Chlorination Revisited: Does Clâ Serve as a Catalyst in the

Subscriber access provided by University of Otago Library

Article #

Chlorination revisited: Does Cl serve as a catalyst in the chlorination of phenols? Stephanie S. Lau, Sonali M. Abraham, and A. Lynn Roberts Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b03539 • Publication Date (Web): 11 Nov 2016 Downloaded from http://pubs.acs.org on November 14, 2016

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 27

Environmental Science & Technology

Chlorination revisited: Does Cl‒ serve as a catalyst in the chlorination of phenols? Stephanie S. Lau,† Sonali M. Abraham,†,‡ and A. Lynn Roberts*, † †

Department of Environmental Health and Engineering, Johns Hopkins University, 313 Ames Hall, 3400 North Charles Street, Baltimore, Maryland 21218, United States

‡

Institute of the Environment and Sustainability, University of California, Los Angeles, La Kretz Hall, 619 Charles E. Young Drive East #300, Los Angeles, California 90024, United States

* Corresponding author contact information: Phone: 410-516-4387; Fax: 410-516-8996; E-mail: [email protected]

1 ACS Paragon Plus Environment


1 2

Abstract The aqueous chlorination of (chloro)phenols are some of the best-studied reactions in the

3

environmental literature. Previous researchers have attributed these reactions to two chlorine

4

species: HOCl (at circum-neutral and high pH) and H2OCl+ (at low pH). In this study, we seek to

5

examine the roles that two largely overlooked chlorine species, Cl2 and Cl2O, may play in the

6

chlorination of (chloro)phenols. Solution pH, chloride concentration, and chlorine dose were

7

systematically varied in order to assess the importance of different chlorine species as

8

chlorinating agents. Our findings indicate that chlorination rates at pH < 6 increase substantially

9

when chloride is present, attributed to the formation of Cl2. At pH 6.0 and a chlorine dose

10

representative of drinking water treatment, Cl2O is predicted to have at best a minor impact on

11

chlorination reactions, whereas Cl2 may contribute more than 80% to the overall chlorination rate

12

depending on the (chloro)phenol identity and chloride concentration. While it is not possible to

13

preclude H2OCl+ as a chlorinating agent, we were able to model our low-pH data by considering

14

Cl2 only. Even traces of chloride can generate sufficient Cl2 to influence chlorination kinetics,

15

highlighting the role of chloride as a catalyst in chlorination reactions.


Page 2 of 27

Page 3 of 27

16 17


Introduction Aqueous chlorine, also known as free available chlorine (FAC), is one of the most widely

18

used disinfectants in drinking water and wastewater treatment processes worldwide. While FAC

19

is critical for protecting consumers from waterborne illnesses, it can react with synthetic

20

compounds and natural organic matter (NOM) to generate disinfection byproducts (DBPs), some

21

of which are known or anticipated to be hazardous to human health.1 The most abundant chlorine

22

species in solutions of FAC are HOCl and OCl‒, with the latter being much less important than

23

the former in electrophilic aromatic substitutions.2 Thus, HOCl (pKa = 7.54 at 25°C, ref. 3) is

24

usually assumed to be the sole chlorinating agent at circum-neutral and high pH.

25

At acidic pH, other chlorinating agents may be present in FAC solutions. The

26

chlorination rates of some organic compounds appear to depend on [H+] at low pH,4-12 a

27

phenomenon that has been attributed to reactions with H2OCl+, the conjugate acid of HOCl with

28

an estimated pKa of -3 to -4 (Fig. 1).13 Due to its positive charge, H2OCl+ has been postulated to

29

be a more reactive electrophile than is HOCl;14 however, no one has managed to detect H2OCl+

30

using established spectrometric methods. The elusive nature of H2OCl+ has led researchers to

31

question whether the high reactivities of some compounds at low pH could be due to reactions

32

with Cl2 (aq) instead. The formation of Cl2 in FAC solutions is favored at low pH and in the

33

presence of chloride:

34 35

HOCl + Cl‒ + H+ ⇌ Cl2 (aq) + H2O

(1)

log KCl2 = 2.72 (ref. 15, 25°C, corrected to 0 M ionic strength using the Davies equation)

36

One study used Raman spectroscopy to pinpoint Cl2 (rather than H2OCl+) as the active low-pH

37

chlorine species involved in the oxidation of several secondary alcohols.16 The importance of Cl2

38

in chlorination reactions has also been shown for several pharmaceuticals,17-21 an herbicide,22 and



Page 4 of 27

39

aromatic ethers.23 Nevertheless, studies from as recent as 2015 and 2016 continue to dismiss Cl2

40

without experimental evidence.24-26 Furthermore, with few exceptions, previous researchers who

41

reported second-order rate constants for Cl2 did not systematically vary [Cl–] in their

42

experiments, potentially yielding estimates with large uncertainties.

43

To explore whether Cl2 has broader significance as a chlorinating agent, we chose to re-

44

examine the reaction kinetics of phenol and its chlorinated derivatives with aqueous chlorine.

45

Phenol is often invoked as a model for phenolic moieties in NOM,27 and (chloro)phenols are

46

widespread micropollutants in the environment.28 Phenol is chlorinated successively via

47

electrophilic aromatic substitution to form 2-chlorophenol (2-CP), 4-chlorophenol (4-CP), 2,4-

48

dichlorophenol (2,4-DCP), 2,6-dichlorophenol (2,6-DCP), and 2,4,6-trichlorophenol (TCP).29

49

Subsequent reactions of TCP with FAC lead to the formation of non-aromatic compounds, with

50

the end products being chloroform6, 30 and trichloroacetic acid.31

51

The reaction kinetics of all six of these (chloro)phenols in the presence of FAC were first

52

studied by Lee and Morris.32 At pH < 6, the reaction rates of some (chloro)phenols increased

53

with decreasing pH. Noting the presence of 1 mM Cl‒ in their reactors, Lee and Morris

54

hypothesized that reactions with Cl2 could explain the observed reaction kinetics at low pH.

55

Gallard and von Gunten,6 however, argued that 1 mM of Cl‒ was too low to generate sufficient

56

Cl2 to influence reaction kinetics. They attributed the high reactivities of some (chloro)phenols at

57

pH < 6 to reactions with H2OCl+ and proposed a model of the form of eq. 2 to describe the

58

apparent second-order rate constant (kapp) for phenol in the presence of FAC:

59

k ୟ୮୮ = k ୌశ, ୅୰୓ୌ ሾH ା ሿ fୌ୓େ୪ f୅୰୓ୌ + k ୌ୓େ୪, ୅୰୓ୌ fୌ୓େ୪ f୅୰୓ୌ + k ୌ୓େ୪, ୅୰୓ష fୌ୓େ୪ f୅୰୓ష

60

where fୌ୓େ୪ represents the fraction of [HOCl]T in the HOCl form; f୅୰୓ୌ and f୅୰୓ష represent the

61

fractions of total phenol ([phenol]T) in the conjugate acid (ArOH) and phenolate (ArO‒) forms, 4 ACS Paragon Plus Environment

(2)

Page 5 of 27


62

respectively. Neither group of researchers conducted additional experiments that would either

63

support or refute the ability of Cl2 to act as a chlorinating agent for (chloro)phenols.

64

Aside from Cl2, Cl2O (aq) is another overlooked chlorine species in aqueous chlorine that

65

has emerged as an important chlorinating agent for some organic compounds.10, 19-21, 23, 33-34 Cl2O

66

exists in equilibrium with HOCl, and its formation is favored at high [FAC]: 2 HOCl ⇌ Cl2O (aq) + H2O

67 68

log KCl2O = -2.06 (ref. 35, corrected to 25°C according to ref. 33)

69

To our knowledge, the influence of Cl2O on the reaction kinetics of all six (chloro)phenols has

70

not been previously investigated.

71

In this study, we systematically varied the solution pH, [Cl‒], and [FAC] in order to

72

examine the influence of Cl2, Cl2O, and HOCl on the reaction kinetics of six (chloro)phenols.

73

Second-order rate constants for Cl2, Cl2O, and HOCl were computed by comparing the

74

experimental rate coefficients to the predicted speciations of different chlorinating agents. An

75

enhancement in chlorination rates in the presence of chloride would implicate Cl2 as a

76

chlorinating agent. One mole of chloride is consumed for each mole of Cl2 formed (eq 1);

77

moreover, one mole of Cl– is regenerated for every mole of Cl2 that reacts with (chloro)phenol

78

(eq. 4). Thus, to the extent that Cl2 represents a potent chlorinating agent, chloride ion could

79

prove of underappreciated significance as a chlorination catalyst. ArOH + Cl2 → Cl‒ArOH + H+ + Cl–

80

81

(3)

(4)

Materials and Methods

82

All aqueous solutions were made using reverse osmosis water that had been distilled in a

83

Barnstead Mega-Pure system (Thermo Scientific) and further purified in a Milli-Q system (EMD

84

Millipore). Commercial sodium hypochlorite (NaOCl) solutions (laboratory grade, 5.65‒6%) 5 ACS Paragon Plus Environment


85

from Fisher Scientific served as the source of FAC and were standardized iodometrically every

86

week according to Standard Methods 4500-Cl B.36 To minimize chlorine demand, all glassware

87

was soaked in a concentrated FAC solution (≥ 0.5 M) for at least 8 hours and thoroughly rinsed

88

with Milli-Q water before use. Complete descriptions of all other reagents are in the Supporting

89

Information (SI).

90

Kinetic Experiments. All experiments were carried out in batch reactors (40-mL amber glass

91

vials with PTFE-lined caps). Once all the reagents had been added, the reactors were kept inside

92

a stainless constant-temperature water bath for the entire duration of the experiments. Each

93

reactor contained 30 mL of reaction solution; the presence of headspace in the reactor did not

94

affect reaction kinetics, implying that the chlorine and (chloro)phenol species do not partition

95

appreciably into the headspace. The ionic strength (0.1 M) was set by adding sodium nitrate

96

(NaNO3). Solution pH was controlled by adding 10 mM of acetate (pH 3‒6), phosphate (pH 6‒

97

8), or carbonate (pH > 8) buffer. No pH buffer was used in experiments at pH < 3. The pH was

98

adjusted using small volumes of HNO3 or NaOH. Measurements revealed that the pH changed

99

by less than 0.05 pH unit over the course of each experiment.

100

Reaction kinetics in the presence of FAC were determined for each (chloro)phenol

101

separately. Stock solutions of (chloro)phenols were made by dissolving the neat compounds in

102

methanol, and spiking solutions of (chloro)phenols were made by diluting the methanolic stocks

103

with water. The initial concentration of each (chloro)phenol was 2 µM in most experiments, and

104

the final methanol content in the reactor was less than 0.05% (v/v). Working solutions of FAC

105

were prepared fresh daily by diluting the commercial NaOCl stock solution with water. Kinetic

106

experiments were carried out under pseudo-first-order conditions in which [FAC] ≈ [FAC]o ≈

107

constant. Values of [FAC]o in the majority of experiments were as follows: 125 µM for phenol, 6 ACS Paragon Plus Environment

Page 6 of 27

Page 7 of 27


108

2-CP, 2,4-DCP, and 2,6-DCP; 160 µM for 4-CP; 185 µM for TCP. To verify that [FAC]

109

decreased by 9, kobs decreases with increasing pH. At pH < 9 and no

156

added chloride, kobs decreases with decreasing pH until pH 4.7, at which point kobs increases with

157

decreasing pH. Adding 1 and 5 mM Cl‒ while maintaining constant ionic strength led to a

158

pronounced increase in kobs at pH < 6. The enhancement in phenol reaction rates with increasing

159

[Cl‒]added at low pH can be attributed to reactions with Cl2―rather than H2OCl+―since [Cl2]

160

depends on both [Cl‒] and [H+] (eq. 1). Chloride addition had no appreciable effect on reaction

161

rates at pH > 6, indicating that the influence of Cl2 on reaction rates at neutral to high pH is

162

negligible.

163

The log kobs vs. pH data for 4-CP, 2,4-DCBP, and TCP are also shown in Fig. 2; those for

164

2-CP and 2,6-DCP can be found in the SI (Fig. S4). The variations in kobs with pH for 2-CP and

165

4-CP are similar to those for phenol, but those for 2,4-DCP, 2,6-DCP, and TCP are markedly

166

different. For the dichlorophenols, kobs reaches a maximum at pH 7–8 and plateaus at low pH.

167

The log kobs vs. pH data for TCP show two distinct regions; reaction rates decrease with

168

increasing pH at pH > 6.5, while revealing little dependence on pH at pH < 6.5. Despite the

169

differences in the shapes of the log kobs vs. pH data, adding 1‒5 mM Cl‒ leads to a significant

170

increase in kobs for all (chloro)phenols at pH < 6 to 7.

171

The emphasis on “added” Cl‒ is necessary because our ion chromatographic

172

measurements confirm the presence of Cl‒ in reactors to which no NaCl was added (SI Fig. S5).

173

For example, we found that a reactor with [FAC]o = 0.125 mM contained 0.17 mM Cl‒, with the

174

Cl‒ coming primarily from the commercial NaOCl stock solutions. The concentration of chloride

175

contributed by our FAC solutions was taken into account in modeling the experimental data. 9 ACS Paragon Plus Environment


176

Effects of Varying [FAC] and [(Chloro)phenol]. For each (chloro)phenol, kinetic experiments

177

with varying [FAC]o were conducted at selected pH values in order to assess the reaction order

178

(n) in [HOCl]. The plots of log kobs vs. log [HOCl]o and their corresponding slopes (i.e., n) for

179

phenol are shown in SI Fig. 6a. In the absence of added Cl‒, n reaches a maximum value (1.71 ±

180

0.05) at pH 4.7 and approaches 1 as the pH increases. When Cl‒ is added at pH 4.7 while

181

maintaining constant ionic strength, the value of n decreases with increasing [Cl‒]added. The log

182

kobs vs. log [HOCl]o data for the other chlorophenols (SI Fig. S6) show trends that are similar to

183

the one for phenol; the value of n in the absence of added Cl‒ is close to 2 at low pH and

184

approaches 1 as the pH increases.

185

The second-order dependence of kobs on [HOCl] could be interpreted as evidence that

186

either Cl2O or Cl2 is the predominant chlorinating agent at low pH. Since [Cl2O] is proportional

187

to [HOCl]2 (eq. 3), a reaction that depends only on Cl2O would have an n of 2. On the other

188

hand, as the concentration of chloride impurity in our FAC solutions was close to that of HOCl at

189

pH < 6.5, the near second-order dependence is also consistent with Cl2 as a reactive species. The

190

dependence of kobs on [Cl2] could masquerade as a second-order dependence on [HOCl] when

191

[Cl‒] ≈ [HOCl] because [Cl2] = KCl2 [HOCl][Cl‒][H+]. Indeed, the apparent second-order

192

dependence on [HOCl] in the chlorination of fluoranthene and naphthalene has been attributed to

193

reactions with Cl2.38 An influence of Cl2 could, therefore, cause n to be greater than 1 at low pH

194

if [HOCl] and [Cl–] are approximately equal (see SI for details). The importance of Cl2

195

diminishes with increasing pH, leading n to approach 1 at higher pH. When NaCl is added to the

196

reactor, the reaction solution is no longer equimolar in [Cl‒] and [HOCl], so the reaction

197

becomes first-order in [HOCl].


Page 10 of 27

Page 11 of 27

198


The reactions under investigation are first-order in [(chloro)phenol], as evidenced by the

199

good adherence to exponential decay of the parent compound under pseudo-first order conditions

200

in which [FAC] ≈ [FAC]o ≫ [(chloro)phenol]o (sample data for phenol are shown in SI Figs. S2

201

and S3). We also used the method of initial rates37 for selected chlorophenols to determine the

202

reaction order in [(chloro)phenol]o. Plots of log (initial rate) vs. log [(chloro)phenol]o for 4-CP

203

and 2,4-DCP result in straight lines with slopes close to 1 (SI Fig. S7), thus confirming that the

204

reactions are first-order in [(chloro)phenol]o.

205

Effects of Other Reactor Constituents. Varying the ionic strength (i.e., [NaNO3]) and

206

concentrations of pH buffers generally had no appreciable effect on chlorination kinetics (data

207

not shown). The only exception is the catalysis of TCP reactions by acetate buffer at pH > 2.8 (SI

208

Fig. S8). We extrapolated the values of kobs to [acetate buffer] = 0 at each pH value. The

209

extrapolated kobs values are the ones shown in Fig. 2d and the ones used in data modeling.

210

Control experiments in which the methanol concentration in the reactor was varied showed that

211

methanol has no appreciable effect on chlorination kinetics at concentrations < 0.1% v/v (data

212

not shown).

213

Data Modeling. Second-order rate constants for the reactions of (chloro)phenols with Cl2, Cl2O,

214

and HOCl were computed by nonlinear least-squares regressions using SigmaPlot 12.5, and the

215

second-order rate constants were estimated from iterative data fittings. Terms that did not

216

improve the model fit were not included in the final model (see SI for details). For phenol, 2-CP,

217

and 4-CP, the final model is

218

k ୭ୠୱ = k ୌ୓େ୪, ୅୰୓ష ሾHOClሿ f୅୰୓ష + k େ୪మ୓, ୅୰୓ୌ ሾClଶ Oሿ f୅୰୓ୌ + k େ୪మ , ୅୰୓ୌ ሾClଶሿ f୅୰୓ୌ + k େ୪మ , ୅୰୓ష ሾClଶ ሿ f୅୰୓ష (6)



Page 12 of 27

219

where f୅୰୓ୌ and f୅୰୓ష represent the fractions of [(chloro)phenol]T in the conjugate acid (ArOH)

220

and phenolate (ArO‒) forms, respectively. As Figure 2a shows, eq. 6 gives a good fit to the log

221

kobs vs. pH data for phenol at [Cl‒]added = 0 and 5 mM, and it is able to predict the log kobs vs. pH

222

data at [Cl‒]added = 1 mM. The model is also able to fit the 4-CP and 2-CP data well at all values

223

of [Cl‒]added (Fig. 2b and SI Fig. S4a, respectively).

224

Despite the similarity in the shapes of the log kobs vs. pH data for 2,4-DCP (Fig. 2c) and

225

2,6-DCP (SI Fig. 4b), the final models that gave the best fits for these compounds are different.

226

The model for 2,4-DCP has three terms (eq. 7), whereas the one for 2,6-DCP has four terms (eq.

227

8). Eq. 8 can also be used to fit the kobs vs. pH data for TCP (Fig. 2d).

228

k ୭ୠୱ = k ୌ୓େ୪, ୅୰୓ష ሾHOClሿ f୅୰୓ష + k େ୪మ ୓, ୅୰୓ୌ ሾClଶ Oሿ f୅୰୓ୌ + k େ୪మ , ୅୰୓ష ሾClଶ ሿ f୅୰୓ష

(7)

229

k ୭ୠୱ = k ୌ୓େ୪, ୅୰୓ష ሾHOClሿ f୅୰୓ష + k େ୪మ୓, ୅୰୓ୌ ሾClଶ Oሿ f୅୰୓ୌ + k େ୪మ୓, ୅୰୓ష ሾClଶ Oሿ f୅୰୓ష + k େ୪మ, ୅୰୓ష ሾClଶ ሿ f୅୰୓ష

(8)

230

The best-fit estimates of the second-order rate constants are listed in Table 1. For each

231

(chloro)phenol, the second-order rate constants for Cl2 and Cl2O are at least as large as, or in

232

many cases orders of magnitude larger than, the rate constants for HOCl. The differences among

233

the rate constants for Cl2 and Cl2O and those for HOCl become more pronounced as the

234

(chloro)phenol becomes more highly chlorinated and, thus, less reactive towards electrophilic

235

aromatic substitution. In general, the second-order rate constants for reactions with the ArO‒

236

form of (chloro)phenols are larger than those for reactions with the ArOH form, which is poorly

237

reactive towards HOCl.

238

To test the validity of the second-order rate constants in Table 1, we used the rate

239

constants to predict the concentration of each (chloro)phenol as a function of time at selected pH

240

values. The model predictions are in close agreement with the measured concentrations for all

241

the (chloro)phenols investigated (SI Fig. S2).


Page 13 of 27

242


To delineate the influence of Cl2 and Cl2O at low pH and in the absence of added Cl‒, we

243

compared the reaction order (n) in [HOCl] computed using the second-order rate constants in

244

Table 1 with the values of n determined from the log kobs vs. log [HOCl] data. An expression for

245

the computed n (ncalc) has been derived with the assumption that [HOCl] = [Cl‒] at pH < 6.5 for

246

the experiments conducted in the absence of added chloride (see SI for details). The close

247

agreement between ncalc and the experimentally determined n (SI Tables S1, S2) is consistent

248

with Cl2 representing the predominant chlorinating agent for (chloro)phenols at the acidic pH

249

values tested.

250

Comparisons with Previous Results. This is the first investigation on the roles of Cl2, Cl2O,

251

and HOCl in the chlorination of all six (chloro)phenols. Our experimental results highlight the

252

role of Cl‒ as a catalyst for the reactions of (chloro)phenols at low pH and reveal the importance

253

of Cl2 as a chlorinating agent. Although Lee and Morris32 previously speculated that Cl2 could

254

influence the reaction kinetics of (chloro)phenols, they did not conduct experiments specifically

255

designed to reveal the importance of Cl2. Grimley and Gordon39 demonstrated that the rate of

256

phenol chlorination at [H+] = 0.023 M (pH < 2) increased with varying [Cl‒] (2.1‒100 mM), but

257

the data set is limited to a single pH value. Our experiments conducted at different values of

258

[Cl‒]added and pH allowed us to compute robust second-order rate constants for Cl2.

259

In their study on (chloro)phenol reactions, Gallard and von Gunten6 considered reactions

260

with both HOCl and H2OCl+ when developing kinetic models for phenol and 4-CP, whereas only

261

the reactions with HOCl were considered for the other (chloro)phenols. The best-fit estimates of

262

kHOCl, ArO- in this study are generally similar to those reported by Gallard and von Gunten. The

263

exception is TCP; the previously reported kHOCl, ArO- (12.84 ± 0.69 M-1 s-1) is more than three

264

times larger than our estimate (3.39 ± 0.47 M-1 s-1). Gallard and von Gunten might have 13 ACS Paragon Plus Environment


265

overestimated kHOCl, ArO- by attributing reactivities of TCP towards Cl2 and Cl2O to reactions with

266

HOCl. Although Gallard and von Gunten reported values of kHOCl, ArOH for phenol and 4-CP, we

267

found that it is not necessary to include a term for the HOCl/ArOH reaction in order for our

268

model (eq. 6) to fit our experimental data.

269

When we considered the presence of trace Cl‒ in the no-NaCl-added reactors and

270

included the kCl2, ArOH and kCl2, ArO- terms in our model for phenol, 2-CP, and 4-CP (eq. 6), we

271

were able to fit the no-added-chloride data at low pH without invoking H2OCl+ as a reactive

272

species. If H2OCl+ were the sole chlorinating agent for (chloro)phenols at low pH, varying

273

[Cl‒]added would not affect the value of kobs because [H2OCl+] depends only on [HOCl] and [H+].

274

We recognize that the presence of Cl2 does not necessarily preclude the involvement of H2OCl+

275

in (chloro)phenol reactions, since the speciations of Cl2 and H2OCl+ parallel each other at low

276

pH (Fig. 1). In order to fully elucidate the importance of H2OCl+, we would need to conduct

277

experiments at [Cl‒] ≪ [FAC]o to minimize the influence of Cl2. As so little chloride is required

278

for appreciable quantities of Cl2 to be generated, however, the relevance of H2OCl+ for

279

(chloro)phenols is dubious. Therefore, while we cannot rule out H2OCl+ as a reactive chlorine

280

species based on our experimental and modeling results, we have shown that it is not necessary

281

to invoke H2OCl+ in order to explain the chlorination kinetics of (chloro)phenols.

282

In their study on the halogenation kinetics of 2,4-DCP, Vikesland et al.34 considered the

283

reactions with Cl2, Cl2O, and HOCl when modeling 2,4-DCP reactivity at pH 5‒10. These

284

researchers computed the following second-order rate constants from the log kobs vs. pH data

285

collected at a single set of [Cl‒] and [FAC]o: kCl2, ArOH = 30 (± 17) M-1 s-1, kHOCl, ArOH =

286

23 (± 13) M-1 s-1, and kHOCl, ArO- = 660 (± 130) M-1 s-1. Vikesland et al. were not able to obtain an

287

estimate of kCl2O, but our modeling results suggest that the influence of Cl2O on 2,4-DCP


Page 14 of 27

Page 15 of 27


288

reactivity should be insignificant at the low [FAC]o (10 µM) used in their experiments. In

289

contrast, our findings reveal that Cl2 reacts primarily with the ArO‒ form―rather than with the

290

ArOH form―of 2,4-DCP. The kHOCl, ArO- computed by Vikesland et al. is about twice as large as

291

our estimate (309 ± 19 M-1 s-1), suggesting that the authors might have overestimated kHOCl, ArO-

292

because they did not consider the reaction of Cl2 with ArO‒. The discrepancies between the

293

findings of Vikesland et al.34 and this study underscore the need to systematically vary solution

294

pH, [FAC]o, and [Cl‒] in order to obtain robust estimates of the second-order rate constants for

295

HOCl, Cl2O, and Cl2.

296

Linear Free Energy Relationships. Hammett-type correlations were constructed for kHOCl, ArO-,

297

kCl2, ArO-, and kCl2O, ArOH using the substituent constants in ref. 40 (Table 1). As shown in Figure

298

3a, the Hammett plots for both the HOCl/ArO‒ and Cl2/ArO‒ reactions have negative slopes (ρ),

299

as expected if the rate-determining step involved electrophilic aromatic substitution.

300

Furthermore, the value of ρ for the HOCl/ArO‒ reaction (-2.09 ± 1.00) is not significantly

301

different from that computed by Gallard and von Gunten6 (-3.00 ± 0.22). Although our ρ value

302

has a relatively large uncertainty, we would expect it to be similar to that determined by Gallard

303

and von Gunten since our estimates of kHOCl, ArO- are generally quite similar.

304

The Hammett plot for the Cl2O/ArOH reaction is distinct from those for the HOCl/ArO‒

305

and Cl2/ArO‒ reactions. As shown in Figure 3b, the Cl2O/ArOH Hammett plot has a small

306

positive ρ value (0.829 ± 0.568), suggesting that the rate-determining step may not be

307

electrophilic aromatic substitution. Small ρ values are often associated with free-radical

308

reactions, as in the case of hydroxyl radicals (•OH),41 and some researchers have proposed that

309

Cl2O reacts with organic compounds in solution via free-radical pathways, with Cl• and ClO•

310

from the photolysis of Cl2O implicated in those reactions.42 Our use of amber glass vials as 15 ACS Paragon Plus Environment


311

reactors, however, would reduce the likelihood of photocatalyzed reactions. Furthermore,

312

analysis of our FAC solutions using electron paramagnetic resonance (EPR) spectroscopy did not

313

provide compelling evidence for the presence of radical species. Even though an EPR signal was

314

observed (SI Fig. S9a) upon adding the spin trap agent 5,5-dimethyl-1-pyrroline-N-oxide

315

(DMPO) to a FAC solution at pH < 3, simulation of the experimental EPR spectrum using

316

WinSim 2002 (SI Fig. S9b) yielded hyperfine splitting constants that are consistent with the one-

317

electron oxidation product of DMPO.43 Therefore, the observed EPR signal was due to an

318

experimental artifact (possibly the oxidation of DMPO by FAC) rather than to the genuine

319

trapping of a radical species.

320

Environmental Implications. Using the second-order rate constants in Table 1, we calculated

321

the pseudo-first-order rate coefficients (kcalc) that would result if we assumed [FAC]o values that

322

are closer to the chlorine doses typically used in drinking water and wastewater treatments. We

323

then calculated the contributions of different reactions to the overall reactivities of the

324

(chloro)phenols as fractions of kcalc . The results for phenol and TCP are shown in SI Figures S10

325

and S11, respectively. For phenol, the Cl2/ArOH reaction predominates at low pH and the

326

HOCl/ArO‒ reaction predominates at circum-neutral and high pH under typical drinking water

327

treatment conditions (SI Fig. S10a), in the presence of excess Cl‒ (SI Fig. 10b), and under typical

328

wastewater treatment conditions (SI Fig. S10c). The importance of the Cl2/ArO‒ reaction

329

increases in the presence of high [Cl‒], a situation that is likely to occur when chlorinating

330

desalinated water or water with FeCl3 added as a coagulant. The Cl2O/ArOH reaction remains

331

unimportant for phenol in all the assumed scenarios. The situation for TCP is different; reactions

332

with Cl2 and Cl2O represent significant fractions of kcalc at circum-neutral pH in all situations (SI

333

Fig. S11). 16 ACS Paragon Plus Environment

Page 16 of 27

Page 17 of 27

334


In accordance with the reactivity-selectivity principle,44 we hypothesized that the less

335

reactive (i.e., the more highly chlorinated) (chloro)phenols would be more selective towards Cl2

336

and Cl2O. Our results show that, at a fixed pH, the contributions of Cl2 and Cl2O towards kcalc

337

relative to the contribution of HOCl increase as the overall reactivity of the (chloro)phenol

338

decreases (SI Fig. S12a). At a higher [Cl‒], the reactions with Cl2 contribute substantially more

339

towards kcalc than they do at a lower [Cl‒], and the importance of Cl2 as a chlorinating agent is

340

the greatest for the most highly chlorinated/least reactive compound (SI Fig. S12b). The

341

contribution of Cl2O, on the other hand, is trivial for most of the (chloro)phenols. The

342

insignificance of Cl2O for (chloro)phenols is in contrast to the findings from previous studies on

343

the chlorination kinetics of dimethenamid,33 aromatic ethers,23 and antipyrine.19

344

One consequence of the tradeoff between reactivity and selectivity is that Cl2 and Cl2O

345

are expected to be more important for the slow-reacting NOM fraction than for a fast-reacting

346

one. The kinetics of DBP formation from NOM can be divided into two phases. The “fast”

347

reactions occur within seconds to minutes; the “slow” reactions occur on the order of hours to

348

days, much longer than the time scale of Cl2 and Cl2O regeneration in aqueous chlorine solutions

349

(see discussion of Cl2 regeneration in SI). One study has demonstrated that the fast-reacting

350

trihalomethane (THM) precursors account for only 15‒30% of all THM precursors in natural

351

waters.27 As most precursors of THMs and perhaps other DBPs belong to the slow-reacting

352

NOM fraction, Cl2 (and possibly Cl2O) may have more influence on the kinetics of DBP

353

formation than is presently recognized.

354

The roles of Cl2 and Cl2O as chlorinating agents are often overlooked because of the low

355

concentrations of these chlorine species under typical water treatment conditions. In most

356

studies, second-order rate constants were computed by dividing the experimental pseudo-first-



357

order rate coefficients by [HOCl]T or [FAC]o , thus assuming that HOCl is the only chlorinating

358

agent present and that the reactions are first-order in [FAC]. Nonetheless, laboratory experiments

359

often employ reaction conditions that favor Cl2 and Cl2O formation, so researchers may need to

360

take these chlorinating agents into account when interpreting their experimental results. In

361

addition, activities such as desalination, water recycling, and hydraulic fracturing can increase

362

[Cl‒] in raw water supplies. Chloride can also accumulate in swimming pools.45 Therefore, Cl2

363

could become an important chlorinating agent when water of impaired quality or water for

364

recreation is chlorinated.

365

Acknowledgements

366

We thank the late Dr. Justine Roth (Johns Hopkins University) for helpful discussions on EPR

367

analysis and for the use of quartz flat cells. We thank Dr. Jung Yoon Lee for help with running

368

the EPR spectrometer at JHU. We are also grateful to Dr. Neil Blough (University of Maryland,

369

College Park) for his help on interpreting EPR spectra. This material is based upon work

370

supported by the U.S. National Science Foundation (CBET-1067391 and the Graduate Research

371

Fellowship to S.S.L. under Grant No. 123825).

372

Associated Content

373

Supporting Information.

374

Experimental details, descriptions of kinetic modeling, and other experimental results can be

375

found in the Supporting Information (SI). This material is available free of charge via the

376

Internet at http://pubs.acs.org.


Page 18 of 27

Page 19 of 27


Literature Cited 1. Richardson, S. D.; Plewa, M. J.; Wagner, E. D.; Schoeny, R.; DeMarini, D. M. Occurrence, genotoxicity, and carcinogenicity of regulated and emerging disinfection by-products in drinking water: A review and roadmap for research. Mutat. Res. - Rev. Mut. Res. 2007, 636, 178-242. 2. Deborde, M.; von Gunten, U. Reactions of chlorine with inorganic and organic compounds during water treatment ‒ Kinetics and mechanisms: A critical review. Water Res. 2008, 42, 1351. 3. Morris, J. C. Acid ionization constant of HOCl from 5 to 35 degrees. J. Phys. Chem. 1966, 70, 3798-3805. 4. Swain, C. G.; Crist, D. R. Mechanisms of chlorination by hypochlorous acid. The last of chlorinium ion, Cl+. J. Am. Chem. Soc. 1972, 94, 3195-3200. 5. Rebenne, L. M.; Gonzalez, A. C.; Olson, T. M. Aqueous chlorination kinetics and mechanism of substituted dihydroxybenzenes. Environ. Sci. Technol. 1996, 30, 2235-2242. 6. Gallard, H.; Von Gunten, U. Chlorination of phenols: Kinetics and formation of chloroform. Environ. Sci. Technol. 2002, 36, 884-890. 7. Deborde, M.; Rabouan, S.; Gallard, H.; Legube, B. Aqueous chlorination kinetics of some endocrine disruptors. Environ. Sci. Technol. 2004, 38, 5577-5583. 8. Gallard, H.; Leclercq, A.; Croué, J.-P. Chlorination of bisphenol A: kinetics and by-products formation. Chemosphere 2004, 56, 465-473. 9. Pinkston, K. E.; Sedlak, D. L. Transformation of aromatic ether- and amine-containing pharmaceuticals during chlorine disinfection. Environ. Sci. Technol. 2004, 38, 4019-4025. 10. Chusaksri, S.; Sutthivaiyakit, S.; Sedlak, D. L.; Sutthivaiyakit, P. Reactions of phenylurea compounds with aqueous chlorine: Implications for herbicide transformation during drinking water disinfection. J. Hazard. Mater. 2012, 209–210, 484-491. 11. Qin, L.; Lin, Y.-L.; Xu, B.; Hu, C.-Y.; Tian, F.-X.; Zhang, T.-Y.; Zhu, W.-Q.; Huang, H.; Gao, N.-Y. Kinetic models and pathways of ronidazole degradation by chlorination, UV irradiation and UV/chlorine processes. Water Res. 2014, 65, 271-281. 12. Lane, R. F.; Adams, C. D.; Randtke, S. J.; Carter Jr, R. E. Chlorination and chloramination of bisphenol A, bisphenol F, and bisphenol A diglycidyl ether in drinking water. Water Res. 2015, 79, 68-78. 13. Arotsky, J.; Symons, M. C. R. Halogen cations. Q. Rev. Chem. Soc. 1962, 16, 282-297.



14. Morris, J. C. The chemistry of aqueous chlorine in relation to water chlorination. In Water Chlorination, Environmental Impact and Health Effects, Jolley, R. L., Ed. Ann Arbor Science Publishers: Ann Arbor, Michigan, 1978; Vol. 1, pp 27-41. 15. Wang, T. X.; Margerum, D. W. Kinetics of reversible chlorine hydrolysis: Temperature dependence and general-acid/base-assisted mechanisms. Inorg. Chem. 1994, 33, 1050-1055. 16. Cherney, D. P.; Duirk, S. E.; Tarr, J. C.; Collette, T. W. Monitoring the speciation of aqueous free chlorine from pH 1 to 12 with Raman spectroscopy to determine the identity of the potent low-pH oxidant. Appl. Spectrosc. 2006, 60, 764-772. 17. Dodd, M. C.; Huang, C. H. Aqueous chlorination of the antibacterial agent trimethoprim: Reaction kinetics and pathways. Water Res. 2007, 41, 647-655. 18. Wang, P.; He, Y.-L.; Huang, C.-H. Reactions of tetracycline antibiotics with chlorine dioxide and free chlorine. Water Res. 2011, 45, 1838-1846. 19. Cai, M.-Q.; Feng, L.; Jiang, J.; Qi, F.; Zhang, L.-Q. Reaction kinetics and transformation of antipyrine chlorination with free chlorine. Water Res. 2013, 47, 2830-2842. 20. Soufan, M.; Deborde, M.; Delmont, A.; Legube, B. Aqueous chlorination of carbamazepine: Kinetic study and transformation product identification. Water Res. 2013, 47, 5076-5087. 21. Cheng, H.; Song, D.; Chang, Y.; Liu, H.; Qu, J. Chlorination of tramadol: Reaction kinetics, mechanism and genotoxicity evaluation. Chemosphere 2015, 141, 282-289. 22. Sivey, J. D.; McCullough, C. E.; Roberts, A. L. Chlorine monoxide (Cl2O) and molecular chlorine (Cl2) as active chlorinating agents in reaction of dmethenamid with aqueous free chlorine. Environ. Sci. Technol. 2010, 44, 3357-3362. 23. Sivey, J. D.; Roberts, A. L. Assessing the reactivity of free chlorine constituents Cl2, Cl2O, and HOCl toward aromatic ethers. Environ. Sci. Technol. 2012, 46, 2141-2147. 24. Abdallah, P.; Deborde, M.; Dossier Berne, F.; Karpel Vel Leitner, N. Kinetics of chlorination of benzophenone-3 in the presence of bromide and ammonia. Environ. Sci. Technol. 2015, 49, 14359-14367. 25. Tawk, A.; Deborde, M.; Labanowski, J.; Gallard, H. Chlorination of the β-triketone herbicides tembotrione and sulcotrione: Kinetic and mechanistic study, transformation products identification and toxicity. Water Res. 2015, 76, 132-142. 26. Gao, Y.; Pang, S.-Y.; Jiang, J.; Ma, J.; Zhou, Y.; Li, J.; Wang, L.-H.; Lu, X.-T.; Yuan, L.-P. Transformation of flame retardant tetrabromobisphenol A by aqueous chlorine and the effect of humic acid. Environ. Sci. Technol. 2016, 50, 9608-9618.


Page 20 of 27

Page 21 of 27


27. Gallard, H.; von Gunten, U. Chlorination of natural organic matter: kinetics of chlorination and of THM formation. Water Res. 2002, 36, 65-74. 28. Czaplicka, M. Sources and transformations of chlorophenols in the natural environment. Sci. Total Environ. 2004, 322, 21-39. 29. Burttschell, R. H.; Rosen, A. A.; Middleton, F. M.; Ettinger, M. B. Chlorine derivatives of phenol causing taste and odor. J. Am. Water Works Assoc. 1959, 51, 205-214. 30. Arnold, W. A.; Bolotin, J.; Von Gunten, U.; Hofstetter, T. B. Evaluation of functional groups responsible for chloroform formation during water chlorination using compound specific isotope analysis. Environ. Sci. Technol. 2008, 42, 7778-7785. 31. Guo, G.; Chen, X. Halogenating reaction activity of aromatic organic compounds during disinfection of drinking water. J. Hazard. Mater. 2009, 163, 1207-1212. 32. Lee, G. F.; Morris, J. C. Kinetics of chlorination of phenol ‒ Chlorophenolic tastes and odors. Int. J. Air Water Pollut. 1962, 6, 419-431. 33. Sivey, J. D.; McCullough, C. E.; Roberts, A. L. Chlorine monoxide (Cl2O) and molecular chlorine (Cl2) as active chlorinating agents in reaction of dimethenamid with aqueous free chlorine. Environ. Sci. Technol. 2010, 44, 3357-3362. 34. Vikesland, P. J.; Fiss, E. M.; Wigginton, K. R.; McNeill, K.; Arnold, W. A. Halogenation of bisphenol-A, triclosan, and phenols in chlorinated waters containing iodide. Environ. Sci. Technol. 2013, 47, 6764-6772. 35. Roth, W. A. Note on thermo-chemistry of chlorine monoxide. Zeitschrift Fur Physikalische Chemie-Abteilung a-Chemische Thermodynamik Kinetik Elektrochemie Eigenschaftslehre 1942, 191, 248-250. 36. Greenberg, A. E.; Clesceri, L. S.; Eaton, A. D. Standard Methods for the Examination of Water and Wastewater. 18 ed.; American Public Health Association, American Water Works Association, Water Environment Federation: Washington, DC, 1992. 37. Zumdahl, S. S. Chemical Principles. 6 ed.; Houghton Mifflin: 2009; pp 723-726. 38. Georgi, A.; Reichl, A.; Trommler, U.; Kopinke, F.-D. Influence of sorption to dissolved humic substances on transformation reactions of hydrophobic organic compounds in water. I. Chlorination of PAHs. Environ. Sci. Technol. 2007, 41, 7003-7009. 39. Grimley, E.; Gordon, G. Kinetics and mechanism of reaction between chlorine and phenol in acidic aqueous solution. J. Phys. Chem. A 1973, 77, 973-978. 40. Barlin, G. B.; Perrin, D. D. Prediction of the strengths of organic acids. Q. Rev. Chem. Soc. 1966, 20, 75-101. 21 ACS Paragon Plus Environment


41. Gligorovski, S.; Strekowski, R.; Barbati, S.; Vione, D. Environmental implications of hydroxyl radicals (•OH). Chem. Rev. 2015, 115, 13051-13092. 42. Renard, J. J.; Bolker, H. I. The chemistry of chlorine monoxide (dichlorine monoxide). Chem. Rev. 1976, 76, 487-508. 43. Evans, J. C.; Jackson, S. K.; Rowlands, C. C.; Barratt, M. D. An electron spin resonance study of radicals from chloramine-T ― 1 : Spin trapping of radicals produced in acid media. Tetrahedron 1985, 41, 5191-5194. 44. Anslyn, E. V.; Dougherty, D. A. Modern Physical Organic Chemistry. University Science: 2006; pp 377-378. 45. E, Y.; Bai, H.; Lian, L.; Li, J.; Blatchley III, E. R. Effect of chloride on the formation of volatile disinfection byproducts in chlorinated swimming pools. Water Res. 2016, 105, 413-420.


Page 22 of 27

Page 23 of 27


Table 1. Second-order rate constants for the reactions of HOCl, Cl2, and Cl2O with the conjugate acid (ArOH) and phenolate (ArO‒) forms of (chloro)phenols. The uncertainties indicate 95% confidence intervals.

Σσo,p

a

pKa

Second-order rate constants (M-1 s-1)

b

‫ ܓ‬۶‫۽‬۱‫ܔ‬, ‫۽ܚۯ‬۶

‫ ܓ‬۶‫۽‬۱‫ܔ‬, ‫۽ܚۯ‬ష

‫ ܓ‬۱‫ܔ‬૛ , ‫۽ܚۯ‬۶

‫ ܓ‬۱‫ܔ‬૛ , ‫۽ܚۯ‬ష

‫ ܓ‬۱‫ܔ‬૛ ‫۽‬, ‫۽ܚۯ‬۶

‫ ܓ‬۱‫ܔ‬૛ ‫۽‬, ‫۽ܚۯ‬ష

0

9.99

‒‒

2.61 (± 0.26) × 104

8.92 (± 0.98) × 104

2.61 (± 0.50) × 109

9.0 (± 3.1) × 104

‒‒

2-CP

0.68

8.56

‒‒

3.17 (± 0.33) × 103

1.76 (± 0.13) × 103

1.12 (± 0.06) × 109

5.9 (± 1.3) × 105

‒‒

4-CP

0.23

9.43

‒‒

3.58 (± 0.23) × 103

2.08 (± 0.17) × 103

1.71 (± 0.13) × 109

9.8 (± 0.6) × 104

‒‒

2,4-DCP

0.91

7.85

‒‒

3.09 (± 0.19) × 102

‒‒

2.69 (± 0.13) × 108

9.9 (± 2.5) × 105

‒‒

2,6-DCP

1.36

6.97

‒‒

1.28 (± 0.09) × 102

‒‒

1.58 (± 0.16) × 108

2.3 (± 0.3) × 106

8.6 (± 2.1) × 107

TCP

1.59

6.15

‒‒

3.39 (± 0.47)

‒‒

9.35 (± 0.67) × 106

9.2 (± 3.3) × 105

3.3 (± 1.0) × 106

Phenol

a b

Calculated using the σp and σo values for chlorine listed in ref. 40 pKa values taken from ref. 6



Figure 1. Speciation of free available chlorine (FAC) at 25°C under typical drinking water chlorination conditions: [FAC]o = 28 µM (2.0 mg/L as Cl2) and [Cl‒] = 0.30 mM (~10 mg/L). The pKa of H2OCl+ is estimated to be between -3 and -4;13 a pKa of -3.5 was assumed in calculating [H2OCl+] for this figure. See SI for details.


Page 24 of 27

Page 25 of 27


Figure 2. Pseudo-first-order rate coefficients (kobs) as function of pH for (A) phenol, (B) 4-CP, (C) 2,4-DCP, and (D) TCP. Reaction conditions: [(chloro)phenol]o = 2 µM; [NaCl]o = 1 or 5 mM (if added); [pH buffer] = 10 mM; ionic strength = 0.1 M; T = 25°C. [FAC]o = 125 µM for phenol and 2,4-DCP; [FAC]o = 160 µM for 4-CP; [FAC]o = 185 µM for TCP. Error bars indicate 95% confidence intervals (smaller than symbols if not shown). Solid lines are fits to a model of the form of eq. 6 (phenol and 4-CP), eq. 7 (2,4-DCP), or eq. 8 (TCP). The dashed lines in (A) and (B) represent the model prediction of kobs based on eq. 6.



Figure 3. Hammett-type linear free energy relationships for (chloro)phenol reactions in FAC correlating (A) kCl2, ArO- and kHOCl, ArO- as well as (B) kCl2, ArOH. ArOH and ArO‒ denote the conjugate acid and phenolate forms, respectively. Error bars (smaller than symbols if not shown) and uncertainties in the equations indicate 95% confidence intervals.


Page 26 of 27

Page 27 of 27


Table of Content (TOC) Art


Chlorination Revisited: Does Clâ Serve as a Catalyst in the

Recommend Documents

Chlorination Revisited: Does Clâ Serve as a Catalyst in the