constants for the reactions of selected contaminants with

63

Mn(III), a proper reference probe should be selected and a kinetic model should be

64

first derived to estimate the second-order reaction rate parameters between this

65

reference compound and Mn(III). It was found that benzene, which could not be

66

oxidized by permanganate even at high-temperature,29 could be rapidly transformed to

67

phenol in the PM/BS process. The generated phenol could be easily detected at

68

λEx/λEm = 272 nm/298 nm with a fluorimeter equipped on the stopped flow

69

spectrometer (SFS), which was negligibly interfered by the presence of other reactants

70

and products in the PM/BS process. Consequently, benzene was chosen as the

71

reference probe to build the competition kinetics method.

72

Although some evidences for the contribution of soluble Mn(III) to the rapid

73

oxidation of organic contaminants in the PM/BS process have been provided in our

74

previous study, the properties of soluble Mn(III) and the concentration of Mn(III)

75

were kept unknown. In the present study, more evidences on the generation and

76

consumption of Mn(III) were provided so as to lay the foundation for constructing a

77

model to identify the second-order rate constants of test compounds toward Mn(III). A

78

kinetic model, in which the generation and utilization of Mn(III) was assumed to be

79

key determinants of oxidation of benzene, was developed to determine the rate

80

parameters for the reaction between Mn(III) and benzene. By using benzene as a

81

reference probe, the second-order rate constants for the reaction of emerging

82

contaminants using aniline and bisphenol A (BPA) as model compounds with Mn(III) 5



83

were indirectly determined based on competition kinetics method.

84

EXPERIMENTAL SECTION

85

Materials. Potassium permanganate (GR grade), phenol (99% pure) and sodium

86

thiosulfate pentahydrate (GR grade) were purchased from the Tianjin Chemical

87

Reagent Co., Ltd. (Tianjin, China). Sodium bisulfite (AR grade) and benzene (AR

88

grade) were obtained from Chinasun Specialty Products Co., Ltd. (Jiangsu, China).

89

BPA (96% pure) and aniline (99% pure) were purchased from Sigma-Aldrich (St.

90

Louis, MO, USA) and methanol (99.9% pure) was supplied by Merck KgaA

91

(Germany). All chemicals were used as received. Crystals of KMnO4 were dissolved

92

in Milli-Q water to prepare a 50 mM stock solution. The stock solution of NaHSO3

93

(250 mM) was freshly prepared for each set of experiments to avoid oxidation by

94

oxygen. The stock solutions of BPA (1.0 mM), aniline (5 mM), sodium thiosulfate

95

pentahydrate (100.0 mM) and benzene (10.0 mM) were prepared in Milli-Q water

96

every day.

97

Stopped-flow Experiments. A stopped-flow spectrophotometer (SFS, Model

98

SX20, Applied Photophysics Ltd., Leatherhead, UK) was employed to conduct the

99

rapid kinetic experiments. A UV-visible spectrophotometer and a fluorimeter were

100

used as the detectors with a 150 W xenon lamp as the light source. An HP computer

101

workstation was employed to control the SFS and acquire the kinetic data. The

102

detailed procedures for the SFS experiments are described in Supporting Information

103

(SI) Text S1. pH was measured by a pH meter with a saturated KCl solution as the 6


Page 6 of 36

Page 7 of 36


104

electrolyte. Daily calibration with proper buffer solution (pH 4.00, 6.86 and 9.18) was

105

performed to ensure its accuracy. All kinetic experiments were carried out in at least

106

triplicate at 18 ± 2 oC, and the data were averaged with the standard deviations < 5%

107

unless otherwise noted.

108

Batch Experiments. To investigate the removal of high concentration of BPA

109

by the PM/BS system, reactions were initiated by quickly spiking permanganate into

110

the solutions containing BPA and bisulfite while being mixed with a magnetic stirrer

111

in glass bottles. Considering the interference of BPA on the analysis of phenol with

112

fluorimeter equipped in SFS, the competition kinetics experiments were also

113

performed in glass bottles to determine the amount of phenol generated in the process

114

of benzene oxidation by PM/BS with the presence of BPA and the details are present

115

in SI Text S2.

116

Chemical Analysis for the Batch Experiments. The concentrations of phenol

117

and BPA in the samples taken from the batch experiments were quantified by UPLC

118

(Waters Co.). Separation was accomplished with a UPLC BEH C18 column (2.1 ×

119

100 mm, 1.7 µm; waters) at 35 ± 1 oC. The mobile phase was comprised by

120

methanol-0.1% formic acid aqueous solution (40:60) for phenol and by

121

methanol-water (60:40) for BPA, respectively. The flow rate was 0.3 mL min-1 and

122

the largest volume injection was 10 µL. The concentration of phenol and BPA were

123

determined by comparing the peak area at 273 nm and 280 nm with that of the

124

corresponding phenol and BPA standard, respectively. The UV-vis spectra of batch 7



125

experiments were determined using Purkinje TU-1902 automatic scanning UV-visible

126

spectrophotometers.

127

RESULTS AND DISCUSSION

128

The Reactions Involved in Organic Contaminants Oxidation in the PM/BS

129

Process. As suggested by our previous study, five major reactions (Eqs 1-5 in Table 1)

130

occurred in the process of organic contaminants degradation by PM/BS.36 No obvious

131

absorbance of manganese intermediates (Mn(VI), Mn(V) and Mn(IV)) during Mn(III)

132

generation was observed, suggesting no generation or extremely rapid generation and

133

disappearance of these manganese intermediates, thus the reaction process from

134

Mn(VII) to Mn(III) was simplified to Eq 1. The generated Mn(III) was then consumed

135

by the following three reactions, including: (i) reduction by HSO3− (Eq 2), (ii)

136

reduction by target compounds (Eq 3), and (iii) disproportionation (Eq 4). When the

137

experiments were performed open to the air, bisulfite could be oxidized by oxygen

138

(Eq 5).39, 40

139

Although it has been reported that Mn(III) stabilized by complexation with

140

pyrophosphate (PP) displays an absorbance peak at 258 nm,43 very little is known

141

about the nature of soluble aquo and/or hydroxo Mn(III) (non-complexed with other

142

ligands) and the reactions of Mn(III) occurring in the PM/BS process. To verify the

143

mechanisms proposed in Eqs 1-5, the time-resolved UV absorbance at various

144

wavelengths was determined at pHini 5.0. As demonstrated in Figure S2, the

145

absorbance at 230-290 nm initially increased sharply and then decreased, forming a 8


Page 8 of 36

Page 9 of 36


146

peak at ~10 ms, followed by a gradual rebound to a plateau after 0.2 s. At the initial

147

stage, MnO4− was reduced to Mn(III) by HSO3− (Eq 1) which accounted for the rapid

148

increase of absorbance at 230-290 nm due to the higher molar absorption coefficient

149

of Mn(III) than MnO4−. After the peak, a rapid decrease in the absorbance at 230-290

150

nm could be attributed to (i) the decrease of MnO4− concentration, (ii) the decrease of

151

generation rate of Mn(III) with the proceeding of the reaction of HSO3− with MnO4−,

152

(iii) the increase of consumption rate of Mn(III) by the reaction between bisulfite and

153

Mn(III). As the reaction proceeded, the concentration of bisulfite decreased and part

154

of Mn(III) disproportionated to Mn2+ and MnO2 accounting for the second raise of

155

absorbance at 230-290 nm as the molar absorption coefficient of MnO2 was greater

156

than that of Mn(III) over this wavelength range.21 The absorbance at 300-350 nm

157

dropped sharply at the beginning, which should be ascribed to the consumption of

158

MnO4− by HSO3−, and then increased gradually, arising from the generation of MnO2.

159

As such, the absorbance over the wavelength of 230-290 nm could be employed to

160

describe the presence of soluble Mn(III).

161

To evaluate the properties of Mn(III) generated at various pH, the change of

162

absorbance at 258 nm with time during the reaction of permanganate with

163

bisulfite/sulfite at a S(IV)/Mn(VII) molar ratio of 5.0 was investigated at pHini 4.5-8.0,

164

as illustrated in Figure 1. It was interesting to find that the profiles of the absorbance

165

at 258 nm at pHini 4.5-6.5 were very similar, i.e., the absorbance increased to an apex,

166

decreased hereafter and then rebound to a plateau. However, the rate of approaching 9



167

to the absorbance apex progressively decreased with increasing pHini. As pHini was

168

elevated to 7.0-8.0, no obvious apex was observed and the absorbance at 258 nm,

169

arising from MnO2 formation, was much greater than that at lower pHini. To identify

170

the role of Mn(III) in PM/BS process at pHini 7.0-8.0, methanol was employed as a

171

probe compound following the practice in our previous study.36 Figure S3 showed that

172

methanol added 20 s after the initiation of the reaction between permanganate with

173

bisulfite had no influence on the amount of generated MnO2, suggesting that the in

174

situ formed MnO2 could not oxidize methanol. However, methanol dosed before the

175

initiation of the reaction suppressed the generation of MnO2 by reducing Mn(III),

176

implying that Mn(III) was responsible for the oxidation of organic contaminant at

177

pHini 8.0. Therefore, the non-appearance of obvious apex at pHini 7.0-8.0 did not deny

178

the generation of Mn(III) in the PM/BS process but it indicated that the

179

disproportionation rate of Mn(III) (Eq 4) at pHini 7.0-8.0 was much greater than that at

180

lower pHini and it was even greater than the reduction of Mn(III) by bisulfite (Eq 2).

181

Our observations are consistent with the previous study that the disproportionation

182

rate of Mn(III) increased with increasing pH.44 In addition, the absorbance at 525 nm

183

in the PM/BS process at pHini 7.0-9.0 did not drop to zero and the steady-state

184

absorbance increased with pHini, implying that the disproportionation of Mn(III) could

185

outcompete the reduction of Mn(III) by HSO3−/SO32− at pHini 7.0-9.0 (Figure S4).

186

There might be a surface association of Mn(III) with in-situ formed MnO2 under

187

neutral and alkaline conditions which is associated with disappearance of apex. 10


Page 10 of 36

Page 11 of 36


188

Consequently, the effective fraction of Mn(III) used for degrading organic

189

contaminants at pHini 7.0-9.0 would be low due to the fast disproportionation rate of

190

Mn(III) over this pHini range. As shown in Figure S5, the removal of BPA in the

191

PM/BS process did drop substantially as pHini increased from 6.5 to 9.0, verifying that

192

the utilization rate of Mn(III) was low when the experiments were initiated at pH ≥

193

7.0.

194

To further authenticate our above inference, the evolutions of absorbance at 258

195

nm during the reaction of MnO4− with HSO3− of various concentrations (250-1000

196

µM) open to the air or under anaerobic conditions were compared at pHini 5.0, as

197

illustrated in Figure S6. Although the oxidation of bisulfite by oxygen (Eq 5) was not

198

so quick as that by permanganate,39 the presence of Mn2+ highly accelerated the

199

consumption of bisulfite by oxygen (Eq 5).40 Therefore, compared to its counterpart

200

without purging with nitrogen, the consumption of HSO3− by oxygen (Eq 5) was

201

inhibited and thus more Mn(III) was reduced by bisulfite to Mn2+ (Eq 2) and the

202

generation of MnO2 was depressed under anaerobic conditions, as shown in Figure

203

S6(A-B). The similar phenomena were observed by increasing the initial HSO3−

204

concentration from 250 to 1000 µM under aerobic conditions (Figure S6(C-D)). All

205

these evidences mentioned above did verify that (i) Mn(III) was the active oxidant in

206

the PM/BS process, (ii) the fraction of Mn(III) effective for decomposing organic

207

contaminant could be decreased by the presence of excessive bisulfite or by the rapid

208

disproportionation of Mn(III) at pHini ≥ 7.0. 11



209

Mn(III) Generation and Utilization Model (MGU model). It was found that

210

benzene could be transformed to phenol by the PM/BS process very rapidly at various

211

pHini levels but the amount of phenol generated at the end of reaction varied with

212

pHini, as demonstrated in Figure 2. Due to the presence of benzene in large excess (2.5

213

mM), the oxidation of phenol generated in this process by Mn(III) could be neglected.

214

This point was testified by spiking 10 µM phenol before the initiation of benzene

215

oxidation by PM/BS at pHini 5.0 and detecting the accumulation of phenol, as shown

216

in Figure S7. The amounts of phenol accumulated in the process of benzene oxidation

217

by PM/BS at pHini 5.0 without and with 10 µM spiked phenol were 11.5 µM and 20.8

218

µM, respectively. Therefore, further oxidation of generated phenol from benzene by

219

Mn(III) was neglected to simplify the mathematical model.

220

With the presence of excess benzene, pH in the PM/BS process kept almost

221

constant when the experiments were initiated at pHini ≤ 7.0 and the residual

222

absorbance at 525 nm was almost zero at pHini ≤ 7.0, indicating that Mn(III)

223

disproportionation was completely inhibited in the presence of excess benzene and no

224

MnO2 was generated at pHini ≤ 7.0 (Figure S8). Therefore, it could be concluded

225

that at pHini ≤ 7.0 permanganate was completely reduced to Mn2+ during the

226

oxidation of benzene in the PM/BS process in two steps: Mn(III) was generated by the

227

reduction of permanganate by bisulfite and then Mn(III) was reduced by benzene and

228

bisulfite to Mn2+. Different from the data shown in Figure S4, no lag phase was

229

observed in the kinetics of permanganate reduction by bisulfite at pHini 5.0-6.0 in the 12


Page 12 of 36

Page 13 of 36


230

presence of excess benzene, which should be ascribed to the stable pH during the

231

reaction. At pHini 7.5-8.0, pH was increased to 8.0-9.0 and the absorbance at 525 nm

232

did not disappear completely at the end of reaction, which should be ascribed to the

233

MnO2 generated from Mn(III) disproportionation. This was confirmed by the

234

concomitant increase in the absorbance at 418 nm at pHini 8.0-9.0 (Figure S9), which

235

also revealed that Mn(III) disproportionated more severely at higher pHini. Therefore,

236

the disproportionation rate of Mn(III) at pHini > 7.0 was greater than the reaction rate

237

of Mn(III) with benzene and sulfite and thus MnO2 was generated, consistent with the

238

data present in Figure 1.

239

As the reaction of permanganate with bisulfite was too fast to be tracked at pHini

240

≤ 4.0, the kinetic fitting was not conducted at pHini ≤ 4.0. The loss of permanganate

241

followed pseudo-first-order kinetics with bisulfite/sulfite in 10-fold excess in the

242

presence of 2.5 mM benzene at pHini 4.5-9.0, suggesting that the reduction of

243

permanganate by bisulfite was first-order with respect to permanganate. The obtained

244

second-order reaction rate constants (k1) of permanganate reduction by bisulfite, listed

245

in Table 2, varied from 2.76×104 to 1.45×105 M-1 s-1, which were close to the value

246

(1.28(± 0.08) × 105 M-1 s-1) reported by Simándi et al.41 although they determined it at

247

pH > 9.0 with NaClO4 as background ion. Furthermore, k1 kept almost constant at

248

pHini 7.5 to 9.0, indicating that the raise of pH to 8.0-9.0 had negligible influence on

249

the rate of permanganate reduction by bisulfite during benzene oxidation by PM/BS

250

when the experiments were initiated at pH 7.5-8.0. 13



Page 14 of 36

251

The above analysis verified that there were five main reactions occurred during

252

benzene oxidation in the PM/BS process (Eqs 1-5): generation of Mn(III) from

253

reaction of permanganate with bisulfite/sulfite, oxidation of excess bisulfite/sulfite by

254

oxygen, consumption of Mn(III) by either reacting with benzene and bisulfite/sulfite

255

or disproportionation. Therefore, a simplified Mn(III) generation and utilization

256

model (MGU model), involving the loss of MnO4− and the generation of phenol, was

257

proposed, as shown in Eqs 6 and 7. In this model, MnO4− was reduced to Mn(III) by

258

HSO3−/SO32− with a rate constant of (Eq 6) and the generated Mn(III) oxidized

259

benzene with a rate constant of (Eq 7). The amount of Mn(III) utilized for the

260

oxidation of benzene to phenol (Eq 7) was assumed to be 1/θ (utilization rate) of total

261

Mn(III) generated from Eq 6, where 1/θ )

(14)

The generation of phenol can be expressed using second order kinetics by Eq 15, [?@0AB] "

= × λ × [Mn(III)] × [Benzene]

288

Thus,

289

[Phenol] = λ × × [Benzene] × -4 [Mn(III)]dt

290

Substitution of Eq 14 into Eq 16 and rearranging,

291

[Phenol] =

292

(10)

The consumption of MnO4− can be expressed using second order kinetics by Eq

275 276

2

"

- .×5 ×[/0100]" & e& -3 .××[/0100]" × -4 × [MnO& dt % ] × [HSO) ] × e 3

(15)

"

C× ×[/0100]×[8: 9 ]3

.× ×[/0100]& ×[;

Modeling the Kinetics of Contaminants Oxidation and the Generation

Recommend Documents