Impact of light and Suwanee River Fulvic Acid on O2 and H2O2

Subscriber access provided by University of Rochester | River Campus & Miner Libraries

Environmental Processes

Impact of light and Suwanee River Fulvic Acid on O2 and H2O2 mediated oxidation of silver nanoparticles in simulated natural waters Hongyan Rong, Shikha Garg, and T. David Waite Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b07079 • Publication Date (Web): 15 May 2019 Downloaded from http://pubs.acs.org on May 16, 2019

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 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 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.

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 37

Environmental Science & Technology

1

Impact of light and Suwanee River Fulvic Acid on O2 and H2O2 mediated

2

oxidation of silver nanoparticles in simulated natural waters

3

Hongyan Rong, Shikha Garg and T. David Waite*

4 5 6

UNSW Water Research Center, School of Civil and Environmental Engineering, The University of New South Wales, Sydney, NSW 2052, Australia

7 8 9 10 11 12

Revised

13

Environmental Science and Technology

14

May 2019

15 16 17 18 19

*Corresponding

author: Tel. +61-2-9385 5060; Email [email protected] 1 ACS Paragon Plus Environment


21

Abstract

22

In this work, we investigate the impact of natural organic matter (NOM) and light on silver

23

nanoparticle (AgNP) dissolution kinetics with particular emphasis on determining the (i)

24

mechanism via which NOM affects the oxidative dissolution of AgNPs, (ii) the role of photo-

25

generated organic radicals and reactive oxygen species (ROS) in oxidative dissolution of

26

AgNPs, and (iii) the mechanism of formation of AgNPs in NOM solution under dark and

27

irradiated conditions. We measured the oxidation of citrate stabilized AgNPs by O2 and

28

hydrogen peroxide (H2O2) in the dark and in irradiated Suwannee River Fulvic Acid (SRFA)

29

solutions at pH 8.0. Results show that the reactivity of AgNPs towards O2 and H2O2 in the dark

30

decreased in the presence of SRFA as a result of blocking of AgNP surface sites through either

31

steric or electrostatic effects. Irradiation promoted dissolution of AgNPs by O2 and H2O2 in the

32

presence of low concentrations (≤ 20 mg.L-1) of SRFA as a result of contribution from photo-

33

generated H2O2 formed on irradiation of SRFA as well as photo-fragmentation of AgNPs.

34

Furthermore, our results show that photo-generated superoxide can induce formation of AgNPs

35

by reducing Ag(I) ions. Based on our experimental results, we have developed a kinetic model

36

to explain AgNP transformation by O2 and H2O2 in the dark and in irradiated SRFA solutions

37

with this model of use in predicting the transformation and fate of AgNPs in natural waters.

38 39 40 41 42

2 ACS Paragon Plus Environment

Page 2 of 37

Page 3 of 37


43

1. Introduction

44

Silver nanoparticles (AgNPs) are widely used in a range of consumer products due to their

45

unique optical and antibacterial properties.1, 2 Widespread use of AgNPs inevitably results in

46

their release to aquatic environments, especially through washing.3 The AgNPs released to the

47

environment may undergo a variety of transformations including oxidation, aggregation, and/or

48

interaction with various inorganic or organic species such as sulfide, chloride and natural

49

organic matter (NOM).4, 5 The interactions of AgNPs with NOM, which is ubiquitous in natural

50

waters, may change AgNP properties and their transformation behaviour.6, 7 Various earlier

51

studies have indicated that the presence of NOM enhances the stability of AgNPs, inhibiting

52

aggregation, either electrostatically or sterically.8-11 Furthermore, the oxygen-mediated

53

dissolution of AgNPs is also inhibited by the presence of NOM, most likely as a result of

54

blocking of the reactive sites on the surface of AgNPs due to NOM sorption.8, 12 It is expected

55

that the reactivity of AgNPs towards other oxidants such as H2O2 and chlorine13, 14 would also

56

be inhibited by the presence of NOM, however clear evidence to support this hypothesis is

57

lacking.

58

It is also reported that the physicochemical properties of AgNPs are influenced by light since

59

AgNPs are photo sensitive.15-17 Earlier studies suggested that fragmentation of AgNPs occurs

60

via photo-ejection of electrons, which accelerates the release of ionic silver.18, 19 Furthermore,

61

organic coatings (such as citrate) on the surface of AgNPs may oxidize on irradiation with

62

resultant formation of electrons which may reduce any silver ions present to form Ag(0) which

63

would then be expected to coalesce to form new AgNPs. Alternatively, these electrons may be

64

stored by AgNPs thereby enhancing the reactivity of AgNPs.20, 21

65



Page 4 of 37

66

While the impact of NOM or light on AgNPs reactivity (towards oxidants) and stability

67

(towards aggregation) appears to be straight-forward, the combined effect of light and NOM is

68

expected to be complex. The interaction of NOM and light generates a suite of oxidants,

69

including reactive oxygen species (ROS, such as superoxide ( O 2 ), hydrogen peroxide (H2O2)

70

and hydroxyl radicals ( HO ))22,

71

and/or peroxyl radical)23-25 which may induce the oxidative dissolution of AgNPs. On the other

72

hand, formation of AgNPs may occur in irradiated NOM solution via pathways including (i)

73

reduction of Ag(I) by photochemically-produced O 2 ,26 (ii) ligand to metal charge transfer

74

(LMCT) within organically complexed Ag(I) species27 and/or (iii) Ag(I) reduction by reduced

75

organic moieties such as hydroquinone/semiquinone species present intrinsically in NOM

76

and/or formed on irradiation.24 Thus, it is expected that the rate and extent of transformation

77

of AgNPs may vary quite significantly in surface waters due to the presence of light and NOM.

78

Although a number of researchers have investigated the effect of light and NOM on the

79

dissolution, aggregation and morphological properties of AgNPs;7,

80

understanding of the interplay between NOM, light and AgNPs is lacking. Thus, in this study,

81

we investigate the effect of Suwannee River Fulvic Acid (SRFA) (a well characterised

82

representative NOM) and solar radiation on the dissolution kinetics of AgNPs at pH 8.0 with

83

the aim to determine (i) the mechanism via which NOM affects the oxidative dissolution of

84

AgNPs, (ii) the role of photo-generated organic radicals and ROS in oxidative dissolution of

85

AgNPs and (iii) the mechanism of formation of AgNPs in NOM solution under dark and

86

irradiated conditions. While we recognize that SRFA is but one example of NOM, SRFA is

87

used in this work because (i) it is widely used and, as noted above, well characterized, and (ii)

88

fulvic acid is reported to be the major reactive component of organic matter in natural waters30-

89

32

90

such as silver than are other constituents of NOM such as humic acids.



23

and organic radicals (triplet NOM, semiquinone radical



8, 28, 29

a comprehensive

and, as such, is more important with regard to the transformations of redox active elements


Page 5 of 37


91

Based on our experimental results, we have developed a mathematical model that satisfactorily

92

describes the results of AgNP dissolution and AgNP formation studies performed here for

93

various concentrations of NOM under dark and irradiated conditions. Insights gained from

94

these studies should be particularly useful in predicting the fate and transformation of AgNPs

95

in natural waters.

96

2. Materials and Methods

97

2.1 Reagents

98

All chemicals used in this work were of analytical grade and supplied by Sigma Aldrich unless

99

stated otherwise. All solutions were prepared in Milli-Q water with resistivity of 18 MΩ∙cm

100

(Millipore). All experiments were performed in 2 mM NaHCO3 solution with pH adjusted to

101

8.0 ± 0.2 using 0.1 M HNO3. Suwannee River Fulvic Acid (SRFA, Standard II) was purchased

102

from the International Humic Substances Society (USA). A 2 g.L-1 of SRFA stock was

103

prepared in Milli-Q water and stored at 4 °C prior to use. A stock solution of 100 µM Amplex

104

Red (AR) and 50 kU.L-1 horseradish peroxidase (HRP) was prepared in Milli-Q water as

105

described in earlier work33 and stored at -85°C prior to use. A stock solution of 3 kU.L-1

106

superoxide dismutase (SOD) was prepared in Milli-Q water and stored at -85 ºC prior to use.

107

A citrate-coated AgNP suspension was prepared as described in our earlier work.13, 14

108

A final AgNP concentration of either 0.35 µM or 5 µM was used in our experiments. The

109

AgNP concentration used here is much higher than the Ag concentration reported in most

110

natural waters (i.e. a few ng.L-1),34-36 but higher Ag concentrations have recently been reported

111

in some rivers37 and near effluent discharge points from industries such as mining and/or legacy

112

photoprocessing.34,

113

natural waters. The H2O2 concentration used in this study ranges between 0.25-7.5 µM which

114

covers the wide range of H2O2 concentration (nM-µM) reported in various surface waters.39, 40

38

Hence, the Ag concentrations used here are representative of some



115

The concentration of SRFA ranges between 5-100 mg.L-1 which is representative of NOM

116

concentrations in surface waters.41-43

117

2.2 Experimental setup

118

For measurement of AgNP oxidation rate in the dark, an appropriate concentration of AgNPs

119

was added to pH 8.0 buffer solution containing H2O2 in the concentration range 0-7.5 µM and

120

the samples were withdrawn at regular intervals for measurement of AgNP concentration. The

121

final dissolved Ag(I) concentration (usually after 2-3 h of commencement of the reaction) was

122

also measured. Photochemical experiments were performed using a Thermo Oriel 150W Xe

123

lamp equipped with air mass filters (AM0 and AM1) to simulate the solar spectrum at the

124

Earth’s surface. The spectral irradiance of the lamp was reported in our earlier work.44 For all

125

experiments, 3.5 mL of sample was irradiated in a 1 cm quartz cuvette and the concentrations

126

of AgNPs and H2O2 were measured at 1, 2, 5 and 10 minutes. The dissolved Ag(I)

127

concentration formed at 10 minutes was also measured.

128

For measurement of AgNP formation on reduction of Ag(I) in irradiated conditions, 3.5 mL of

129

pH 8.0 solution containing Ag(I) and SRFA (5-100 mg.L-1) was irradiated and the dissolved

130

Ag(I) remaining and/or AgNP absorbance spectra measured after 10 minutes. In order to

131

 illustrate the role of O 2 in reduction of Ag(I), pH 8.0 solutions containing 5 µM Ag(I) and 100

132

mg.L-1 SRFA were irradiated in the presence of varying SOD (an enzyme which catalyses the

133

 decay of O 2 to O2 and H2O2) concentration and the absorbance spectra of AgNPs formed

134

measured after 10 minutes.

135


Page 6 of 37

Page 7 of 37


136

2.3 Analytical Methods

137

2.3.1 AgNP measurement

138

AgNP specific surface plasmon resonance (SPR) was monitored using a UV-visible

139

spectrophotometer (Cary 50, Varian Inc.). The concentration of AgNPs in the sample was

140

determined by measuring the SPR peak since there was no significant (p > 0.1, using single-

141

tailed student’s t-test) shift in the SPR peak over the experimental duration investigated here

142

unless specifically stated. The agreement between the measured dissolved Ag(I) concentration

143

and decrease in the AgNP concentration following AgNP oxidation (discussed in detail in

144

Section 3.1) also confirms that the AgNPs concentrations quantified here via SPR peak

145

absorbance are correct. A molar absorptivity of 14,300 M-1 cm-1 at 392 nm was obtained based

146

on the measured absorbance of AgNP stock solution which is similar to the value reported in

147

our earlier work.45

148

2.3.2 Ag (I) determination

149

Dissolved Ag(I) released on AgNP oxidation in the absence and presence of SRFA was

150

measured using ICP-MS (Agilent) after removing AgNPs using Amicon ultrafilters with a

151

membrane NMWL (Nominal Molecular Weight Limit) of 3 kDa (Millipore). A 4 mL aliquot

152

of the sample was centrifuged at 4000 rpm for 45 min and the filtrate was used for Ag(I)

153

measurement with or without 5-fold dilution in pH 8.0 buffer. Standard Ag(I) samples in the

154

concentration range of 0-1.9 µM in the absence and presence of SRFA were prepared using the

155

same procedure as described above. In the presence of 0.5 M NaCl, > 99% of Ag(I) exists as

156

1 n dissolved AgCln (2 ≤ n ≤ 4) species in the absence of SRFA (Tables S1 and S2) which can

157

also be measured by ICP-MS. In the presence of SRFA and absence of NaCl , >97% of silver

158

exists as Ag+ (Table S1) at an initial AgNP concentration of 5 µM thus ICP-MS can be used to

159

determine dissolved Ag(I) in SRFA solution. At lower Ag concentration (i.e. 0.35 µM), >80%

160

of silver forms Ag-SRFA complex (Table S2). Note that at lower Ag concentration (i.e. 0.35 7 ACS Paragon Plus Environment


161

µM), the total measured silver concentration (i.e. AgNP concentration remaining + dissolved

162

Ag(I) concentration formed) was nearly 20% lower than the total silver added for all conditions

163

investigated here suggesting that there is some loss of Ag (present as Ag-SRFA complex)

164

possibly during filtration of the samples prior to dissolved Ag(I) measurement. Thus, in our

165

analysis of AgNPs dissolution results for low AgNPs concentration, only AgNPs

166

concentrations obtained from the SPR peak measurements were used.

167

2.3.3 Morphology and size of AgNPs

168

The changes in the morphology and size of citrate stabilized AgNPs following exposure to

169

SRFA and/or following oxidation by H2O2 were characterized by transmission electron

170

microscopy (TEM) using a FEI Tecnai G20 microscope housed within the Mark Wainright

171

Analytical Centre at the University of New South Wales. The changes in the size of 5 µM

172

AgNPs in pH 8.0 suspension under various solution conditions were also followed by dynamic

173

light scattering (DLS) using a Malvern nanosizer (Malvern Zetasizer Nano S) employing the

174

settings and procedures described in our earlier work.46

175

2.4 Speciation and kinetic modelling

176

The speciation of silver in the absence and presence of 0.5 M NaCl and/or SRFA in pH 8.0

177

solution was calculated using MINTEQ (version 3.1, Jon Petter Gustafsson, KTH, Sweden). In

178

the presence of SRFA, the Stockholm Humic Model was used to represent SRFA.

179

Kinetic modelling was undertaken using KinTek Explorer. KinTek Explorer is a simulation

180

program that enables prediction of the concentration of reactants and products as a function of

181

time based on numerical integration of the rate equations appropriate to a hypothesized reaction

182

mechanism.47 The rate constants used for various reactions in modelling were either obtained

183

from literature and/or measured experimentally. Agreement (or lack thereof) of the predicted

184

concentrations of reactants and products with measured concentrations for the same entities


Page 8 of 37

Page 9 of 37


185

provides a measure of the veracity of the hypothesized reaction set and/or the rate constants

186

used.

187

3. Results and Discussion

188

3.1 AgNP dissolution by O2 and H2O2 in dark SRFA solution

189

As shown in Figure 1a, significant decrease in AgNP concentration was observed within 4

190

hours when 0.35 µM AgNPs were added to pH 8.0 solution with the rate and extent of decrease

191

in AgNP concentration decreasing with increase in SRFA concentration. In contrast, no

192

decrease in AgNPs concentration was observed in the solution containing initial AgNP

193

concentration of 5 µM (Figure 1b). It is important to note, however, that the time scales shown

194

here (i.e. 30 min) are much smaller than those used in Figure 1a (300 min). The decrease in

195

concentration of 0.35 µM AgNPs suggests that slow oxidation of AgNPs by dioxygen occurs.

196

The decrease in the rate of AgNPs oxidation by dioxygen with increase in the SRFA

197

concentration is consistent with the earlier reports investigating the impact of NOM on AgNP

198

oxidation.48, 49 Addition of H2O2 to AgNP solution at pH 8.0 significantly increases the rate

199

and extent of dissolution of 0.35 µM and 5.0 µM AgNPs (Figure 1c and 1d) with the extent of

200

AgNP dissolution increasing with increase in H2O2 concentration (Figure S1). The presence of

201

SRFA decreased the rate and extent of AgNPs dissolution by H2O2 (Figure 1c and 1d) with the

202

dissolved Ag(I) concentration formed decreasing with increase in SRFA concentration (Figure

203

2) with this result consistent with effect of SRFA observed on O2-mediated AgNPs dissolution

204

(Figure 1a). The oxidation of citrate-coated AgNPs observed here, suggests that O2 and H2O2

205

are able to access the active sites on the AgNPs surface, even in the presence of citrate, with

206

this conclusion in agreement with the weak binding of citrate to the AgNP surface reported in

207

various earlier studies.50, 51 The impact of O2 and H2O2 on the citrate coating is expected to be

208

negligible since citrate is stable in the presence of O2 and H2O2.



209

The decrease in AgNP oxidation rate (by both O2 and H2O2) in the presence of SRFA may

210

result from a decrease in the accessibility of reactive surface sites of AgNPs as a result of

211

coating of AgNPs by SRFA. Alternatively, it is also possible that reduction of dissolved Ag(I)

212

(formed on AgNP oxidation) may occur by the reducing organic moieties that are intrinsically

213

present in NOM25 thereby resulting in an apparent decrease in the dissolution of AgNPs in the

214

presence of SRFA. However no AgNP SPR peak formation or decrease in dissolved Ag(I)

215

concentration (data not shown) was observed in dark solutions containing SRFA and Ag(I) at

216

the time scales investigated here suggesting that thermal reduction of Ag(I) by SRFA is not

217

important. This observation further supports the conclusion that coating of AgNPs by SRFA is

218

responsible for the decreased oxidation of AgNPs by O2 and H2O2 in dark SRFA solutions. The

219

adsorption of SRFA on citrate-coated AgNPs surface is reported to occur via replacement of

220

citrate on the surface of AgNPs by SRFA, particularly as a result of the affinity of the reduced

221

sulfur and nitrogen containing functional groups present in SRFA for elemental Ag.8,52-54 The

222

decrease in the reactivity of AgNPs in the presence of SRFA, which may occur either due to

223

steric hindrance and/or electrostatic repulsion, is in agreement with the results of various earlier

224

studies.7, 55, 56

225

3.2 AgNPs dissolution by O2 and H2O2 in irradiated SRFA solution

226

Compared to non-irradiated SRFA solutions, AgNPs undergo faster dissolution in irradiated

227

SRFA solutions (Figure 3a) with the rate of oxidation of 0.35 µM AgNPs 17.4%, 22.9% and

228

26.7% higher in irradiated solutions than in non-irradiated solutions containing 0, 5 and 10

229

mg.L-1 SRFA respectively. Furthermore, our results show that the rate and extent of oxidation

230

of 0.35 µM AgNPs in irradiated SRFA solution is nearly independent of SRFA concentration.

231

The rate of AgNPs oxidation in the presence of H2O2 is also higher in irradiated SRFA solution

232

compared to that measured in the dark (see Table S3 for comparison of AgNP oxidation rates


Page 10 of 37

Page 11 of 37


233

under dark and irradiated conditions). The faster dissolution of 5 and 0.35 µM AgNPs in

234

irradiated SRFA solution (in the concentration range 0-20 mg.L-1) compared to that observed

235

in dark solution may possibly occur due to

236

(i)

19

237 238

(ii)

alteration in the morphology of AgNPs on irradiation thereby affecting their reactivity towards O2 and H2O2;28, 29

239 240

fragmentation of AgNPs caused by photo-ejection of electrons as reported earlier;18,

(iii)

oxidation of AgNPs by photo-generated oxidants (i.e. H2O2 and/or oxidizing

241

organic moieties e.g. semiquinone-type radicals)22,

242

SRFA.

57

formed on irradiation of

243

As shown in Table S3, significant increase in the AgNPs oxidation rate (17.4 ± 5.2% and 32.3

244

± 5.0 % for 0.35 µM AgNPs in the absence and presence of H2O2 respectively) is observed in

245

irradiated solution compared to non-irradiated solution even in the absence of SRFA with this

246

result supporting the hypothesis that a small proportion of AgNPs present may undergo

247

oxidation either as a result of the photo-ejection of electrons and/or an increase in the reactivity

248

of AgNPs towards O2 and H2O2 on irradiation. It has been reported that in the case of citrate-

249

coated AgNPs, citrate is oxidized on irradiation with release of electrons with these electrons

250

either stored by AgNPs and/or active in reducing Ag(I) to form secondary AgNPs, thereby

251

enhancing the chemical reactivity of AgNPs.15,

252

duration (≤ 10 min) employed here, the extent of oxidation of citrate is expected to be negligible.

253

Hence, the hypothesis that photo-fragmentation of AgNPs occur on irradiation, as described in

254

eqs 1-3, appears to be more reasonable though the possibility of changes in the reactivity of

255

AgNPs on irradiation cannot be completely rejected based on the results presented here.

256

 (Ag) n  2h   (Ag)n  eaq

20, 21

However, due to the small irradiation

(1)



Page 12 of 37

257

 (Ag)n  eaq   (Ag) n

(2)

258

(Ag)n   (Ag) n1  Ag 

(3)

259

While both of these possibilities can explain the increase in AgNP oxidation rate in the absence

260

of SRFA on irradiation as well as the increase in AgNP oxidation rate under irradiated

261

conditions in the presence of SRFA for higher AgNP concentrations, these processes cannot

262

account for the observed lack of dependence of AgNPs oxidation rate on SRFA concentration

263

under irradiated conditions at the lower AgNPs concentration examined (Figure 3a and 3c). As

264

shown in Figure 3, under irradiated conditions, the oxidation rate of 5 µM AgNPs decreases

265

with increase in SRFA concentration however the oxidation rate of 0.35 µM AgNPs is

266

independent of SRFA concentration in the absence and presence of H2O2. This discrepancy in

267

the observed effect of SRFA at varying Ag concentrations supports the hypothesis that oxidants

268

generated on SRFA irradiation play a role in AgNP oxidation at lower AgNP concentrations

269

with the effect of SRFA on AgNP reactivity countered by the increase in AgNP oxidant

270

concentration. In contrast, at the higher AgNP concentrations, these photo-generated oxidants

271

do not have much impact due to their low concentration with the effect of SRFA mainly

272

associated with changes in the reactivity of AgNP as a result of coating of the nanoparticle

273

surface by SRFA. The involvement of oxidants generated on SRFA irradiation in AgNP

274

oxidation is supported by the measurement of AgNP oxidation rate when 0.35 µM AgNP is

275

added to SRFA that was previously irradiated for 10 min prior to AgNP addition in the dark.

276

As shown in Figure S2, the measured AgNP oxidation rate in previously-irradiated SRFA

277

solution is slightly higher than that measured in a dark solution, especially during the initial

278

stages, supporting the conclusion that oxidants are generated on irradiation of SRFA.

279

3.2.1 Nature of the photo-generated oxidant

280

As discussed above, organic oxidants (semiquinone radical, triplet NOM and/or peroxyl

281

radicals) and/or inorganic oxidants (such as H2O2) generated on irradiation of SRFA appear to 12 ACS Paragon Plus Environment

Page 13 of 37


282

play an important role in AgNP oxidation, at least at lower AgNP concentrations. As reported,

283

H2O2 is generated on SRFA photolysis and is a known oxidant of AgNPs.22, 23, 58 Comparison

284

of the measured AgNP oxidation rate in the presence of a H2O2 concentration comparable to

285

that generated on SRFA irradiation (Figure S3) with the measured AgNP oxidation rate in

286

irradiated SRFA solution (Figure S4) shows that the rate and extent of AgNP oxidation in

287

irradiated SRFA solution is similar to that measured in dark SRFA solution containing a

288

comparable concentration of H2O2. This result supports the hypothesis that photogenerated

289

H2O2 is involved in AgNP oxidation under irradiated conditions. Furthermore, the longevity of

290

the entities involved in AgNP oxidation (as indicated by the data shown in Figure S2) negates

291

 the involvement of short-lived oxidizing entities such as triplet NOM, singlet oxygen, O 2

292

and/or peroxyl radicals and supports the conclusion that H2O2, a stable end product of SRFA

293

photolysis,22 accounts for the observed increase in rate and extent of AgNP oxidation in the

294

presence of light and SRFA.

295

3.3 Changes in particle size during AgNP dissolution

296

TEM images of the AgNP suspensions used here (see Figure S5a) show the presence of

297

relatively uniform spherical particles with a mean diameter of 14.3 ± 4.3 nm. AgNP size

298

measured by DLS (see Figure S6a) is consistent with the TEM results confirming the presence

299

of relatively monodisperse particles with a z-averaged mean hydrodynamic diameter of 14.2 ±

300

1.9 nm. According to both the DLS results (Figure S7) and TEM images, the presence of SRFA

301

and/or H2O2 (in the dark) appears to result in no significant (p> 0.5 using single tailed student’s

302

t-test) change in the average particle size, at least over the 10 minutes of DLS analysis, though

303

some larger particles are formed following dissolution in the presence of H2O2 possibly due to

304

the coalescence of particles.



305

On irradiation, the mean size of the particles is seen to decrease as they undergo oxidative

306

dissolution (see Figures S6 and S7). As can be seen from Figure S6, the particles remain

307

relatively monodisperse throughout the dissolution process. These results are consistent with a

308

dissolution mechanism wherein the entire ensemble of AgNP particles dissolves uniformly

309

with resultant decrease in particle size over time. Since the rate of AgNP oxidation is expected

310

to be proportional to the surface site concentration, the smaller sized particles would be

311

expected to undergo slower oxidation as a result of the smaller surface area available and the

312

associated smaller concentration of accessible Ag surface sites. This is consistent with the clear

313

decrease in the AgNP oxidation rate observed in Figure 3, especially after the initial 2 min of

314

irradiation. Thus, we suggest that dissolution of the entire ensemble of particles occurs over

315

time with dissolution rate proportional to the concentration of surface sites available for

316

reaction with oxidant.

317

3.4 Effect of chloride addition on AgNPs dissolution by O2 and H2O2

318

As chloride is present in natural waters and can impact the reactivity (towards oxidation) and

319

stability (towards aggregation) of AgNPs as well as the speciation of Ag(I), we investigated

320

the effect of chloride addition on AgNP dissolution in dark and irradiated solutions containing

321

SRFA. According to previous studies, in the presence of nucleophilic Cl‒, a negative shift in

322

the AgNP reduction potential occurs which might be expected to result in an increase in the

323

rate of oxidation of AgNPs by dioxygen.59, 60 However, as shown in Figure 4, no significant (p

324

> 0.05, using single tailed student’s t-test) effect of NaCl addition on 5 µM AgNP dissolution

325

was observed in the absence of H2O2 under either dark or irradiated conditions with this lack

326

of effect possibly due to the combined effect of NaCl on AgNP dissolution and aggregation of

327

particles (Figure S8). The rate and extent of oxidative dissolution of 5 µM AgNP by H2O2 in

328

both dark and irradiated solutions significantly decreased in the presence of NaCl. The

329

discrepancy between the impact of NaCl addition on AgNP oxidation by O2 and H2O2 results 14 ACS Paragon Plus Environment

Page 14 of 37

Page 15 of 37


330

from the fact that AgNPs oxidation by dioxygen is negligible (at the higher AgNP

331

concentration on the time scales investigated here) and thus the decrease in the oxidation rate

332

of AgNP as a result of aggregation in the presence of NaCl cannot be detected. Rapid

333

aggregation of AgNPs however is more discernible in the presence of H2O2 as a result of the

334

rapid oxidation of AgNPs by H2O2. Note that since a significant fraction of the concentration

335

of Ag(I) formed on dissolution of 0.35 µM AgNPs is lost during preparation of samples for

336

measurement, effect of NaCl addition on AgNPs dissolution cannot be determined under this

337

condition.

338

While only the impact of NaCl addition is investigated here, the presence of other anions (such

339

as SO42-) and cations (such as Mg2+ and Ca2+) will also increase the rate of aggregation of

340

AgNPs as a result of changes in the ionic strength as reported earlier7, 51, 61, 62 which will

341

consequently impact the oxidation of AgNPs.

342

3.5 AgNP formation on reduction of Ag(I) in irradiated SRFA solution

343

The absorbance spectra of solutions containing varying concentrations of Ag(I) and SRFA after

344

10 min of irradiation are shown in Figures 5a and 5b with clear evidence of absorbance in the

345

region of the SPR peak. No absorbance was observed in dark solutions containing Ag(I) and

346

SRFA after 10 min (data not shown). The absorbance peak at ~440 nm is presumably due to

347

the formation of AgNPs and is in agreement with the results of earlier studies which showed

348

that the AgNPs formed in solutions of Ag(I) and NOM possess specific absorbance at 420-450

349

nm.26,

350

concentrations of Ag(I) and SRFA (Figures 5a and 5b) indicates that the formation of AgNPs

351

on irradiation is due to Ag(I) reduction by SRFA. The formation of AgNPs on reduction of

352

Ag(I) is also supported by the observation that decrease in dissolved Ag(I) concentration occurs

353

during irradiation (Figure 5c). The fact that the size of the particles formed increased from ~80

354

nm to over 320 nm in irradiated solution containing Ag(I) and 100 mg.L-1 SRFA (Figure 5d)

63

The observation that the absorbance peak is higher in the presence of higher



355

suggests that the AgNPs formed via photo-reduction of Ag(I) aggregate rapidly. The photo-

356

formation of AgNPs in irradiated SRFA solution may occur via the following pathways:

357

 (i) reduction of Ag(I) by O 2 generated on irradiation of SRFA; and/or

358

(ii) LMCT in SRFA-Ag(I) complex; and/or

359

(iii) reduction of Ag(I) by reduced organic moieties formed on irradiation of SRFA (note

360

that Ag(I) reduction by reduced organic moieties intrinsically present in SRFA is not

361

important as discussed earlier in §3.1).

362

As shown in Figure S9, the presence of 100 kU.L-1 SOD decreased the extent of SPR peak

363

formation with no peak detected when 200 kU.L-1 SOD was added with this result supporting

364

 the hypothesis that photo-formation of AgNPs in irradiated SRFA solution occurs via O 2 -

365

mediated reduction of Ag(I). Note that even though 100 kU.L-1 SOD should be sufficient to

366

 outcompete all reactions of O 2 occurring in the solution phase, it is possible that the O 2 -

367

mediated AgNP formation (which ensues on the AgNP surface via a charging process64 at

368

diffusion limited rate) is not completely outcompeted by the rapid disproportionation of O 2 in

369

the presence of 100 kU.L-1 SOD with even higher SOD concentrations required to achieve

370

27 complete inhibition of O 2 -mediated AgNP formation. While Hou et al had earlier suggested

371

that LMCT is the main pathway of AgNPs formation in irradiated Aldrich humic acid (ALHA)

372

solution based on positive correlation between concentration of Ag(I) complexed by ALHA in

373

dark control samples and AgNPs formation on irradiation, it appears to play an insignificant

374

role in SRFA solution possibly due to the low Ag binding capacity of SRFA (Table S1)

375

compared to ALHA.8, 63 The absence of any role of photo-generated reduced organic moieties

376

such as semiquinones in Ag(I) reduction is possibly due to the low steady-state concentration


Page 16 of 37

Page 17 of 37


377

and low reduction potential (~0. 59 V) of these entities25 compared to that of Ag(I) (-1.8 V)26

378

in the experimental system investigated here.

379

Comparison of the measured AgNP formation rate with the AgNP oxidation rate in irradiated

380

SRFA solutions further supports the conclusion that the decrease in the AgNP dissolution rate

381

in the presence of SRFA is due to reduced AgNP reactivity as a result of SRFA surface coating

382

rather than AgNP reformation via SRFA-mediated Ag(I) reduction. Based on the decrease in

383

dissolved Ag(I) concentration after 10 minute irradiation (Figure 5c), we calculate the photo-

384

reformation rate of AgNPs in the presence of 5 µM Ag(I) and 100 mg.L-1 to be 0.11 µM.min-

385

1.

386

appears to be at least four times lower (

Impact of light and Suwanee River Fulvic Acid on O2 and H2O2

Recommend Documents