A New Reaction Pathway for Bromite to Bromate in the Ozonation of

Subscriber access provided by CMU Libraries - http://library.cmich.edu

Article

A New Reaction Pathway for Bromite to Bromate in the Ozonation of Bromide Alexandra Fischbacher, Katja Löppenberg, Clemens von Sonntag, and Torsten C Schmidt Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.5b02634 • Publication Date (Web): 15 Sep 2015 Downloaded from http://pubs.acs.org on September 21, 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 19

1 2

Environmental Science & Technology

A New Reaction Pathway for Bromite to Bromate in the Ozonation of Bromide

3 4

Alexandra Fischbacher,a Katja Löppenberg,a Clemens von Sonntag†,a,b and Torsten C.

5

Schmidta,c*

6

(a) Instrumental Analytical Chemistry, University of Duisburg-Essen, Universitätsstraße 5,

7

45141 Essen, Germany

8

(b) Max-Planck-Institut für Bioanorganische Chemie, Stiftstr. 34-36, 45413 Mülheim an der

9

Ruhr, Germany

10

(c) Centre for Water and Environmental Research (ZWU), University of Duisburg-Essen,

11

Universitätsstraße 2, 45141 Essen, Germany

12

*

13

E-Mail: [email protected]

Corresponding author: Torsten C. Schmidt Phone: +49 201-183-6772

14 15

Fax: +49 201-183-6773 †

Clemens von Sonntag passed away April 5, 2013

16 17

Graphical abstract

18 19 20

1 ACS Paragon Plus Environment


21

Page 2 of 19

Abstract

22

Ozone is often used in drinking water treatment. This may cause problems if the water to be

23

treated contains bromide, as its reaction with ozone leads to bromate formation, which is

24

considered to be carcinogenic. Bromate formation is a multistep process, resulting from the

25

reaction of ozone with bromite. Although, this process seemed to be established, it has been

26

shown that ozone reacts with bromite not by the previously assumed mechanism via O-transfer

27

but via electron transfer. Besides bromate, the electron transfer reaction also yields O3• ͞ , the

28

precursor of OH radicals. The experiments were set-up in a way, that OH radicals are not

29

produced from ozone self-decomposition, but solely by the electron transfer reaction. This study

30

shows that hydroxyl radicals are indeed generated using tBuOH as OH radical scavenger and

31

measuring its product formaldehyde. HOBr and bromate yields were measured in systems with

32

and without tBuOH. As OH radicals contribute to bromate formation, higher bromate and HOBr

33

yields were observed in absence of tBuOH than in its presence, where all OH radicals are

34

scavenged. Based on the results presented here a pathway from bromide to bromate revised in

35

the last step was suggested.

36

Introduction

37

Ozone is often used for disinfection and oxidation purposes and has a wide application in

38

drinking water treatment. However, its use in water treatment may cause problems when the

39

water to be treated contain bromide at concentrations higher than 50 µg/L 1, as the reaction of

40

ozone with bromide leads to bromate formation. Bromide concentrations in natural waters

41

usually range between 4 and 1000 µg/L (in some cases, e.g., oceans even more)2 and thus the

42

critical value for bromide of 50 µg/L is environmentally relevant. Therefore, the bromide

43

concentration must be controlled before applying an ozone-based treatment step. Bromate is

44

considered a potential carcinogen, with a threshold limit value in drinking water of 10 µg/L in

45

America and Europe3, 4. Therefore, it is of great interest to elucidate the mechanism of bromate

46

formation in drinking water treatment. The low drinking water standard may not be justified

47

anymore, as preliminary experiments conducted in real and synthetic gastric juices predict a

48

short lifetime for bromate in human stomach.5-7 It was shown that at H+, Cl

49

concentrations typically found in gastric juices, 99 % of bromate is reduced within its retention

50

time of 20 - 30 minutes in the stomach. However, as long as the low drinking water standard is


͞

and H2S

Page 3 of 19


51

still present, bromate is one of the most relevant by-products formed in ozone-based processes.

52

Furthermore, once formed, it is neither biodegradable nor removable by conventional treatment.

53

Bromate formation is a multi-step process as has been so far described in reactions (1) – (4) 8.

54

Br ͞ + O3 → BrO ͞ + O2

k = 160 M-1 s-1

(1)

55

BrO ͞ + O3 → BrO2 ͞ + O2

k = 100 M-1 s-1

(2)

56

BrO ͞ + O3 → Br ͞ + 2 O2

k = 330 M-1 s-1

(3)

57

BrO2 ͞ + O3 → BrO3 ͞ + O2

k > 1 × 105 M-1 s-1

(4)

58

This mechanistic concept has been studied in detail and seems to be well established.

59

However, it has been suggested that bromite reacts with ozone by electron transfer (reaction 5) 9

60

and not by O-transfer (reaction 4), as previously assumed. The authors investigated the reaction

61

of bromite with ozone and postulated that electron transfer provides a lower kinetic barrier than

62

breakage of the O–O bond, and, therefore, the reaction (4) is unlikely to proceed despite

63

comparable reaction rate constants. The authors provided experimental evidence for the electron

64

transfer reaction by measuring the intermediate BrO2• (reaction 5). However, a further

65

experimental or theoretical corroboration is missing.

66 67 68 69 70

BrO2 ͞ + O3 → BrO2• + O3• ͞

k = 8.9 × 104 M-1 s-1

(5)

The electron transfer (5) yields BrO2• and O3• ͞ , the precursor of •OH (overall reaction 6; for details see10). O3• ͞ + H2O → •OH + O2 + OH ͞

(6)

BrO2• disproportionates into bromite and bromate (reaction 7) 11, 12.

71

2 BrO2• ⇌ Br2O4

k = 1.4 × 109 M-1 s-1

(7a)

72

Br2O4 + OH ͞ → BrO2 ͞ + BrO3 ͞ + H+

k = 7 × 108 M-1 s-1

(7b)

73

In the ozone-based reaction pathway from bromide to bromate, •OH can also act as an

74

oxidation agent and contribute to bromate formation.13 In a cascade of reactions (8 – 12),

75

bromide reacts with •OH yielding relevant intermediates (e.g. HOBr) that can further react with

76

ozone or •OH giving bromate (reactions 13 – 16 and 7). 3 ACS Paragon Plus Environment


Page 4 of 19

77

Br ͞ + •OH ⇌ BrOH• ͞

k = 1.06 × 1010 M-1 s-1

(8) 14

78

BrOH• ͞ + H+ ⇌ Br• + H2O

k = 4.4 × 1010 M-1 s-1

(9) 14

79

Br• + Br ͞ → Br2• ͞

k ≃ 1010 M-1 s-1

(10) 14

80

Br2• ͞ + Br2• ͞ → Br ͞ + Br3 ͞

k = 1.8 × 109 M-1 s-1

(11) 15

81

Br3 ͞ + H2O → HOBr + 2 Br ͞ + H +

82

HOBr + •OH → BrO• + H2O

k = 2 × 109 M-1 s-1

(13) 16

83

OBr ͞ + •OH → BrO• + OH ͞

k = 4.5 × 109 M-1 s-1

(14) 11

84

2 BrO• + H2O → OBr ͞ + BrO2 ͞ + 2 H+

k = 4.9 ± 1.0 × 109 M-1 s-1 (15) 11

85

BrO2 ͞ + •OH → BrO2• + OH ͞

k = 1.8 × 109 M-1 s-1

(12)

(16) 11

86

The impact of •OH on the bromate formation is negligible if H2O2 is present.17 As H2O2 is

87

always in large excess over •OH, the formed HOBr is not oxidized further but reduced to

88

bromide (reaction 17) 18, outcompeting the further oxidation. HOBr + HO2 ͞ → Br ͞ + H2O + O2

89

k = 7.6 × 108 M-1 s-1

(17)

90

As •OH contribute to bromate formation, an indication for •OH formed in the system could be

91

the comparison of the bromate formation with and without a suitable •OH scavenger. If •OH are

92

present, the bromate yield should be higher than in systems without •OH. The same would apply

93

to HOBr formation. Due to the fast reaction of bromide with •OH, higher HOBr concentrations

94

should be found in the absence of an •OH scavenger. A prerequisite for formation of •OH is the

95

intermediate ozonide radical anion (O3• ͞ ). If the experimental conditions are chosen in a way

96

that no ozone self-decomposition occurs, O3• ͞ is only generated in pure water if reaction (5)

97

occurs.

98

Ozone self-decomposition in water is initiated by the reaction of ozone with hydroxide ions

99

(reaction 18). Therefore, it is strongly pH dependent. At pH 6.5, the second-order rate constant

100

for ozone decomposition in water is 2.42 M-1 s-1, while the second order rate constant for the

101

reaction of bromide with ozone is 160 M-1 s-1. When ozone decomposes, •OH are generated

102

(reactions 18 – 21). The ozone concentration was chosen not to be higher than the bromide


Page 5 of 19


103

concentration, so no ozone is present in excess. The fraction of ozone self-decomposition was

104

calculated for the worst-case scenario in which the ozone concentration was equal to the

105

bromide concentration and was only around 1.5 %. If reactions of ozone with other bromine

106

species are considered (alongside bromide), the fraction would be even smaller. Therefore, self-

107

decomposition of ozone can be excluded as an •OH source in the present study.19

108

O3 + OH ͞ → HO2 ͞ + O2

(18)

109

O3 + HO2 ͞ → O2•͞ + •OH + O2

(19)

110

O3 + O2•͞ → O3•͞ + •O2

(20)

111

O3• ͞ + H2O → •OH + O2 + OH ͞

(21)

112

Even though the electron transfer reaction was suggested about a decade ago based on UV-Vis

113

spectroscopic observation of BrO2•, in the context of water treatment the previously established

114

concept of the O-transfer reaction in the last step of the pathway is still prevalent20, 21. It is the

115

intention of the present paper to determine the •OH yield and based on these data to provide

116

further evidence of electron transfer occurring in the oxidation of bromite on route to bromate

117

formation. This seems essential to establish a correct mechanistic concept for bromate formation

118

and has consequently already been picked up in the recent textbook by von Sonntag and von

119

Gunten 22, that, however, does not provide any experimental details.

120

Material and Methods

121

Ozone stock solutions (0.9 mM) were prepared by generating ozone with an oxygen-fed

122

ozonator (BMT 802X, BMT Messtechnik, Berlin) and passing it through ice-cooled (4 °C) ultra

123

pure water (Purelab Ultra, Elga LabWater, Celle). The ozone concentration of these stock

124

solutions was determined spectrophotometrically using ε(260 nm) = 3300 M-1 cm-1.23

125

•

126

Hydroxyl radicals (•OH) are highly reactive and therefore short-lived, so they cannot be

127

quantified by direct measurement. As direct measurement of •OH is difficult an appropriate

128

indirect method must be selected. The indirect quantification is performed by scavenging •OH

129

by a suitable probe compound and determinatingthe formed products. Such a probe compound

130

must react rapidly with •OH and barely with other substances present in the sample solution.

OH quantification



Page 6 of 19

131

Furthermore, the yield of the products from this reaction must be known, so that the •OH yield

132

can be derived from these results. In the present system, tBuOH was selected as a scavenger, as

133

it complies with the requirements. It reacts readily with •OH (k = 6 × 108 M-1 s-1)

134

with ozone (k ~ 3 × 10-3 M-1 s-1) 25. In all experiments, the concentrations of tBuOH (180 mM)

135

and bromide (0.6 mM) were kept constant, while ozone concentration was varied (0.15 - 0.6

136

mM). The tBuOH concentration was chosen to be 200 times higher than that of bromide for

137

assuring that 99 % of the formed •OH are scavenged by tBuOH and do not react with any

138

intermediates present in the solution (the calculation of the •OH fraction reacting with tBuOH is

139

shown in SI). As nearly all •OH react with tBuOH their quantification is possible. One of the

140

main products of the reaction of •OH with tBuOH is formaldehyde, which can be determined

141

after derivatization as its 2,4-dinitrophenylhydrazone by HPLC-UV at λ = 350 nm

142

derivatization, 0.1 mL perchloric acid (1 M in acetonitrile) and 0.2 mL 2,4-DNPH solution (6

143

mM in acetonitrile) were added to 1.7 mL of sample. After standing for 45 minutes in the dark,

144

the solutions were measured by HPLC-UV (Agilent 1100, Agilent Technologies, Santa Clara,

145

CA; column: NUCLEODUR 5µm C18 ec 250 x 3 mm (Macherey-Nagel), eluent: gradient

146

acetonitrile/water pH 3 within 20 minutes from 45 to 80 % acetonitrile at 0.5 mL/min). The

147

retention times for 2,4-DNPH and derivatized formaldehyde were 5.5 and 9.3 minutes,

148

respectively. The quantification was performed using formaldehyde standards derivatized in the

149

same way as the samples. To control whether DNPH was contaminated with formaldehyde,

150

blank samples were measured containing perchloric acid, DNPH solution and deionized water.

151

The contamination of the DNPH solution was negligible. As the samples were always in the

152

same concentration range as the standards, the impact of an incomplete derivatization can also

153

be neglected.

24

but barely

26

. For the

154

As an alternative probe compound methanol was tested, as it has been applied as •OH

155

scavenger in ozone based systems27. It reacts fast with •OH (k = 9.7 × 108 M-1 s-1) 24 and barely

156

with ozone (k ~ 2 × 10-2 M-1 s-1) 25. To obtain formaldehyde as the main product, HPO42 ͞ and/or

157

OH

158

experiments were conducted in 2 mM H2PO4 ͞ /HPO42

159

concentration of methanol (470 mM) and bromide (0.6 mM) were kept constant while ozone

160

was varied between 0.15 and 0.6 mM. The fraction of •OH reacting with methanol is calculated

161

to be 99 % in the given system (the calculation of the •OH fraction reacting with methanol is

162

shown in SI). Formaldehyde, the product of this reaction, was determined as described above.

͞

are required as the reaction is base-catalyzed (reactions 28 – 31). ͞

28

Therefore, all

buffered water (pH 7 and 8). The


Page 7 of 19


163

Quantification of bromate and HOBr

164

Bromate was determined by ion chromatography (883 Basic IC Plus, Metrohm, Filderstadt)

165

using a Metrosep A Supp 4 – 250/4 column and an eluent consisting of 0.17 mM NaHCO3 and

166

0.18 mM Na2CO3 at 1 ml/min. Retention times for bromate and bromide were 12 and 19.3 min,

167

respectively.

168

HOBr stock solution was prepared from a 1 mM pH 4 buffered ozone solution (100 mM

169

phosphate buffer) by addition of 0.8 mM potassium bromide solution29. This solution was stored

170

for 24 h in the dark and residual ozone was removed by purging the solution with nitrogen gas

171

for 15 minutes. The first prepared HOBr solution was standardized iodometrically and by direct

172

photometric determination of hypobromite at 329 nm (ε = 332 L mol-1 cm-1) after adjusting the

173

pH to 11 by sodium hydroxide.30 Both methods gave consistent results and further

174

standardization of HOBr solutions was only done photometrically before every use.

175

A photometric determination of HOBr formed in the samples was not possible due to other

176

possible constituents of the samples, which might absorb light in the same range. Therefore,

177

HOBr formed in the samples was determined by measuring its derivatization products with

178

phenol, which yields 2- and 4-bromophenol.31, 32 The pH for derivatization was varied from 2 to

179

5.

180

Calibration for 2- and 4-bromophenol was prepared from purchased ≥98 % pure 2- and 4-

181

bromophenol, respectively. In order to investigate the reaction of HOBr with phenol and

182

determine the bromophenol yield per HOBr, standards with defined HOBr concentrations were

183

prepared from the produced HOBr stock solution. These solutions were buffered at pH 4 (100

184

mM phosphate buffer). 15 µL phenol (0.5 M) was added to 1 mL of the buffered standards.

185

After 2 h (after the reaction was completed) the solutions were measured by HPLC-UV.

186

2- and 4-bromophenol were determined using HPLC-UV at a wavelength of 271 nm and a

187

flow rate of 0.4 mL min-1 (column: NUCLEODUR 5µm C18 ec 250 x 3 mm (Macherey-Nagel)

188

eluent: gradient acetonitrile/water within 18 minutes from 30 to 70 % acetonitrile). The retention

189

times for phenol, 2- and 4-bromophenol were 9.2, 14.1 and 15.2 minutes, respectively.

190

Results and Discussion

191

•

OH Quantification by tBuOH



Page 8 of 19

192

The reaction of tBuOH with •OH initiates a radical chain reaction 33. In the presence of O2, a

193

cascade of reactions leads to stable products. One of these products that can be easily quantified

194

is formaldehyde.

195

to the generation of O3 from O2), the reactions yielding formaldehyde dominate here.

Since in ozonation processes O2 is always present in excess over O3 (due

50

100 80

[CH2O] / µM

40

60 30 40 20

10 100

[•OH] / µM

33, 34

20

200

300

400

500

600

0 700

[O3] / µM

196 197 198

Figure 1 Formation of formaldehyde (⊠, ■ and □ correspond to 3 independent experiments

199

and cannot be shown as mean values as the ozone concentrations differ slightly) in the

200

reaction of •OH and tBuOH (left y-axis). •OH are generated in the reaction of ozone

201

with bromite. The shaded area represents the possible range of •OH yield (right y-

202

axis). The upper limit corresponds to the highest possible yield, if •OH yield is 3.3

203

times the formaldehyde yield. The lower limit corresponds to the lowest possible •OH

204

yield, if •OH yield is twice of the formaldehyde yield. Circles and the linear trend line

205

correspond to the possible calculated •OH yields using a correction factor of

206

2.29±0.06 (details see text).


Page 9 of 19


207

The results obtained from the tBuOH/•OH system show that •OH were formed (Figure 1) as

208

indicated by reaction (6). As the formation of bromate from bromide occurs not via one step

209

(reactions 1 – 5), the yields are not linear. The formaldehyde yield always depends on the

210

system generating •OH, as superoxide (O2•͞ ) is formed in the sequence of reactions of tBuOH

211

with •OH.34 It reacts fast with ozone giving rise to additional •OH (reaction 22) that contribute

212

to higher formaldehyde yields. This additional formaldehyde source has to be taken into account

213

when calculating the •OH yield. The formaldehyde yield is about 30±4 % of •OH (•OH yield is

214

3.3 times the formaldehyde yield), if no ozone is present. In the presence of ozone it is estimated

215

to be around 50 % (•OH yield is 2 times the formaldehyde yield). 34 If the yield is not corrected,

216

it leads to overestimation of •OH yields. In this system, the •OH yield is just an approximate

217

value, as O2•͞ reacts not only with ozone, but also with other species present in the sample

218

(reactions 22 – 27). HOBr must be the most relevant bromine species reacting with O2•͞ due to

219

its high concentration in the sample and its fast reaction rate with O2• ͞ . As the pH of the solution

220

is 6.5 and the pKa of HOBr is 8.69 35, the slower reaction with OBr ͞ can be neglected, as under

221

these conditions HOBr (99%) is present in excess over OBr ͞ (1%). O2• ͞ reacts with HOBr (k =

222

3.5 × 109 M-1 s-1) even faster than with ozone, hence HOBr acts here as competitor for ozone

223

and makes it difficult to define the formaldehyde yield per •OH, as it is not possible to use the

224

known correction factor for the formaldehyde yield in ozone-based processes.

225

O3 + O2• ͞ → O3• ͞ + O2

226

HOBr + O2• ͞

→ O2 + OH ͞

227

OBr ͞ + O2• ͞

→ O2 + 2 OH ͞ HO

Br ͞

Br ͞

+ Br2• ͞ + Br2• ͞

k = 1.6 × 109 M-1 s-1

(22)

k = 3.5 × 109 M-1 s-1

(23)

k > 2 × 108 M-1 s-1

(24)

2

228

Br2• ͞ + O2• ͞ → O2 + 2 Br ͞

k = 1.7 × 108 M-1 s-1

(25)

229

Br3• ͞ + O2• ͞ → O2 + Br ͞ + Br2• ͞

k = 1.5 × 109 M-1 s-1

(26)

230

Br2• ͞ + Br2• ͞ → Br ͞ + Br3 ͞

k = 1.8 × 109 M-1 s-1

(27)

231

The •OH yield is shown in Figure 1. Due to the reaction of O2• ͞ with ozone and HOBr, it is

232

not possible to give an exact •OH yield per ozone. The shaded area indicates the possible range

233

for the yield. The lower line gives the lowest limit considering solely the reaction of O2• ͞ with

234

ozone and the upper line indicates the highest yield excluding the reaction of O2• ͞ with ozone.



Page 10 of 19

235

As the system is very complex it cannot be surely predicted to what extent O2• ͞ reacts with

236

ozone and to what extent with bromine species. If one assumes that the most important bromine

237

species reacting with superoxide is HOBr, a possible •OH yield can be calculated. Therefore, the

238

fractions of superoxide reacting with ozone and HOBr were calculated, from which the

239

correction factors were adapted for every ozone concentration. The calculated correction factor

240

is 2.29 ±0.06 (•OH yield is 2.29 times the formaldehyde yield), if the only bromine species

241

reacting with O2• ͞ is HOBr. The calculated possible •OH yields are shown in Figure 1 (white

242

circles and corresponding linear trend line).

243

The exact quantification of •OH would be possible using the DMSO/•OH system, which gives

244

methanesulfinic acid, that is formed in 92 % yield.36 However, this system could not be used

245

here, as high DMSO concentrations would have to be present to scavenge all •OH, but at the

246

same time it would compete with bromide for O3. To comply with the prerequisites of a •OH

247

scavenger the DMSO concentration must be at least 100 times higher than the bromide

248

concentration. Under these conditions, the fraction of •OH reacting with DMSO would be 99 %,

249

but the fraction of ozone reacting with bromide only 45 %.

250

•

251

The reaction of methanol with •OH yields formaldehyde (ractions 28 – 31).28 The

252

formaldehyde yield per •OH is unity. Once a •OOCH2OH is formed it reacts with HPO42 ͞ and

253

OH ͞ to formaldehyde (reaction 30 and 31).

OH Quantification by MeOH

254

•

OH + CH3OH → •CH2OH + H2O

k = 9.7 × 108 M-1 s-1

(28)

255

•

CH2OH + O2 → •OOCH2OH

k = 4.5 × 109 M-1 s-1

(29)

256

•

OOCH2OH + HPO42 ͞ → HCHO + O2• ͞ + H2PO4 ͞

k ≈ 2 × 106 M-1 s-1

(30)

257

•

OOCH2OH + OH ͞ → HCHO + O2• ͞ + H2O

k = 1.8 × 1010 M-1 s-1

(31)

258

In the present system the formaldehyde yield and thus the •OH yield is 83 % with respect to

259

the applied ozone (Figure 2). Such a high •OH yield could not exclusively emerge from the

260

reaction of ozone with bromide. Thus, there must be another relevant source of •OH in the

261

system. In the sequence of the reactions between methanol and •OH, O2• ͞ is formed. Once it is

262

formed, it reacts readily with ozone (reaction 22) and outcompetes the reaction of bromide with


Page 11 of 19


͞

giving •OH, that react with methanol

263

ozone. Hence, nearly all ozone is consumed by O2•

264

yielding formaldehyde. This undesirable reaction leads to an overestimated •OH yield.

265

Therefore, methanol cannot be used for the quantification of •OH in ozone-based reactions.

0.5 y = 0.828x + 0.04 [CH2O] / mM

0.4

0.3

0.2

0.1 0.1

0.2

0.3

268 269

0.5

0.6

[O3] / mM

266 267

0.4

Figure 2 Formation of formaldehyde in the reaction of •OH with MeOH buffered at pH 7 (■) and 8 (□) with 2 mM H2PO4 ͞ /HPO42 ͞ . Formation of Bromate and HOBr

270

Since •OH contribute to bromate formation (reactions 8 – 16) scavenging of •OH must lead to

271

a reduced bromate yield.13 This could be observed in experiments with and without the addition

272

of tBuOH as an •OH scavenger (Figure 3). Such a significant difference between the bromate

273

yield in presence and absence of tBuOH is a further evidence for the •OH formation in the

274

multistep reaction of ozone with bromide to bromate. The bromate yield without tBuOH is

275

around twice its yield in the presence of tBuOH.



Page 12 of 19

30

[BrO−3 ] / µM

20

10

0 100

200

300

276

400 500 [O3] / µM

600

700

277

Figure 3 Formation of bromate in presence of tBuOH as •OH scavenger (squares) and in

278

absence of tBuOH (circles) in the reaction of ozone with bromide plotted against the

279

ozone concentration (n = 3).

280

HOBr is the most important and stable intermediate in the multi-step process en route to

281

bromate. Based on published results presenting the mechanism of the formation of 2- and 4-

282

bromophenol as an example for the formation of brominated phenolic compounds in natural

283

waters in ozone treatment a well working method for HOBr quantification was developed

284

HOBr was quantified indirectly by measuring 2- and 4-bromophenol, the products from its

285

reaction with phenol. The pH was varied from 2 to 5 and the highest yields were obtained at pH

286

4. The slope of the graph in Figure 4 gives the yield as 0.99 with an error of 0.026. Hence, the

287

bromophenol yield with respect to the applied HOBr is ~100 %.


32

.

Page 13 of 19


180

[Bromophenols] / µM

160

y = 0.99x + 2.04

140 120 100 80 60 40 20 0 0

20

40

60

80

100

120

140

160

180

[HOBr] / µM 288 289 290

Figure 4 2-bromophenol (open circles), 4-bromophenol (open triangles) and sum of both

291

bromophenols (solid squares) yielded from the derivatization of HOBr with phenol at

292

pH 4 (n = 3). Relative standard deviations are indicated but often smaller than symbol

293

size. The dashed line indicates a 100 % yield of bromophenols per HOBr.

294

The ratio of 4-bromophenol and 2-bromophenol formed in the reaction of hypobromous acid

295

with phenol is 1.64 at pH 4. This first appears to be contrary to the results of Acero et al. as they

296

reported a ratio of 0.53 for experiments conducted at pH 8-9

297

bromophenols is strongly pH dependent as Tee et al

298

bromophenols is para substituted at pH 0. This means that at low pH 4-bromophenol is

299

prevalent, and at high pH 2-bromophenol is dominant. As the ratio of 2- and 4-bromophenol is

300

pH dependant, it is highly recommended measuring both bromophenols and using their sum

301

concentration to determine the HOBr concentration. HOBr was measured after the reaction was

302

completed and all ozone was consumed. Figure 5 shows the bromophenol yields in the samples.

303

The sum of 2- and 4-bromophenol corresponds to the HOBr formed in the reaction of bromide

304

and ozone. The HOBr concentration in the samples in the absence of tBuOH is higher than in its

31

32

. However, the ratio of the two

published that 95% of formed



Page 14 of 19

305

presence, as •OH are not scavenged and react rapidly with bromide that is present at high

306

concentration yielding HOBr (reaction 8). Furthermore, if no tBuOH is added, no O2• ͞ is formed

307

in the sequence of the reaction between tBuOH and •OH that could react with HOBr (reactions

308

23 and 24). Thus, the lower HOBr yield observed in the presence of tBuOH provides further

309

evidence for •OH formed from the reaction of bromite with ozone via electron transfer.

310

0.18

[Bromophenols] / mM

0.15 R2 = 0.733 0.12 0.09 R2 = 0.977 0.06 0.03 0.00 0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

[O3] / mM 311 312 313

Figure 5 Formation of bromophenols at pH 4 from the reaction of phenol and HOBr formed as

314

intermediate from bromide and ozone. The sum of 2- and 4-bromophenol formed in

315

the presence of tBuOH (solid squares) and in the absence of tBuOH (open squares) as

316

a function of the applied ozone concentration (n = 3).

317

From the results discussed above, it can be concluded that in the sequence of the ozone

318

reactions from bromide to bromate, •OH must be generated. Thus, the original concept of

319

bromate formation has to be revised. Based on the obtained results the known pathway from

320

bromide to bromate was modified (Figure 6). It was updated at the last ozone reaction step. The

321

O-transfer reaction was replaced by the electron transfer reaction.


Page 15 of 19


322 323

Figure 6 (a) Old pathway for the bromate formation in the reaction of ozone with bromide

324

established by U. von Gunten1,

325

of bromide with ozone based on this study adapted from U. von Gunten’s reaction

326

scheme1,

327

reaction of ozone on the right hand side.

328

13

13

and (b) new recommended pathway for the reaction

. Note the difference between the old and new mechanism in the last

Practical Implications

329

The present paper shows that in the multi-step reaction of ozone with bromide to bromate, •OH

330

are generated. This is due to the fact that the last of these steps is not as previously assumed an

331

O-transfer but an electron transfer reaction. The electron transfer reaction also yields bromate,

332

but this new fact could be essential for reaction modeling. Until now, the modeling of the

333

reaction of bromide with ozone does not give good results, as there are still too many not well

334

known reactions included (U. von Gunten 2015, personal communication). In systems with low

335

•

OH scavenger capacity (e.g., waters with low dissolved organic matter content) OH radical



Page 16 of 19

336

formation in the electron transfer step might enhance the bromate yield, as the •OH formed on

337

site may contribute to bromate formation. Under which conditions this is indeed relevant in

338

natural waters requires further investigation.

339

Supporting Information

340

Additional information on the calculation of the •OH fraction reacting with a probe compound as

341

noted in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

342

Acknowledgement

343

We dedicate this paper to the memory of our co-author and dear friend Clemens von Sonntag

344

who was an inspiration to us all. We thank Dr. Holger Lutze and Prof. Urs von Gunten for

345

fruitful discussion and Alua Suleimenova for reviewing the manuscript.

346 347 348

References

349

Formation in Presence of Bromide, Iodide or Chlorine. Water Res. 2003, 37, (7), 1469-1487.

350

2.

351

Ozonation of Bromide-Containing Waters. Aqua (Oxford) 1992, 41, (5), 299-304.

352

3.

353

Disinfection Byproducts. In Federal Register, Environmental Protection Agency: 1994; Vol. 59,

354

(145).

355

4.

356

European Communities, Council Directive 98/83/EC: 1998.

357

5.

358

Bromate Ion in Simulated Gastric Juices. Ozone Sci. Eng. 2006, 28, (3), 165-170.

359

6.

360

Reaction of Bromate Ion with Synthetic and Real Gastric Juices. Toxicology 2006, 221, (2-3),

361

225-228.

362

7.

363

Simulated Gastric Juice. J. Am. Water Works Ass. 2010, 102, (11), 77-86.

364

8.

365

Formation of Hypobromous Acid and Bromate. Environ. Sci. Technol. 1983, 17, (5), 261-267.

1.

von Gunten, U., Ozonation of Drinking Water: Part II. Disinfection and By-Product

von Gunten, U.; Hoigne, J., Factors Controlling the Formation of Bromate During

Drinking Water; National Primary Drinking Water Regulations: Disinfectants and

On the Quality of Water Intended for Human Consumption. In Official Journal of the

Keith, J. D.; Pacey, G. E.; Cotruvo, J. A.; Gordon, G., Preliminary Data on the Fate of

Keith, J. D.; Pacey, G. E.; Cotruvo, J. A.; Gordon, G., Experimental Results from the

Cotruvo, J. A.; Keith, J. D.; Bull, R. J.; Pacey, G. E.; Gordon, G., Bromate Reduction in

Haag, W. R.; Hoigne, J., Ozonation of Bromide-Containing Waters: Kinetics of


Page 17 of 19


366

9.

Nicoson, J. S.; Wang, L.; Becker, R. H.; Hartz, K. E. H.; Muller, C. E.; Margerum, D.

367

W., Kinetics and Mechanisms of the Ozone/Bromite and Ozone/Chlorite Reactions. Inorg.

368

Chem. 2002, 41, (11), 2975-2980.

369

10.

370

Peroxide (Peroxone Process): A Revision of Current Mechanistic Concepts Based on

371

Thermokinetic and Quantum-Chemical Considerations. Environ. Sci. Technol. 2010, 44, (9),

372

3505-3507.

373

11.

374

Compounds; the Spectra and Reactions of BrO and BrO2. Proc. R. Soc. London, Ser. A 1968,

375

304, (1479), 427-439.

376

12.

377

Free Radicals in Aqueous Solution: Quantum-Chemical Calculations. Environ. Sci. Technol.

378

2011, 45, (21), 9195-9204.

379

13.

380

Containing Waters: Interaction of Ozone and Hydroxyl Radical Reactions. Environ. Sci.

381

Technol. 1994, 28, (7), 1234-1242.

382

14.

383

Pulse Radiolytic Investigation. J. Phys. Chem. 1972, 76, (3), 312-319.

384

15.

385

Deaerated Aqueous Bromide Solutions. J. Phys. Chem. 1966, 70, (7), 2092-2099.

386

16.

387

Aqueous Solution - Reactions of Cl- and Br-Atoms. Ber. Bunsen-Ges. Phys. Chem 1985, 89,

388

(3), 243-245.

389

17.

390

Bromate Formation Mechanisms. Environ. Sci. Technol. 1998, 32, (1), 63-70.

391

18.

392

Hypobromous Acid: Implication on Water Treatment and Natural Systems. Water Res. 1997,

393

31, (4), 900-906.

394

19.

395

Influence of pH and Temperature. Russ. J. Phys. Chem. A 2009, 83, (8), 1295-1299.

396

20.

397

Formation During Ozonation. Environ. Technol. 2012, 33, (15), 1747-1753.

Merenyi, G.; Lind, J.; Naumov, S.; von Sonntag, C., Reaction of Ozone with Hydrogen

Buxton, G. V.; Dainton, F. S., The Radiolysis of Aqueous Solutions of Oxybromine

Naumov, S.; von Sonntag, C., Standard Gibbs Free Energies of Reactions of Ozone with

von Gunten, U.; Hoigne, J., Bromate Formation During Ozonization of Bromide-

Zehavi, D.; Rabani, J., Oxidation of Aqueous Bromide Ions by Hydroxyl Radicals -

Matheson, M. S.; Mulac, W. A.; Weeks, J. L.; Rabani, J., The Pulse Radiolysis of Klaning, U. K.; Wolff, T., Laser Flash-Photolysis of HClO, ClO-, HBrO, and BrO- in

von Gunten, U.; Oliveras, Y., Advanced Oxidation of Bromide-Containing Waters:

von Gunten, U.; Oliveras, Y., Kinetics of the Reaction Between Hydrogen Peroxide and

Ershov, B. G.; Morozov, P. A., The Kinetics of Ozone Decomposition in Water, the

Antoniou, M. G.; Andersen, H. R., Evaluation of Pretreatments for Inhibiting Bromate



Page 18 of 19

398

21.

Han, Q.; Wang, H. J.; Dong, W. Y.; Liu, T. Z.; Yin, Y. L., Suppression of Bromate

399

Formation in Ozonation Process by Using Ferrate(VI): Batch Study. Chem. Eng. J. 2014, 236,

400

110-120.

401

22.

402

Treatment - From Basic Principles to Applications. IWA Publishing: London, 2012.

403

23.

404

Bands of Ozone in Aqueous Solutions. Anal. Chem. 1983, 55, (1), 46-49.

405

24.

406

Constants for Reactions of Hydrated Electrons, Hydrogen-Atoms and Hydroxyl Radicals

407

(.OH/.O-) in Aqueous-Solution. J. Phys. Chem. Ref. Data 1988, 17, (2), 513-886.

408

25.

409

Compounds in Water-I Non: Dissociating Organic Compounds. Water Res. 1983, 17, (2), 173-

410

183.

411

26.

412

Automobile Exhaust with an Improved 2,4-Dinitrophenylhydrazine Method. J. Chromatogr.

413

1982, 247, (2), 297-306.

414

27.

415

Propagation Reactions on the Hydroxyl Radical Formation in Ozonation and Peroxone

416

(Ozone/Hydrogen Peroxide) Processes. Water Res. 2015, 68, 750-758.

417

28.

418

Quantum Yield and Applications for Polychromatic UV Actinometry in Photoreactors. Environ.

419

Sci. Technol. 2007, 41, (21), 7486-7490.

420

29.

421

Determination of Reactive Bromine and Chlorine Species with ABTS. Water Res. 2000, 34,

422

(18), 4343-4350.

423

30.

424

Hypobromous Acid Reactions with Iodide and with Sulfite and the Hydrolysis of Bromosulfate.

425

Inorg. Chem. 1991, 30, (18), 3538-3543.

426

31.

427

Phenols and Phenoxide Ions in Aqueous Solution. Diffusion-Controlled Rates. J. Am. Chem.

428

Soc. 1989, 111, (6), 2233-2240.

von Sonntag, C.; von Gunten, U., Chemistry of Ozone in Water and Wastewater

Hart, E. J.; Sehested, K.; Holcman, J., Molar Absorptivities of Ultraviolet and Visible

Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B., Critical-Review of Rate

Hoigne, J.; Bader, H., Rate Constants of Reactions of Ozone with Organic and Inorganic

Lipari, F.; Swarin, S. J., Determination of Formaldehyde and other Aldehydes in

Liu, Y.; Jiang, J.; Ma, J.; Yang, Y.; Luo, C.; Huangfu, X.; Guo, Z., Role of the

Goldstein, S.; Aschengrau, D.; Diamant, Y.; Rabani, J., Photolysis of Aqueous H2O2:

Pinkernell, U.; Nowack, B.; Gallard, H.; Von Gunten, U., Methods for the Photometric

Troy, R. C.; Margerum, D. W., Non-Metal Redox Kinetics: Hypobromite and

Tee, O. S.; Paventi, M.; Bennett, J. M., Kinetics and Mechanism of the Bromination of


Page 19 of 19


429

32.

Acero, J. L.; Piriou, P.; von Gunten, U., Kinetics and Mechanisms of Formation of

430

Bromophenols During Drinking Water Chlorination: Assessment of Taste and Odor

431

Development. Water Res. 2005, 39, (13), 2979-2993.

432

33.

433

Radical-Induced Oxidation of 2-Methyl-2-Propanol in Oxygenated Aqueous Solution - Product

434

and Pulse Radiolysis Study. J. Phys. Chem. 1979, 83, (7), 780-784.

435

34.

436

Determination of .OH, O2.−, and Hydroperoxide Yields in Ozone Reactions in Aqueous Solution.

437

J. Phys. Chem. B 2003, 107, (30), 7242-7253.

438

35.

439

Ohio, 1971-1972.

440

36.

441

Radicals with Sulfoxides. J. Chem. Soc., Perkin Trans. 2 1980, (1), 146-153.

Schuchmann, M. N.; von Sonntag, C., Radiation-Chemistry of Alcohols 22. Hydroxyl

Flyunt, R.; Leitzke, A.; Mark, G.; Mvula, E.; Reisz, E.; Schick, R.; von Sonntag, C.,

Weast, R. C., Handbook of Chemistry and Physics. 52nd ed.; The Chemical Rubber Co.:

Veltwisch, D.; Janata, E.; Asmus, K. D., Primary Processes in the Reaction of OH-

442 443


A New Reaction Pathway for Bromite to Bromate in the Ozonation of

Recommend Documents