Photochemical Transformation of Four Ionic Liquid Cation Structures

Sep 28, 2017 - Bacteria use expanded genetic code. The genome of every cell on Earth uses four DNA bases—adenine, thymine, cytosine, and guanine—t...
0 downloads 10 Views 524KB Size
Subscriber access provided by Gothenburg University Library

Article

Photochemical transformation of four ionic liquid cation structures in aqueous solution Sarah G. Pati, and William A. Arnold Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b04016 • Publication Date (Web): 28 Sep 2017 Downloaded from http://pubs.acs.org on September 29, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

Environmental Science & Technology

Photochemical transformation of four ionic liquid cation structures in aqueous solution Sarah G. Pati and William A. Arnold∗ Department of Civil, Environmental, and Geo- Engineering, University of Minnesota, 500 Pillsbury Drive SE, Minneapolis, Minnesota 55455-0116, United States E-mail: [email protected] Phone: (612) 625-8582. Fax: (612) 626-7750

1 ACS Paragon Plus Environment

Environmental Science & Technology

Abstract

1

2

Ionic liquids (ILs) are a new class of solvents expected to be used increasingly by

3

the chemical industry in the coming years. Given their slow biodegradation and limited

4

sorption affinities, IL cations have a high potential to reach aquatic environments. We

5

investigated the fate of ILs in sunlit surface water by determining direct and indirect

6

photochemical transformation rates of imidazolium, pyridinium, pyrrolidinium, and

7

piperidinium cations. The photodegradation of all investigated IL cations was faster

8

in solutions containing dissolved organic matter (DOM) than in ultrapure water, illus-

9

trating the importance of indirect photochemical processes. Experiments with model

10

sensitizers and DOM isolates revealed that reactions with hydroxyl radicals dominated

11

the transformation of tested IL cations. Bimolecular reaction rate constants with hy-

12

droxyl radicals ranged from (2.04 ± 0.37) · 109 M−1 s−1 to (8.47 ± 0.97) · 109 M−1

13

s−1 and showed an increase in rate constants with increasing carbon side-chain length.

14

Consequently, average estimated half-lives of IL cations in sunlit surface water ranged

15

from 32 ± 4 days to 135 ± 25 days, highlighting the potential of IL cations to become

16

persistent aquatic contaminants.

2 ACS Paragon Plus Environment

Page 2 of 27

Page 3 of 27

Environmental Science & Technology

17

Introduction

18

Ionic liquids (ILs) are a relatively new class of solvents and expected to be used increasingly

19

by the chemical industry in the coming years as replacements for volatile organic solvents. 1,2

20

Due to their low volatility, non-flammability, and thermal stability, ILs have the potential

21

to be applied in a variety of processes ranging from synthesis and catalysis to energy and

22

remediation technologies as well as analytical applications. 2–4 Most ILs are composed of

23

an organic or inorganic anion and a quaternary ammonium cation, such as imidazolium,

24

pyridinium, pyrrolidinium, piperidinium, or tetraalkylammonium (cations used in this study

25

are shown in Figure 1). IL cations typically have asymmetric N -alkyl substituents with

26

side-chain lengths between 2 and 10 C atoms making them generally hydrophilic in nature. 5

27

Given their ionic structure and hydrophilic properties, IL cations are not likely to be readily

28

removed from aqueous systems through sorption and sedimentation. Consequently, if their

29

consumption volume increases, IL cations have the potential to become mobile environmen-

30

tal contaminants that can reach surface waters through accidental release into industrial

31

wastewater streams. 4

32

Biodegradation studies of IL cations have found that these chemicals, particularly those

33

with aromatic rings, are not readily biodegradable in conventional wastewater treatment sys-

34

tems. 6–8 For example, Romero et al. 8 showed that only 2−10% of imidazolium-based ILs are

35

removed after 10 days of incubation with a wastewater microbial consortia. Consequently,

36

the removal of IL cations will likely be minimal during biological wastewater treatment, and

37

IL cations could enter aquatic environments via industrial wastewater effluents. While ex-

38

posure of surface waters to IL cations is likely to increase with usage volumes, the potential

39

fate and impact of these compounds on aquatic environments is still unclear. Recent toxicity

40

studies demonstrate a significant variability of effects on aquatic organisms by different ILs

41

with EC50 -values ranging from nM to mM concentrations, 4,9,10 whereas fate processes of IL

42

cations in surface waters have received very little attention. Physical processes are unlikely to

43

impact the fate of IL cations in surface waters given the generally weak adsorption affinities 3 ACS Paragon Plus Environment

Environmental Science & Technology

N

N N

1-ethyl-3-methylimidazolium (C2-imidazolium)

N

N N

C2H5

Page 4 of 27

C4H9

1-butyl-3-methylimidazolium (C4-imidazolium)

N

C6H13

1-hexyl-3-methylimidazolium (C6-imidazolium)

N N

C8H17

1-octyl-3-methylimidazolium (C8-imidazolium)

N

N

C10H21

1-decyl-3-methylimidazolium (C10-imidazolium)

N

N C4H9

C4H9

1-butylpyridinium (C4-pyridinium)

1-butyl-1-methylpyrrolidinium (C4-pyrrolidinium)

C4H9 1-butyl-1-methylpiperidinium (C4-piperidinium)

Figure 1. Structures of all IL cations investigated in this study with abbreviations in parenthesis. 44

of imidazolium, pyridinium, and tetraalkylammonium cations for aquatic sediments, natural

45

soils, and bacterial cells. 11–13 Consequently, chemical and biological transformation processes

46

are key to assessing IL cations in the environment. Studies on the biological transformation

47

of IL cations in natural systems are scarce. The persistence of quaternary ammonium com-

48

pounds towards degradation in sewage sludge communities, however, suggests that biological

49

transformations will play a minor role. 14,15 The importance of abiotic transformation pro-

50

cesses, in particular the direct and indirect photolysis, of IL cations in sunlit surface waters

51

is still essentially unexplored.

52

Due to the lack of efficient removal from sorption and biodegradation, photochemical

53

transformation processes could have a significant effect on the fate of IL cations in surface

54

water. So far, only Calza and co-workers 16,17 have investigated the direct and indirect

55

photochemical transformation of 1-ethyl-1-methylimidazolium and three pyridinium cations

56

under environmental conditions. They concluded that direct photochemical transformation

4 ACS Paragon Plus Environment

Page 5 of 27

Environmental Science & Technology

57

will play a major role in most cases and that half-lives of 1-ethyl-1-methylimidazolium in

58

surface waters can be up to one month. Unfortunately, experiments to study the effect

59

of indirect photochemical transformations were only performed with model sensitizers. In

60

natural waters, the presence of dissolved organic matter (DOM) will influence the relative

61

importance of direct and indirect photochemical processes. Additionally, indirect processes

62

may be enhanced via association with DOM, as shown for charged amines. 18

63

To improve the general assessment of the fate of IL cations in sunlit surface water,

64

this study aimed to determine the direct and indirect photochemical transformation rates

65

of a set of IL cations with imidazolium, pyridinium, pyrrolidinium, and piperidinium core

66

structures and varying side-chain length. We performed laboratory-scale irradiation exper-

67

iments with model sensitizers for different photochemically-produced reactive intermediates

68

(PPRIs), namely hydroxyl radicals, singlet oxygen, and carbonate radicals, as well as exper-

69

iments with DOM isolates and natural water samples. From this data set, we determined

70

bimolecular reaction rate constants of the selected IL cations with hydroxyl radicals, the

71

relative contribution of direct and indirect photochemical transformation of IL cations in

72

natural water samples, and estimations of half-lives in sunlit surface waters under environ-

73

mental conditions.

74

Experimental Section

75

All chemicals used as well as details on organic matter isolate solutions and natural water

76

samples are described in the Supporting Information (SI). IL cations will be abbreviated

77

with the number of C-atoms in the side-chain and the name of the ring structure, e.g.,

78

C2 -imidazolium for 1-ethyl-3-methylimidazolium (see Figure 1).

5 ACS Paragon Plus Environment

Environmental Science & Technology

79

Irradiation experiments in solutions with dissolved organic matter

80

Photolysis experiments were performed in solutions containing DOM to determine (i) the

81

relevance of direct photochemical transformation of IL cations in natural waters, (ii) the im-

82

portance of PPRIs for the indirect photochemical transformation of IL cations, and (iii) any

83

influence of association with dissolved organic matter on the photochemical transformation

84

rates of IL cations. All experiments were run in an Atlas Suntest CPS+ solar simulator

85

equipped with a xenon arc lamp and a 290-nm cutoff filter. As in previous studies, 19–21 test

86

tubes were irradiated at an angle of approximately 30° from horizontal. A light intensity of

87

765 W m−2 and 10-mL quartz test tubes (13 mm o.d., 11 mm i.d.) sealed with aluminum

88

foil and rubber caps were used for all experiments with DOM.

89

Experiments with Suwannee River Fulivc Acid (SRFA), Suwannee River Natural Organic

90

Matter (SRNOM), Mississippi river water and wastewater effluent were conducted with C6 -

91

imidazolium (see SI for details on DOM isolates and water samples). In addition, experiments

92

with Suwannee River Humic Acid (SRHA) were performed with all imidazolium as well

93

as with C4 -pyridinium cations. DOM isolates were diluted with phosphate buffer (5 mM

94

NaH2 PO4 , pH 7.0) until the transmittance at 290 nm was above 50%. Each test tube was

95

filled with 6 mL of a buffered DOM isolate solution or a natural water sample and spiked

96

with 10 µM of one IL and, in experiments assessing the role of hydroxyl radical, 10 µM

97

benzoic acid. Control experiments accounting for direct photochemical transformation and

98

non-photochemical loss were run for all IL cations. Direct photolysis controls were performed

99

in ultrapure water containing the IL and benzoic acid. Dark controls were run in test tubes

100

wrapped in aluminum foil containing SRHA solutions with the IL and benzoic acid. An

101

experiment with the hydroxyl radical quencher 2-propanol was performed in buffered solution

102

containing SRFA, 10 µM C6 -imidazolium, 10 µM benzoic acid, and 26 mM 2-propanol.

103

During all experiments, sub-samples of 0.5 mL were withdrawn from the test tubes at regular

104

time intervals and transferred into 1.5-mL amber glass HPLC vials with crimp caps. For

105

samples containing benzoic acid, 10 µL 2 M HCl was added to each HPLC vial to decrease 6 ACS Paragon Plus Environment

Page 6 of 27

Page 7 of 27

Environmental Science & Technology

106

the pH to ≤ 2. Pseudo-first order rate constants were determined as the slopes of linear

107

regressions of log-normalized concentrations, ln (c/c0 ), vs. reaction time. Differences in

108

pseudo-first order rate constants between DOM experiments and direct photolysis controls

109

were attributed to indirect photochemical transformations and differences in pseudo-first

110

order rate constants between direct photolysis and dark controls were attributed to direct

111

photochemical transformation.

112

Irradiation experiments with model sensitizers

113

Experiments to determine the reactivity of IL cations towards hydroxyl radicals, singlet

114

oxygen, and carbonate radicals were conducted in 10-mL borosilicate glass tubes (13 mm o.d.,

115

11 mm i.d.) sealed with aluminum foil and rubber caps according to previously published

116

procedures. 22–25 All experiments were run in duplicate in the solar simulator at a light

117

intensity of 765 W m−2 (hydroxyl and carbonate radical experiments) or 350 W m−2 (singlet

118

oxygen experiments). For hydroxyl radical experiments, test tubes were initially filled with

119

6 mL phosphate buffer (10 mM NaH2 PO4 , pH 7.0), 10 µM imidazolium or pyridinium IL,

120

10 µM benzoic acid, and 1 mM H2 O2 . Control experiments were run without H2 O2 in the

121

solar simulator for all IL cations. A dark control in the presence of H2 O2 was performed

122

with C6 -imidazoium. Equivalent experiments with C4 -pyrrolidinium or C4 -piperidinium were

123

performed with initial IL concentrations of 20 µM. Singlet oxygen experiments were run in

124

a similar way with test tubes containing 6 mL phosphate buffer (10 mM NaH2 PO4 , pH 7.0),

125

10 µM imidazolium or pyridinium IL, and 10 µM rose bengal. Control experiments were run

126

without rose bengal. Carbonate radical experiments were performed in test tubes filled with

127

6 mL carbonate buffer (0.7 M NaHCO3 , pH 10), 10 µM C6 -imidazolium, and 1 mM H2 O2 .

128

Temporally resolved samples were taken as described above, but shorter time intervals were

129

used due to faster reactions. Bimolecular reaction rate constants of IL cations with hydroxyl

130

radicals (ki,HO• ) were derived from linear regressions of log-normalized concentrations of IL

131

cations vs. benzoic acid as shown in eq 1 7 ACS Paragon Plus Environment

Environmental Science & Technology

 ln

[IL] [IL]0

 =

ki,HO• kBZA,HO•

 · ln

[BZA] [BZA]0

Page 8 of 27

 (1)

132

where kBZA,HO• is the bimolecular reaction rate constant of benzoic acid at pH 7 with hydroxyl

133

radicals ((5.93 ± 0.55) · 109 M−1 s−1 ), 26 [IL] and [BZA] are concentrations of an IL cation and

134

benzoic acid measured in the same sub-sample, and [IL]0 and [BZA]0 are initial concentrations

135

of the IL cation and benzoic acid, respectively. The ratio of measured pseudo-first order rate

136

constant and bimolecular reaction rate constant of benzoic acid and furfuryl alcohol was

137

used to calculate steady-state concentrations of hydroxyl radicals ([OH]ss ) and singlet oxygen

138

([1 O2 ]ss ), respectively. A bimolecular reaction rate constant of 8.3 · 107 M−1 s−1 was used

139

for the reaction of furfuryl alcohol with singlet oxygen. 22,27 No probe compound was used to

140

quantify the steady-state concentration of carbonate radicals. Carbonate radicals are more

141

stable in water than hydroxyl radicals and steady-state concentrations of carbonate radicals

142

are expected to be slightly higher than steady-state concentrations of hydroxyl radicals under

143

the same experimental conditions. 28,29

144

Analytical methods

145

Ultraviolet-visible light absorption spectra were measured with a Shimadzu UV-1601PC

146

spectrophotometer using 1-cm quartz cuvettes. Concentrations of imidazolium and pyri-

147

dinium cations and benzoic acid were determined by high-pressure liquid chromatography

148

(HPLC) on an Agilent 1100 LC system with a variable wavelength detector set to 210 nm.

149

For most analyses, an Eclipse XDB C-18 column (4.6 × 150 mm, 3.5 µm, Agilent) was used

150

with isocratic mixtures or step gradients of methanol and water (both containing 10 mM

151

ammonium acetate and 0.1% acetic acid) at a flow rate of 1 mL min−1 . Eluent mixtures

152

were chosen to enable quantification of IL cations and benzoic acid in the same run, which

153

was possible for all compounds except for C2 -imidazolium. C2 -imidazolium was measured

8 ACS Paragon Plus Environment

Page 9 of 27

Environmental Science & Technology

154

separately on a Zorbax SB-C8 column (4.6 × 150 mm, 3.5 µm, Agilent) with 1 ml min−1

155

of 50% acetonitrile and 50% 10 mM ammonium acetate in water. Injection volumes were

156

100 µL for all measurements. C4 -pyrrolidinium and C4 -piperidinium were quantified with a

157

Hewlett Packard 1050 HPLC system coupled to a Hewlett Packard 1100 MSD mass spec-

158

trometer operated in positive electrospray ionization and selective ion monitoring mode. A

159

BetaSil C-18 column (2.1 × 50 mm, 5 µm, Thermo Scientific) was used with an 8-min linear

160

gradient from 70% to 0% 2 mM ammonium acetate in water and 30% to 100% methanol at

161

a flow rate of 200 µL min−1 . Injection volumes were 5 µL and the monitored mass-to-charge

162

(m/z) ratios were 142 and 156. All standards and samples containing C4 -pyrrolidinium were

163

amended with 10 µM C4 -piperidinium as an internal standard and vice versa.

164

Results and Discussion

165

Photochemical transformation of C6 -imidazolium in natural water

166

samples

167

The photochemical transformation of C6 -imidazolium, an IL cation with a medium-sized

168

alkyl side-chain, was studied under simulated sunlight in the presence and absence of DOM.

169

Figure 2 shows the time course of normalized C6 -imidazolium concentrations during irra-

170

diation in ultapure water and in three DOM-containing samples. The photodegradation

171

half-life of C6 -imidazolium was 345 ± 20 h in ultrapure water, 65 ± 7 h in wastewater ef-

172

fluent, 40 ± 3 h in Mississippi river water, and 24 ± 1 h in a buffered solution containing

173

SRFA. The significant increase of transformation rates in the presence of DOM indicates

174

an important contribution of indirect photochemical processes towards the transformation

175

of C6 -imidazolium. Estimates from comparing pseudo-first order reaction rate constants

176

suggest that the contribution of indirect photochemical transformation ranged between 81%

177

and 94%.

178

Because of the small contribution of direct photolysis, the differences in photochemical 9 ACS Paragon Plus Environment

Environmental Science & Technology

-1

1.0

kobs = 0.0020 ± 0.0003 h

0.8 -1

kobs = 0.013 ± 0.002 h

c/c0

0.6

0.4

0.2

ultrapure water wastewater effluent Mississippi river water SRFA solution

0.0 0

10

20

-1

kobs = 0.023 ± 0.001 h

-1

kobs = 0.029 ± 0.001 h

30 40 time (h)

50

60

Figure 2. Time-trend of normalized concentrations (c/c0 ) of C6 -imidazolium during irradiation experiments in ultrapure water (green circles), wastewater effluent (yellow triangles), Mississippi river water (turquoise diamonds), and buffered SRFA solution (red squares). Solid lines represent nonlinear regressions of first-order reaction kinetics with the fitting parameter kobs . The faster rate of loss in the organic matter-containing solutions indicates a role for indirect photolysis.

179

transformation rates observed in the three DOM-containing samples cannot be explained

180

by light screening. In fact, the light transmission was lower for the SRFA solution than

181

for the Mississippi river and wastewater effluent samples (see Figure S2). The differences in

182

photochemical transformation rates were likely caused by varying steady-state concentrations

183

of PPRIs formed in the presence of different DOM sources, but photodegradation rates of C6 -

184

imidazolium did not correlate with dissolved organic carbon (DOC) concentrations (SRFA:

185

7.0 mg/L, Mississippi: 7.0 mg/L, effluent: 10.4 mg/L) or pH values (SRFA: 7.0, Mississippi:

186

8.3, effluent: 8.2). It has previously been shown, however, that the quality of DOM has

187

an important influence on PPRI steady state concentrations. 21 Considering that the three

188

types of DOM used in our experiments have very different sources, the composition of DOM

189

molecules is likely very different in these three samples.

190

Role of PPRIs in the photochemical transformation of IL cations

191

As described above, transformation of C6 -imidazolium under simulated sunlight indicates an

192

important role of indirect photochemical processes. To identify which PPRIs contribute to 10 ACS Paragon Plus Environment

Page 10 of 27

Page 11 of 27

Environmental Science & Technology

193

increased transformation rates of C6 -imidazolium in solutions containing DOM, we performed

194

irradiation experiments with model sensitizers to generate specific PPRIs. Figure 3a shows

195

normalized concentrations of C6 -imidazolium in the presence of hydroxyl radicals, singlet O2 ,

196

and carbonate radicals. No significant change in C6 -imidazolium concentration was observed

197

over the course of 50 min in experiments with singlet O2 and carbonate radicals. In fact, none

198

of the imidazolium or pyridinium cations showed significant reactivity towards singlet O2 (see

199

Figure S3). Singlet O2 is known to efficiently react with compounds containing imidazole

200

rings, such as the amino acid histidine, 30 however, the presence of a positive charge in the

201

ring seems to hinder the same reaction with imidazolium cations. Carbonate radicals react

202

selectively with organic compounds, particularly with aromatic amines, phenolates, and keto

203

groups. 28 The fact that C6 -imidazolium does not contain such functional groups explains the

204

lack of reactivity towards carbonate radicals.

205

In contrast to the persistence towards singlet O2 and carbonate radicals, a significant

206

decrease in concentration was observed in the presence of hydroxyl radicals for all tested IL

207

cations (see Figures S3-4). The half-life of C6 -imidazolium transformation in the presence of

208

hydroxyl radicals ([OH]ss = (4.6 ± 1.0) · 10−14 M) as shown in Figure 3a was 49.8 ± 0.8 min.

209

Despite the fact that hydroxyl radicals react unselectively with most organic compounds,

210

distinct variations in transformation rates were observed for the different IL cations (see

211

Figures S3-4). All imidazolium cations were transformed faster than C4 -pyridinium, C4 -

212

pyrrolidinium, and C4 -piperidinium, which had the longest half-life of 196±6 min. Half-lives

213

of the imidazolium cations ranged from 77 ± 6 min for C2 -imidazolium to 32.9 ± 0.2 min

214

for C10 -imidazolium. These results suggest that (i) imidazolium rings were more reactive

215

towards hydroxyl radicals than the other ring structures and (ii) elongation of the side-chain

216

increases the reactivity towards hydroxyl radicals.

217

In addition to reactive oxygen species and radical PPRIs, excited triplet states of DOM

218

play an important role in the indirect photochemical transformation of many organic com-

219

pounds. 31 In experiments with the model DOM isolates SRFA, SRHA, and SRNOM, photo-

11 ACS Paragon Plus Environment

Environmental Science & Technology

a)

Page 12 of 27

1.0

0.8

c/c0

0.6

0.4

-1

kobs = 0.016 ± 0.001 min

0.2

hydroxyl radicals singlet oxygen carbonate radicals

0.0 0

b)

10

20 30 time (min)

40

50

1.0

0.8

c/c0

0.6

0.4 SRFA + 2-PrOH ultrapure water SRNOM SRFA SRHA

0.2

0.0 0

10

20 30 time (h)

-1

kobs = 0.027 ± 0.002 h

-1

kobs = 0.029 ± 0.001 h

-1

kobs = 0.029 ± 0.001 h

40

50

Figure 3. (a) Time-trend of normalized concentrations (c/c0 ) of C6 -imidazolium during irradiation experiments with model sensitizers producing hydroxyl radicals (blue circles), singlet oxygen (red triangles), and carbonate radicals (green squares). Solid lines represent nonlinear regressions of first-order reaction kinetics with the fitting parameter kobs . Steady-state concentrations of the reactive species were approximately 5 · 10−14 M for hydroxyl radicals, 2 · 10−11 M for singlet oxygen, and ≥ 10−13 M for carbonate radicals. Transformation of C6 -imidazolium would be detected with bimolecular reaction rate constants of ≥ 106 M−1 s−1 and ≥ 108 M−1 s−1 for singlet oxygen and carbonate radicals, respectively. (b) Time-trend of normalized concentrations (c/c0 ) of C6 -imidazolium during irradiation experiments with SRFA (red squares), SRFA and 2-propanol (green triangles), SRNOM (blue diamonds), SRHA (yellow circles), and ultrapure water (green circles). Solid lines represent nonlinear regressions of firstorder reaction kinetics with the fitting parameter kobs . The difference between transformation rates in ultrapure water and SRFA solutions containing 2-propanol illustrates the effect of light screening by DOM on the direct photolysis of C6 -imidazolium.

12 ACS Paragon Plus Environment

Page 13 of 27

Environmental Science & Technology

220

chemical transformation of C6 -imidazolium was observed at comparable rates with half-lives

221

of 20.3 ± 2.5 to 23.7 ± 0.8 h (see Figure 3b). Because reactions with singlet O2 and carbonate

222

radicals can be excluded, only reactions with hydroxyl radicals and excited triplet states of

223

DOM can account for C6 -imidazolium photodegradation. To verify whether only hydroxyl

224

radicals or both PPRIs were responsible for C6 -imidazolium photodegradation, a second irra-

225

diation experiment with SRFA was performed with 26 mM 2-propanol, which is widely used

226

to quench hydroxyl radicals. 19,29 As apparent from Figure 3b, addition of 2-propanol to the

227

SRFA solution completely inhibited transformation of C6 -imidazolium, which suggests that

228

in DOM-containing solutions, only hydroxyl radicals play a significant role in the indirect

229

photolysis of IL cations. These results are, in part, contrasting to previous results pub-

230

lished for C2 -imidazolium and C4 -pyridinium, where reactions were reported to occur with

231

hydroxyl radicals, singlet O2 , and triplet excited states of antraquinone-2-sulfonic acid. 16,17

232

While reaction rates with triplet excited states of antraquinone-2-sulfonic acid cannot be

233

compared directly to the DOM used in our study, the bimolecular reaction rate constants

234

with singlet oxygen determined in these previous studies were small (0.9 − 1.5 · 106 M−1

235

s−1 ) 16,17 compared to other organic contaminants and associated with large errors.

236

Bimolecular reaction rates of IL cations with hydroxyl radicals

237

We determined bimolecular reaction rate constants of IL cations with hydroxyl radicals

238

(ki,OH ) in competition experiments with benzoic acid using either hydrogen peroxide or SRHA

239

as a source for hydroxyl radicals. Dark controls with both hydrogen peroxide and SRHA

240

showed no significant changes in IL cation concentrations (see Figures S3 and S5-10). Figure

241

4a shows log-normalized concentrations of the 5 imidazolium cations vs. log-normalized

242

concentrations of benzoic acid. From the slope of the linear regressions shown in Figure 4a

243

and Figures S5-10 and the bimolecular reaction rate constant of benzoic acid, ki,OH -values

244

were calculated for all IL cations. ki,OH -values of imidazolium cations increased with the

245

number of C atoms in the side-chain from (2.57 ± 1.10) · 109 M−1 s−1 for C2 -imidazolium to 13 ACS Paragon Plus Environment

Environmental Science & Technology

246

(8.47 ± 0.97) · 109 M−1 s−1 for C10 -imidazolium (see Table 1). This trend is illustrated in

247

Figure 4b where ki,OH -values are plotted against the number of aliphatic C-atoms in each

248

molecule. Values of ki,OH for C4 -pyridinium, C4 -pyrrolidinium, and C4 -piperidinium were all

249

approximately 2 · 109 M−1 s−1 and thus smaller than the values of any imidazolium cation

250

(see Table 1 and Figure 4b).

251

Reactions of hydroxyl radicals with organic compounds are unspecific and occur at various

252

functional groups. H-atom abstraction from aliphatic C–H bonds and OH-group addition

253

to aromatic rings are commonly observed. 26 The trend of increasing ki,OH -values with side-

254

chain length observed for the imidazolium cations suggests that H-atom abstractions at C–H

255

bonds in the side-chain contributes significantly to the overall reaction of IL cations with

256

hydroxyl radicals. This trend is, however, less pronounced than in the case of simple alkanes

257

where only H-atom abstraction at C–H bonds occurs. ki,OH -values from Buxton et al. 26

258

show a stronger increase per additional C-atom for the alkane series from methane to octane

259

than ki,OH -values for imidazolium cations (see Figure 4b). The fact that the three other

260

IL cations all have smaller ki,OH -values than the smallest imidazolium cation suggests that

261

reactions with hydroxyl radicals might occur both at the side-chain and the ring in the case

262

of imidazolium cations. In agreement with our findings, reaction products identified for

263

N -alkylpyridinium cations under photocatalytic conditions with TiO2 were predominantly

264

side-chain hydroxylations. 32

265

While bimolecular reaction rate constants determined by Calza and co-workers for C2 -

266

imidazolium (2·1010 M−1 s−1 ) 16 and C4 -pyridinium (3·108 M−1 s−1 ) 17 with hydroxyl radicals

267

are in agreement with our results considering experimental uncertainties, other studies using

268

H2 O2 and UV-irradiation or Fenton reactions have observed very different trends in reactiv-

269

ity of imidazolium cations towards hydroxyl radicals. Siedlecka and Stepnowski 33 and Step-

270

nowski and Zaleska 34 have investigated the removal rates of C4 -, C6 -, and C8 -imidazolium in

271

Fenton and UV/H2 O2 systems, respectively, and observed increasing removal with decreas-

272

ing side-chain length. Although these results contrast with the trend in bimolecular reaction

14 ACS Paragon Plus Environment

Page 14 of 27

Page 15 of 27

Environmental Science & Technology

a)

-1.2

C2-imidazolium C4-imidazolium C6-imidazolium C8-imidazolium C10-imidazolium

ln(c/c0) for IL cations

-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.0

b)

12

-1.2

8

9

-1

-1

-0.4 -0.6 -0.8 -1.0 ln(c/c0) for benzoic acid

alkanes imidazolium cations C4-pyridinium C4-pyrrolidinium C4-piperidinium

10

kHO• (10 M s )

-0.2

6 4 2 0 0

c)

-1.2

4 6 8 aliphatic C-atoms

10

12

H2O2 in buffer SRFA in buffer Mississippi water Waste water effluent

-1.0 ln(c/c0) for C6-imidazolium

2

-0.8 -0.6 -0.4 -0.2 0.0 0.0

-0.2

-0.4 -0.6 -0.8 -1.0 ln(c/c0) for benzoic acid

-1.2

Figure 4. (a) Log-normalized concentrations of imidazolium cations vs. benzoic acid from irradiation experiments with H2 O2 . (b) Second order reaction rate constants with hydroxyl radicals of imidazolium cations (green circles), C4 -pyridinium (red diamonds), C4 -pyrrolidinium (blue triangle), C4 -piperidinium (purple hexagon), and alkanes (yellow squares). Error bars are 95%-confidence intervals and values for alkanes are from Buxton et al. 26 (c) Log-normalized concentrations of C6 -imidazolium vs. benzoic 15 ACSHParagon Plus Environment acid from irradiation experiments with O (blue circles), SRFA (red squares), Mississippi river water 2 2 (turquoise hexagons) and wastewater effluent (yellow triangles).

Environmental Science & Technology

273

rate constants observed in our study, it is difficult to compare the results from our experi-

274

ments with these two studies. First, experimental procedures were different in that different

275

light sources, different initial IL cation concentrations, and different hydroxyl radical sources

276

were used. Second, removal percentages and bimolecular reaction rate constants cannot be

277

compared directly without knowledge of the steady-state concentrations of hydroxyl rad-

278

icals, which have not been determined by Siedlecka and Stepnowski 33 or Stepnowski and

279

Zaleska 34 .

280

In addition to experiments with the model sensitizer H2 O2 , we performed a series of com-

281

petition experiments with C6 -imidazolium and benzoic acid in various solutions containing

282

DOM of different origins. Different distributions of C6 -imidazolium and benzoic acid can be

283

expected in solutions containing DOM because of the opposite charges of these molecules. IL

284

cations are permanently positively charged and have been shown to strongly associate with

285

the negatively charged moieties of DOM. 35,36 For benzoic acid, however, no interaction with

286

DOM is expected because at pH 7 benzoic acid is predominantly (99.8%) present as the nega-

287

tively charged benzoate. The formation mechanism of hydroxyl radicals in DOM-containing

288

waters is still incompletely understood, but DOM will act both as an important source and

289

the predominant scavenger of hydroxyl radicals. 29,37 Thus, unlike excited triplet states of

290

DOM and singlet oxygen, which are concentrated around DOM molecules, hydroxyl radicals

291

could be homogeneously distributed in DOM-containing solutions. As a result, identical

292

relative transformation rates of IL cations and benzoic acid in experiments with and without

293

DOM are an indication that indeed only reactions with hydroxyl radicals are relevant for

294

IL cations in DOM-containing solutions. As shown in Figure 4c, the relative transformation

295

rates of C6 -imidazolium and benzoic acid varied only minimally in experiments with H2 O2 ,

296

SRFA, Mississippi river water, and wastewater effluent. Likewise, no significant differences

297

were observed between experiments with H2 O2 and SRHA for all other IL cations except

298

for C8 -imidazolium (see Figures S5-10). ki,OH -values were, however, identical within error

299

(overlapping confidence intervals) in all experiments (see Table 1), even for C8 -imidazolium,

16 ACS Paragon Plus Environment

Page 16 of 27

Page 17 of 27

Environmental Science & Technology

300

suggesting that transformation of IL cations in natural sunlit surface waters are estimated

301

accurately from laboratory experiments with H2 O2 and average steady-state concentrations

302

of hydroxyl radicals. Table 1. Bimolecular reaction rate constants of IL cations with hydroxyl radicals in experiments with either H2 O2 or SRHA, ratios of pseudo-first order reaction rate constants from direct photolysis controls (kdir ) and experiments with SRHA (kSRHA ), and estimated half-lives in surface water. a

IL cation

ki,HO• (H2 O2 ) ki,HO• (SRHA) (109 M−1 s−1 )

kdir b kSRHA

T1/2 c

(%)

(days)

(109 M−1 s−1 )

C2 -imidazolium

2.57 ± 1.10

2.91 ± 0.52

13 ± 24

107 ± 46

C4 -imidazolium

3.98 ± 0.92

3.30 ± 1.70

7±6

69 ± 16

C6 -imidazolium

5.81 ± 0.94

6.30 ± 0.82 d

7±3

47 ± 8

C8 -imidazolium

8.05 ± 1.14

6.26 ± 1.31

19 ± 15

34 ± 5

C10 -imidazolium

8.47 ± 0.97

8.42 ± 1.29

10 ± 7

32 ± 4

C4 -pyridinium

2.04 ± 0.37

1.77 ± 0.60

19 ± 19

135 ± 25

C4 -pyrrolidinium

2.22 ± 0.77

124 ± 43

C4 -piperidinium

2.14 ± 0.96

129 ± 58

a

Errors are given as 95%-confidence intervals;

b

Rates are corrected for concentration changes in dark controls (see Figures S5-10);

c

Calculated from ki,HO• (H2 O2 ) assuming [OH]ss = 10−16 M and 7 hours of sunshine per day;

d

ki,HO• (SRFA) = 6.10 ± 0.95 · 109 M−1 s−1 , ki,HO• (SRNOM) = 6.05 ± 0.65 · 109 M−1 s−1 , ki,HO• (Mississippi river water) = 6.27 ± 0.87 · 109 M−1 s−1 ,

303

ki,HO• (wastewater effluent) = 6.71 ± 0.88 · 109 M−1 s−1

17 ACS Paragon Plus Environment

Environmental Science & Technology

304

Predicted fate of ionic liquid cations in sunlight surface water

305

We estimated the relative contribution of direct and indirect photochemical degradation of

306

imidazolium and pyridinium cations under environmental conditions from the comparison

307

of pseudo-first order reaction rate constants in experiments with SRHA and ultrapure water

308

kdir ) under simulated sunlight. Note that because of the slow direct photolysis of IL ( kSRHA

309

cations, the resulting relative contributions of direct photolysis to the overall photochemical

310

transformation are associated with large uncertainties (see Table 1). These values should

311

be considered as an upper limit because direct photolysis in DOM-containing waters will be

312

smaller than in ultrapure water due to light screening (see Figure 3b). All reaction rates

313

were corrected for changes in concentrations observed in dark controls, which were generally

314

negligible or slightly increasing over time likely due to evaporation of small amounts of

315

water. In the case of C8 - and C10 -imidazolium, however, concentrations decreased slightly

316

over time in dark controls (see Figures S9-10) indicating a slow thermal decay of the cations

317

caused by the elevated temperatures during irradiation experiments. Rate constants of

318

imidazolium cations in the presence of SRHA increased with the number of C atoms in

319

the side-chain as shown in Figure 5 in agreement with the fact that bimolecular reaction

320

rate constants of imidazolium cations with hydroxyl radicals increased accordingly. Rate

321

constants in ultrapure water, however, showed no relationship with the side-chain length and

322

the direct photolysis rate constant of C4 -pyridinium was within the range of the imidazolium

323

cations.

324

The relative importance of indirect photochemical processes, which is dominated by reac-

325

tions with hydroxyl radicals, was smallest for C8 -imidazolium (81 ± 15%) and C4 -pyridinium

326

(81 ± 19%) and largest for C4 - and C6 -imdazolium with 93 ± 6% and 93 ± 31%, respectively

327

(see Table 1). These results show that reactions with hydroxyl radicals dominate the fate

328

of IL cations in sunlit surface waters independent of side-chain length and ring structure.

329

Direct photochemical transformation of C4 -pyrrolidnium and C4 -piperidinium are expected

330

to be negligible due to a lack of absorption of these compounds at wavelengths above 200 18 ACS Paragon Plus Environment

Page 18 of 27

Page 19 of 27

Environmental Science & Technology

331

nm (see Figure S1). Our findings are in contrast to results from Calza et al., 16,17 which indi-

332

cated that direct photolysis of C2 -imidazolium and C4 -pyridinium corresponds to 60 − 90%

333

and 20 − 90%, respectively, of the total photodegradation in surface waters. Direct pho-

334

tochemical transformation rates were very slow in both this and previous studies, which is

335

explained by the minimal overlap between the absorption spectra of imidazolium and pyri-

336

dinium cations with the solar emission spectrum. In fact, we could only observe absorption

337

of C2 -imidazolium between 290 and 400 nm when elevated concentrations (> 5 mM) were

338

used (see Figure S1). It should be noted, however, that molar absorption coefficients at 300

339

nm amounted only to approximately 1.0 M−1 cm−1 for C2 -imidazolium determined in this

340

study and 1.5 M−1 cm−1 for C4 -pyridinium determined by Calza et al. 17 . Slightly different

341

absorption spectra were observed for C2 -imidazolium in this study and by Calza et al. 16

342

but molar absorptivity is small in both cases and should not lead to drastically different

343

results. The fact that estimated contributions of direct photochemical transformation are

344

significantly different in this and previous studies is explained by the different experimental

345

approaches. Calza et al. 16,17 used a modeling approach to extrapolate direct and indirect

346

photochemical transformation rates determined in model experiments without DOM to envi-

347

ronmental conditions, which lead to the reported high contributions of direct photolysis. In

348

our study, however, transformation rates of IL cations in solutions with and without DOM

349

are compared under the same irradiation conditions, revealing the dominant contribution of

350

indirect photolysis for all tested IL cations.

351

Because the overlap between the absorption spectra of IL cations and the emittance

352

spectrum of the lamp in the solar simulator were very small, we refrained from determining

353

quantum yields for the tested IL cations. Instead, bimolecular reaction rate constants with

354

hydroxyl radicals were used to extrapolate our results to relevant environmental conditions.

355

Table 1 shows estimations of half-lives for all tested IL cations for sunlit surface waters

356

assuming a steady-state concentration of hydroxyl radicals of 1 · 10−16 M and an average

357

of 7 hours of daily sunshine. Steady-state hydroxyl radical concentrations in sunlit surface

19 ACS Paragon Plus Environment

Environmental Science & Technology

1.4 1.2

-1

k (days )

1.0 0.8 0.6 0.4 0.2

C4-pyridinium

C10-imidazolium

C8-imidazolium

C6-imidazolium

C4-imidazolium

C2-imidazolium

0.0

Figure 5. Pseudo-first order reaction rate constants of imidazolium and pyridinium cations in buffered SRHA solution (red striped bars) and ultrapure water (blue solid bars). Error bars are 95%-confidence intervals.

358

waters can vary substantially (10−17 − 10−15 M). 29,38 Our results should reflect an average

359

estimate of the half-lives in the photic zone of a surface water, but lower hydroxyl radical

360

concentrations would lead to (dramatically) longer half-lives. The fastest transformation

361

is expected for C10 -imidazolium with an estimated half-live of 32 ± 4 days (Table 1). C4 -

362

pyridinium, C4 -pyrrolidinium, and C4 -piperidinium are the most recalcitrant IL cations with

363

estimated half-lives between 124 ± 43 and 135 ± 25 days (Table 1). The estimated half-

364

life of C2 -imidazolium was 107 ± 46 days (see Table 1), which is significantly larger than

365

values determined by Calza et al. 16 (5 − 30 days). In their follow-up study, Calza et al. 17

366

have shown, however, that their modeled half-lives in surface water can vary drastically

367

depending on water depth, DOC concentrations, and type of DOM. The differences in half-

368

lives determined in this and previous studies are a consequence of the different experimental

369

and modeling approaches. Nevertheless, with half-lives on the order of days to months, a

370

common conclusion is reached that photochemical transformation of IL cations in surface

20 ACS Paragon Plus Environment

Page 20 of 27

Page 21 of 27

371

Environmental Science & Technology

waters will be slow compared to other organic contaminants.

372

The fact that IL cations are poorly or only slowly removed by both biological treatment

373

in wastewater treatment systems and photodegradation processes in sunlit surface waters

374

highlights the potential of these compounds to become persistent aquatic contaminants. If

375

the production volumes and usage of ILs by the chemical industry increase as predicted, care

376

should be taken to avoid release of IL cations into the environment.

377

Acknowledgement

378

This work was supported by the Swiss National Science Foundation (Early Postdoc.Mobility

379

fellowship 168888 to SGP) and the Joseph T. and Rose S. Ling Professorship (to WAA).

380

Thanks to Yousof Aly and Michael McCarty for their support with LC/MS analysis and

381

Meghan O’Connor for providing the wastewater effluent sample.

382

Supporting Information Available

383

Chemicals, organic matter isolates, natural water samples, and additional figures. This

384

information is available free of charge via the Internet at http://pubs.acs.org.

21 ACS Paragon Plus Environment

Environmental Science & Technology

385

386

387

388

389

References (1) Chiappe, C.; Pieraccini, D. Ionic liquids: Solvent properties and organic reactivity. J. Phys. Org. Chem. 2005, 18, 275–297. (2) Plechkova, N. V.; Seddon, K. R. Applications of ionic liquids in the chemical industry. Chem. Soc. Rev. 2008, 37, 123–150.

390

(3) Ho, T. D.; Zhang, C.; Hantao, L. W.; Anderson, J. L. Ionic liquids in analytical chem-

391

istry: Fundamentals, advances, and perspectives. Anal. Chem. 2013, 86, 262–285.

392

(4) Amde, M.; Liu, J.-F.; Pang, L. Environmental Application, Fate, Effects, and Concerns

393

of Ionic Liquids: A Review. Environ. Sci. Technol. 2015, 49, 12611–12627.

394

(5) Ropel, L.; Belv`eze, L. S.; Aki, S. N.; Stadtherr, M. A.; Brennecke, J. F. Octanol–water

395

partition coefficients of imidazolium-based ionic liquids. Green Chem. 2005, 7, 83–90.

396

(6) Jordan, A.; Gathergood, N. Biodegradation of ionic liquids – a critical review. Chem.

397

Soc. Rev. 2015, 44, 8200–8237.

398

(7) Docherty, K. M.; Dixon, J. K.; Kulpa Jr, C. F. Biodegradability of imidazolium and

399

pyridinium ionic liquids by an activated sludge microbial community. Biodegradation

400

2007, 18, 481–493.

401

402

(8) Romero, A.; Santos, A.; Tojo, J.; Rodriguez, A. Toxicity and biodegradability of imidazolium ionic liquids. J. Hazard. Mater. 2008, 151, 268–273.

403

(9) Bubalo, M. C.; Radoˇsevi´c, K.; Redovnikovi´c, I. R.; Halambek, J.; Srˇcek, V. G. A brief

404

overview of the potential environmental hazards of ionic liquids. Ecotoxicol. Environ.

405

Saf. 2014, 99, 1–12.

406

407

(10) Pham, T. P. T.; Cho, C.-W.; Yun, Y.-S. Environmental fate and toxicity of ionic liquids: A review. Water Res. 2010, 44, 352–372. 22 ACS Paragon Plus Environment

Page 22 of 27

Page 23 of 27

408

409

Environmental Science & Technology

(11) Beaulieu, J.; Tank, J.; Kopacz, M. Sorption of imidazolium-based ionic liquids to aquatic sediments. Chemosphere 2008, 70, 1320–1328.

410

(12) Stepnowski, P.; Mrozik, W.; Nichthauser, J. Adsorption of alkylimidazolium and

411

alkylpyridinium ionic liquids onto natural soils. Environ. Sci. Technol. 2007, 41, 511–

412

516.

413

(13) Gorman-Lewis, D. J.; Fein, J. B. Experimental study of the adsorption of an ionic liquid

414

onto bacterial and mineral surfaces. Environ. Sci. Technol. 2004, 38, 2491–2495.

415

(14) Modelli, A.; Sali, A.; Galletti, P.; Samor`ı, C. Biodegradation of oxygenated and non-

416

oxygenated imidazolium-based ionic liquids in soil. Chemosphere 2008, 73, 1322–1327.

417

(15) Neumann, J.; Grundmann, O.; Th¨oming, J.; Schulte, M.; Stolte, S. Anaerobic

418

biodegradability of ionic liquid cations under denitrifying conditions. Green Chem.

419

2010, 12, 620–627.

420

(16) Calza, P.; Vione, D.; Fabbri, D.; Aigotti, R.; Medana, C. Imidazolium-Based Ionic

421

Liquids in Water: Assessment of Photocatalytic and Photochemical Transformation.

422

Environ. Sci. Technol. 2015, 49, 10951–10958.

423

(17) Calza, P.; No, G.; Fabbri, D.; Santoro, V.; Minero, C.; Vione, D.; Medana, C. Pho-

424

toinduced transformation of pyridinium-based ionic liquids, and implications for their

425

photochemical behavior in surface waters. Water Res. 2017, 122, 194 – 206.

426

(18) Chu, C.; Lundeen, R. A.; Remucal, C. K.; Sander, M.; McNeill, K. Enhanced Indirect

427

Photochemical Transformation of Histidine and Histamine through Association with

428

Chromophoric Dissolved Organic Matter. Environ. Sci. Technol. 2015, 49, 5511–5519.

429

(19) Zeng, T.; Arnold, W. A. Pesticide photolysis in prairie potholes: Probing photosensi-

430

tized processes. Environ. Sci. Technol. 2013, 47, 6735–6745.

23 ACS Paragon Plus Environment

Environmental Science & Technology

431

(20) Karpuzcu, M. E.; McCabe, A. J.; Arnold, W. A. Phototransformation of pesticides in

432

prairie potholes: effect of dissolved organic matter in triplet-induced oxidation. Environ.

433

Sci.: Processes Impacts 2016, 18, 237–245.

434

(21) McCabe, A. J.; Arnold, W. A. Seasonal and spatial variabilities in the water chemistry

435

of prairie pothole wetlands influence the photoproduction of reactive intermediates.

436

Chemosphere 2016, 155, 640–647.

437

(22) Latch, D. E.; Stender, B. L.; Packer, J. L.; Arnold, W. A.; McNeill, K. Photochemical

438

fate of pharmaceuticals in the environment: cimetidine and ranitidine. Environ. Sci.

439

Technol. 2003, 37, 3342–3350.

440

441

(23) Larson, R. A.; Zepp, R. G. Reactivity of the carbonate radical with aniline derivatives. Environ. Toxicol. Chem. 1988, 7, 265–274.

442

(24) Janssen, E. M.; Marron, E.; McNeill, K. Aquatic photochemical kinetics of benzotriazole

443

and structurally related compounds. Environ. Sci.: Processes Impacts 2015, 17, 939–

444

946.

445

(25) Burns, J. M.; Cooper, W. J.; Ferry, J. L.; King, D. W.; DiMento, B. P.; McNeill, K.;

446

Miller, C. J.; Miller, W. L.; Peake, B. M.; Rusak, S. A.; Rose, A. L.; David, W. T.

447

Methods for reactive oxygen species (ROS) detection in aqueous environments. Aquat.

448

Sci. 2012, 74, 683–734.

449

(26) Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. Critical review of rate

450

constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals in

451

aqueous solution. J. Phys. Chem. Ref. Data 1988, 17, 513–886.

452

453

(27) Haag, W. R.; Hoign´e, J.; Gassman, E.; Braun, A. M. Singlet oxygen in surface waters– Part I: Furfuryl alcohol as a trapping agent. Chemosphere 1984, 13, 631–640.

24 ACS Paragon Plus Environment

Page 24 of 27

Page 25 of 27

Environmental Science & Technology

454

(28) Canonica, S.; Kohn, T.; Mac, M.; Real, F. J.; Wirz, J.; von Gunten, U. Photosensitizer

455

method to determine rate constants for the reaction of carbonate radical with organic

456

compounds. Environ. Sci. Technol. 2005, 39, 9182–8.

457

(29) Rosario-Ortiz, F. L.; Canonica, S. Probe compounds to assess the photochemical ac-

458

tivity of dissolved organic matter. Environ. Sci. Technol. 2016, 50, 12532–12547.

459

(30) Boreen, A. L.; Edhlund, B. L.; Cotner, J. B.; McNeill, K. Indirect photodegradation

460

of dissolved free amino acids: the contribution of singlet oxygen and the differential

461

reactivity of DOM from various sources. Environ. Sci. Technol. 2008, 42, 5492–5498.

462

(31) Arnold, W. A.; McNeill, K. Analysis, Fate and Removal of Pharmaceuticals in the

463

Water Cycle; Petrovi´c M., Barcel´o D. (Eds.), Elsevier: Amsterdam, 2007; pp 361–385.

464

(32) Calza, P.; Fabbri, D.; No`e, G.; Santoro, V.; Medana, C. Assessment of the photo-

465

catalytic transformation of pyridinium-based ionic liquids in water. J. Hazard. Mater.

466

2018, 341, 55–65.

467

(33) Siedlecka, E. M.; Stepnowski, P. The effect of alkyl chain length on the degradation of

468

alkylimidazolium- and pyridinium-type ionic liquids in a Fenton-like system. Environ.

469

Sci. Pollut. Res. 2009, 16, 453–458.

470

(34) Stepnowski, P.; Zaleska, A. Comparison of different advanced oxidation processes for

471

the degradation of room temperature ionic liquids. J. Photochem. Photobiol. A 2005,

472

170, 45–50.

473

(35) Zhang, Z.; Liu, J.-F.; Cai, X.-Q.; Jiang, W.-W.; Luo, W.-R.; Jiang, G.-B. Sorption to

474

dissolved humic acid and its impacts on the toxicity of imidazolium based ionic liquids.

475

Environ. Sci. Technol. 2011, 45, 1688–1694.

476

(36) Liu, X.-Y.; Chen, W.; Qian, C.; Yu, H.-Q. Interaction between Dissolved Organic

25 ACS Paragon Plus Environment

Environmental Science & Technology

477

Matter and Long-Chain Ionic Liquids: A Microstructural and Spectroscopic Correlation

478

Study. Environ. Sci. Technol. 2017, 51, 4812–4820.

479

(37) Brezonik, P. L.; Fulkerson-Brekken, J. Nitrate-induced photolysis in natural waters:

480

controls on concentrations of hydroxyl radical photo-intermediates by natural scaveng-

481

ing agents. Environ. Sci. Technol. 1998, 32, 3004–3010.

482

(38) Sulzberger, B.; Canonica, S.; Egli, T.; Giger, W.; Klausen, J.; von Gunten, U. Oxidative

483

transformations of contaminants in natural and in technical systems. CHIMIA 1997,

484

51, 900–907.

26 ACS Paragon Plus Environment

Page 26 of 27

Page 27 of 27

485

Environmental Science & Technology

Graphical TOC Entry ionic liquids from wastewater

photochemical reactions with OH-radicals

HO

N

N

R

R

N N

HO

N

R

HO HO

HO

T1/2 > 30 days HO

R

486

27 ACS Paragon Plus Environment