Assessment of N-Oxide Formation during Wastewater Ozonation

Dec 12, 2016 - Gaëlle Guillet , Julia L.A. Knapp , Sylvain Merel , Olaf A. Cirpka , Peter Grathwohl , Christian Zwiener , Marc Schwientek. Science of...
0 downloads 0 Views 2MB Size
Subscriber access provided by UNIV OF NEWCASTLE

Article

Assessment of N-oxide Formation during Wastewater Ozonation Sylvain Merel, Sascha Lege, Jorge Eduardo Yanez Heras, and Christian Zwiener Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b02373 • Publication Date (Web): 12 Dec 2016 Downloaded from http://pubs.acs.org on December 13, 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 30

Environmental Science & Technology

1

Assessment of N-oxide Formation during

2

Wastewater Ozonation

3 4

Sylvain Merel, Sascha Lege, Jorge E. Yanez Heras, Christian Zwiener*

5 6

Environmental Analytical Chemistry, Center for Applied Geosciences, Eberhard Karls

7

University Tübingen, Hölderlinstraße 12, 72074 Tübingen, Germany.

8 9 10

* Corresponding author:

11

Prof. Dr. Christian Zwiener

12

Environmental Analytical Chemistry

13

Eberhard Karls University Tübingen

14

Hölderlinstraße 12

15

72074 Tübingen, Germany

16

Phone: (+49) 7071-29-74702

17

Email: [email protected]

18 19

ACS Paragon Plus Environment

1

Environmental Science & Technology

Page 2 of 30

20

ABSTRACT

21

Worldwide, ozonation of secondary wastewater effluents is increasingly considered in order to

22

decrease the load of organic contaminants before environmental discharge. However, despite the

23

constantly growing knowledge of ozonation over the last few years, the characterization of

24

transformation products (TPs) is still a major concern, particularly because such TPs might

25

remain biologically active. It has been shown for selected tertiary amine pharmaceuticals that

26

they react with ozone and form the corresponding N-oxides. This study therefore applies liquid

27

chromatography-high resolution mass spectrometry (LC-HRMS) to assess the overall N-oxide

28

formation during the pilot-scale ozonation of a secondary wastewater effluent from a major city

29

in Germany. Sample analysis by LC-HRMS revealed the occurrence of 1,229 compounds,

30

among which 853 were precursors attenuated by ozone and 165 were TPs. Further examination

31

of precursors and TPs using Kendrick mass and Kendrick mass defect analysis revealed 34 pairs

32

of precursors and products corresponding to a mono-oxygenation. Among these, 27 pairs (16%

33

of all TPs) were consistent with N-oxides since the TP had a higher retention time than the

34

precursor, a characteristic of these compounds. Using high resolution tandem mass spectrometry,

35

10 of these N-oxides could be identified and were shown to be stable during a subsequent

36

filtration step.

37 38 39 40 41

ACS Paragon Plus Environment

2

Page 3 of 30

42

Environmental Science & Technology

Table of content illustration

43

44 45 46

INTRODUCTION

47

Nowadays tens of thousands of chemicals are commonly used by industries in order to satisfy

48

the growing requirement of our society in terms of pharmaceuticals and personal care products,

49

food production, housing or transportation. Generally the number of these chemicals is

50

increasing and a large fraction of them will be collected in wastewater during the course of their

51

life cycle. However, since conventional wastewater treatment plants are not designed to remove

52

such trace organic contaminants, most of them are poorly attenuated and therefore discharged

53

into receiving waters. This is creating a major environmental and public health concern.1 For

54

instance, pharmaceuticals, personal care products and their metabolites have become ubiquitous

55

environmental contaminants1,

56

favor the emergence of bacteria resistant to antibiotics.5, 6 In addition, the effects of long term

57

exposure to low doses remain largely unknown with respect to ecological and human health.

2

that can affect wildlife through endocrine disruption3,

4

or can

58

The reduction of water contamination mostly involves the improvement of wastewater

59

treatment to ensure a more efficient removal of trace organic contaminants. A major trend

ACS Paragon Plus Environment

3

Environmental Science & Technology

Page 4 of 30

60

consists in upgrading conventional wastewater treatment plants with additional sorption or

61

oxidation processes. Among several potential options, the implementation of ozonation is

62

particularly considered since this is a well-established process in drinking water treatment and it

63

has a well-proven high efficiency.7 Indeed, ozone is a strong oxidant that reacts quickly with

64

aromatic, amine- and sulfur-containing compounds.8 This comprises a large number of soluble

65

organic contaminants, like antibiotics,9-12 antidepressants,13 antineoplastic drugs14 and other

66

pharmaceuticals.15-19 However, ozonation forms multiple transformation products8,

67

that have to be characterized since they could be equally or more toxic than the parent

68

compound.14

9, 13, 16, 20-25

69

Among the transformation products formed during wastewater ozonation are those resulting

70

from the reaction of ozone with nitrogen-containing compounds. While anilines are primarily

71

degraded via an attack on the aromatic ring,8 ozone mostly reacts with other amine moieties via

72

adduct formation at the lone electron pair with a subsequent O2 loss, leaving a zwitterion called

73

N-oxide.8 Such N-oxides are quickly transformed into hydroxylamines for primary and

74

secondary amine moieties,8 as it has been reported for metoprolol.19 In case of tertiary amines the

75

N-oxides are rather stable. N-oxide formation has been increasingly reported for ozonation of

76

common wastewater contaminants such as the antidepressants citalopram26 and desvenlafaxine,13

77

the analgesic tramadol27 and the antibiotics clarithromycin10 and levofloxacin.11 However,

78

according to experimental conditions, N-oxides 1) can be stable and remain biologically

79

active,28-31 2) are reduced to the original compound or 3) are further oxidized via N-dealkylation

80

or ozone reaction with other functional groups in the molecule.8, 32 Further sources of N-oxides

81

in wastewater can be from human metabolism of pharmaceuticals33 and from activated sludge

82

processes.34 However the potentially increased formation of N-oxides during wastewater

ACS Paragon Plus Environment

4

Page 5 of 30

Environmental Science & Technology

83

ozonation should be carefully assessed, particularly when such treatment strategy becomes

84

progressively more common. In water treatment, ozonation is generally followed by a

85

biologically active filtration step, which has the potential to further reduce the concentration of

86

biodegradable DOC fractions and TPs formed during ozonation.

87

The formation of oxygenated species can be easily observed by high resolution mass

88

spectrometry and Kendrick mass defect analysis. Indeed, this approach was initially developed to

89

observe CH2 homologue series35 and adapted in atmospheric science to characterize the reaction

90

of ozone with isoprene36 and in water science to detect the occurrence of polymer surfactants

91

such as alkylphenol ethoxylates.37 Generally LC retention times of oxygenated TPs are shorter

92

than the parent compounds on reversed-phase columns. In contrast, N-oxides exhibit higher

93

retention times than their precursors,

94

hydroxylated compounds. The objective of this study is to gain insight into transformation

95

processes and TP formation during ozonation of wastewater treatment plant effluents. LC-HRMS

96

data combined with Kendrick mass defect analysis and retention time shifts will be used to

97

investigate removal of precursor compounds and formation of oxygenated TPs with a focus on

98

N-oxides during pilot-scale wastewater ozonation and subsequent filtration.

38-41

which can be further used to distinguish them from

99 100 101

MATERIALS AND METHODS

102

Chemicals. Analytical standards for lidocaine and amisulpride were purchased from TCI

103

chemicals, while tramadol was from LGC Standards. Additional analytical standards for

104

sulpiride, tiapride, clarithromycin N-oxide and tiapride N-oxide were from Sigma-Aldrich, and

ACS Paragon Plus Environment

5

Environmental Science & Technology

Page 6 of 30

105

so was formic acid used for LC. Water and acetonitrile used as LC eluents were purchased from

106

Fischer Scientific at the highest purity available.

107 108

Sample Collection. This study was performed at a major wastewater treatment plant in Berlin,

109

Germany, with a dry weather capacity higher than 40,000 m3/d. The raw wastewater effluent

110

passes through bar screens, mechanical primary treatment and biological secondary treatment.

111

The secondary wastewater effluent with concentration of dissolved organic carbon (DOC) of 13

112

mg/L and of chemical oxygen demand (COD) of 36 mg/L was partially diverted to a pilot-scale

113

ozone reactor and a flocculation filter. The mean residence time in the ozone reactor was about

114

15 min during which an ozone dose of 6.4 mg/L was applied to achieve an O3/DOC ratio of 0.5.

115

After the ozone reactor FeCl3 was added as coagulant (3.5 mg/L calculated as Fe). The

116

subsequent flocculation filter with 1.2 m anthracite and 0.6 m sand was run at a filtration velocity

117

of 5.4 m/h. The mean hydraulic residence time in the filter was therefore of about 20 min.

118

Ozonation and filtration resulted in a DOC removal of 2.8 mg/L. The removal of the dissolved

119

fraction of COD during ozonation was about 5 mg/L and during filtration about 4.3 mg/L. The

120

betablocker metoprolol was reduced to 41 % after ozonation and to 21 % after filtration. The

121

data reveal further removal in the filtration process, whereas a combination of biodegradation,

122

sorption and flocculation is responsible. COD and micropollutants removal during ozonation is

123

due to direct and radical mediated oxidation processes.

124

To achieve the objectives of this study, time-proportional 22 h mixed samples (2 x 50 mL)

125

were collected after the biological secondary treatment, after ozonation and after filtration while

126

accounting for the respective hydraulic retention time.

127

ACS Paragon Plus Environment

6

Page 7 of 30

Environmental Science & Technology

128

Liquid Chromatography. Each sample collected was aliquoted for triplicate analysis. After

129

centrifugation at 6,000 rpm for 10 minutes in order to remove suspended solids, samples were

130

analyzed by liquid chromatography and in triplicates. Using a 1260 infinity device from Agilent

131

Technologies and without previous enrichment, 100 µL of sample were injected on a reversed

132

phase C18 column (Agilent Poroshell 120 with 2.1 mm internal diameter, 100 mm length and 2.7

133

µm particle size) at a temperature of 40°C. Analytes were eluted with a gradient of 0.4 ml/min of

134

water and acetonitrile (both acidified with 0.1% formic acid) similar to that previously described

135

in peer-reviewed literature for non-target screening of contaminants in wastewater.13,

136

Briefly, the fraction of acetonitrile was kept at 5% for 1 min, linearly increased to 90% within

137

the next 14 minutes, then set to 100% and held for 3 minutes. Before analyzing the next sample,

138

fraction of acetonitrile was returned to 5% and the column was allowed to equilibrate for 7 min.

42, 43

139 140

Time of Flight Mass Spectrometry. Following chromatographic separation, the compounds

141

occurring in the sample were detected by high resolution mass spectrometry, using a 6550

142

quadrupole-time of flight (QTOF) instrument from Agilent Technologies. Analytes were ionized

143

by positive electrospray (ESI+), applying 12 L/min sheath gas (N2) at 400°C, 16 L/min drying

144

gas (N2) at 150°C, 35 psi as nebulizer pressure, 3 kV as capillary voltage and 300 V as nozzle

145

voltage. Initially, only MS data were acquired by monitoring m/z in the range 50-1,000 at a rate

146

of three spectra/sec. In a second step, MS/MS data were acquired for the selected pairs of

147

precursor/N-oxide. The precursor ion [M+H]+ was selected by the quadrupole with an isolation

148

width of ± 0.65 u, and successively fragmented using N2 as collision gas with three different

149

collision energies set at 10 V, 20 V and 40 V.

150

ACS Paragon Plus Environment

7

Environmental Science & Technology

Page 8 of 30

151

Data Processing. The large amount of data obtained from high resolution mass spectrometry

152

was processed as described in a previous study43 and through five successive phases summarized

153

in Figure S1 of the supporting information. Briefly, in the first phase, a deconvolution of the total

154

ion chromatogram of each sample was performed using the “Molecular Feature Extraction”

155

algorithm from the MassHunter software (Agilent Technologies). The algorithm reported a

156

compound if a chromatographic peak is recognized with a specific accurate mass (m/z) and with

157

at least two isotopes (e.g. from

158

charge carriers. The algorithm binned ionization isomers (e.g. ions [M+H]+ and [M+Na]+) and

159

isotope signals to a single compound characterized by its accurate mass, abundance, retention

160

time and isotope pattern. In the second phase, these lists of compounds obtained from the

161

algorithm were imported into the “Mass Profiler Professional” software (Agilent Technologies)

162

for further statistical analysis. While compounds that did not occur in three out of three replicates

163

were discarded, a T-test (p-value < 0.05) was performed to compare the samples collected before

164

and after ozonation. Compounds with a lower peak area after ozonation were considered as

165

precursors while those with a higher peak area were considered as transformation products (TPs).

166

In the third phase the measured accurate masses of precursors and TPs were converted to

167

Kendrick masses (KM) with oxygen as the reference element (Eq. 1) and the respective Kendrick

168

mass defect (KMD) was calculated (Eq. 2). Then, precursors and their oxygenated species were

169

distinguished among all pairs of precursors/TPs since their KM was shifted by a multiple of 16

170

(Eq. 3) while their KMD was the same (Eq. 4). Subsequently, N-oxides were distinguished from

171

the remaining oxygenated species through their increased retention time compared to their

172

respective precursors. In the fourth phase, precursors and their N-oxides were tentatively

173

identified based on their accurate mass and a compound database (STOFF-IDENT,44 and Agilent

13

C) to limit artifacts. H and Na were considered as potential

ACS Paragon Plus Environment

8

Page 9 of 30

Environmental Science & Technology

174

ForTox library). Finally, the fifth phase confirmed the identification by high resolution tandem

175

mass spectrometry.

176

KM = measured accurate mass * 16 / 15.99491

(Eq. 1)

177

KMD = KM – nominal KM

(Eq. 2)

178

KMTP = n*16 + KMprecursor

(Eq. 3)

179

KMDTP = KMDprecursor ± 2 mDa

(Eq. 4)

180

Where: KM is the Kendrick mass (oxygen as reference element) calculated from the measured

181

accurate mass; KMD is the Kendrick mass defect; nominal KM is the nominal Kendrick mass

182

(KM rounded to the closest integer); KMTP is the Kendrick mass of the transformation product;

183

KMprecursor is the Kendrick mass of the precursor; KMDTP is the Kendrick mass defect of the

184

transformation product and KMDprecursor is the Kendrick mass defect of the precursor.

185 186

RESULTS AND DISCUSSION

187

Assessment of matrix effects during analysis. The extent of matrix effects on quantification

188

was assessed by spiking three stable isotopes of caffeine, diclofenac and amitriptyline into all

189

samples. The results (Fig. S2) show the high stability of the retention time throughout the

190

analysis and recoveries in the range between 85% and 115% for all compounds and across all

191

samples. Therefore, matrix effects could be neglected and signal abundances of single

192

compounds can be directly compared.

193

Overall impact of ozonation on water composition. Ozonation of the secondary wastewater

194

effluent had a significant impact on the overall water composition assessed by non-targeted

195

screening of organic contaminants, as shown in Figure 1. Indeed, a total of 1,229 compounds

196

were detected, among which 1,097 occurred in the secondary effluent and 599 after ozonation.

ACS Paragon Plus Environment

9

Environmental Science & Technology

Page 10 of 30

197

The distribution of these compounds in a Venn diagram further reveals that 630 compounds were

198

attenuated below detection limits by ozonation while 132 products were formed. The remaining

199

467 compounds which co-occurred before and after ozonation represent 1) contaminants that

200

were not attenuated, 2) contaminants partially attenuated or 3) compounds already occurring

201

before ozonation and which are further increased. For instance, the antibiotic clarithromycin and

202

its main human metabolite, N-desmethyl-clarithromycin,45 both occur in wastewater.42 However,

203

the concentration of N-desmethyl-clarithromycin could increase after ozonation since it is also

204

formed by the reaction of ozone with the residual clarithromycin.10 Similarly, citalopram N-

205

oxide and N-desmethyl-citalopram are two major human metabolites of the antidepressant

206

citalopram which can also be formed by ozonation of the residual drug.26

207

The comparison of samples before and after ozonation by a T-test (p-value < 0.05) statistically

208

subdivided the overall 1,229 compounds into 853 precursors (decreasing abundance), 165 TPs

209

(increasing abundance) and 211 resistant compounds (stable abundance). Previous research on

210

wastewater analysis with a similar non-target screening method reported several thousands of

211

statistically significant compounds, with a number of TPs similar or higher than the number of

212

precursors.43 However, the lower number of compounds reported in this study resulted from the

213

absence of sample enrichment. In fact, sample enrichment was avoided on purpose in order to

214

minimize bias due to any loss of analyte during solid phase extraction or increased ion

215

suppression due to the higher concentration of contaminants in the extracts. As a result, this

216

study focuses exclusively on the most abundant organic contaminants and their related ozonation

217

TPs which were particularly scrutinized in order to identify precursors forming potential

218

oxygenated species and most particularly N-oxides.

219

ACS Paragon Plus Environment

10

Page 11 of 30

Environmental Science & Technology

220

Formation of oxygenated species. The precursors and TPs found by the T-Test were paired

221

according to their respective KM and KMD for oxygen homologs. Related pairs of precursors

222

and oxygenated species are characterized by the same KMD. This led to the overall observation

223

of 60 potential oxygenated species for 99 precursors (Table 1 and Figure S3). Among the

224

oxygenated species formed during ozonation of the secondary wastewater effluent, the majority

225

(56%) are mono-oxygenated species. These TPs are consistent with mono-hydroxylation which

226

is a common degradation pathway during ozonation of a wide range of wastewater contaminants,

227

particularly those with reactive aromatic rings.8 For instance, hydroxylation resulting from

228

ozonation has been observed for the antibiotics amoxicillin46 and sulfamethoxazole,47 the anti-

229

inflammatory drug diclofenac,18 the beta blocker metoprolol,23 or the antibacterial triclosan.21

230

However, mono-oxygenated species can also result from N-oxide formation due to oxidation of

231

compounds with a tertiary amine moiety,8 as previously described for several pharmaceuticals

232

commonly found in wastewater.9, 10, 13, 27, 48 Also poly-oxygenated species were formed (Table 1).

233

These TPs are consistent with multiple hydroxylation23 or a combination of hydroxylation with

234

N-oxide formation. However, this study will focus on assessing N-oxide formation among mono-

235

oxygenated species.

236 237

Assessment of N-oxide formation. The application of a retention time filter indicated that

238

79% (27 out of 34) of mono-oxygenated species formed during ozonation of the secondary

239

wastewater are potential N-oxides, which are characterized by retention times higher than the

240

precursors (Table 1). Moreover, these potential N-oxides also represent 16% (27 out of 165) of

241

the overall ozonation TPs. Therefore, additional analysis was performed in an attempt to confirm

242

the identity of such N-oxides.

ACS Paragon Plus Environment

11

Environmental Science & Technology

Page 12 of 30

243

The precursors identified with a confidence level of 1 or 249, 50 were the antipsychotic drugs

244

sulpiride, tiapride and amisulpride, the antihistamine diphenhydramine, the antidepressants

245

venlafaxine and citalopram, the pain medication tramadol, the local anesthetic lidocaine, the

246

antibiotic clarithromycin and its metabolite clindamycin sulfoxide (Table 2). Identification was

247

based on matches with authentic standards or the ForensicTox MS/MS library (Figures S4-S14).

248

For these ten compounds, the occurrence of the corresponding N-oxides after ozonation could be

249

verified with a confidence level 1 (Figure 2, Figures S15-S17), 2a or 2b (Figures S18-S30).

250

Whenever neither standards nor MS/MS spectra were available, for example for sulpiride N-

251

oxide, the identity of the compound was deduced by diagnostic evidence based on MS/MS

252

fragments as shown in Figure 3. While three product ions were common to sulpiride and its N-

253

oxide and confirmed their similar structure, two other product ions from sulpiride associated with

254

the tertiary amine moiety had their m/z shifted by 15.995 with sulpiride N-oxide, which is

255

consistent with the addition of oxygen on the amine. Finally, a last piece of evidence supporting

256

the identification of N-oxides is the calculated LogD values. For all the precursors previously

257

identified, LogD values were calculated at pH 2.7 (pH of the eluent) according to the public web

258

resource developed by ChemAxon,51 and so was done for the corresponding N-oxides and

259

hydroxyl products. While higher LogDs are associated with higher retention times on reversed-

260

phase separation, the positive LogD shifts of the N-oxides and the negative LogD shifts of the

261

hydroxylated compounds tend to confirm the identification of TPs as N-oxides rather than

262

hydroxylated products.

263

Overall, 37% of the potential N-oxides from the mono-oxygenated TPs (10 out of 27) could be

264

identified. The other 17 compounds remain potential N-oxides but no tentative identity could be

265

assigned. This is mostly attributed to the inability to identify a precursor based on MS/MS

ACS Paragon Plus Environment

12

Page 13 of 30

Environmental Science & Technology

266

spectra in libraries. In fact, secondary wastewater effluent is a complex matrix that is likely to

267

contain unknown tertiary amines, such as metabolites of pharmaceuticals not yet identified or

268

included in compound libraries. Besides the N-oxides identified in this study, previous research

269

has also shown the potential N-oxide formation from other common wastewater contaminants,

270

including cetirizine, fexofenadine, hydrochlorothiazide, pargyline, deprenyl and pheniramine.20,

271

34

272

pre-concentration) their N-oxide formation could not be assessed during pilot-scale ozonation.

273

However, N-oxides associated with these non-detected compounds further strengthen the

274

relevance of N-oxide formation during ozonation.

Since these precursors were not detected in this study (potentially due to the absence of sample

275

The 10 N-oxide precursors identified in this study were almost totally transformed during

276

ozonation (Table 2). While their N-oxides were not detected in the secondary wastewater

277

effluent (Figure 4), they were detected with a substantial abundance after ozonation (peak area

278

normalized to the peak area of the precursor). Some of these N-oxides were previously reported

279

to be formed during other processes, such as activated sludge treatment.34 However ozone tends

280

to yield a higher formation of N-oxides. For instance, while venlafaxine and lidocaine formed

281

respectively 5% and 19% of N-oxide during wastewater treatment with activated sludge, they

282

formed respectively 16% and 38% of N-oxide after ozonation.

283

The assumption of N-oxides representing 16% of the overall ozonation TPs should be

284

considered as conservative. Indeed, N-oxides might also occur for poly-oxygenated species. For

285

instance, clindamycin sulfoxide N-oxide observed after ozonation can be a mono-oxygenated

286

species of the precursor clindamycin sulfoxide, but it can be also considered as a di-oxygenated

287

species of the precursor clindamycin (Fig. S27). However, N-oxides among poly-oxygenated

ACS Paragon Plus Environment

13

Environmental Science & Technology

Page 14 of 30

288

species were not further investigated in this study and will require further research since the rule

289

of increasing retention time cannot be applied in this case.

290 291

Fate of N-oxides during filtration. After ozonation, an additional treatment by filtration was

292

applied, which resulted in further changes in the water composition presented in a heat map of a

293

hierarchical cluster analysis (Fig. S31). The clusters show compounds which are removed during

294

filtration like ozonation products and the beta-blocker atenolol (known for its biodegradability on

295

trickling filters),52 those which are released into the water and a large fraction of ozonation

296

products which reveal stable during filtration. Compound and COD removal of 4.3 mg/L suggest

297

biodegradation processes during filtration, but they work with limited efficiency as the stability

298

of ozonation products reveal. Also N-oxides formed during ozonation were very poorly retained

299

or transformed during the filtration process (Table 2). Therefore, the effect of filtration on the

300

overall water composition should be further studied in more detail. However, this is out of the

301

scope of this study on N-oxides as ozonation TPs.

302 303

Implications for wastewater treatment. Advanced oxidation processes are increasingly

304

considered to upgrade conventional wastewater treatment and to decrease the load of

305

contaminants discharged to the environment. This study confirms that ozonation of secondary

306

wastewater effluents efficiently decreases the concentration of common micro-pollutants.

307

However, a substantial fraction of these compounds containing a tertiary amine moiety appears

308

to react with ozone to yield quaternary N-oxides by a formal addition of oxygen. In this study,

309

potential N-oxides represented 16% (27 out of 165) of the total TPs observed after ozonation.

310

Ten N-oxides were identified, while 17 remain potential N-oxides without any tentative

ACS Paragon Plus Environment

14

Page 15 of 30

Environmental Science & Technology

311

identification. There were no further accurate mass matches of precursors in compound libraries,

312

which underlines the need for enhanced compound libraries.53 The estimation of N-oxides

313

representing 16% of the overall TPs is conservative since it does not account for N-oxides in

314

poly-oxygenated species and those eluting earlier than their precursors. These cases were shown

315

to occur and will be examined in future studies.

316

The hydroxylation of water contaminants, often considered in degradation studies, was shown

317

to represent only a small fraction of mono-oxygenated ozonation TPs compared to N-oxide

318

formation based on our analysis using KMD and positive RT shifts. A sub-structure search for

319

compounds with a tertiary amine moiety in the publicly available database STOFF-IDENT44

320

resulted in over 1,300 hits for possible N-oxide precursors. This reveals a rather large potential

321

of wastewater contaminants to form N-oxides. However, N-oxide formation is pH dependent,

322

since ozone does not react rapidly with a protonated tertiary amine.17, 19, 24, 25 Therefore, among

323

all potential compounds in the wastewater effluent, only those with a deprotonated tertiary amine

324

at the pH of wastewater would be the principal N-oxide precursors during ozonation.

325

Since the N-oxides were not removed by a following filtration step in this work and since they

326

can be reduced to reform the native compound,34 these observations have significant implications

327

for the application of ozonation in wastewater treatment. While ozonation efficiently eliminates

328

the native contaminant, the N-oxides formed should be further characterized since some were

329

also shown to remain biologically active.

330

analyzed by suitable bioassays to assess the impact of ozonation on the biological activity. In

331

addition, the formation of N-oxides should be further assessed with higher ozone doses as they

332

might be further degraded. Finally, the occurrence and fate of N-oxides in river water should be

333

particularly considered when ozonation is applied in wastewater treatment.

28-31

Therefore, ozonated samples should also be

ACS Paragon Plus Environment

15

Environmental Science & Technology

Page 16 of 30

334 335

ASSOCIATED CONTENT

336

Supporting Information. The supporting information includes 31 figures showing a workflow

337

for compound identification, the investigation of matrix effects, a Kendrick mass plot, mass

338

spectra and fragmentation pathways of precursors and N-oxides, and the effect of ozonation and

339

filtration on water composition by cluster analysis. This material is available free of charge via

340

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

341 342 343

ACKNOWLEDGMENT

344

The authors thank the staff of Berlin Wasserbetriebe and Kompetenz Zentrum Wasser Berlin,

345

Germany, for providing the samples used in this study. The German Federal Environmental

346

Foundation (Deutsche Bundesstiftung Umwelt, DBU) is gratefully acknowledged for funding the

347

Ph.D. scholarship of Sascha Lege (Nr. 20013/277) and the German Research Foundation (DFG)

348

for funding the project ZW 73/14.

349 350

ABBREVIATIONS

351

LC, liquid chromatography; MSMS, tandem mass spectrometry; TP, transformation product;

352

DOC, dissolved organic carbon.

353 354 355

REFERENCES

ACS Paragon Plus Environment

16

Page 17 of 30

356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399

Environmental Science & Technology

(1) Petrie, B.; Barden, R.; Kasprzyk-Hordern, B. A review on emerging contaminants in wastewaters and the environment: Current knowledge, understudied areas and recommendations for future monitoring. Water Res. 2015, 72, 3-27. (2) Mompelat, S.; Le Bot, B.; Thomas, O. Occurrence and fate of pharmaceutical products and by-products, from resource to drinking water. Environ. Int. 2009, 35 (5), 803-814. (3) Baynes, A.; Green, C.; Nicol, E.; Beresford, N.; Kanda, R.; Henshaw, A.; Churchley, J.; Jobling, S. Additional treatment of wastewater reduces endocrine disruption in wild fish-a comparative study of tertiary and advanced treatments. Environ. Sci. Technol. 2012, 46 (10), 5565-5573. (4) Orlando, E. F.; Ellestad, L. E. Sources, concentrations, and exposure effects of environmental gestagens on fish and other aquatic wildlife, with an emphasis on reproduction. Gen. Comp. Endocrinol. 2014, 203, 241-249. (5) Makowska, N.; Koczura, R.; Mokracka, J. Class 1 integrase, sulfonamide and tetracycline resistance genes in wastewater treatment plant and surface water. Chemosphere 2016, 144, 16651673. (6) Manaia, C. M.; Macedo, G.; Fatta-Kassinos, D.; Nunes, O. C. Antibiotic resistance in urban aquatic environments: Can it be controlled? Appl. Microbiol. Biotechnol. 2016, 100 (4), 1543-1557. (7) Zwiener, C.; Frimmel, F. H. Oxidative treatment of pharmaceuticals in water. Water Res. 2000, 34 (6), 1881-1885. (8) Hübner, U.; von Gunten, U.; Jekel, M. Evaluation of the persistence of transformation products from ozonation of trace organic compounds – a critical review. Water Res. 2015, 68, 150-170. (9) Hamdi El Najjar, N.; Touffet, A.; Deborde, M.; Journel, R.; Leitner, N. K. V. Levofloxacin oxidation by ozone and hydroxyl radicals: Kinetic study, transformation products and toxicity. Chemosphere 2013, 93 (4), 604-611. (10) Lange, F.; Cornelissen, S.; Kubac, D.; Sein, M. M.; von Sonntag, J.; Hannich, C. B.; Golloch, A.; Heipieper, H. J.; Möder, M.; von Sonntag, C. Degradation of macrolide antibiotics by ozone: A mechanistic case study with clarithromycin. Chemosphere 2006, 65 (1), 17-23. (11) Witte, B. D.; Langenhove, H. V.; Hemelsoet, K.; Demeestere, K.; Wispelaere, P. D.; Van Speybroeck, V.; Dewulf, J. Levofloxacin ozonation in water: Rate determining process parameters and reaction pathway elucidation. Chemosphere 2009, 76 (5), 683-689. (12) Sein, M. M.; Schmidt, T. C.; Golloch, A.; von Sonntag, C. Oxidation of some typical wastewater contaminants (tributyltin, clarithromycin, metoprolol and diclofenac) by ozone. Water Sci. Technol. 2009, 59 (8), 1479-1485. (13) Lajeunesse, A.; Blais, M.; Barbeau, B.; Sauvé, S.; Gagnon, C. Ozone oxidation of antidepressants in wastewater –treatment evaluation and characterization of new by-products by lc-qtofms. Chem. Cent. J. 2013, 7 (1), 1-11. (14) Lin, A. Y.-C.; Hsueh, J. H.-F.; Hong, P. K. A. Removal of antineoplastic drugs cyclophosphamide, ifosfamide, and 5-fluorouracil and a vasodilator drug pentoxifylline from wastewaters by ozonation. Environ. Sci. Pollut. Res. 2015, 22 (1), 508-515. (15) Antoniou, M. G.; Hey, G.; Rodríguez Vega, S.; Spiliotopoulou, A.; Fick, J.; Tysklind, M.; la Cour Jansen, J.; Andersen, H. R. Required ozone doses for removing pharmaceuticals from wastewater effluents. Sci. Total Environ. 2013, 456–457, 42-49.

ACS Paragon Plus Environment

17

Environmental Science & Technology

400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445

Page 18 of 30

(16) Ikehata, K.; Jodeiri Naghashkar, N.; Gamal El-Din, M. Degradation of aqueous pharmaceuticals by ozonation and advanced oxidation processes: A review. Ozone: Sci. Eng. 2006, 28 (6), 353-414. (17) Faber, H.; Lutze, H.; Lareo, P. L.; Frensemeier, L.; Vogel, M.; Schmidt, T. C.; Karst, U. Liquid chromatography/mass spectrometry to study oxidative degradation of environmentally relevant pharmaceuticals by electrochemistry and ozonation. J. Chromatogr. A 2014, 1343, 152159. (18) Sein, M. M.; Zedda, M.; Tuerk, J.; Schmidt, T. C.; Golloch, A.; Sonntag, C. v. Oxidation of diclofenac with ozone in aqueous solution. Environ. Sci. Technol. 2008, 42 (17), 6656-6662. (19) Benner, J.; Ternes, T. A. Ozonation of metoprolol: Elucidation of oxidation pathways and major oxidation products. Environ. Sci. Technol. 2009, 43 (14), 5472-5480. (20) Borowska, E.; Bourgin, M.; Hollender, J.; Kienle, C.; McArdell, C. S.; von Gunten, U. Oxidation of cetirizine, fexofenadine and hydrochlorothiazide during ozonation: Kinetics and formation of transformation products. Water Res. 2016, 94, 350-362. (21) Chen, X.; Richard, J.; Liu, Y.; Dopp, E.; Tuerk, J.; Bester, K. Ozonation products of triclosan in advanced wastewater treatment. Water Res. 2012, 46 (7), 2247-2256. (22) Tay, K. S.; Rahman, N. A.; Abas, M. R. B. Degradation of deet by ozonation in aqueous solution. Chemosphere 2009, 76 (9), 1296-1302. (23) Tay, K. S.; Rahman, N. A.; Abas, M. R. B. Ozonation of metoprolol in aqueous solution: Ozonation by-products and mechanisms of degradation. Environ. Sci. Pollut. Res. 2013, 20 (5), 3115-3121. (24) Tekle-Röttering, A.; Jewell, K. S.; Reisz, E.; Lutze, H. V.; Ternes, T. A.; Schmidt, W.; Schmidt, T. C. Ozonation of piperidine, piperazine and morpholine: Kinetics, stoichiometry, product formation and mechanistic considerations. Water Res. 2016, 88, 960-971. (25) Tekle-Röttering, A.; von Sonntag, C.; Reisz, E.; Eyser, C. v.; Lutze, H. V.; Türk, J.; Naumov, S.; Schmidt, W.; Schmidt, T. C. Ozonation of anilines: Kinetics, stoichiometry, product identification and elucidation of pathways. Water Res. 2016, 98, 147-159. (26) Hörsing, M.; Kosjek, T.; Andersen, H. R.; Heath, E.; Ledin, A. Fate of citalopram during water treatment with o3, clo2, uv and fenton oxidation. Chemosphere 2012, 89 (2), 129-135. (27) Zimmermann, S. G.; Schmukat, A.; Schulz, M.; Benner, J.; Gunten, U. v.; Ternes, T. A. Kinetic and mechanistic investigations of the oxidation of tramadol by ferrate and ozone. Environ. Sci. Technol. 2012, 46 (2), 876-884. (28) Gottfries, C. G. Clinical research with imipramine-n-oxide in mental depression. Nordic journal of psychiatry 1968, 22 (2), 167-74. (29) Dencker, S. J. Clinical trial with imipramine n-oxide and amitriptyline n-oxide. Nordic journal of psychiatry 1971, 25 (5), 463-470. (30) Godt, H. H.; Fredslund-Andersen, K.; Edlund, A. H. Amitriptyline n-oxide. A new anti depressant: A clinicall double-blind trial in comparison with amitriptyline. Nordic journal of psychiatry 1971, 25 (3), 237-246. (31) Raffa, R. B.; Haslego, M. L.; Maryanoff, C. A.; Villani, F. J.; Codd, E. E.; Connelly, C. D.; Martinez, R. P.; Schupsky, J. J.; Buben, J. A.; Wu, W. N.; Takacs, A. N.; Mckown, L. A. Unexpected antinociceptive effect of the n-oxide (rwj 38705) of tramadol hydrochloride. J. Pharmacol. Exp. Ther. 1996, 278 (3), 1098-104. (32) Iobbi-Nivol, C.; Pommier, J.; Simala-Grant, J.; Méjean, V.; Giordano, G. High substrate specificity and induction characteristics of trimethylamine-n-oxide reductase of escherichia coli. Biochim. Biophys. Acta 1996, 1294 (1), 77-82.

ACS Paragon Plus Environment

18

Page 19 of 30

446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491

Environmental Science & Technology

(33) Gago-Ferrero, P.; Schymanski, E. L.; Bletsou, A. A.; Aalizadeh, R.; Hollender, J.; Thomaidis, N. S. Extended suspect and non-target strategies to characterize emerging polar organic contaminants in raw wastewater with lc-hrms/ms. Environ. Sci. Technol. 2015, 49 (20), 12333-12341. (34) Gulde, R.; Meier, U.; Schymanski, E. L.; Kohler, H.-P. E.; Helbling, D. E.; Derrer, S.; Rentsch, D.; Fenner, K. Systematic exploration of biotransformation reactions of aminecontaining micropollutants in activated sludge. Environ. Sci. Technol. 2016, 50 (6), 2908-2920. (35) Kendrick, E. A mass scale based on CH2 = 14.0000 for high resolution mass spectrometry of organic compounds. Anal. Chem. 1963, 35 (13), 2146-2154. (36) Nguyen, T. B.; Bateman, A. P.; Bones, D. L.; Nizkorodov, S. A.; Laskin, J.; Laskin, A. High-resolution mass spectrometry analysis of secondary organic aerosol generated by ozonolysis of isoprene. Atmos. Environ. 2010, 44 (8), 1032-1042. (37) Thurman, E. M.; Ferrer, I.; Blotevogel, J.; Borch, T. Analysis of hydraulic fracturing flowback and produced waters using accurate mass: Identification of ethoxylated surfactants. Anal. Chem. 2014, 86 (19), 9653-9661. (38) Christopher, L. J.; Cui, D.; Li, W.; Barros, A.; Arora, V. K.; Zhang, H.; Wang, L.; Zhang, D.; Manning, J. A.; He, K.; Fletcher, A. M.; Ogan, M.; Lago, M.; Bonacorsi, S. J.; Humphreys, W. G.; Iyer, R. A. Biotransformation of [14c]dasatinib: In vitro studies in rat, monkey, and human and disposition after administration to rats and monkeys. Drug Metab. Dispos. 2008, 36 (7), 1341-1356. (39) Dumasia, M. C.; Teale, P. N-deethylation and n-oxidation of etamiphylline: Identification of etamiphylline-n-oxide in greyhound urine by high performance liquid chromatography-mass spectrometry. J. Pharm. Biomed. Anal. 2005, 36 (5), 1085-1091. (40) Sun, X.; Niu, L.; Li, X.; Lu, X.; Li, F. Characterization of metabolic profile of mosapride citrate in rat and identification of two new metabolites: Mosapride n-oxide and morpholine ringopened mosapride by uplc–esi-ms/ms. J. Pharm. Biomed. Anal. 2009, 50 (1), 27-34. (41) Tomar, R. S.; Joseph, T. J.; Murthy, A. S. R.; Yadav, D. V.; Subbaiah, G.; Krishna Reddy, K. V. S. R. Identification and characterization of major degradation products of risperidone in bulk drug and pharmaceutical dosage forms. J. Pharm. Biomed. Anal. 2004, 36 (1), 231-235. (42) Hernández, F.; Ibáñez, M.; Gracia-Lor, E.; Sancho, J. V. Retrospective lc-qtof-ms analysis searching for pharmaceutical metabolites in urban wastewater. J. Sep. Sci. 2011, 34 (24), 3517-3526. (43) Merel, S.; Anumol, T.; Park, M.; Snyder, S. A. Application of surrogates, indicators, and high-resolution mass spectrometry to evaluate the efficacy of UV processes for attenuation of emerging contaminants in water. J. Hazard. Mater. 2015, 282, 75-85. (44) Stoffident Stoffident. http://bb-x-stoffident.hswt.de/login. (45) Baumann, M.; Weiss, K.; Maletzki, D.; Schüssler, W.; Schudoma, D.; Kopf, W.; Kühnen, U. Aquatic toxicity of the macrolide antibiotic clarithromycin and its metabolites. Chemosphere 2015, 120, 192-198. (46) Andreozzi, R.; Canterino, M.; Marotta, R.; Paxeus, N. Antibiotic removal from wastewaters: The ozonation of amoxicillin. J. Hazard. Mater. 2005, 122 (3), 243-250. (47) Gómez-Ramos, M. d. M.; Mezcua, M.; Agüera, A.; Fernández-Alba, A. R.; Gonzalo, S.; Rodríguez, A.; Rosal, R. Chemical and toxicological evolution of the antibiotic sulfamethoxazole under ozone treatment in water solution. J. Hazard. Mater. 2011, 192 (1), 1825.

ACS Paragon Plus Environment

19

Environmental Science & Technology

492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510

Page 20 of 30

(48) Lester, Y.; Mamane, H.; Zucker, I.; Avisar, D. Treating wastewater from a pharmaceutical formulation facility by biological process and ozone. Water Res. 2013, 47 (13), 4349-4356. (49) Schymanski, E. L.; Singer, H. P.; Slobodnik, J.; Ipolyi, I. M.; Oswald, P.; Krauss, M.; Schulze, T.; Haglund, P.; Letzel, T.; Grosse, S.; Thomaidis, N. S.; Bletsou, A.; Zwiener, C.; Ibáñez, M.; Portolés, T.; de Boer, R.; Reid, M. J.; Onghena, M.; Kunkel, U.; Schulz, W.; Guillon, A.; Noyon, N.; Leroy, G.; Bados, P.; Bogialli, S.; Stipaničev, D.; Rostkowski, P.; Hollender, J. Non-target screening with high-resolution mass spectrometry: Critical review using a collaborative trial on water analysis. Anal. Bioanal. Chem. 2015, 407 (21), 6237-6255. (50) Schymanski, E. L.; Jeon, J.; Gulde, R.; Fenner, K.; Ruff, M.; Singer, H. P.; Hollender, J. Identifying small molecules via high resolution mass spectrometry: Communicating confidence. Environ. Sci. Technol. 2014, 48 (4), 2097-2098. (51) ChemAxon Chemicalize. http://www.chemicalize.org/ (52) Kasprzyk-Hordern, B.; Dinsdale, R. M.; Guwy, A. J. The removal of pharmaceuticals, personal care products, endocrine disruptors and illicit drugs during wastewater treatment and its impact on the quality of receiving waters. Water Res. 2009, 43 (2), 363-380. (53) Zedda, M.; Zwiener, C. Is nontarget screening of emerging contaminants by lc-hrms successful? A plea for compound libraries and computer tools. Anal. Bioanal. Chem. 2012, 403 (9), 2493-2502.

511 512

ACS Paragon Plus Environment

20

Page 21 of 30

513

Environmental Science & Technology

FIGURES

514 515

Figure 1. Overall impact of ozonation on water composition.

516 517 518 519 520 521 522

ACS Paragon Plus Environment

21

Environmental Science & Technology

Page 22 of 30

523 524

Figure 2. Identification of N-oxides by LC-MS/MS analysis based on analytical standards

525

(Retention time differences between analytes and standards were below 0.05 min; identification

526

is based on a minimum of 7 common MS/MS fragments for analytes and standards).

ACS Paragon Plus Environment

22

Page 23 of 30

Environmental Science & Technology

527 528

Figure 3. Identification of sulpiride N-oxide by diagnostic evidence based on MS/MS fragments.

529

ACS Paragon Plus Environment

23

Environmental Science & Technology

Page 24 of 30

530 531

Figure 4. Extracted ion chromatograms show the relative abundance of selected precursors and

532

N-oxides before ozonation, after ozonation and after biofiltration.

533 534 535 536 537 538 539 540 541 542 543

ACS Paragon Plus Environment

24

Page 25 of 30

Environmental Science & Technology

544

TABLES

545

Table 1. Assessment of the formation potential of oxygenated species and N-oxides during

546

wastewater ozonation. Precursors and oxygenated species are paired based on the same KMD

547

without restriction, precursor/(N-)oxide pairs were selected by KMD and a positive retention

548

time shift.

Number of Precursors → Oxygenated species oxygenation (without restriction) 1 42 → 34

Precursors → (N-)Oxides (RTprecursor < RToxide) 27 → 27

2

26 → 21

5→7

3

19 → 12

1→1

4

18 → 11

4→3

5

3→2

-

All

99 → 60

39 → 37

549 550

ACS Paragon Plus Environment

25

Environmental Science & Technology

Page 26 of 30

Table 2. Overview on identification and occurrence of precursors and their respective N-oxides.

Precursor

N-oxide Relative abundance in water*

Identity

Relative abundance in water*

Identity

Name

Formula

Level**

Untreated

Ozonated

Filtered

Level**

LogD shift

RT-shift (min)

Untreated

Ozonated

Filtered

Sulpiride

C15H23N3O4S

1

100%

n.d.

n.d.

2b

1,5

1,73

n.d.

45%

45%

Tiapride

C15H24N2O4S

1

100%

n.d.

1%

1

1,51

0,57

n.d.

50%

51%

Lidocaine

C14H22N2O

1

100%

n.d.

6%

2a

1,52

0,91

n.d.

38%

39%

Amisulpride

C17H27N3O4S

1

100%

n.d.

1%

2a

1,5

0,34

n.d.

20%

20%

Tramadol

C16H25NO2

1

100%

11%

11%

1

1,5

0,27

n.d.

31%

32%

Venlafaxine

C17H27NO2

2a

100%

15%

15%

2a

1,5

0,38

n.d.

16%

16%

Diphenhydramine

C17H21NO

2a

100%

n.d.

n.d.

2b

1,5

0,34

n.d.

141%

144%

Citalopram

C20H21FN2O

2a

100%

5%

21%

2b

1,5

0,19

n.d.

82%

88%

Clarithromycin

C38H69NO13

1

100%

n.d.

1%

1

1,51

0,32

n.d.

61%

63%

Clindamycin sulfoxide

C18H33N2O6SCl

2b

100%

n.d.

n.d.

2b

1,51

0,35

n.d.

21%

29%

* Peak area normalized to the abundance of the precursor in the untreated water, where n.d. stands for not detected. ** According to peer-reviewed literature49, 50. Level 1 implies identification with an analytical standard showing less than 0.05 min difference in retention time and a minimum of 2 common product ions. Level 2a implies identification with a spectral library and a minimum of 3 common product ions. Level 2b implies identification by diagnostic evidence based on mass fragments with a minimum of 3 annotated product ions.

ACS Paragon Plus Environment

26

Page 27 of 30

Environmental Science & Technology

Figure 1. Overall impact of ozonation on water composition. 76x74mm (96 x 96 DPI)

ACS Paragon Plus Environment

Environmental Science & Technology

Figure 2. Identification of N-oxides by LC-MS/MS analysis based on analytical standards (Retention time differences between analytes and standards were below 0.05 min; identification is based on a minimum of 7 common MS/MS fragments for analytes and standards). 160x111mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 28 of 30

Page 29 of 30

Environmental Science & Technology

Figure 3. Identification of sulpiride N-oxide by diagnostic evidence based on MS/MS fragments. 160x198mm (96 x 96 DPI)

ACS Paragon Plus Environment

Environmental Science & Technology

Figure 4. Extracted ion chromatograms show the relative abundance of selected precursors and N-oxides before ozonation, after ozonation and after biofiltration. 160x98mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 30 of 30