The Control of NDMA, HNM, and THM Precursors by Nanofiltration

Mahmut Selim Ersan, David A.Ladner, Tanju Karanfil*

Department of Environmental Engineering and Earth Sciences

Clemson University

342 Computer Court

Anderson, SC 29625

United States

Submitted to Water Research

September October XX, 2015

*Corresponding author: email: tkaranf@clemson.edu; phone: +1-864-656-1005; fax: +1-864-656-0672

1

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

ABSTRACT

Nanofiltration (NF) membranes are susceptible a promising tool forto removinge precursors of

disinfection by products from different source waters prior to chloramination, ozonation, and

chlorination thereby reducing the formation of disinfection by-productsbyproducts. However, it

has also been proven that virgin NF membranes can leach NDMA precursors (Ersan et al., 2015).

Otherwise, tTheir effectiveness of NF in removing those precursors may be altered by

background water characteristics such as water type, pH, ionic strength, and Ca2+ concentration

as well as membrane characteristics. Therefore, in this study, the removal efficienciesy of N-

nitrosodimethylamine (NDMA), halonitromethane (HNM), and trihalomethane (THM)

precursors by nanofiltration (NF) membranes were investigated for different source water types

such as surface water, wastewater impacted surface water, and municipal and industrial

wastewater treatment effluents. Results showed that the overall precursor removal efficiencies

ranged between 57%-83% for NDMA, 72%-97% for THM, and 48%-87% for HNM precursors.

The effects of pH (6-9), ionic strength (0.05-0.005 M) and Ca2+ (6-60 mg/L) were also studied

for municipal and industrial wastewater treatment plant effluents under eight different water

chemistry conditions. The results revealed that, while the Ca2+ concentrations showed an effect

on the removal of NDMA precursors, pH changes altered the removal of both NDMA and THM

precursors from municipal WWTP effluent; however, change in conditions did not alter the

removal of disinfection by-productbyproduct precursors from industrial WWTP effluent. The

change in solution chemistry also had a remarkable effect on membrane flux patterns. Besides,

results from membrane type effect experiments (under varying pH and ionic strength) also

showed that type of the membrane had an impact on the removal of all the DBPs whilest the pH

showed an impact on NDMA precursor removal, and also ionic strength altered the removal of

2

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

THM precursors. Overall, in this study, results suggested that the removal of NDMA, HNM and

THM precursors by NF membrane have been altered by change in pH , ionic strength and Ca2+

concentrations as well as type of NF membrane.

Keywords: Disinfection Byp-Product Precursors, NDMA, THM, HNM, Nanofiltration, pH,

Ionic Strength, Ca2+

1. INTRODUCTION

Disinfection is a necessary n inevitable process in drinking water treatment to eliminate

pathogenic microorganisms from drinking water. However, an unintended consequence is the

formation of disinfection by-productsbyproducts (DBPs) due to the reaction between precursors

and oxidants. Due to various health effects posed by the DBPs, the US Environmental Protection

Agency (USEPA) has been imposing increasingly stringent regulations for DBPs under the

Disinfectants/DBP Rule (D/DBPR) (USEPA, 2013). The Stage 2 D/DBPR requires water

utilities to comply with the maximum contaminant levels of for TTHM4

(bBromodichloromethane [(BDCM]), bBromoform [(CHBr3]), cChloroform [(CHCl3]),

Ddibromochloromethane [(DBCM])) and HAA5 (dDibromoacetic acid [(DBAA]),

dDichloroacetic acid [(DCAA]), mMonobromoacetic acid [(MBAA]), mMonochloroacetic acid

[(MCAA)], tTrichloroacetic acid ([TCAA)]) at 80 µg/L and 60 µg/L, respectively, as locational

running annual average in distribution systems locations. As a result, water utilities have in some

cases been switcswitched tohing alternative disinfectants (e.g., chloramine, chlorine dioxide, and

ozone instead of conventional chlorine) to reduce the formation of THM and HAA, which are the

major DBPs of chlorination. Nevertheless, other unregulated DBPs such as NDMA (N-

nitrosodimethylamine) and HNMs (hHalonitromethanes) have been observed during

chloramination and ozonation-chlorination, respectively (Mitch et al., 2003; Chen and Valentine,

3

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

2006; Nawrocki and Andrzejewski, 2011; Krasner et al., 2006; Hu et al., 2010). NDMA is

reported as a potential human carcinogen and has been detected mainly in distribution systems

during chloramination of finished waters (Russell et al., 2012). Due to health concerns associated

with NDMA, the California Department of Health Services set an NDMA action level of 10 ng/L

(CDPH, 2010). USEPA will be is likely to consider NDMA for regulation during the six year

review of the D/DBPR. It was has been reported that regarding to emerging DBPs also revealed

that HNMs are one of the most cytotoxic and genotoxic classes among the emerging DBPs, and

have orders of magnitude higher toxicity than the regulated THMs and HAAs (Plewa et al.,

2004).

One of the best approaches to control the formation of DBPs during water treatment is to

eliminate their precursors before oxidant addition. Various techniques including enhanced

coagulation, powdered and granular activated carbon adsorption, ion exchange, nanofiltration

and reverse osmosis have been commonly used to reduce the formation of disinfection by-

products subsequent to oxidation with different disinfectants for this purpose (Uyak et al., 2007;

Boyer et al., 2005; Miyashita et al., 2009; Schmidt et al., 2008; Snyder et al., 2006a; Krauss et

al., 2010). Among all techniques used, nanofiltration (NF) is becoming an increasingly popular

treatment technology that can remove DBP precursors effectively while also , meanwhile

removing the solids, particles, and pathogens. In addition, its operation cost is relatively lower

than reverse osmosis units.

Although the removal efficiency of NDMA precursors by NF membranes (e.g. DMA, MEA,

DEA, DPA, DMS, DEET, Erythromycin, ranitidine, DMAE, MDPA, DMHA) has been studied

for selected model precursors (Miyashita et al., 2009; Schmidt et al., 2008; Snyder et al., 2006a;

Krauss et al., 2010), there is no such information about their NF removal from natural source

4

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

waters with different water characteristics. Natural organic matter (NOM) is known as one of the

main precursor of THMs, which is generally associated with higher molecular weight molecules

as opposed to NDMA precursors (Gerecke et al., 2003; Mitch et al., 2004; Krasner et al., 2013),

which are mostly related to small molecular weight compounds (Gerecke et al.,2003; Krasner et

al., 2009; Mitch et al., 2004; Pehlivanoglu-Mantas et al., 2008). There are intensive studies

available in the literature about the removal of NOM from laboratory simulated and natural

waters by nanofiltration (Schafer et al., 1998; Cho et al., 1998; Hong et al., 1997; Braghetta et

al., 1997). These studies have been shown that the change in solution characteristics had an

impact on size and structure of molecules as well as the conformation of the membrane surface;

both effects are important in conformation thus effects their understanding NOM removal from

the water (Ghosh et al., 1980). Even though several studies examined the removal of THM

precursors by NF due to regulatory concern, there is lack of information regarding the removal of

THM precursors and unregulated DBPs (e.g., NDMA, HNMs).

Overall, to the best of our knowledge, there is no such bench scale study have been conducted to

investigate the removal of selected disinfection by product precursors by nanofiltration under

different type of background conditions. This study is also novel as being the first study in the

literature in which provides realistic NDMA precursor removal efficiencies due to NDMA

precursor leaching potential of Nanofiltration membranes (Ersan et. al, 2015).

The objective of this study was to examine the removal of selected unregulated (i.e., NDMA and

HNMs) and regulated THM precursors by NF membrane filtration under different background

water chemistry conditions. The removal of NDMA, HNM, and THM precursors by NF

membrane filtration from two municipal wastewater treatment plant effluents, one industrial

wastewater effluent, one wastewater impacted surface water, and one surface source water have

5

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

been studied. Experiments were also conducted at different pH (6-9), ionic strength (0.005-0.05

M), Ca2+ concentrations (6 - 60 mg/L), and different nanofiltration membranes to assess the

effectiveness of NF filtration for DBP precursor control.

2. MATERIALS AND METHODS

2.1. Filtration Experiments

In this study, three commercially available flat sheet polyamide based thin film composite NF

membranes, TS80, ESNA 1-LF2 and NF270 (TriSep Corp., Hydranautics, and FilmTec Corp.)

were tested (Table 1). In each experiment, membranes were initially cleaned following a

procedure explained elsewhere (Ersan et. al, 2015) to prevent the leaching of NDMA precursors

from membranes. All experiments were carried out at 25±1 0C and at a pressure of 34.5 bar (500

psi) using a cross-flow filtration cell (Figure S1). The effective area of the membrane cell was

140 cm2. Fig. S1 shows a schematic diagram of the filtration module used in this study. Treated

water samples were collected in 500 mL amber glass bottles from the permeate side of the

membranes and tested for desired analysis.

The removal efficiency, R (%) was calculated as

R (% )=(1−C p

C f)∗100 Eq.1

where Cp is the concentration of permeate and Cf is the concentration of feed.

During filtration experiments, flux (J) data were also obtained and normalized according to

Equation 2 to neglect flux changes with changing background water characteristics and coupon

to coupon flux variability, respectively.

Normalized Flux (J ¿¿ Normalized )=¿¿ Eq.2

6

116

117

118

119

120121122123

124

125

126

127

128

129

130

131

132

133

134135136

137

138

139

140

2.2. Water Samples

NDMA, HNM and THM precursor removal experiments were conducted using five different

source waters (Table 2): treated effluents of two municipal wastewater treatment plants (WWTP-

1, WWTP-2), a textile wastewater treatment plant (IWWTP) effluent, a wastewater impacted

surface water (ISW), and a surface water (SW) which serves as a source water for a drinking

water treatment plant. All collected water samples were immediately filtered with pre-washed

0.2 µm cartridge filters (Whatman™ Polycap™ TC) upon arrival at the laboratory to eliminate

the biological activity, and were stored at 4oC in the dark until the experiments were performed.

The effects of pH, ionic strength, and Ca2+concentration on DBP precursor removal were

investigated with the experiments conducted for municipal and industrial WWTP effluents.

Using factorial design, eight different experiments were performed per selected wastewater

effluents for TS80 membrane (Table S1 and S2), and also twelve experiments were conducted

for municipal WWTP effluent using TS80, NF270, and ESNA membranes (Table S3).

According to experimental conditions, feed solution pH was adjusted to the target pH levels by

addition of HCl or NaOH while target total ionic strengths were achieved by addition of NaCI

(Fisher Scientific, Pittsburgh, PA), and Ca2+ were adjusted with CaCl2 addition (Fisher

Scientific, Pittsburgh, PA) .

2.4. NDMA, THM, and HNM formation potential tests

Formation potential (FP) test has been employed in source waters to identify the concentration of

NDMA, THM, and HNM precursors. For NDMA FP tests, a chloramine stock solution (2500

mg/L) was prepared by titrating ammonium sulfate stock solution (5000 mg/L) with sodium

hypochlorite solution (5000 mg/L) at Cl:N mass ratio of 4:1 at pH 7.8. NDMA FP tests were

7

141142143144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161162163

164

165

166

performed by treating permeate samples with excess chloramine (100 mg/L as Cl2) at pH 7.8 in

the presence of 20 mM phosphate buffer during 5 days. HNM FP tests were conducted by

sequential application of ozone and chlorine to the samples. Ozone was generated with a bench

scale ozone generator (Model GTC-1B, Griffin Technics, NJ) fed with ultra-high purity oxygen

gas within 50 min at a concentration of 25-30 mg O3/L. In each time, the concentration of ozone

stock solution was measured with Hach DR\890 colorimeter. Due to being an unstable gas,

ozone used immediately following to generation. Ozone stock solution were introduced to the

water samples at O3: DOC mass ration of 1:1 at pH 7.8 in the presence of 20 mM phosphate

buffer for 5 min. Afterwards, chlorine stock solution were spiked to the ozonated samples with a

dose of 50 mg Cl2/L for 24 h, while for THM FP test samples were treated with excess chlorine

(50 mg/L) at pH 7.8 for five days. Formation potential tests were conducted for each sample with

two measurements. The bars in data plot represent mean values, and the error bars show the data

range.

2.4. Statistical Analysis

In this study, since running replicates of experiments is not feasible due to lack of sources and

time, the experiments were conducted using an unreplicated 23 factorial treatment structure with

the factors: Ca2+ (6 or 60 mg/L), pH (6 or 9) and ionic strength (0.005 M or 0.05 M). Because of

the experiments were unreplicated, there were no degrees of freedom for estimating experimental

error. To obtain a surrogate estimation of the experimental error, the mean squares for two- and

three-factor interactions were combined. This approach provides a conservative procedure in that

p-values computed using this approach can be larger than the true p-value and may reject the null

hypothesis less often than it should be rejected. Therefore, if the null hypothesis is rejected, there

8

167

168

169

170

171

172

173

174

175

176

177

178

179

180181

182

183

184

185

186

187

188

189

is an indication of a significant effect due to the factor. Statistical analyzes is performed using

SAS 9.3 software.

2.5. Analytical methods

There are number of analysis have been conducted during this study. All analytical methods are

described in supporting information section (S1) and summarized in Table S4.

3. RESULTS AND DISCUSSION

3.1 Characteristics of Water Sources

The selected characteristics of source waters used in this study were given in Table 2. DOC,

SUVA254 and conductivity ranged between 2.2 - 26 mg/L, 0.9 - 4.6 L/mg-m, and 8x10-4 – 1x10-2

M, respectively. The two municipal wastewater effluents had distinctly higher DON and

conductivity levels as compared to other waters. pH values of all the waters ranged between 7.2

– 8.0.

The DBP FPs of the samples ranged from 112 to 1078 ng/L for NDMA, 7 to 46 µg/L for HNM,

and 230 to 578 µg/L for THM in the water sources (Table 2). In general, among the all studied

water sources, municipal wastewater effluents had higher NDMA and HNM formation potentials

as compared to non-impacted and impacted surface water sources. This may be attributed to the

abundance of reactive NDMA and HNM precursors in wastewater sources rather than surface

waters. THM FP values for sources varied one source to another and, in most of the cases, it

correlated well with DOC levels (~90%). Industrial wastewater effluent had high DOC, likely

leading to the high THM formation potentials. To evaluate the effect of source water

characteristics on the removal of NDMA, HNM, and THM precursors, two different water

samples were collected from the effluent of a municipal WWTP (WWTP-1) and industrial

WWTP (IWWTP). The selected water sources showed a noticeable difference for all measured

9

190

191

192193194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

parameters (Table 3). As municipal wastewater effluent had almost eight-fold higher NDMA FP

than industrial wastewater effluent, industrial wastewater effluent was containing much higher

HNM and THM precursors.

Lastly, another set of municipal WWTP (WWTP-1) samples were collected from the same

location but in different time and filtered through three different NF membranes (TS80, NF270,

and ESNA) to investigate the effect of membrane type on the removal of NDMA and THM

precursors. The characteristics of municipal WWTP effluent is given in Table 4. Although the

general water characteristics and NDMA FP of municipal WWTP showed more or less

similarities with the previous collected municipal WWTP (WWTP-1), the observed THM level

was two-fold higher.

3.2 Removal of NDMA, HNM and THM Precursors

The removal efficiencies of NF membrane for NDMA, HNM, and THM precursors are

shown in Figure 1. The removal efficiency of NDMA, HNM and THM precursors ranges

between 57-83%, 48-87%, and 72-97 %, respectively. In general, the removal of THM

precursors was higher than those of NDMA and HNM precursors. This was attributed to the

higher molecular weight of THM precursors (i.e., NOM) than NDMA precursors which has been

linked to small molecular weight compounds in anthropogenic origin (Gerecke et al., 2003;

Mitch et al., 2004; Krasner et al., 2013).

Previous studies have shown that the removal of THM precursors by NF membranes ranged

41-98% (Angeles et al., 2008; Lin. et al., 2007; Chellam, 2000); which was comparable with the

values observed in this study. In all studied water sources, the removal of HNM precursors by

10

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

the NF membrane was always lower than THM precursors, and also for WWTP-1, it was lower

than NDMA precursor removal.

Higher NDMA precursor removals were observed for the effluents of two municipal

wastewater treatment plants than surface water and industrial wastewater effluents. This suggests

that wastewater effluents and impacted surface waters may either contain higher molecular

weight / sized precursors or some particle-associated NDMA precursors (Mitch et al., 2004). The

lower NDMA precursor removal, which was observed for industrial wastewater, may be also

associated with small molecular sized NDMA precursors in which available in treated effluent.

In this study, the observed THM precursor removals by NF membrane follow the order of

IWWTP > WWTP-2 > ISW > WWTP-1 > SW.

In essence, their removal efficiencies by NF membranes highly depend on the characteristics

of background solution (such as pH, ionic strength, solute concentration etc.) as well as

characteristics of precursors and membrane type (Ghosh et al., 1980). For the studied water

sources, IWWTP had the lowest SUVA254 value and the highest pH value with a conductivity of

4.8x10-3 M. Therefore, the observed higher removal efficiency (97%) may be experienced by two

mechanism: (i) due to adsorption of hydrophilic constituents onto hydrophilic membrane surface

(electrostatic interaction), (ii) at high pH value NOM molecules uncoils and rejected by the NF

membrane due to their size (size exclusion). The THM precursors in WWTP-2 were also

removed with a high percentage (94%). The observed high removal for WWTP-2 may be

associated with either the size of the THM precursors or the transphilic characteristics of the

studied water. Even though WWTP-2 had the highest conductivity in studied waters, which may

causes less flux, the availability of transphilic constituents in the water reduced the developing of

fast flux decline. NF membrane also showed high removal efficiency for ISW (92%). The high

11

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

THM precursor removal in ISW was also obtained due to repelling of hydrophobic NOM

molecules from membrane surface. Because conductivity in ISW was very low, the observed

flux decline that occurs during the experiment was insignificant due to limited interaction

between hydrophilic membrane surface and hydrophobic NOM molecules. Depending on the

degree of the treatment, molecular size distribution of NOM may vary greatly from very small to

big size molecules (Angeles et al., 2008). The removal of THM precursors from non-impacted

surface water exhibited limited removal (71%) by NF membrane. Non-impacted source water

had low DOC (2.2 mg/L) and pH was around 7.7. Since NOM molecules are subjected to

conformational changes because of change in solution pH, at pH 7.7, NOM molecules are

expected to be more expanded form due to increased charge repulsion thus removed by size

exclusion. However, specific to this case some of the NOM molecules were not removed by NF

membrane, due to presence of small molecular sized THM precursors in SW source. Therefore,

the removal of NOM molecules by NF are controlled by their molecular sizes as well as their

charge densities.

HNM precursor removals from different sources varied according to following sequence

WWTP-2> IWWTP> ISW> SW> WWTP-1. The difference in removal efficiencies under the

same experimental conditions proves that the size and types of HNM precursor varies. In

addition, it is noticeable that even in some sources HNM removal efficiencies are much lower

than NDMA precursor removal efficiencies, in other sources HNM removals were as high as

THM precursor removals. This finding reveals that the size of HNM precursors also varies

sources to sources.

3.3 Effects of pH on the removal of NDMA, HNM, and THM Precursors

12

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

Figure 2 displays the effect of pH on the removal of DBP precursors under eight different

conditions for municipal and industrial WWTP effluent. NF membrane showed different removal

patterns for NDMA, THM, and HNM precursors. While the observed removal efficiencies for

municipal WWTP effluent ranged between 69 -79% for NDMA, 78-93% for THM, and 40-64 %

for HNM precursors , the removal of DBP precursors from industrial WWTP effluent ranged

from 52-66 % for NDMA, 94-98 % for THM, and 92-96 % for HNM precursors. Investigation

of these eight different conditions for each WWTP effluents exhibited that increasing the pH

from 6 to 9 increased NDMA precursor removal, but decreased THM precursor removal. There

was no significant change observed in HNM precursor removal for municipal WWTP. In

addition, there was no significant affect for treatment of DBP precursors in industrial WWTP.

Statistical analysis of the results also revealed the significance of pH changes on NDMA

precursor removal (p-value ≤ 0.0036) in municipal WWTP effluent, whereas the pH effect is

found to be insignificant (p-value ≤ 0.0973) for industrial WWTP effluent. This difference may

be attributed to the size/charge diversity of the NDMA precursors in different sources. In

general, NDMA precursors were known as amine based compounds and thus under pH 6

condition they will be present as protonated forms, which increases their interaction with

negatively charged membrane surface (Mitch et al., 2008). Therefore, when the NDMA

precursors closely interact with the membrane surface, due to their low molecular weights

(Gerecke et al., 2003; Krasner et al., 2009; Mitch et al., 2004; Pehlivanoglu-Mantas et al., 2008),

they may diffuse through the NF membrane. Whereas, at higher pH (i.e., 9), amine based

compounds become slightly deprotonated or neutral which may increase their electrostatic

repulsion from the negatively charged membrane surface, and therefore would result in slightly

higher removal efficiencies. However, the removal of NDMA precursors from NF membranes

13

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

were still limited either due to their size and/or lack of interactions with membrane surface. In

addition, the collected flux data in this study suggested that an increase in pH from 6 to 9

decreased the membrane flux for WWTP effluent sources (Fig. S1-S2), which is also consistent

with the findings of Kilduff et al., (2004). The flux decline, with the change in pH, may have

been caused by conformational changes in membrane properties and/or physicochemical

properties of NDMA precursors, thus expected to alter the removal of NDMA precursors.

However, in the present study, pH change did not make a significant effect for the filtration of

industrial WWTP but some effect has been observed for municipal WWTP effluent.

In contrast to NDMA precursor removals, THM precursors were removed effectively (>80%) in

all eight different conditions for various studied sources. The examination of eight different

background solutions for THM precursor removal in municipal and industrial WWTP effluents

showed that THM precursor removals from municipal WWTP have been slightly affected

(~10%) by the change of pH from 6 to 9 (Fig. 2-A), while no change was observed for industrial

WWTP effluent (Fig. 2-B). Single replicate analysis results also revealed that, whilst the effect

of pH change is significant for municipal WWTP filtration (p-value ≤0.0045) the effect was

insignificant for industrial WWTP effluent (p-value ≤ 0.6326). This was attributed to the

difference of THM precursors in industrial WWTP from municipal effluent in terms of their size

and physicochemical characteristics. Since at low pH values most of the time THM precursors

and membrane surface carries weakly negative charge (Hong et al., 1997), the observed higher

removal efficiency in municipal WWTP effluent, at pH 6, may be caused by compaction of the

membrane. Whereas the observed lower removal efficiency, at pH 9, may be explained by

increased interaction of THM precursors with the membrane surface by virtue of lower flux (Fig.

2-A). Under hard water and high ionic strength conditions (Fig. 2-A), at pH 9, the removal of

14

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

THM precursors were enhanced ~5% due to presence of Ca2+ and Na+ ions which increases the

compaction of the membrane by neutralizing the surface charge and thus a higher removal

efficiency was observed by NF. Although the removal patterns of HNM precursors were

different for municipal and industrial WWTP effluents (~55% vs 95%), the change in pH from 6

to 9 was not significant (p-value ≤ 0.4293 vs 0.6523). This suggests that, depending on the

source water characteristics, the molecular size of HNM precursors could be as low as NDMA

precursors or as high as THM precursors.

3.4 Effect of ionic strength on the removal of NDMA, HNM, and THM Precursors

The effect of ionic strength on the removal of NDMA, THM, and HNM precursors from

treated municipal and industrial WWTP effluent sources are also shown in Fig. 2 (A and B),

respectively. In both water types and the all studied conditions, the flux patterns have been

altered by change in ionic strength from 0.005M to 0.05M (Fig. S1 & S2). According to the

collected flux data, the membrane matrix was in more expanded state at low ionic strength and

less expanded state at high ionic strength conditions with respect to intramembrane electrostatic

repulsion forces which was in agreement with previous findings (Braghetta et. al., 1997; Hong et.

al., 1997). Although there were changes in membrane matrix and precursors charge density, the

removal of NDMA, THM, and HNM precursors were not affected by increase in ionic strength

in both examined waters. Statistical analysis of the results also has shown that the effect of ionic

strength on the removal of DBPs is insignificant (0.05≤p-value).

3.5 Effects of Ca2+ on the removal of NDMA, HNM, and THM Precursors

15

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

The effect of Ca2+ on the removal of NDMA, HNM, and THM precursors was studied in treated

municipal and industrial effluent under eight different background conditions and results were

given in Fig. 2 (A and B). It is plausible that the presence of Ca 2+ ions in source waters may

interfere with the removal of DBP precursors by either changing the precursor stability in the

water or altering the flux of the membrane filters (Kabsch-Korbutowicz et al., 1999). Under all

studied conditions for treated industrial WWTP effluent, results revealed that increasing the Ca2+

concetration of source water did not show any significant change (0.05 ≤ p-values) on the

removal of all DBP precursors. However, in treated municipal WWTP effluent, while the

removal of THM and HNM precursor were not significantly affected by change in background

Ca2+ concentration (0.05 ≤ p-values), the removal of NDMA precursors by NF membrane have

been increased due to increase in background Ca2+ concentrations. Statistical analysis of

examined conditions for municipal WWTP effluent suggested that increase in Ca2+ concentration

had a significant effect on the removal of NDMA precursors (p-value ≤ 0.0013). For the

examined water conditions, when the Ca2+ content of source waters were increased, a severe flux

decline were observed due to formation and accumulation of Ca2+-NOM complexes onto NF

membrane surface (Fig. S1 and S2). Therefore, the abundance of positively charged divalent ions

on the membrane surface may act as a barrier for the small sized NDMA precursors thus favors

their rejection by NF membrane.

3.6 Effects of membrane type on the removal of NDMA and THM Precursors

The effect of membrane type on NDMA and THM precursor removals have been

investigated using three different commercially available NF membrane under varying pH (6-9)

and ionic strength conditions (0.005M - 0.05M) (Table S3).

16

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

The results showed that while the removal efficiency for NDMA precursors ranged

within 78-82%, 51-58%, and 57-62%, for TS80, NF270 and ESNA, respectively, the removal of

THM precursor ranged between 83-87%, 72-80%, and 87-91% for TS80, NF270 and ESNA,

respectively (Fig. 3).

The removal of DBPs from municipal WWTP effluent have shown to be varied one

membrane to another due to difference in membrane physicochemical characteristics (Table 1).

Of the membranes studied, the removal of NDMA precursors decreased with the

increasing average molecular weight cut off and negative surface charge, which was attributed to

low molecular weight nature of NDMA precursors, while the removal of THM precursors (i.e.,

NOM) was not effected from the membrane characteristics. Statistical analysis of all the results

also revealed that membrane type is an important factor in which significantly effects the

removal of NDMA and THM precursor (p-value≤0.0001 and p-value ≤ 0.0001, respectively). As

far as water chemistry effects, while the pH has a significant effect on the removal of NDMA

precursors (p-value ≤ 0.006) ,the effect of pH and ionic strength on THM precursor removal was

insignificant (p-value ≤ 0.0829 and p-value ≤ 0.2929, respectively).

4. Conclusions

The removal efficiency of NDMA, HNM, and THM precursors by NF membrane are

source water specific due to varying types of precursors in different sources.

The molecular size/weights of HNM precursors can be as low as NDMA precursors or as

high as THM precursors depending on the source waters characteristics.

17

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

The observed stable DBP precursor removal efficiencies under different water conditions

suggested the tolerance of the nanofiltration membranes on varying background

conditions.

While the background water characteristics changed the flux patterns significantly,

change in flux did not lead to a notable effect on the removal of DBP precursors.

The statistical analysis of the results suggested that change in pH, ionic strength, and Ca2+

concentration alters the removal of NDMA, HNM, and THM precursors, however this

change would not be considered as significant for water utilities.

The removal efficiencies of DBP precursors were influenced by membrane type along

with the characteristics of the background water.

18

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

Table 1. Membrane Characteristics

Designation ESNA 1-LF2 TS80 NF270

Manufacturer Hydranautics Trisep Dow Filmtech

Compositiona Polyamide Polyamide Polyamide

MWCO (Dalton)a 200 150 200-400

NaCl Rejection (%)a 80 80-90 99

pH range (25oC)a 2-10 2-11 2-11

Typical Flux (L.m2.hr-1.psi-1)a 21 / 75 34 / 110 72-98 / 130

Surface Charge (at pH:7)b Negative Slightly Negative Highly Negative

a Specified by the manufacturer. b Information obtained from the literature.

19

414

415

416

417

418

419

420

421

422

423

424

425

426

20

427

Table 2. Water Characteristics of Source Waters

Parameter SW ISW WWTP-1 WWTP-2 IWWTP

UV254nm 0.06 0.30 0.13 0.11 0.24

DOC (mg/L) 2.2 6.5 5.3 4.5 26

SUVA254 (L/mg.m) 2.63 4.55 2.45 2.40 0.91

DON (mg/L) 0.6 0.6 4.9 9.7 0.6

NH4 (mg/L) 0.025 0.075 4 0.025 2

Ca2+(mg/L) 5 4 5 20 2

pH 7.7 7.2 7.6 7.4 8

Ionic Strength (Molar as NaCl) 8x10-4 9x10-4 3.1x10-3 1x10-2 4.8x10-3

NDMAFP (ng/L) 112 148 1078 629 193

HNMFP (µg/L) 7 14 27 46 16

THMFP (µg/L) 230 578 244 351 1092

SW: Surface Water; ISW: Impacted Surface Water; WWTP-1: Municipal Wastewater Treatment Plant-1; WWTP-2: Municipal Wastewater Treatment Plant-2; IWWTP: Industrial Wastewater Treatment Plant.

21

428429

430

Table 3. Water Characteristics for pH, Ionic Strength, and Ca2+ Experiments

Parameter WWTP-1 IWWTP

UV254nm 0.0499 0.337

DOC (mg/L) 1.9 9.7

SUVA254 (L/mg.m) 2.6 3.5

DON (mg/L) 2.3 1

*Ca2+ (mg/L) 6 (6– 60*)

*pH 7.5 (6* – 9*)

*Ionic Strength (Molar as NaCl) 0.0016 (0.005* - 0.05*)

NDMA FP (ng/L) 576 71

HNM FP (µg/L) 12 52

THM FP (µg/L) 139 695

*Values were adjusted according to desired levels

WWTP-1: Municipal Wastewater Treatment Plant-1

IWWTP: Industrial Wastewater Treatment Plant

22

431

432

433

434

435

436

437

438

439

440

441

442

Table 4. Water Characteristics for pH, Ionic Strength, and Ca2+ Experiments using

different NF membranes

Parameter WWTP-1

UV254nm 0.0966

DOC (mg/L) 4.7

SUVA254 (L/mg.m) 2.1

DON (mg/L) 3.4

Ca2+ (mg/L) 5

*pH 7.2 (6* - 9*)

*Ionic Strength (Molar as NaCl) 0.0032 (0.005*-0.05*)

NDMA FP (ng/L) 544

THM FP (µg/L) 308

*Values were adjusted according to desired levels

WWTP-1: Municipal Wastewater Treatment Plant-1

23

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

SW: Surface Water; ISW: Impacted Surface Water; WWTP-1: Municipal Wastewater Treatment Plant-1; WWTP-2: Municipal Wastewater Treatment Plant-2; IWWTP: Industrial Wastewater Treatment Plant.

Figure 1. The removal NDMA, HNM, and THM precursors from different source

waters

24

461

462

463464465

466

467

468

469

470

471

472

Figure 2. Effect of Water Chemistry on the removal of NDMA, HNM, and THM

precursors -A) Municipal Wastewater Effluent B) Industrial Wastewater Effluent

25

473

474

475

476

477

478

Figure 3. Effect of Membrane Type on the removal of NDMA and THM precursors

from Municipal Wastewater Treatment Plant Effluent

26

479

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

References

Angeles, R.; Manuel, R.; Victor, M. L.; Daniel, P. Removal of natural organic matter and THM

formation potential by ultra- and nanofiltration of surface water. Water Res., 2008, 42, 714 –

722.

Bellona, C., Drewes, J. E., Xu, P., and Amy, G. “Factors affecting the rejection of organic

solutes during NF/RO treatment—A literature review.” Water Res., Vol.38, No: 12, 2795–2809,

2004.

Boyer T.H., Singer, P.C., Bench-scale testing of a magnetic ion exchange resin for removal of

disinfection by-product precursors, Water Res., Vol.39, 1265–1276, 2005.

Braghetta B. A., DiGiano F. A., and Ball W.P. Nanofiltration of Natural Organic Matter: pH and

Ionic Strength Effects. Journal of Environmental Engineering, Vol. 123: 623-641, July, 1997.

CDPH, (2010). Drinking Water Notification Levels and Response Levels: an Overview.

California Department of Public Health. Drinking Water Program.

<http://www.cdph.ca.gov/certlic/drinkingwater/Documents/Notificationlevels/

notificationlevels.pdf> (accessed 24.01.2014).

Chellam S., Effects of Nanofiltration on Trihalomethane and Haloacetic Acid Precursor Removal

and Speciation in Waters Containing Low Concentrations Of Bromide Ion. Environ. Sci.

Technol. 2000, 34, 1813-1820.

Cho J., Amy G., and Pellegrino J. Membrane Filtration of Natural Organic Matter: Initial

Comparison of Rejection and Flux Decline Characteristics with Ultrafiltration and Nanofiltration

Membranes. Water Res. Vol. 33, No. 11, 2517-2526, 1999.

27

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

522

Choi, J. H.; Valentine, R. L. (2002). Formation of N-nitrosodimethylamine (NDMA) from

Reaction of Monochloramine: a New Disinfection Byproduct. Water Res., 36 (4), 817–824.

Confalonieri, U., Menne B., Akhtar, R., Ebi, K.L.,  Hauengue, M., Kovats, R.S., Revich, B. and

Woodward, A. 2007. Human health. Climate Change 2007: Impacts, Adaptation and

Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the

Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK, 391-

431.

Gerecke, A. C.; Sedlak, D. L. Precursors of N-nitrosodimethylamine in natural waters. Environ.

Sci. Technol. 2003, 37 (7), 1331–1336.

Hong S., Elimelech M. Chemical and Physical Aspects of Natural Organic Matter (NOM)

Fouling of Nanofiltration Membranes. Journal of Membrane Science 132 (1997) 159-181.

K. Ghosh, M. Schnitzer, Macromolecular structures of humic substances, Soil Sci. 129 (5)

(1980) 266–276.

Kilduff J.E., Mattaraj S., Belfort G. Flux Decline During Nanofiltration of Naturally-Occurring

Dissolved Organic Matter: Effects of Osmotic Pressure, Membrane Permeability, and Cake

Formation. Journal of Membrane Science 239 (2004) 39–53.

Krasner, S. W.; Westerhoff, P.; Chen, B.; Rittmann, B. E.; Nam, S.-N.; Amy, G. Impact of

wastewater treatment processes on organic carbon, organic nitrogen, and DBP precursors in

effluent organic matter. Environ. Sci. Technol. 2009, 43 (8), 2911–2918.

28

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

Krasner, S.W., Mitch, W.A., McCurry, D.L., Hanigan, D., Westerhoff, P. Formation, precursors,

control, and occurrence of nitrosamines in drinking water: A review. Water Res. 47 (13), 4433-

4450, 2013.

Krasner, S.W., Weinberg, H.S., Richardson, S.D., Pastor, S.J., Chinn, R., Sclimenti, M.J.,

Onstad, G.D., Thruston, A.D., 2006. Occurrence of a new generation of disinfection byproducts.

Environ. Sci. Technol. 40, 7175–7185.

M. Kabsch-Korbutowicz, K. Majewska-Nowak, T. Winnicki, Analysis of membrane fouling in

the treatment of water solutions containing humic acids and mineral salts, Desalination 126

(1999) 179–185.

Martin Krauss, Philipp Longr´ee , Emmanuel Vanhoutte, Johan Cauwenberghs, and Juliane

Hollender Assessing the Fate of Nitrosamine Precursors in Wastewater Treatment by

Physicochemical Fractionation. Environmental Science &Technology 2010, 44, 7871-7877.

Mitch, W. A.; Sedlak, D. L. (2002). Factors Affecting the Formation of NDMA during

Chlorination. Environ. Sci. Technol., 36, 588–595.

Mitch, W. A.; Sedlak, D. L. Characterization and fate of N-nitrosodimethylamine precursors in

municipal wastewater treatment plants. Environ. Sci. Technol. 2004, 38 (5), 1445– 1454.

Oxenford, J. L. 1996 Disinfection By-ProductsByproducts: Current Practices and Future

Directions, In: Disinfection By-ProductsByproducts in Water Treatment: The Chemistry of Their

Formation and Control, Roger, A. M. and Gary, L. A. (eds.) CRC press. Inc. pp 3-16.

29

548

549

550

551

552

553

554

555

556

557

558

559

560

561

562

563

564

565

566

567

568

569

570

571

572

Pehlivanoglu-Mantas, E.; Sedlak, D. L. Measurement of dissolved organic nitrogen forms in

wastewater effluents: Concentrations, size distribution and NDMA formation potential. Water

Res. 2008, 42 (14), 3890–3898.

Plewa, M.J., Wagner, E.D., Jazwierska, P., Richardson, S.D., Chen, P.H., McKague, A.B., 2004.

Halonitromethane drinking water disinfection byproducts: chemical characterization and

mammalian cell cytotoxicity and genotoxicity. Environ. Sci. Technol. 38, 62–68.

Richardson, S.D., Simmons, J.E., Rice, G., 2002. Disinfection byproducts: the next generation.

Environ. Sci. Technol. 36, 198A–205A.

Roberson, J.A. (2011). Regulatory options for nitrosamines. In: Proceedings of AWWA Annual

Conference and Exposition, June 12-16, Washington, DC.

Russell, C.G., Blute, N.K., Via, S., Wu, X.Y., Chowdhury, Z., 2012. Nationwide Assessment of

Nitrosamine Occurrence and Trends. J. Am. Water Works Assoc. 104 (3), E205-E217.

Schaifer A.I., Fane A.G, Waite T.D. Nanofiltration of Natural Organic Matter: Removal, Fouling

and the Influence of Multivalent Ions. Desalination 118 (1998) 109-122.

Snyder, S. A., S. Adham, A. M. Redding, F. S. Cannon, J. De Carolis, J. Oppenheimer, E. C.

Wert, and Y. Yoon. Role of Membranes and Activated Carbon in the Removal of Endocrine

Disruptors and Pharmaceuticals. Desalination 2006, 156-181.

USEPA. Drinking water contaminants. United State Environmental Protection Agency; 2013

[http://water.epa.gov/drink/contaminants/index.cfm#List. [Accessed March 5, 2013].

30

573

574

575

576

577

578

579

580

581

582

583

584

585

586

587

588

589

590

591

592

593

594

595

596

597

598

599

Uyak V., Yavuz S.,Toroz I., Ozaydin S.,Genceli E.A., Disinfection by-productsbyproducts

precursors removal by enhanced coagulation and PAC adsorption, Desalination 216 (2007) 334–

344.

W.A.Mitch, I.M.Schreiber, Degradation of tertiaryalkylamines during

chlorination/chloramination: implications for formation of aldehydes,nitriles,halonitroalkanesand

nitrosamines, Environ.Sci.Technol.42 (2008)4811–4817.

Weinberg, H.S., Krasner, S.W., Richardson, S.D., Thruston, A.D., 2002. The Occurrence of

Disinfection By-ProductsByproducts (DBPs) of Health Concern in Drinking Water: Results of a

nationwide DBP occurrence study. U.S. EPA, EPA/600/R-02/068.

Yi-Li L., Pen-Chi C., E.-E. Chang, Removal of Small Trihalomethane Precursors from Aqueous

Solution by Nanofiltration Journal of Hazardous Materials 146 (2007) 20–29.

Yu Miyashita, Sang-Hyuck Park, Hoon Hyung, Ching-Hua Huang, and Jae-Hong Kim. Removal

of N-Nitrosamines and Their Precursors by Nanofiltration and Reverse Osmosis Membranes.

Journal of Environmental Engineering, Vol. 135, No. 9, September, 2009.

31

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620

621

622

623

624

625

626