Impact of combining chlorine dioxide and chlorine on ...

JES-01150; No of Pages 8 J O U RN A L OF E N V I RO N ME N TA L S CI EN CE S X X (2 0 1 7 ) XX X–XXX

Available online at www.sciencedirect.com

ScienceDirect www.elsevier.com/locate/jes

Q4 Q33Q2 4 5 6

F

Daekyun Kim1 , Nuray Ates2 , S. Sule Kaplan Bekaroglu3 , Meric Selbes1 , Tanju Karanfil1,⁎

O

2

Impact of combining chlorine dioxide and chlorine on DBP formation in simulated indoor swimming pools

1. Department of Environmental Engineering and Earth Sciences, Clemson University, Anderson, SC 29625, USA 2. Department of Environmental Engineering, Erciyes University, Kayseri 38039, Turkey 3. Water Institute, Suleyman Demirel University, Isparta 32260, Turkey

R O

1Q1

7 9

AR TIC LE I NFO

ABSTR ACT

11

Article history:

The main objective of this study was to assess the combined use of chlorine dioxide (ClO2) 16

12

Received 6 February 2017

and chlorine (Cl2) on the speciation and kinetics of disinfection by-product (DBP) formation 17

13

Revised 16 April 2017

in swimming pools using synthetic pool waters prepared with a body fluid analog (BFA) 18

14

Accepted 20 April 2017

and/or fresh natural water. At 1:25 (mass ratio) of ClO2 to Cl2, there was no significant 19

15

Available online xxxx

reduction in the formation of trihalomethanes (THMs) and haloacetic acids (HAAs) for both 20

36

Keywords:

ratio of ClO2 to Cl2 increased to 1:1, substantial decreases in both THMs and HAAs were 22

37

Chlorine dioxide

observed in the natural water, while there was almost no change of DBP formations in the 23

38

Chlorine

BFA solution. Haloacetonitriles and halonitromethane levels in both water matrices 24

39

Disinfection by-products

remained similar. In the presence of bromide, the overall DBP formation increased in both 25

40

Body fluid analog

BFA solution and natural water. For the DBP formation kinetics, after 72 hr of contact time, 26

41

Natural organic matter

very low formation of THMs and HAAs was observed for the use of ClO2 only. Compared to 27

E

D

P

10

E

C

T

BFA solution and natural water compared to the application of Cl2 alone. When the mass 21

42

R

Cl2 control, however, applying the 1:1 mixture of ClO2/Cl2 reduced THMs by >60% and HAAs 28 by > 50%. Chlorite was maintained below 1.0 mg/L, while the formation of chlorate 29

R

significantly increased over the reaction time. Finally, in a bench-scale indoor pool 30 experiment, applying ClO2 and Cl2 simultaneously produced less THMs compared to Cl2 31

44 45 43 46

N C O

control and kept chlorite at < 0.4 mg/L, while HAAs and chlorate accumulated over 4-week 32 © 2017 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. 34

Introduction

49

Chlorine (Cl2 or HOCl) is the most commonly used oxidant and/or disinfectant in swimming pools (Kanan et al., 2015). In the United States (US), Cl2 is continuously applied to pool waters to satisfy the oxidant and/or disinfectant demand. Typically, swimming pools maintain a free available Cl2 (FAC) concentration between 1 and 5 mg/L to control pathogens. During each swimming event, swimmers release organic matter (e.g., saliva, sweat, mucus, urine, and skin particles)

51 52 53 54 55 56

U

48 47

50

33

operation period.

Published by Elsevier B.V. 35

and synthetic chemicals such as sunscreen, cosmetics, and soap residues that increase the total organic carbon (TOC) concentrations in pool waters. Cl2 reacts with such organic matter and produces undesirable reaction by-products known as disinfection by-products (DBPs). A survey of 23 public indoor swimming pools in the US showed that TOC ranged from 3 to 23.6 mg/L with a median of 7.1 mg/L (Kanan and Karanfil, 2011). Such high levels of TOC accumulated in swimming pools requires high Cl2 demands which may result in elevated formation of chlorinated DBPs. Actually the DBP

⁎ Corresponding author. E-mail: [email protected] (Tanju Karanfil).

http://dx.doi.org/10.1016/j.jes.2017.04.020 1001-0742/© 2017 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V.

Please cite this article as: Kim, D., et al., Impact of combining chlorine dioxide and chlorine on DBP formation in simulated indoor swimming pools, J. Environ. Sci. (2017), http://dx.doi.org/10.1016/j.jes.2017.04.020

57 58 59 60 61 62 63 64 65 66

2

81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126

F

1. Materials and methods

127 128 129 130 131 132 133 134 135 136

138 137

O

80

139

1.1. Preparation of BFA and synthetic pool water

R O

79

BFA was prepared using distilled and deionized water (DDW) following the recipe shown in Table 1 (Goeres et al., 2004; Kanan and Karanfil, 2011). Urea, creatinine, uric acid, free or combined amino acids are the most abundant compounds in human body fluids that are assumed to have an impact on the Cl2 demand and DBP formation in swimming pools (Hureiki et al., 1994; Li and Blatchley, 2007). Albumin protein, a common model protein in nature, was used in preparing BFA to simulate the human body proteins, peptides, and free amino acids that are likely to occur in swimming pool water. A synthetic swimming pool water was prepared by adding calcium chloride and sodium bicarbonate for adjusting hardness (120 mg/L) and alkalinity (200 mg/L), respectively, in DDW. Using sodium bisulfate and sodium carbonate, the pH of synthetic swimming pool water was maintained about 7.5. Finally, predetermined amounts of BFA were added to achieve different target TOC concentrations (e.g., 1 or 3 mg/L). To examine the effect of high total dissolved solid (TDS) on DBP formation in bench-scale swimming pool model experiments, sodium chloride was added to achieve 1200 mg/L as Cl− of TDS.

140

1.2. Preparation of ClO2 and Cl2 stock solutions

160

Sodium hypochlorite (NaOCl, 5% available Cl2) was used as free Cl2 source. The Cl2 stock solutions were always standardized according to N,N-diethyl-phenylenediamine (DPD) titrimetric method. ClO2 was generated by slow acidification of sodium chlorite solution with sulfuric acid (Jones et al., 2012).

161

P

78

D

77

E

76

water disinfected with combined ClO2 and Cl2. To simulate swimmers' body excretions, the body fluid analog (BFA) proposed by Goeres et al. (2004) was used, and the formation and speciation of DBPs including THMs, HAAs, haloacetonitriles (HANs), and HNMs as well as chlorite and chlorate were investigated. Understanding the DBP formation kinetics for the use of the mixture of ClO2 and Cl2 is important for developing strategies to control the formation of currently regulated DBPs and/or emerging DBPs to reduce the exposure of swimmers to DBPs.

T

75

C

74

E

73

R

72

R

71

O

70

C

69

concentrations in the investigated pools were far higher than the drinking water regulation values in the US or swimming pool regulations in other countries. Trihalomethanes (THMs) ranged between 26 and 213 μg/L with an average of 80 μg/L, and haloacetic acids (HAAs) ranged between 173 and 9005 μg/L with an average of 1541 μg/L (Kanan, 2010). Such DBPs formed in swimming pools can be accidentally ingested, inhaled or absorbed through skin. More risks of DBP exposure from inhalation and dermal pathways have been shown during swimming, showering and bathing than ingestion drinking water (Caro and Gallego, 2007; Villanueva et al., 2007; Kanan, 2010). Chlorine dioxide (ClO2) is a strong oxidant rather than a chlorinating reagent. ClO2 oxidation is through one-electron exchange mechanism, and it attacks the electron-rich centers of organic molecules. As one electron is transferred, ClO2 is reduced to chlorite (ClO−2). The US Environmental Protection Agency (EPA) set the maximum residual disinfectant level (MRDL) of ClO2 at 0.8 mg/L, and the maximum contaminant concentration (MCL) of ClO−2 at 1.0 mg/L in drinking water (USEPA, 2006). During oxidation reactions with organic matter, approximately 50–70% of ClO2 is typically converted to ClO−2 and the remainder is converted to chlorate (ClO−3) and chloride (Cl−) (USEPA, 1999). As a result, ClO2 concentrations greater than about 1.4 mg/L are not common during drinking water treatment practices, not to violate the MCL for ClO−2. ClO2 may also form other by-products in addition to ClO−2 and ClO−3. ClO2 produces much less THMs and HAAs than Cl2 during drinking water treatment (Aieta and Berg, 1986; Hua and Reckhow, 2007; Gates et al., 2009). In a study on Suwannee River fulvic acid treated with ClO2 and Cl2, Zhang et al. (2000) reported that the formation of total organic halide (TOX) was much lower from ClO2 than Cl2. Although it is well established that ClO2 can reduce THMs and trihalogenated HAA formations compared to Cl2, it may lead to some formations of dihalogenated HAAs and organic halides (Aieta and Berg, 1986; Werdehoff and Singer, 1987; Zhang et al., 2000). Generally, halonitromethanes (HNMs) have not been detected in ClO2 treated waters (Zhang et al., 2000). On the other hand, Hua and Reckhow (2007) observed ClO2 producing higher percentage of unidentified TOX than Cl2. The mixture of ClO2 and Cl2 has been investigated with some surface waters and source waters of drinking water utilities to see its impacts on the formation of DBPs. With different mixing ratios, a series of bench- and full-scale testing showed that applying combined ClO2 and Cl2 reduced THM formation in most source waters. As the mass ratios of ClO2 to Cl2 increased from 0:2 to 8:2, the reduction in THM formation reached up to 77% in a lake water (Rav-Acha et al., 1985). Li et al. (1996) have also shown that a combination of ClO2 and Cl2 reduced the formation of THMs in the presence of bromide, and that THM formation potentials (FPs) decreased with an increase in the ratio of ClO2 to Cl2. In a full-scale water treatment plant research, mixed ClO2 with Cl2 at 1:1 (mass ratio) followed by chlorination reduced THM formation by about one half compared to Cl2 only (Rittmann et al., 2009). Today, little is known for the impact of combining ClO2 and Cl2 on the formation of DBPs in swimming pools. The objective of this study was to examine the DBP formation kinetics and DBP speciation in a synthetic pool

N

68

U

67

J O U RN A L OF E N V I RO N ME N TA L S CIE N CE S X X (2 0 1 7 ) XXX –XXX

Table 1 – Body fluid analog (BFA) components.

141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159

162 163 164 165

t1:1

Ingredients

mg/L

Urea Creatinine Uric acid Lactic acid (85%) Albumin Glucuronic acid Ammonium chloride Sodium chloride Sodium sulfate Sodium bicarbonate Potassium phosphate Potassium sulfate

62.6 4.3 1.5 3.3 9.7 1.2 7.0 22.1 35.3 6.7 11.4 10.1


t1:3 t1:2

t1:4 t1:5 t1:6 t1:7 t1:8 t1:9 t1:10 t1:11 t1:12 t1:13 t1:14 t1:15 t1:16

3

J O U RN A L OF E N V I RO N ME N TA L S CI EN CE S X X (2 0 1 7 ) XX X–XXX

204

To simulate indoor swimming pools, a bench scale swimming pool model was used. Many public swimming pools are rectangular shaped with dimensions of 25–50 m length, 10– 50 m width, and 1–5.5 m depth. In this study, a 50 cm (L) × 25 cm (W) × 6 (H) cm aquarium with 7.5 L water capacity was used to scale down a typical swimming pool (Fig. S1). This model pool's top was open and water was gently mixed with magnetic stir bars. Initially, 6 L of the synthetic pool water was prepared in a pool at high TDS (1200 mg/L Cl−). ClO2 and Cl2 were added simultaneously twice a day to maintain their residuals at 1 mg/L and 2.5 mg/L, respectively. BFA (0.1 mg/L TOC) was spiked in the pool every day to simulate the swimmers' activity and reached 3 mg/L TOC by the end of 4-week experiment. Samples were taken twice a week to determine DBP levels. The DBP formation from a control pool operated with Cl2 alone was also examined for the purpose of comparison.

205

1.5. Analytical methods

206

THMs, HANs and HNMs were analyzed using EPA 551.1 with minor modifications. HAAs were measured in accordance with the method SM 6251 B. All DBPs extracted with methyl tert-butyl ether were analyzed using an Agilent 6890 gas chromatography equipped with micro electron capture detector (μ-ECD). DB-1 column (J&W Scientific 30 m × 0.25 mm × 1.00 μm) was used for the analysis of HAAs, HANs and THMs, and DB-5 column (J&W Scientific 30 m × 0.32 mm × 0.25 μm) was used for chromatographic separation of HNMs. The details of analytical methods can be found elsewhere (Hu et al., 2010; Karanfil et al., 2011). The minimum reporting level (MRL) for all THM, HAA, and HAN species was 1.0 μg/L, while it was 0.7 μg/L for HNMs. The concentrations of ClO2 as well as Cl2, ClO−2, and ClO−3 were determined with Amperometric Titration (SM 4500-ClO2-E) (APHA/AWWA/WEF, 2005). In selected experiments, ClO2 HACH

178 179 180 181 182 183 184 185 186

190 191 192 193 194 195 196 197 198 199 200 201 202 203

207 208 209 210 211 212 213 214 215 Q5 216 217 218 219 220

C

177

E

176

R

175

R

174

N C O

173

225 224

2.1. Impact of combined ClO2 and Cl2 on DBP formation in BFA 226 solution and natural water 227 The Cl2 demand for 1 mg/L TOC BFA solution was 10 mg/L for 24 hr of contact time, which was attributed to the demand of nitrogenous and carbonaceous compounds present in the BFA components (Table S2). This demand is consistent with those reported from a previous chlorination study where the same BFA was investigated (Hureiki et al., 1994; Li and Blatchley, 2007; Hong et al., 2008; Kanan and Karanfil, 2011). Since there was no detectable bromide present in DDW, only chlorinated DBPs (i.e., chloroform [TCM] of THMs, dichloroacetic acid [DCAA] and trichloroacetic acid [TCAA] of HAAs) were observed (Table S3). Due to trace levels of bromide present in the ingredients of BFA, dichlorobromomethane (DCBM) was detectable, but its concentration remained below the MRL. In swimming pools, there is also natural organic matter (NOM) that comes from the filling or make-up water and serves as DBP precursors. Since NOM and body fluids are different in their chemical composition and properties, the DBP formation experiments were also conducted with natural fresh water samples collected from Lake Hartwell located in South Carolina for the purpose of comparison. Although the Cl2 demand for the lake water was much lower (2 mg/L) than that (10 mg/L) of BFA solutions, application of 25 mg/L Cl2 resulted in almost twice of THMs and HAAs formed in natural raw water, indicating that NOM contains more reactive THM and HAA precursors than the BFA ingredients (Fig. 1). There was no apparent difference in the formation of THMs and HAAs between the uses of ClO2 + Cl2 (at 1:25 mass ratio) and Cl2 alone, due to very low ClO2/Cl2 ratio. ClO2 and chlorite residuals after 24 hr were 0.4 mg/L and 0.3 mg/L, respectively, for both water matrices. The formation of HAN (~1 μg/L) and HNM (< 2 μg/L) was quantifiable, but significantly lower than THMs and HAAs.

U

172

2. Results and discussion

F

189

171

O

1.4. DBP formation in a bench-scale swimming pool model

170

R O

188

169

P

187

The DBP formation experiments were conducted with the BFA solution and natural fresh water (TOC was adjusted to 1 mg/L for both water matrices) for the simultaneous application of ClO2 and Cl2 at two mass ratios (1:25 and 1:1 of ClO2 to Cl2). The DBP formation was investigated after a 24-hr reaction period at pH 7.5 with and without bromide. For the kinetic experiments, the DBP formation from 3 mg/L TOC BFA solutions at high TDS (1200 mg/L Cl−) was monitored for 72 hr for three disinfection scenarios (120 mg/L Cl2 alone, 120 mg/L ClO2 alone, and 120 mg/L ClO2 + 120 mg/L Cl2). High oxidant concentrations were used to avoid oxidant depletion for 3 mg/L TOC BFA solutions during 72 hr of the reaction time. All experiments were conducted in 250-mL amber glass bottles closed tightly with Teflon-faced septa screw caps. Headspace free samples spiked with predetermined disinfectants were mixed well using a magnetic stirrer and stored in a water bath at 26°C. In all experiments, each bottle was opened and ClO2 and Cl2 residuals, ClO−2, ClO−3, and pH were immediately measured after designated reaction times. Accurate volume of a sample was rapidly transferred to extraction vials to determine four classes of DBPs (i.e., THM, HAA, HAN, and HNM).

168

D

167

test kit (Method 10,126) was also used to confirm some of 221 the measurements. A summary of the analytical methods and 222 instrumentation used in the study is presented in Table S1. 223

E

1.3. DBP formation and kinetic experiments

T

166

Fig. 1 – THM and HAA formation in 1 mg/L TOC BFA solutions (a) and 1 mg/L TOC raw lake water (b) after 24 hr contact time at 26°C and pH 7.5 (25 mg/L Cl2 and 1 mg/L ClO2 were applied). THM: trihalomethane; HAA: haloacetic acid; TOC: total organic carbon; BFA: body fluid analog.


228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258

4

273 274 275

F

O

272

R O

271

P

270

D

269

E

268

shows DBP formation from BFA solutions (1 mg/L TOC) for 24 hr contact time with Cl2 alone and 1:1 mixture of ClO2/Cl2. The residual ClO−2 was below the MCL (Table S4). For the formation of THMs, HAAs, HANs, and HNMs in BFA solutions, there was no significant reduction in DBP formation when the 1:1 mixture of ClO2/Cl2 was used compared to Cl2 alone. In the presence of bromide, more brominated species formed, which increased the total THM and HAA formation, while the changes in HAN and HNM formation were negligible (Table S5). However, apparent impacts of combined ClO2 and Cl2 on reducing DBP formation from BFA were not observed. In contrast, the DBP formation from 1 mg/L TOC raw natural water was reduced by applying the 1:1 mixture of ClO2/Cl2 (Fig. 2b). In the absence of bromide, THM and HAA formation decreased about 44% and 20%, respectively, compared to Cl2 alone, while ClO−2 levels were maintained below the MCL (Tables S6 and S7). Observed THM levels after 24 hr of reaction time decreased from 29 μg/L to 18 μg/L when ClO2/Cl2 mass ratios increased from 1:25 to 1:1 (Figs. 1 and 2b), indicating that THM precursors present in NOM are deactivated effectively by combining ClO2 with Cl2. Increasing ClO2/Cl2 ratios also reduced HAA formation from 41 μg/L (at 1:25) to 24 μg/L (at 1:1). The reduction rates of THM and HAA formation will vary depending on the properties of organic matter which comes from both body fluids and filling waters in swimming pools. In the presence of bromide, the formation of THM and HAA increased due to brominated species. Decreases in THM and HAA formation from raw natural water by ClO2/Cl2 were 33% and 23%, respectively, compared to Cl2 alone. However, due to their relatively low concentrations, deactivation of HAN and HNM precursors by ClO2/Cl2 was not apparent whether bromide was added or not. Since there is neither regulatory requirement for DBPs nor systematic treatment for periodic dilution by adding make-up water in the US indoor pools, relatively high concentration of organic matter may develop and react with disinfectants enhancing DBP formation. Although the impact of combining ClO2 and Cl2 on reducing DBP formation from 1 mg/L TOC BFA was insignificant, THM and HAA formation from NOM present in raw natural water decreased considerably. Therefore, these results suggest that operating swimming pools at low Cl2 residual concentrations with simultaneously adding ClO2 may maintain the desired microbial water quality while decreasing THM and HAA formation, which will be primarily due to less THM and HAA formation from the filling water (Aieta and Berg, 1986; Hua and Reckhow, 2007; Gates et al., 2009).

T

267

C

266

E

265

R

264

R

263

O

262

C

261

In previous studies on DBP formation in drinking water treatment plants, the use of ClO2/Cl2 at mass ratios of 1:1 or 1:0.66 reduced THM formation while ClO−2 concentrations were maintained below the regulatory level (Rittmann et al., 2009). To further examine the impact of ClO2/Cl2 mixture for reducing DBP formation in swimming pools, experiments were conducted with higher ClO2/Cl2 ratio: 1 mg/L of ClO2 and 1 mg/L of Cl2 were spiked in 1 mg/L TOC BFA solutions. Since the Cl2 demand for the BFA solution (1 mg/L TOC) was 10 mg/L, residual concentrations of ClO2 and Cl2 were measured every hour during the initial 12 hr, and ClO2 and Cl2 were added in the reaction bottles, as needed, to maintain the same ratio of 1 mg/L ClO2 and 1 mg/L Cl2. At time = 12 hr, ClO2 and Cl2 were raised to 1.4 mg/L (still 1:1 ratio) and the reaction underwent for another 12 hr. For the purpose of comparison, the same experiments were conducted with Cl2 alone. All experiments were carried out at 26°C and pH 7.5 in the absence and presence of bromide (100 μg/L). Fig. 2a

N

260

U

259


Fig. 2 – THM, HAA, HAN, and HNM formation from 1 mg/L TOC BFA solutions (a) and 1 mg/L TOC natural raw water (b) after 24 hr contact time at 26°C and pH 7.5 in the absence and presence of bromide, 100 μg/L. Residual Cl2 and ClO2 were monitored every hour and maintained at 1:1 by adding oxidants, as needed, until time = 12 hr, and then Cl2 and ClO2 concentrations were raised to 1.4 mg/L to complete 24 hr of reaction. THM: trihalomethane; HAA: haloacetic acid; HAN: haloacetonitrile; HNM: halonitromethane; TOC: total organic carbon; BFA: body fluid analog.

276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321

2.2. DBP formation kinetics in the synthetic swimming pool at 322 1:1 ClO2/Cl2 323 To further assess the efficacy of using a mixture of ClO2 and Cl2, the DBP formation kinetics and speciation were investigated in the synthetic pool water spiked with BFA (3 mg/L TOC) under relatively high TDS (1200 mg/L Cl−) conditions. Since the Cl2 demand for 3 mg/L TOC BFA was 50 mg/L, excess amounts (i.e., 120 mg/L) of ClO2 and Cl2 (at 1:1 mass ratio) were applied for the DBP formation kinetic studies to avoid oxidant depletion during 72 hr of the reaction time. The residual ClO2 concentrations were 0.2–0.5 mg/L at time = 72 hr (Table 2). To examine the bromide effect, kinetic experiments were performed at


324 325 326 327 328 329 330 331 332 333

5

J O U RN A L OF E N V I RO N ME N TA L S CI EN CE S X X (2 0 1 7 ) XX X–XXX

Table 2 – Changes in residual ClO2, ClO−2, ClO−3, and Cl2 during 72 hr of the reaction time for three different disinfection scenarios (ClO2 alone, Cl2 alone, and ClO2 + Cl2) in the syntheric pool water spiked with 3 mg/L TOC BFA at ambient bromide and 200 μg/L bromide added.

(mg/L)

t2:7 t2:8 t2:9 t2:10 t2:11 t2:12 t2:13 t2:14 t2:15 t2:16 t2:17 t2:18 t2:19 t2:20 t2:21 t2:22 t2:23 t2:24 t2:25 t2:26

Ambient Br−

t2:27 t2:28

TOC: total organic carbon; BFA: body fluid analog.

339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356

2ClO2 þ HOCl þ H2 O → 2ClO−3 þ Cl− þ 3Hþ 359 358 357 360 361 362 363 364

120.0 102.5 99.5 95.0 85.0 75.0 70.7 70.0 70.0 70.0 120.0 99.5 95.0 87.5 80.0 76.0 70.7 70.0 70.0 70.0

T

C

E

338

ambient and 200 μg/L bromide concentrations. However, since the high TDS condition led to a relatively high background bromide (up to 100 μg/L due to max. 0.01% bromide impurity in sodium chloride) in the synthetic pool water, brominated DBPs were observed even for ambient bromide condition (Table S8). The kinetic experiments with ClO2 alone and Cl2 alone were also performed as controls to compare with the use of ClO2/Cl2 mixture. When ClO2 alone was used as a disinfectant, 80% or more of ClO2 was consumed within 48 hr, but no or very low ClO−2 was observed at the end of the experiment (i.e., at time = 72 hr) regardless of added bromide concentrations. Up to 20 mg/L of ClO−2 appeared and disappeared between time = 8 hr and time = 48 hr, indicating that ClO2 decomposes to ClO−2 and ClO−3 (Table 2). ClO−2 can react with HOCl to form ClO2 (Csordás et al., 2001). As for the use of the 1:1 mixture, ClO−2 was not detected during 72 hr of the reaction time. The reduction rate of ClO2 in the mixture was very fast compared to the scenario with ClO2 alone, and high concentrations of ClO−3 which is not currently regulated in the US were produced very quickly within the initial 30 min, indicating that ClO2 reacts with Cl2 to form ClO−3, and this reaction is relatively fast (Eq. (1)) (Csordás et al., 2001; Wang and Margerum, 2002).

R

337

0.0 2.3 2.5 2.6 2.8 7.9 25.5 35.9 76.0 93.9 0.0 1.4 1.5 1.7 1.8 6.5 42.1 62.9 82.3 82.8

R

336

0.0 0.0 0.0 0.0 0.3 0.8 17.7 12.2 3.4 0.0 0.0 1.7 2.1 1.7 2.5 19.4 14.8 10.5 0.3 0.6

N C O

335

120.0 107.5 110.4 108.3 108.8 94.0 52.3 43.8 20.2 10.3 120.0 107.1 106.0 103.7 108.3 69.1 26.3 11.4 4.3 5.2

ClO2 + Cl2

(mg/L) Cl2 (mg/L) ClO2 (mg/L) Cl2 (mg/L) ClO−2 (mg/L) ClO−3 (mg/L)

U

334

200 μg/L Br−

0 0.5 1 2 4 8 16 24 48 72 0 0.5 1 2 4 8 16 24 48 72

Cl2 alone ClO−3

120.0 25.3 18.5 16.0 11.8 8.0 1.7 1.7 0.8 0.5 120.0 20.2 16.0 10.1 7.6 1.7 0.4 0.4 0.4 0.2

120.0 63.8 44.3 37.1 34.6 33.9 32.8 31.0 29.7 28.1 120.0 64.2 51.9 47.3 43.9 41.7 36.1 33.9 28.1 24.7

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

F

ClO2 (mg/L)

ClO−2

O

t2:6

ClO2 alone

R O

Time (hr)

P

t2:5

0.0 124.2 137.1 144.6 153.9 155.4 157.8 168.3 167.7 172.7 0.0 131.1 139.6 142.3 148.3 149.9 151.9 153.0 165.4 168.0

D

t2:4

E

t2:1 t2:2 t2:3

ð1Þ

Therefore, it is noted that less Cl2 may be available as an oxidant or disinfectant when Cl2 is applied with ClO2 simultaneously. Although the formation of ClO−2 was well below the MCL, ClO−3 levels were very high (reached up to 170 mg/L) for the simultaneous use of Cl2 and ClO2. Although these experiments were conducted at very high oxidant levels, these results show

the possibility of continuous accumulation of ClO−3 during the use of ClO2 or ClO2 simultaneously with Cl2. Fig. 3 shows the total THM and HAA formation as a function of time in the synthetic swimming pool water spiked with 3 mg/L TOC BFA. For the use of ClO2 alone, DBP formation was drastically suppressed with only 6–7 μg/L of THMs and 9 μg/L of HAAs observed after 72 hr of contract time regardless of background bromide concentrations. It has been known that Cl2 reacts with precursor compounds by oxidation and electrophilic substitution to yield THMs and HAAs, while ClO2 forms a few or no halogenated DBPs by oxidizing (not chlorinating) organic precursors (Richardson et al., 2000; LeChevallier and Au, 2004; Gagnon et al., 2005). For the use of Cl2 alone, THM and HAA formations reached up to 53–71 μg/L and 121–135 μg/L, respectively, during 72 hr reaction. The presence of additional bromide (200 μg/L) increased the THM formation by 34% and HAA formation by 12%, while the bromide effect was not critical on the other two disinfection scenarios (i.e., ClO2 alone and the use of ClO2/Cl2 mixture). Higher HAA formation than THM from 3 mg/L TOC BFA was consistent with previous chlorination results from the same BFA solution at 1 mg/L TOC (Kanan and Karanfil, 2011). Approximately 50% of the total THMs formed within 10 hr, while 50% of HAAs formed within 4 hr indicating that the formation of HAAs by chlorination is faster than the formation of THMs (Table S8). In contrast, the use of combined ClO2 with Cl2 reduced THM formation by 62–69% and HAA formation by 45–48% compared to the use of Cl2 alone. More DBP reductions were observed when 200 μg/L of bromide was added. The results suggest that THM formation may be preferentially reduced by 1:1 mixture of ClO2/ Cl2. Therefore, using ClO2 and Cl2 together as disinfectants and conventionally treated water as filling or make-up water will potentially reduce the formation of THMs and HAAs in indoor


365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396

6

E

C

T

E

D

P

R O

O

F


400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416

O

C

399

swimming pools. However, since the retention time of water in swimming pools which depends on the frequency of water replacement varies, further practical studies would be required.

2.3. DBP formation in a bench-scale swimming pool model operated with ClO2 and Cl2

N

398

The batch experiments discussed in the previous sections were informative to gain insight into the DBP formation patterns. However, to examine the DBP formation in a swimming pool condition, additional experiments were conducted in a benchscale swimming pool model. To simulate a typical swimming pool, BFA (0.1 mg/L TOC) was spiked in the synthetic pool water (6 L) every day to simulate the swimmer activities, and ClO2 and Cl2 concentrations were measured 2 times a day for 4 weeks and then added to maintain their residuals at 1 mg/L and 2.5 mg/L, respectively (Fig. S2). The pools were not operated during the weekends. Samples were taken twice a week to determine THM, HAA, HAN and HNM levels. For the last sample taken after 4 week of period, ClO−2 and ClO−3 levels were also measured. Fig. 4 shows the total THMs, HAAs, and HANs in a swimming pool model operated with ClO2 and Cl2. HANs were

U

397

R

R

Fig. 3 – Formation of THMs (a) and HAAs (b) as a function of the reaction time in the synthetic pool water spiked with 3 mg/L TOC BFA (initial doses: 120 mg/L of ClO2, 120 mg/L of Cl2, with ambient bromide and 200 μg/L bromide at 26°C). THM: trihalomethane; HAA: haloacetic acid; TOC: total organic carbon; BFA: body fluid analog.

significantly low compared to THMs and HAAs, and HNM levels were below the detection limit. TCM and DCAA were most abundant THM and HAA species, respectively (Table S9). While the total HAA concentrations increased gradually along with reaction time, formation of the total THM was slow down especially after 12 days for both ClO2/Cl2 and Cl2 alone operations. At the end of the experiment, the formation of HAAs (58 μg/L) was much higher than THMs (17 μg/L), which is consistent of the kinetic results. Measured ClO−2 and ClO−3 in the synthetic swimming pool water operated for 4 weeks were 0.4 mg/L and 12.9 mg/L, respectively. The control swimming pool model operated with Cl2 alone (2.5 mg/L) for 4 weeks generated 24 μg/L of THMs and 59 μg/L of HAAs. Low levels of HANs (