Effects of C60 on the Photochemical Formation of Reactive Oxygen

Oct 6, 2016 - Buckminsterfullerenes (C60) are widely used nanomaterials that are present in surface water. The combination of C60 and humic acid (HA) ...
0 downloads 9 Views 769KB Size
Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Effects of C60 on the Photochemical Formation of Reactive Oxygen Species (ROS) from Natural Organic Matter Lijuan Yin, Huaxi Zhou, Lushi Lian, Shuwen Yan, and Weihua Song Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b04488 • Publication Date (Web): 06 Oct 2016 Downloaded from http://pubs.acs.org on October 11, 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 29

Environmental Science & Technology

1

2

3

4

Effects of C60 on the Photochemical Formation of Reactive Oxygen Species

5

(ROS) from Natural Organic Matter

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

Lijuan Yin, Huaxi Zhou, Lushi Lian, Shuwen Yan, and Weihua Song*

Department of Environmental Science & Engineering, Fudan University, Shanghai 200433, P. R. China

Resubmitted to Environ. Sci. & Technol.

*Corresponding author: Email: [email protected]; Tel: (+86)-21-6564-2040

30 31 32 33

1

ACS Paragon Plus Environment

Environmental Science & Technology

34

Abstract

35

Buckminsterfullerenes (C60) are widely used nanomaterials that are present in surface water. The

36

combination of C60 and humic acid (HA) generates reactive oxygen species (ROS) under solar

37

irradiation, but this process is not well understood. Thus, the present study focused on the

38

photochemical formation of singlet oxygen (1O2), hydroxyl radical (HO•)-like species, superoxide

39

radicals (O2•−), hydrogen peroxide (H2O2), and triplet excited states (3C60*/3HA*) in solutions

40

containing both C60 and HA. The quantum yield coefficients of excited triplet states (ƒTMP) and

41

apparent quantum yields of ROS were measured and compared to the calculated values, which were

42

based on the conservative mixing model. Although C60 proved to have only a slight impact on the

43

1

44

of H2O2 followed the conservative mixing model due to the reaction of C60•− with HO2•/O2•−, and the

45

biomolecular reaction rate constant has been be measured as (7.4 ± 0.6) × 106 M-1 s-1. The apparent

46

ƒTMP was significantly lower than the calculated value, indicating that the steric effect of HA was

47

significant in the reaction of 3C60* with the TMP probe. In contrast, C60 did not have an effect on the

48

photochemical formation of HO• from HA, suggesting that HO• is elevated from the hydrophilic

49

surface of HA. The aforementioned results may be useful for predicting the photochemical influence

50

of C60 on aqueous environments.

O2 formation from HA, C60 played a key role in the inhibition of O2•−. The photochemical formation

51

2

ACS Paragon Plus Environment

Page 2 of 29

Page 3 of 29

52

Environmental Science & Technology

Introduction

53

Since its discovery in 1985, buckminsterfullerene (C60) has received extensive attention due to

54

its unique photophysical and photochemical properties.1, 2 For instance, C60 presents strong UV-Vis

55

absorbance due to the presence of extended conjugated double bonds. Product use, disposal, and/or

56

weathering may release fullerenes into aquatic environments. Despite being exceedingly

57

hydrophobic and virtually non-wettable (the solubility of C60 in water is 18.2 MΩ) produced by a Milli-Q

122

water purification system was used for the preparation of all solutions.

carboxylate

luciferin

analog

trifluoromethanesulfonate

(MCLA)

and

(AE)

were

123

Preparation and measurement of C60. Ultrasound energy (40 KHz, 180 W) was applied to a

124

sealed bottle containing a heterogeneous mixture of 20 mL of toluene, 5.0 mg of C60 and 180 mL of

125

ultrapure water for 24 h using an ultrasonicator (Xinzhi®, China). Detailed information on the

126

preparation of C60 can be found in the literature.37 The stock solution was filtered through 0.45 µm 5

ACS Paragon Plus Environment

Environmental Science & Technology

127

PTFE filters.38, 39 The residual concentration of toluene in the stock solution was measured as 0.53

128

µM (Section S1 of Supporting Information, SI), which has negligible effect on the photochemical

129

formation of HO• in the following studies. The concentration of C60 was determined by UV

130

absorbance at 334 nm, in accordance with a previous oxidation-extraction protocol.40 A 1.5-mL

131

aliquot of fullerene was collected (3 replicas) and placed into a 4-mL extraction vessel containing

132

150 µL of 1.0 M magnesium perchlorate. Next, a 1.5-mL aliquot of toluene was added to the

133

extraction vessel, which was subsequently vortexed for several seconds and vibrated for 30 min at

134

200 rpms on an orbital shaker. The solution was placed in a freezer at −80 °C for 30 min to freeze the

135

aqueous layer in order to decant the remaining toluene layer. The aqueous layer was allowed to melt,

136

and another 1.5-mL aliquot of toluene was added to the extraction vessel. Three sequential

137

extractions were performed using the aforementioned procedure. The extracted fullerene was stirred

138

and subsequently determined by monitoring the UV absorbance (Cary 60, Agilent). The

139

intensity-average particle size distribution of aggregates obtained from different concentrations of

140

C60 in the presence of HA was characterized by dynamic light scattering (DLS) on a Zetasizer Nano

141

ZS90 instrument (Malvern, U.K.).

142

Photochemical experiments. A solar simulator (Suntest XLS+, Atlas) equipped with a 1700W

143

xenon lamp and a filter was employed for all of the photochemical experiments. A temperature

144

control unit (Suncool) is applied to fix the temperature at 25 ± 1°C. The absolute irradiance spectrum

145

of the lamp and sunlight was recorded on a spectrometer (USB-4000, Ocean Optics Inc.), as

146

displayed in Figure S1 of SI. Light intensity on the surface of the solutions was measured as 1.36 ×

147

10-8 Einstein s-1 cm-2 in the range of 290 to 400 nm. All of the samples were placed in specially made

148

cylindrical quartz containers (diameter 9.5 cm, height 4.5 cm, thickness 2 mm), which allowed the

149

full transmission of UV light and were irradiated for a given amount of time. The total organic

150

carbon (TOC) of the solutions was determined using a TOC analyzer (Shimadzu, TOC CPH/CN). All

151

of the solutions were buffered to pH 7.3 using 5.0 mM phosphate buffer to eliminate pH effects. The 6

ACS Paragon Plus Environment

Page 6 of 29

Page 7 of 29

Environmental Science & Technology

152

phosphate buffer presented negligible effects on the photochemical formation of ROS and triplet

153

states from C60 solutions, as shown in Figure S2 of SI. The concentration of dissolved oxygen was

154

measured as 6.4 ± 0.1 mg L-1 and kept constant during the irradiation experiments.

155

The steady state concentrations of 1O2, HO•, and 3HA*. To study the steady state

156

concentrations of HO•, 1O2, and 3HA* and/or 3C60* in the bulk solution, TA, FFA and TMP were

157

employed as chemical probes, respectively.41, 42 As shown in Figure S3 of SI, no significant direct

158

photodegradation has been observed for these chemical probes in DI-H2O under our irradiated

159

condition. For HO•, varied concentrations of TA were employed to selectively quench the radicals,

160

yielding the product 2HTA, which was measured by HPLC with fluorescence detection (λexcitation =

161

315 nm, λemission = 425 nm).43 The observed formation rates of 2HTA were converted to steady state

162

concentrations of HO• by dividing by the initial TA concentration, reaction yield and hydroxyl radical

163

reaction rate constant. More details can be found in Section S2 of SI. It should be noted that TA can

164

also be hydroxylated by other hydroxylating reactive species presented in photosensitized NOM

165

solutions. At this point distinguishing them is challenge, therefore we actually measured HO•-like

166

species in this manuscript.

167

The steady state concentration of 1O2 was measured using FFA, a common probe, which was

168

irradiated with SRHA solutions in the solar simulator. As shown in Figure S5 of SI, varying

169

concentrations of FFA (20 to 300 µM) were engaged to trap 1O2. Finally 20 µM of FFA has been

170

applied. The loss of FFA was determined using an UHPLC (λ = 219 nm) equipped with a Gemini C18

171

column (4.6 × 250 mm, 5 µm). Further details are provided in Figure S6 of SI. The observed FFA

172

degradation rates were divided by the 1O2 rate constant (k 1O2, FFA = 8.3 × 107 M-1 s-1) to obtain the

173

steady state concentrations of 1O2.44

174

TMP (35 µM) was employed to characterize the triplet states of HA (3HA*) and C60 (3C60*).45

175

The analysis was carried out using an UHPLC equipped with a photodiode array detector (λ = 280

176

nm). The corresponding TMP degradation profile is shown in Figure S7 of SI. Previous researchers 7

ACS Paragon Plus Environment

Environmental Science & Technology

Page 8 of 29

177

have estimated that the second-order rate of the reaction of 3HA* and TMP is k3HA*,TMP = 3 × 109 M-1

178

s-1 45and that the reaction rate of 3C60* and TMP is k3C60*,TMP = (2.8 ± 0.7) × 109 M-1 s-1, as detailed in

179

Section S4 of the SI.

180

Determination of the steady state concentrations of O2•− and the generation rates of H2O2.

181

Due to the low concentration of superoxide, 2-methyl-6-[p-methoxyphenyl]-3,7-dihydroimidazo

182

[1,2-a]pyrazin-3-one (MCLA) was employed as a chemiluminescent probe, which selectively reacted

183

with O2•− in a flow injection analysis (FIA) system (Waterville Analytical). MCLA radicals resulting

184

from the initial reduction by HO2• reacted rapidly with a second superoxide molecule to form a

185

dioxetanone, which decayed spontaneously, emitting a photon with a wavelength of approximately

186

380 nm.46 The MCLA reagent for the analysis of superoxide in water was prepared according to the

187

protocol of Rose and Waite.47 The standard curve correlating the signal intensity to the concentration

188

of O2•− is illustrated in Section S5 of SI. The FIA system was also employed to measure the

189

generation

190

trifluoromethanesulfonate (AE),48 which reacts stoichiometrically with H2O2 to form a stable

191

intermediate.48 Subsequent addition of an alkaline reagent results in the formation of a second

192

intermediate, which spontaneously decays, emitting photons with a wavelength of approximately 470

193

nm.49 The rate of H2O2 generation in the samples is shown in Figure S10 of SI.

of

H2O2,

utilizing

10-methyl-9-(p-formylphenyl)-acridinium

carboxylate

194

Determination of quantum yields. The losses of TMP and FFA, which were used as probes

195

to determine the quantum yield coefficients for 3HA*/3C60* (ƒTMP) in the electron transfer pathway

196

and the apparent quantum yield for 1O2 (Φ1O2) in the energy transfer pathway, respectively, were

197

determined following standard approaches (as described in Section S6 of SI).50 The ƒTMP and Φ1O2

198

of solutions of C60/HA are shown in Table S1 of SI. For this study, the wavelength range used to

199

calculate the quantum yields was from 290 to 400 nm. The apparent quantum yields of HO• (ΦHO•),

200

O2•− (ΦO2•−) and H2O2 (ΦH2O2) were also calculated based on similar approaches.51

201 8

ACS Paragon Plus Environment

Page 9 of 29

202

Environmental Science & Technology

Results and Discussions

203

The particle size of C60. The intensity-averaged particle size distributions (PSD) of various

204

amounts of C60 dispersed in water are presented in Figure S11 of SI. The average hydrodynamic

205

radius of C60 at a concentration of 0.80 mgC L-1 was 42.6 nm, which was similar to that observed in

206

the presence of 5.26 mgC L-1 of HA. In mixtures of C60 and HA, as the concentration of C60

207

decreased from 0.80 to 0.20 mgC L-1, significant changes in the PSDs were not observed,

208

demonstrating that the effects of the PSDs on the experimental results could be neglected. Although

209

previous studies have shown that HA reduces the particle size and increases the colloidal stability of

210

C60 as the concentration increases, the dispersive effect of HA on C60 in the present study was

211

negligible due to the low C60 concentrations.28, 49,

212

numerous C60 molecules were present in the C60 aggregates.52 Considering that the molecular weight

213

of HA ranged from a few hundred to 1×105 Daltons (Da)53 and that C60 acted as a large inner core,

214

numerous relatively small HA molecules were adsorbed onto the C60 surface, which was consistent

215

with previous results.29, 30

50

Because C60 possesses a diameter of 7 Å,

216

Photochemical production of singlet oxygen. Singlet oxygen, the lowest excited state of

217

molecular oxygen, is one of the most active intermediates in chemical and biochemical reactions.38, 39

218

As shown in Figure 1a, as the concentration of C60 increased from 0.01 to 0.80 mgC L-1, the steady

219

state concentration of 1O2 increased, indicating that aqueous solutions of C60 produce significant

220

amounts of 1O2 under solar irradiation. This result was consistent with the results of previous studies,

221

which reported that the apparent quantum yield of 1O2 production by aqu/nC60 was 0.015.54, 55

222

Previous studies also showed that the initial rate of formation (∆[1O2]/∆t) increased as the C60

223

concentration increased from 1 to 10 mg L-1.56 Under sunlight irradiation, C60 was photochemically

224

excited to the singlet state (1C60*), which was subsequently converted to the triplet state (3C60*)

225

through intersystem conversion. As 3C60* returned back to the ground state, producing singlet oxygen

226

(Scheme 1), the energy in 3C60* was transferred to dissolved oxygen.55, 57 9

ACS Paragon Plus Environment

Environmental Science & Technology

Page 10 of 29

(Insert Figure 1, Scheme 1)

227 228

In HA-enriched solutions, the excited triplet states of phenyl ketone and quinone moieties

229

were primarily responsible for the photochemical formation of 1O2.58-60 After combining C60 with HA,

230

the steady state concentration of 1O2 increased slightly with an increase in the C60 concentration

231

(from 0.01 to 0.80 mgC L-1), as shown in Figure 1a. We examined the apparent quantum yields of the

232

combination of HA and C60 to explore the effect of C60 on the formation rate of 1O2 from HA. Based

233

on the conservative mixing model, the calculated quantum yield, Φ cal, of a mixture can be obtained

234

from the ratio of the overall production of 1O2 to the rate of light absorption, assuming that the two

235

photoreactive components do not interact:41

236

, =

         ( (   ) )

(1)

237

Where ΦHA and ΦC60 are the apparent quantum yield for HA and C60, αHA and αC60 are the specific

238

absorption coefficients of HA and C60 (L mgC-1 cm-1), DOCHA and DOCC60 are the organic matter

239

concentrations of HA and C60 (mgC L-1), and z is the optical path length (cm).

240

The measured quantum yields of 1O2 for the combination of HA and C60 were acquired based

241

on the previous approach (see SI for details on the calculation). As summarized in Figure 1b, the

242

quantum yields of 1O2 for HA/C60 were in accordance with the trends expected for conservative

243

mixing at low C60 concentrations. As the concentration of C60 increased, the difference between the

244

measured and calculated quantum yields increased slightly, which suggested that the interactions

245

between C60 and HA led to a minor reduction in the photochemical formation of 1O2. Previous

246

studies have shown that dissolved humic acid readily adsorbed onto C60, which enhanced the

247

stability of pristine nC60 due to steric hindrance effects.30, 61 In this study, the average hydrodynamic

248

radius of nC60 cluster was measured as 42.6 nm, and shown in Figure S11 of SI. Meanwhile the

249

hydrodynamic radius for HA was estimated in the range of 0.5 to 3.3 nm.62 Since the average

250

hydrodynamic radius of nC60 cluster did not change significantly when NOM mixed with aqueous

251

C60, we could propose that nC60 acted as a core, and HA was distributed on the surface of C60 with 10

ACS Paragon Plus Environment

Page 11 of 29

Environmental Science & Technology

252

the hydrophobic portion toward the inside, and the hydrophilic region toward the outside. The

253

photochemically induced 1O2 from 3C60* was slightly scavenged by adsorbed HA on the surface of

254

C60, reducing the amount of 1O2 released into the bulk solution. The observed scavenging effect

255

would be even more noticeable when higher concentrations of C60 are employed.

256

Effect of C60 on the photo-formation of superoxide radicals. The reduction of molecular

257

oxygen can result in the generation of superoxide anion radicals (O2•−). Figure 2a shows that C60

258

alone cannot produce O2•− under simulated solar irradiation. Previous reports have demonstrated that

259

C60 aggregates do not yield O2•− due to the lack of electron donors in C60 solutions.56, 63, 64 Moreover,

260

the mechanism for photochemical formation of O2•− by HA is still under investigation. One possible

261

pathway is that the intramolecular electron transfer between electron donors and acceptors inside

262

1

263

to produce O2•−.20 Herein, the photochemical formation of O2•− from solutions of HA was observed

264

under simulated solar irradiation, and the steady state concentration was 3.1 × 10-8 M. In contrast,

265

when C60 was combined with HA, the [O2•−]ss decreased dramatically as the concentration of C60

266

increased, as illustrated in Figure 2a. When the concentration of C60 increased to 0.80 mgC L-1, the

267

inhibition ratio drastically increased to 92%. The apparent quantum yield of O2•− in a mixture of HA

268

and C60 was also compared to the calculated quantum yield, which was based on the conservative

269

mixing model. As shown in Figure 2b, the experimental quantum yields of O2•− were significantly

270

lower than the calculated values.

271

NOM*/3NOM* produces reducing radical intermediates (HA•−), which is further oxidized by oxygen

(Insert Figure 2)

272

To determine the cause of this discrepancy, we suggested there are two possible mechanisms.

273

One is that C60 may act as an electron acceptor, drawing electrons from HA•− and yielding C60•−.65, 66

274

Then the competition between C60 and O2 for the abstraction of an electron from HA•− led to a

275

decrease in the [O2•−]ss. The other possibility is that C60 reacts with O2•− directly, decreasing the

276

[O2•−]ss. Therefore an experiment was carried out to explore the reaction of C60 with O2•−. A standard 11

ACS Paragon Plus Environment

Environmental Science & Technology

Page 12 of 29

277

solution of superoxide (with a known concentration) was generated by the photolysis of a 1.0 mM

278

borate buffer solution (pH 12) containing 41 mM acetone, 12 M absolute ethanol, and 15 µM

279

DTPA.67 Then it was spiked into aqueous C60 solutions, and immediately delivered into FIA system

280

using a peristaltic pump to real-time monitoring of the changes of [O2•−]ss. Figure 3 demonstrated

281

that the signal of O2•− immediately dropped with a multiple exponential decay when O2•− were spiked

282

in aqueous C60 solutions. To extrapolate back to time zero by using the fitting curve, the initial

283

attenuation of O2•− could be obtained. Moreover the linear relationship of the initial attenuation of

284

O2•− with the concentrations of C60 was observed and shown in Figure S12 of SI. Therefore we

285

suggested that O2•− quickly reacts with C60, as eq 2.

286

O2•− + nC60



nC60•− + O2

(2)

287

The molecular ratio of C60 to the attenuation of O2•− kept constant (the slope of Figure S12 of SI,

288

27±1), indicating that C60 act as clusters, and the average of n=27. This reaction rate constant cannot

289

be measured due to the experimental setup. However we have examined the decay rates of peroxide

290

(Figure 3), which involved two processes in aqueous C60 solutions. One is the disproportionation of

291

peroxides, as eq 3 & 4. .

292

O2•− + H+ ⇋ HO2•

(3)

293

2 × HO2• → H2O2 + O2

(4)

294 295

Combined eq 3 & 4 to generate eq 5: "#

O2•− + O2•− $%& H2O2 + O2 kobs

(5)

296

Therefore the second-order decay of O2•− is apparent pH depended because the actual reacting

297

species is its conjugate acid, HO2•. Since the concentration of HO2• decreases by 1 order of

298

magnitude per pH unit above the pKa for HO2•/O2•− (4.8), the rate constant for the apparent reaction

299

in eq 5 thus varies with pH as68: Kobs = 5 × 1012[H+] M-1 s-1

300

The other possible process is that peroxide is quenched by C60•−, according to eq 6 & 7.

301

C60•− + HO2• →

C60

+ HO2−

(6) 12

ACS Paragon Plus Environment

Page 13 of 29

Environmental Science & Technology

HO2− + H+ ⇋

302 303

H2O2

(7)

Combine both of the decay processes, the decay of O2•− could be described as eq 8: −

304



()O2 (.

• "

• 63HO• 6 = /012 )O2 - + /" 3C5 "

(8)

305

 Where 3HO•" 6=8 3O•

" 63H 6, therefore eq 8 converts to eq 9

306



307

Since [C60•−] and pH kept constant during the decay of O2•−, eq 9 can be simplified and integrated as

308

eq 10

309

(39• : 6 (.







"  = /012 3O•

" 6 + /" 3;5 638 3O" 63H 6]

?@