Sunlight affects aggregation and deposition of graphene oxide in the aquatic environment

Accepted Manuscript Sunlight Affects Aggregation and Deposition of Graphene Oxide in the Aquatic Environment Indranil Chowdhury, Wen-Che Hou, David Goodwin, Matthew Henderson, Richard G. Zepp, Dermont Bouchard PII:

S0043-1354(15)00221-3

DOI:

10.1016/j.watres.2015.04.001

Reference:

WR 11227

To appear in:

Water Research

Received Date: 17 December 2014 Revised Date:

5 March 2015

Accepted Date: 1 April 2015

Please cite this article as: Chowdhury, I., Hou, W.-C., Goodwin, D., Henderson, M., Zepp, R.G., Bouchard, D., Sunlight Affects Aggregation and Deposition of Graphene Oxide in the Aquatic Environment, Water Research (2015), doi: 10.1016/j.watres.2015.04.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

1

Sunlight Affects Aggregation and Deposition of

3

Graphene Oxide in the Aquatic Environment

RI PT

2

4

Indranil Chowdhury*,†, Wen-Che Hou†,┴, David Goodwin‡, Matthew Henderson§, Richard G.

6

Zepp§, Dermont Bouchard*,§ †

7

National Research Council Associate, National Exposure Research Laboratory, Ecosystems Research Division, U. S. Environmental Protection Agency, Athens, GA, 30605

8

9

M AN U

SC

5

┴

Department of Environmental Engineering, National Cheng Kung University, Tainan City, Taiwan, 70101 ‡

11

13

§

Department of Chemistry, Johns Hopkins University, Baltimore, MD

National Exposure Research Laboratory, Ecosystems Research Division, U. S. Environmental

EP

12

TE D

10

Protection Agency, Athens, GA, 30605

*Address correspondence to either author. Phone: 706-355-8341 (I.C.); 706-355-8333 (D.C.B.)

15

E-mail: [email protected] (I.C.); [email protected] (D.C.B.)

16 17

AC C

14

Submitted to Water Research, December 2014

18 19

1

ACCEPTED MANUSCRIPT

Graphical Abstract

20

M AN U

SC

RI PT

21

22 23 24 25

TE D

26 27 28

31 32 33 34

AC C

30

EP

29

35 36 37

2

ACCEPTED MANUSCRIPT

Abstract

39

In this study, we investigate the role of simulated sunlight on the physicochemical properties,

40

aggregation, and deposition of graphene oxide (GO) in aquatic environments. Results show that

41

light exposure under varied environmental conditions significantly impacts the physicochemical

42

properties and aggregation/deposition behaviors of GO. Photo-transformation has negligible

43

effects on GO surface charge, however, GO aggregation rates increase with irradiation time for

44

direct photo-transformation under both aerobic and anaerobic conditions.

45

conditions, photo-reduced GO has a greater tendency to form aggregates than under aerobic

46

conditions. Aggregation of photo-transformed GO is notably influenced by ion valence, with

47

higher aggregation found in the presence of divalent ions versus monovalent, but adding natural

48

organic matter (NOM) reduces it. QCM-D studies show that deposition of GO on surfaces coated

49

with organic matter decreases with increased GO irradiation time, indicating a potential increase

50

in GO mobility due to photo-transformation. General deposition trends on Suwannee River

51

Humic Acid (SRHA)-coated surfaces are control GO > aerobically photo-transformed GO ≈

52

anaerobically photo-transformed GO. The release of deposited GO from SRHA-coated surfaces

53

decreases with increased irradiation time, indicating that photo-transformed GO is strongly

54

attached to the NOM-coated surface.

56 57

TE D

M AN U

SC

Under anaerobic

EP

AC C

55

RI PT

38

Keywords: Graphene oxide, Photo-transformation, Sunlight, Transport, Aggregation.

58 59 60 61

3

ACCEPTED MANUSCRIPT

62

1. Introduction

63

Graphene, an atomically thin, two-dimensional carbon-based nanomaterial, has been

65

receiving great attention recently in research and applications due to its unique electronic, optical

66

and mechanical properties (Dreyer et al. 2010, Geim 2009, Geim and Novoselov 2007). Potential

67

applications of graphene-based nanomaterials include numerous electronic devices, energy

68

storage, biomedicine, drug delivery, imaging, and even in various environmental pollution

69

management strategies (Compton and Nguyen 2010, Segal 2009, Stankovich et al. 2006, Yang et

70

al. 2013a). Besides exceptional properties and potential applications, usage of graphene-based

71

materials is rapidly increasing due to simple mechanical and chemical process for preparation of

72

bulk quantities of graphene (Compton and Nguyen 2010). To facilitate the aqueous dispersion of

73

graphene, the surface is often modified to generate graphene oxide (GO). Among different

74

graphene materials, GO has been found to be the most toxic (Akhavan and Ghaderi 2010, Duch

75

et al. 2011, Hu and Zhou 2013), which indicates how important it is to understand GO

76

environmental exposures for safe implementation (Hu and Zhou 2013).

EP

TE D

M AN U

SC

RI PT

64

The environmental fate and transport of GO has been investigated and the studies

78

(Chowdhury et al. 2013a, Wu et al. 2013) show that GO can be highly stable against aggregation

79

and deposition in the natural aquatic environment, and it is thus likely that GO remains in the

80

water column where interacting sunlight can result in transformations that alter its properties.

81

Moreover, nitrate and natural organic matter (NOM) in the natural environment can facilitate

82

highly reactive hydroxyl radical formation under sunlight which can further react with graphene

83

nanomaterials.

AC C

77

4

ACCEPTED MANUSCRIPT

84

Transformation is a major factor controlling nanomaterials’ fate in the environment (Lowry et al. 2012).

86

nanomaterials react in natural waters. Previous studies (Hou and Jafvert 2009, Hou et al. 2010,

87

Kong et al. 2013, Kong et al. 2009) have shown that sunlight exposure can photo-chemically

88

transform fullerene (C60), and its derivative fullerol, into CO2 and products with significant

89

oxygen-containing functionalities. For other nanomaterials such as carbon nanotubes, photo-

90

transformation is strongly dependent on the involvement of reactive oxygen species (ROS) such

91

as hydroxyl radical (Chen and Jafvert 2010, Hou et al. 2014). Transport properties of

92

nanomaterials have also been found to be affected by exposure to sunlight (Cheng et al. 2011, Qu

93

et al. 2010). While nC60’s stability has been reported to increase after UVA irradiation due to an

94

increased negative charge (Qu et al. 2010), oxidized multi-walled carbon nanotubes form

95

aggregates under UVC irradiation due to loss of functional groups (Bitter et al. 2014).

SC

M AN U

Recent studies have shown that graphene is photoreactive (Gengler et al. 2013, Koinuma et

TE D

96

Sunlight photolysis is one of the primary routes by which carbonaceous

RI PT

85

al. 2012, Matsumoto et al. 2011, Matsumoto et al. 2010, Zhou et al. 2012).

One study

98

(Matsumoto et al. 2011) reported that UV irradiation can reduce GO which creates many holes

99

and defects due to photoreactions of oxygen-containing functional groups and carbon.

100

(Matsumoto et al. 2011) and (Koinuma et al. 2012) reported formation of nanopores in the

101

vicinity of oxygen-containing functional groups as a result of GO photoreactions in the presence

102

of oxygen under UV irradiation. Another study (Gengler et al. 2013) observed an ultrafast

103

photo-induced chain reaction responsible for GO reduction during UV pulsing that photo-ionizes

104

the solvent, liberating hydrated electrons which trigger reduction (Gengler et al. 2013). Our

105

recent study (Hou et al. 2015) on the transformation of GO under simulated sunlight found that

106

GO readily photo-reacts under simulated sunlight exposure, forming fragmented photoproducts

AC C

EP

97

5

ACCEPTED MANUSCRIPT

similar to reduced GO (rGO) as well as lowmolecular-weight species. We further showed that

108

GO photo-reactivity involves simultaneous formation of oxidative and reductive transient

109

species, and is also dependent on dissolved oxygen level.

RI PT

107

The objective of this study was to determine the effect of sunlight on the aggregation and

111

deposition behavior of GO nanomaterials in aquatic environments. To our knowledge, this is the

112

first study investigating effects of sunlight on the processes that are important in determining GO

113

transport in the environment. Our findings indicate that photo-transformation can significantly

114

alter transport characteristics of GO which are also a function of oxygen level during GO photo-

115

transformation.

116

118

2. Materials and Methods

2.1 Photo-transformed GO Sample Preparation

TE D

117

M AN U

SC

110

119

Materials. GO was obtained from Cheap Tubes Inc. (Brattleboro, VT) in a 2 mg/mL

121

dispersion in pure water; according to the manufacturer, GO was synthesized using the modified

122

Hummer’s Method (Hummers and Offeman 1958). All other chemicals used are the highest

123

purity available from Sigma Aldrich (St. Louis, MO). All aqueous samples were prepared using

124

water purified with an Aqua Solutions 2121BL system (≥ 18.0 MΩ).

AC C

EP

120

125

Irradiation. The photo-transformation of GO was carried out in an Atlas SunTest CPS+

126

solar simulator equipped with a 1 kW xenon arc lamp; details have been described elsewhere

127

(Hou et al. 2015) and described briefly in SI. Briefly, photoreactions of GO in water were

6

ACCEPTED MANUSCRIPT

128

examined under two distinct reaction conditions (air equilibrated (aerobic) and oxygen deficient

129

(anaerobic)) over a period of time (0-200 h).

131

RI PT

130

2.2 Characterization of Photo-transformed GO

A range of characterization was conducted to determine the physicochemical properties

133

of photo-transformed GO. Physical dimensions were determined using a Veeco Multimode

134

Atomic Force Microscopy (AFM) with a Nanoscope V controller and an E scanner (Bruker AXC

135

Inc., Madison, WI). Images were taken under ScanAsyst-Air mode with a Silicon Nitride

136

cantilever (ScanAsyst-Air, Bruker AXC Inc., Madison, WI). AFM images were further analyzed

137

for size distribution using Nanoscope Analysis software (Bruker AXC Inc., Madison, WI).

138

Surface functional groups on photo-transformed GO were determined by X-ray photoelectron

139

spectroscopy (XPS). Electrokinetic and hydrodynamic properties of photo-transformed GO were

140

measured with a ZetaSizer Nano ZS (Malvern Instruments, Worcestershire, U.K.) as a function

141

of irradiation time and NaCl concentration at pH 5.5 ± 0.3. Prior to measurement of

142

electrophoretic mobility and hydrodynamic size, the sample was vortexed for 10 s. All solutions

143

were filtered through a 100-nm filter (Anotop 25, Whatman, Middlesex, UK).

144

describe the standard deviation based on at least three replicates.

146 147

M AN U

TE D

EP

Error bars

AC C

145

SC

132

2.3 Aggregation Study of Photo-transformed GO

148

Change of GO hydrodynamic diameter (Dh) as a function of ionic strength (IS), ion

149

valence, and presence of organic matter was measured with time-resolved dynamic light

150

scattering (TR-DLS) (Bouchard et al. 2012). Suwannee River Humic Acid standard II (SRHA)

7

ACCEPTED MANUSCRIPT

(International Humic Substances Society, St Paul, MN) was used as model NOM and a SRHA

152

stock solution was prepared by accepted procedures (Chen and Elimelech 2008),(Chowdhury et

153

al. 2012). Equal volumes (750 µL) of GO suspension and electrolyte solution (NaCl, CaCl2)

154

were pipetted into a DLS glass cuvette (Malvern Instruments, Worcestershire, U.K.) to achieve

155

specific electrolyte and GO concentrations. For all photo-transformed GO, the concentration

156

used in the aggregation study was 2.5 mg/L total organic carbon (TOC) in order to achieve

157

adequate signal from the light scattering unit. The cuvette was immediately placed in the DLS

158

instrument after vortexing for 10 s. Intensity of scattered light was measured at 173° and the

159

autocorrelation function was allowed to accumulate for 15 s during the aggregation study. The

160

Dh measurements were conducted over periods of 30 to 300 min. Initial aggregation was defined

161

as the period from experiment initiation (t0) to the time when measured Dh values exceeded

162

1.50Dh,initial (Bouchard et al. 2012).

163

proportional to the initial rate of increase of Dh with time (Chen and Elimelech 2006):

166 167 168



→

M AN U

SC



TE D



 

(1)

where N0 is the initial particle concentration.

EP

165




 ∝  

Initial aggregation rate constants (ka) for the GO are

2.4 Deposition and Release Study of Phototransformed GO

AC C

164

RI PT

151

169

GO deposition on, and release from, silica surfaces was investigated using QCM-D (E4,

170

Q-Sense, Västra Frölunda, Sweden), following procedures reported elsewhere (Chen and

171

Elimelech 2006, Chowdhury et al. 2013b, Xiaojun Chang 2013, Yi and Chen 2011).

172

deposition experiments were conducted on bare and SRHA-coated silica crystal sensors (QSX

173

303, Q-Sense). Though mineral surfaces and silica crystal sensors of QCM-D are not fully 8

All

ACCEPTED MANUSCRIPT

174

comparable due to differences in surface heterogeneity and roughness, initial deposition

175

measurements from QCM-D studies can provide fundamental information regarding

176

interactions of photo-transformed GO and mineral surfaces. Details regarding SRHA-coated

177

surface preparation have been described in our previous publication (Chowdhury et al. 2014)

178

and also appear in Appendix A. For all experiments, the flow rate was maintained at 0.1

179

mL/min, in parallel flow configuration, using a peristaltic pump (Ismatec SA, Switzerland) to

180

maintain laminar flow in the module. Solutions inside the chamber were maintained at 25 ±

181

0.2°C. For all photo-transformed GO, the concentration used in the deposition experiment was 1

182

mg/L TOC in order to achieve sufficient deposition of all photo-transformed GO. Ionic strength

183

was maintained at 30 mM NaCl.

186

RI PT

SC

M AN U

185

With deposition, the crystal sensors’ overtone frequencies decrease (i.e., become more negative), following the Sauerbrey relationship (Sauerbery 1959): ∆  

TE D

184

the

∆

(2)



where ∆m is the deposited mass, ∆fn is the shift in overtone frequency, n is the overtone number

188

(1, 3, 5, 7, and …), and C is the crystal constant (17.7 ng/Hz·cm2 for the 5 MHz crystal).

189

Deposition can also increase the crystal’s ability to dissipate energy which can be measured

190

simultaneously with the dissipation unit (D): 

AC C

191

EP

187

 !" #$!"%&

(3)

192

where Edissipation is the energy dissipated in one oscillation and Estored is the total energy stored in

193

the oscillator.

9

ACCEPTED MANUSCRIPT

The Sauerbrey equation is not directly applicable for calculating deposited mass from frequency

195

shift, using Equation 1, since deposited nanomaterials do not form a homogeneous rigid layer on

196

the QCM-D crystals (Reviakine et al. 2011, Sauerbery 1959). Relative deposition behaviors of

197

nanomaterials, however, can be determined from the frequency shifts monitored by QCM-D by

198

calculating deposition rate and attachment efficiency (Chen and Elimelech 2008). Here, shifts in

199

frequency and dissipation were monitored at the third overtone. Initial deposition rates rf are

200

defined as rates of frequency shift in a time period t, respectively: ∆ )



*→

(

(4)

202 203

SC

'  (

3. Results and Discussion

204

M AN U

201

RI PT

194

3.1 Physicochemical Properties of Photo-transformed GO

206

3.1.1 Physical Dimensions and Surface Functionality

TE D

205

207

AFM images of photo-transformed GO for selected conditions are presented in Figures

209

1A-1C and size distributions from the images in Figures 1D-1F. Size distributions from AFM

210

imaging showed that GO broke into smaller fragments from effective diameters of ~200 nm to

211

~120 nm after 11 h of irradiation under aerobic conditions (Figure 1E). XPS results of photo-

212

transformed GO from our previous study (Hou et al. 2015) showed GO was reduced during

213

photo-transformation under aerobic conditions (Table 1), as evidenced by the O:C ratio decrease

214

from 58% to 48% after 11 h irradiation.

AC C

EP

208

10

ACCEPTED MANUSCRIPT

215

AFM imaging also showed GO was breaking into smaller fragments during irradiation

216

under anaerobic conditions (Figure 1C). The photo-fragmentation mechanism as discussed in

217

our earlier report (Hou et al. 2015)

218

photoreactive GO driven by electron-hole pair formation under sunlight conditions.

219

Photoreactivity of GO leads to the formation of CO2 and fragmented reduced GO-like materials

220

as observed by AFM imaging. After 11 h of irradiation, the effective diameter of GO decreased

221

from ~200 nm to ~130 nm which is very similar to the trend observed for GO photo-transformed

222

in the presence of oxygen. XPS results showed the O:C ratio decreased from 58% to 50% after

223

11 h irradiation under anaerobic conditions, similar to the trend for aerobic conditions (Table 1).

224

Overall, AFM and XPS measurements indicated size and surface functional groups of GO

225

remained quite similar for aerobic and anaerobic conditions as a function of irradiation time.

RI PT

SC

M AN U

TE D

226

227

involves concurrent oxidation and reduction of the

3.1.2 Electrokinetic and Hydrodynamic Properties

228

Electrophoretic mobilities (EPMs) and hydrodynamic diameters of GO photo-

230

transformed under aerobic conditions are summarized in Figure 2. The Dh of GO decreased from

231

~200 nm to ~100 nm as irradiation time increased (Figure 2A). The rate of Dh decrease was

232

higher initially, but after 200 h of irradiation there was no further size reduction. This may be

233

due to the breakup of GO flakes during irradiation since AFM imaging confirmed that GO broke

234

into smaller fragments during photo-transformation (Figure 1). A recent study (Andryushina et

235

al. 2014) on GO photo-chemically reduced under UV light observed a similar decrease in

236

hydrodynamic diameter. EPMs of GO as a function of time (Figure 2B) decreased slightly at first

237

and then remained quite stable throughout the rest irradiation. GO was also highly negatively

AC C

EP

229

11

ACCEPTED MANUSCRIPT

charged (EPM < -3 x 10-8 m2V-1s-1 or zeta potential < -40 mV) throughout the irradiation period,

239

indicating GO is completely dispersible throughout since colloids with zeta potentials of less

240

than or equal to -40 mV are considered electrostatically stable, according to ASTM (Li et al.

241

2008, Si and Samulski 2008).

RI PT

238

Hydrodynamic diameters and EPM of GO photo-transformed under anaerobic conditions

243

as a function of irradiation time are presented in Figures 2A and 2B, respectively. Consistent

244

with AFM measurements, the hydrodynamic size of GO decreased from ~200 nm to ~100 nm

245

with irradiation time; the change in hydrodynamic size as a function of irradiation time was

246

quite similar for GO photo-transformed under both aerobic and anaerobic conditions and trends

247

in GO EPM values over time were also similar.

M AN U

SC

242

Figures 2C and 2D present the EPM and hydrodynamic size of aerobically photo-

249

transformed GO, respectively, as a function of NaCl concentration. As expected with colloidal

250

materials, absolute values of EPM of photo-transformed GO decreased with increased NaCl

251

concentration due to charge-screening (Elimelech 1995).

252

transformed GO yielded similar EPM values as a function of NaCl concentration, indicating that

253

irradiation time did not affect surface charge of GO. There were differences in EPM values over

254

the environmentally important range of 1 to 10 mM NaCl (Figure 2C), however, they did not

255

result in hydrodynamic diameter differences (Figure 2D). Above 10 mM NaCl, hydrodynamic

256

diameters of photo-transformed GO were notably affected by NaCl concentration, and increased

257

irradiation time produced larger hydrodynamic size as a function of NaCl concentration.

Above 10 mM NaCl, all photo-

AC C

EP

TE D

248

258

EPM values of GO as a function of NaCl concentration are quite similar under aerobic

259

and anaerobic conditions (Figure 2C), however, hydrodynamic diameters of GO photo-

12

ACCEPTED MANUSCRIPT

transformed are significantly larger under anaerobic conditions (Figure 2D). Irradiation time

261

significantly impacts the hydrodynamic size of photo-transformed GO under anaerobic

262

conditions: larger GO hydrodynamic diameter was observed with increased irradiation time,

263

although EPM remained quite similar.

RI PT

260

264 265

3.2 Aggregation Kinetics of Photo-transformed GO

267

SC

266

3.2.1 Aggregation of GO Photo-transformed under Aerobic Conditions

M AN U

268

Aggregation kinetics of GO photo-transformed under aerobic conditions in the presence

270

of 50 mM NaCl, with and without SRHA, are presented in Figure S1. Initial aggregation rates

271

(ka) were determined from aggregation kinetics of photo-transformed GO (Figure 3). Generally,

272

aggregation of photo-transformed GO increased with irradiation time under aerobic conditions.

273

In the presence of 50 mM NaCl, the initial aggregation rate of control GO is 2.3 nm/min (Table

274

S1), but as irradiation time increased from 0 h to 187 h, initial rates increased from ~2 nm/min to

275

~10 nm/min (Table S1). Although EPM values of GO remained insensitive to irradiation time in

276

the presence of 50 mM NaCl (Figure 2C), the five-fold increase in aggregation from irradiation

277

may be related to changes in surface functional groups of GO; XPS analysis showed those

278

groups decreased with irradiation time (Table 1). An overall decrease in oxygen-containing

279

functionalities (58% to 36%) and increase in graphitic carbon contents (46% to 73%), after 187 h

280

of sunlight exposure, were observed (Hou et al. 2015). Because photo-transformed GO has

281

fewer surface functional groups, a lower NaCl concentration is required to destabilize it. Another

AC C

EP

TE D

269

13

ACCEPTED MANUSCRIPT

282

study (Yi and Chen 2011) on multi-walled carbon nanotubes showed that increasing their surface

283

oxidation resulted in greater stability in the presence of NaCl. In the presence of 1 mM CaCl2, the initial aggregation rate of control GO is 11.6 nm/min

285

(Figure 3). As irradiation time increases from 0 h to 187 h, initial aggregation rates increase

286

from ~10 nm/min to ~90 nm/min, indicating the increased rates are significantly higher in CaCl2

287

than in NaCl. This may be due to the binding capacity of Ca2+ ions with GO functional groups

288

(Chowdhury et al. 2013a, b). We note a drop in aggregation rates as irradiation time increased

289

from 61 h to 187 h and that the decrease is more pronounced in the presence of CaCl2 which may

290

be due to the photo-transformation mechanism of GO discussed in Section 3.4.

M AN U

SC

RI PT

284

Adding SRHA increased stability of photo-transformed GO primarily due to steric

292

effects, similar to the trend we observed previously (Chowdhury et al. 2013a). In 50 mM NaCl

293

with SRHA, negligible aggregation was observed until 11 h of irradiation. Aggregation rates of

294

GO increased to 0.2 nm/min at 187h of irradiation, which is at least one order of magnitude

295

lower than in the absence of SRHA. A similar trend is observed in 1 mM CaCl2 with SRHA,

296

although the decrease in aggregation rates is more significant than in 50 mM NaCl.

299

EP

298

3.2.2 Aggregation of GO Photo-transformed under Anaerobic Conditions

AC C

297

TE D

291

300

In the presence of 50 mM NaCl, initial aggregation rates of GO photo-transformed under

301

anaerobic conditions increased from ~2 nm/min to ~170 nm/min as irradiation time increased

302

from 0 h to 61 h (Figure 3). Aggregation rates of GO photo-transformed under anaerobic

303

conditions were at least one order of magnitude higher than those photo-transformed under 14

ACCEPTED MANUSCRIPT

aerobic conditions, which indicates transformed GO’s aggregation behavior is notably dependent

305

on oxygen level during photo-transformation. At a shorter irradiation time (3 h), however, ka of

306

the anaerobic sample is about seven times higher than the aerobic, while at high irradiation times

307

(11 h and 61 h), ka of the anaerobic sample is much higher (>15 times). This may be due to

308

differences in the photo-transformation mechanism under anaerobic and aerobic conditions

309

discussed in Section 3.4. EPM values of GO photo-transformed under aerobic and anaerobic

310

conditions are very similar, and AFM images showed that sizes of both are also very similar.

311

XPS results showed the oxygen-containing functional groups of GO decreased from ~60% to

312

~40% after 61 h of sunlight exposure for both aerobic and anaerobic conditions (Table 1) (Hou et

313

al. 2015). The higher aggregation of GO phototransformed under anaerobic conditions may be

314

due to lower molecular-weight photoproducts as discussed in Section 3.4.

M AN U

SC

RI PT

304

In the presence of 1 mM CaCl2, aggregation rates of anaerobically photo-transformed GO

316

increased from ~10 nm/min to ~200 nm/min as irradiation time increased from 0 h to 61 h

317

(Figure 3). At a low irradiation time (3h), aggregation rates of photo-transformed GO were quite

318

similar for aerobic and anaerobic samples, however, rates of anaerobically photo-transformed

319

GO are at least two times higher than aerobically photo-transformed GO at higher irradiation

320

times (11h and 61h). This is similar to the trend observed in the presence of NaCl. Overall,

321

aggregation rates of GO are notably higher in the presence of CaCl2.

AC C

EP

TE D

315

322

Adding SRHA reduced the aggregation of anaerobically photo-transformed GO under all

323

conditions. In the presence of 50 mM NaCl, aggregation rates decreased more than 100 times in

324

the presence of SRHA at 11 h irradiation time, however, the rate of decrease was significantly

325

lower at 61 h (Table S1). In the presence of 1mM CaCl2, a similar trend was observed with

326

SRHA. However, the overall decrease in rates of aggregation due to addition of SRHA were 15

ACCEPTED MANUSCRIPT

327

notably lower in CaCl2 than in NaCl solutions. This may be due to the binding capacity of Ca2+

328

ions with functional groups of SRHA and GO (Chowdhury et al. 2013a, b).

330

RI PT

329

3.3 Deposition and Release of Photo-transformed GO

Deposition and release of photo-transformed GO was investigated on silica and SRHA-

332

coated surfaces. Figure S2 shows a negligible amount of photo-transformed GO was deposited

333

on silica surfaces, indicating that its interactions with silica surface will be limited in aquatic

334

environments. The QCM-D sensor frequency shifts at the third overtone during deposition of

335

photo-transformed GO on SRHA surface are shown in Figure 4A; deposition rates (rf) were

336

calculated from initial slopes of the frequency shifts at the third overtone (Figure 4B). All photo-

337

transformed GO showed noticeably lower deposition on SRHA-coated surfaces than control GO,

338

indicating photo-transformation will decrease deposition of GO on many environmental surfaces.

339

Generally, deposition trends on the SRHA surface are control GO > aerobically photo-

340

transformed GO ≈ anaerobically photo-transformed GO. Deposition and release processes of

341

photo-transformed GO are discussed in the following sections.

M AN U

TE D

EP

343

3.3.1 Deposition and Release of GO Photo-transformed under Aerobic Conditions

AC C

342

SC

331

344

Frequency shifts of aerobically photo-transformed GO are always lower than control GO

345

(Figure 4A), showing deposition of aerobically photo-transformed GO on SRHA-coated surface

346

is lower. As irradiation time increased, deposition of GO on the SRHA surface decreased. The

347

deposition rate of control GO on SRHA-coated surface is 2.06 ± 0.17 Hz/min (Figure 4); as

348

irradiation time increased from 0 h to 61 h, the rf value decreased to 1.19 ± 0.09 Hz/min. XPS

349

measurements show the amount of C-O functional groups was reduced from ~45% to ~15% as 16

ACCEPTED MANUSCRIPT

irradiation time increased from 0 h to 187 h, which may contribute to the decreased interaction of

351

GO with the SRHA surface (Hou et al. 2015). A recent study (Chowdhury et al. 2014) showed

352

that interactions of GO with NOM-coated surfaces are primarily governed by functional groups

353

of GO and NOM. Oxygen-containing functional groups of GO can bind with NOM functional

354

groups via hydrogen bonds, Lewis acid-base and π-π interactions (Hartono et al. 2009, Yang et

355

al. 2013b). Hence, reduction of functional groups from GO due to sunlight exposure can result

356

in decreased interactions with NOM-coated surfaces.

SC

RI PT

350

Release results are summarized in Table 1. For the GO control sample, about 14% of the

358

deposited GO was released after introduction of DI water. With increased irradiation time,

359

release of deposited GO decreased. After 3 h of irradiation, a negligible amount (<1%) of

360

deposited GO was released from the SRHA surface, indicating aerobically photo-transformed

361

GO is more strongly attached to the SRHA surface than control GO.

TE D

M AN U

357

362

363

3.3.2 Deposition and Release of GO Photo-transformed under Anaerobic Conditions

EP

364

The deposition rate of anaerobically photo-transformed GO on SRHA surface decreases

366

as irradiation time increased from 3 h to 61 h (Figure 4B). However, deposition rates of

367

anaerobically photo-transformed GO on the SRHA surface are quite similar to aerobically photo-

368

transformed GO for similar irradiation times (Figure 4). Such trends indicate deposition of

369

photo-transformed GO is not a function of oxygen level during irradiation. XPS results showed

370

that changes in functional groups of GO during photo-transformation are quite similar under

371

aerobic and anaerobic conditions (Table 1). Since the interactions of GO with NOM-coated

AC C

365

17

ACCEPTED MANUSCRIPT

372

surfaces mainly depend on these functional groups, similar deposition behavior was observed for

373

GO photo-transformed under aerobic and anaerobic conditions. Release results indicate that <1 % of anaerobically photo-transformed GO is released

375

from the SRHA surface upon introduction of DI water (Table 1). This shows anaerobically

376

photo-transformed GO essentially is irreversibly attached to the SRHA surface. Moreover, at the

377

3 h irradiation time we observed release of aerobically photo-transformed GO, but no release of

378

the anaerobically photo-transformed material, indicating the latter is more strongly attached to

379

the SRHA surface.

M AN U

SC

RI PT

374

380 381

3.4 Mechanisms

Overall we found exposure to sunlight can significantly influence the physicochemical

383

properties of GO and its subsequent transport in the environment. Aggregation of GO increased

384

with higher irradiation time in both aerobic and anaerobic conditions due to successive reduction

385

of GO with increased irradiation time as confirmed by XPS results (Table 1). XPS results

386

showed irradiation mainly reduced C-O functional groups, while no major changes were

387

observed in –COOH and C=O functional groups (Hou et al. 2015). C-O functional groups are

388

primarily located on the basal plane of GO (Dreyer et al. 2010, Gao et al. 2009), indicating that

389

aggregation of photo-transformed GO may be dominated by plate-plate interactions instead of

390

edge-edge, due to reduced functional groups from the basal plane. While it is not entirely

391

possible to separate between plate-plate and edge-edge interactions from aggregation data using

392

dynamic light scattering, another recent study (Wu et al. 2013) also showed that aggregation of

393

GO can be dominated by edge-edge and plate-plate interactions depending on the aquatic

394

chemistry. Besides reduced functional groups, AFM imaging (Figure 1) showed that GO breaks

AC C

EP

TE D

382

18

ACCEPTED MANUSCRIPT

into smaller fragments during irradiation which is further confirmed by hydrodynamic size

396

measurements (Figure 2). Smaller GO fragments possess higher diffusion rates, which can

397

potentially increase inter-particle collision and contribute to larger aggregates of photo-

398

transformed GO (Elimelech 1995).

RI PT

395

A notable difference between GO photo-transformed under aerobic versus anaerobic

400

conditions was observed -- overall, aggregation rates were significantly higher for GO

401

transformed under anaerobic conditions.

402

measurements between conditions, however, indicating electrostatic interactions may not be

403

responsible for this difference. XPS spectra were also quite similar for GO photo-transformed

404

under both conditions providing evidence that differences in surface functional groups had no

405

role in higher aggregation rates under anaerobic conditions. AFM imaging also showed that GO

406

flake sizes are quite comparable for GO photo-transformed under both aerobic and anaerobic

407

conditions.

SC

399

TE D

M AN U

No notable difference was observed in EPM

Our previous study (Hou et al. 2015) found concurrent formations of CO2 and chemically

409

reduced photoproducts which indicates GO photoreactivity results in its disproportionation, a

410

redox reaction in which a species is simultaneously reduced and oxidized to form two different

411

products. Thus, photo-transformed samples are heterogeneous mixtures of GO, photo-reduced

412

GO, as well as low molecular weight species formed during photo-transformation. We showed

413

that aerobic and anaerobic samples exhibit distinct photochemical behaviors in terms of CO2

414

formation and light absorbance, especially after initial irradiation. CO2 formation and light

415

absorbance change remain similar under both conditions within the first 10 h, but are three times

416

more and five times less, respectively, in aerobic samples with longer irradiation. The strong

417

light absorbance of anaerobic samples after initial irradiation agrees with their reduced CO2

AC C

EP

408

19

ACCEPTED MANUSCRIPT

formation, showing that greater light-absorbing photoproduct mass accumulates in anaerobic

419

samples. It also implies that highly oxidized, low molecular weight species may be formed

420

during photo-transformation in aerobic samples, which potentially could influence aggregation

421

of photo-transformed GO (Bai et al. 2014, Radich et al. 2014, Zhou et al. 2012). One recent

422

study (Bai et al. 2014) shows that these low molecular weight species can adsorb to the GO sheet

423

forming an additional layer. Bai et al. (2014) observed an increase in thickness of GO sheets due

424

to sorption of low molecular weight species. Figure 5 presents the thickness of GO sheets as a

425

function of irradiation time. Under both aerobic and anaerobic conditions, the thickness of GO

426

sheets increases with longer irradiation time, which is similar to the trend observed by Bai et al.

427

(2014) for the photo-Fenton reaction of GO. The increase in GO thickness may be due to

428

sorption of low molecular weight species formed in our study. Low molecular weight species in

429

the anaerobic samples, on the other hand, are less oxidized (less CO2 formation) which may

430

contribute to aggregation. The presence of less oxidized lowmolecular weight species thus may

431

result in higher aggregation rate for anaerobic samples; additionally, at short irradiation times (3

432

h), ka of the anaerobic sample is about seven times higher than the aerobic while at high

433

irradiation times (11 h and 61 h), ka of the anaerobic sample is more than 15 times higher. This

434

may be due to differences in lower molecular-weight product oxidation level as reflected by CO2

435

formation. We observed reduced aggregation of aerobic samples at higher irradiation (187 h)

436

which may be due to higher amount of carboxyl functional groups for longer irradiated samples

437

as observed from XPS results (Hou et al. 2015).

AC C

EP

TE D

M AN U

SC

RI PT

418

438 439 440 441 442

20

ACCEPTED MANUSCRIPT

443

4. Conclusions This study shows sunlight exposure can significantly change the physicochemical

445

properties of GO, which subsequently can affect GO aggregation/deposition and therefore

446

transport in the aquatic environment. Direct photo-transformation can increase GO aggregation

447

rate in aqueous suspension, suggesting that it will reduce this material’s stability in the

448

environment. Deposition studies show photo-transformation can reduce deposition of GO on

449

NOM-coated surfaces (the most common environmental surface), indicating that photo-

450

transformed GO will be highly mobile in the environment. However, deposition of photo-

451

transformed GO on NOM-coated surfaces is found to be highly irreversible, indicating that

452

photo-transformation will reduce remobilization of GO in the aquatic environment. Photo-

453

transformation also displayed a high dependence on oxygen level:

454

transformed GO exhibited higher aggregation, indicating GO photo-transformed under low

455

oxygen levels will have limited transportin the environment. However, future research is needed

456

to understand the complex roles of pH, NOM and other natural colloids in the fate of photo-

457

transformed GO. In general, transformation of GO by sunlight exposure can notably impact this

458

emerging material’s fate in the environment. Based on our findings, it is quite possible that GO

459

will undergo transformation in water treatment plants and landfills due to sunlight and UV

460

exposure. Since photo-transformed GO has higher aggregation rates, coagulants in the treatment

461

plant will readily settle out the transformed GO. Hetero-aggregation of photo-transformed GO

462

with NOM-coated surfaces will be low, indicating that removal of transformed GO via filter

463

media (sand) may be less effective in the treatment plant. Furthermore, formation of low

464

molecular weight species (polyaromatic hydrocarbons) will add additional complexity to the fate

465

of photo-transformed GO in treatment plants, which will require future research.

anaerobically photo-

AC C

EP

TE D

M AN U

SC

RI PT

444

21

ACCEPTED MANUSCRIPT

466

Acknowledgments

468

Funding for Indranil Chowdhury was provided by a National Research Council (NRC) Research

469

Associateship Award at EPA. Additional financial support for Wen-Che Hou provided by the

470

Ministry of Science and Technology (MOST) of Taiwan (for Hou) under grant number MOST

471

103-2221-E-006-015-MY3 is acknowledged. This paper has been reviewed in accordance with

472

the USEPA’s peer and administrative review policies and approved for publication. Mention of

473

trade names or commercial products does not constitute endorsement or recommendation for use.

474

M AN U

SC

RI PT

467

Appendix A: Supplementary Data

476

Additional information regarding irradiation study, SRHA-coated surface preparation for QCM-

477

D study, aggregation of photo-transformed GO and QCM-D data for silica surface is available in

478

Appendix A.

TE D

475

479

482

483

AC C

481

EP

480

484

485

22

ACCEPTED MANUSCRIPT

References.

487

Akhavan, O. and Ghaderi, E. (2010) Toxicity of Graphene and Graphene Oxide Nanowalls

488

Against Bacteria. ACS Nano 4(10), 5731-5736.

489

Andryushina, N., Stroyuk, O., Yanchuk, I. and Yefanov, A. (2014) A dynamic light scattering

490

study of photochemically reduced colloidal graphene oxide. Colloid and Polymer Science

491

292(2), 539-546.

492

Bai, H., Jiang, W., Kotchey, G.P., Saidi, W.A., Bythell, B.J., Jarvis, J.M., Marshall, A.G.,

493

Robinson, R.A.S. and Star, A. (2014) Insight into the Mechanism of Graphene Oxide

494

Degradation via the Photo-Fenton Reaction. The Journal of Physical Chemistry C 118(19),

495

10519-10529.

496

Bitter, J.L., Yang, J., Beigzadeh Milani, S., Jafvert, C.T. and Fairbrother, D.H. (2014)

497

Transformations of oxidized multiwalled carbon nanotubes exposed to UVC (254 nm)

498

irradiation. Environmental Science: Nano 1(4), 324-337.

499

Bouchard, D., Zhang, W., Powell, T. and Rattanaudompol, U.s. (2012) Aggregation Kinetics and

500

Transport of Single-Walled Carbon Nanotubes at Low Surfactant Concentrations. Environmental

501

Science & Technology 46(8), 4458-4465.

502

Chang, X. and Bouchard, D.C. (2013) Multiwalled Carbon Nanotube Deposition on Model

503

Environmental Surfaces. Environmental Science & Technology 47(18), 10372-10380.

504

Chen, C.-Y. and Jafvert, C.T. (2010) Photoreactivity of Carboxylated Single-Walled Carbon

505

Nanotubes in Sunlight: Reactive Oxygen Species Production in Water. Environmental Science &

506

Technology 44(17), 6674-6679.

507

Chen, K.L. and Elimelech, M. (2006) Aggregation and Deposition Kinetics of Fullerene (C60)

508

Nanoparticles. Langmuir 22(26), 10994-11001.

AC C

EP

TE D

M AN U

SC

RI PT

486

23

ACCEPTED MANUSCRIPT

Chen, K.L. and Elimelech, M. (2008) Interaction of Fullerene (C60) Nanoparticles with Humic

510

Acid and Alginate Coated Silica Surfaces: Measurements, Mechanisms, and Environmental

511

Implications. Environmental Science & Technology 42(20), 7607-7614.

512

Cheng, Y., Yin, L., Lin, S., Wiesner, M., Bernhardt, E. and Liu, J. (2011) Toxicity Reduction of

513

Polymer-Stabilized Silver Nanoparticles by Sunlight. The Journal of Physical Chemistry C

514

115(11), 4425-4432.

515

Chowdhury, I., Cwiertny, D.M. and Walker, S.L. (2012) Combined Factors Influencing the

516

Aggregation and Deposition of nano-TiO2 in the Presence of Humic Acid and Bacteria.

517

Environmental Science & Technology 46(13), 6968-6976.

518

Chowdhury, I., Duch, M.C., Mansukhani, N.D., Hersam, M.C. and Bouchard, D. (2013a)

519

Colloidal Properties and Stability of Graphene Oxide Nanomaterials in the Aquatic Environment.

520

Environmental Science & Technology 47(12), 6288-6296.

521

Chowdhury, I., Duch, M.C., Mansukhani, N.D., Hersam, M.C. and Bouchard, D. (2013b)

522

Deposition and Release of Graphene Oxide Nanomaterials Using a Quartz Crystal Microbalance.

523

Environmental Science & Technology 48(2), 961-969.

524

Chowdhury, I., Duch, M.C., Mansukhani, N.D., Hersam, M.C. and Bouchard, D. (2014)

525

Interactions of Graphene Oxide Nanomaterials with Natural Organic Matter and Metal Oxide

526

Surfaces. Environmental Science & Technology 48(16), 9382-9390.

527

Compton, O.C. and Nguyen, S.T. (2010) Graphene Oxide, Highly Reduced Graphene Oxide, and

528

Graphene: Versatile Building Blocks for Carbon-Based Materials. Small 6(6), 711-723.

529

Dreyer, D.R., Park, S., Bielawski, C.W. and Ruoff, R.S. (2010) The chemistry of graphene

530

oxide. Chemical Society Reviews 39(1), 228-240.

AC C

EP

TE D

M AN U

SC

RI PT

509

24

ACCEPTED MANUSCRIPT

Duch, M.C., Budinger, G.R.S., Liang, Y.T., Soberanes, S., Urich, D., Chiarella, S.E.,

532

Campochiaro, L.A., Gonzalez, A., Chandel, N.S., Hersam, M.C. and Mutlu, G.M. (2011)

533

Minimizing Oxidation and Stable Nanoscale Dispersion Improves the Biocompatibility of

534

Graphene in the Lung. Nano Letters 11(12), 5201-5207.

535

Elimelech, M., Gregory, J., Jia, X., Williams, R.A. (1995) Particle Deposition and Aggregation:

536

Measurement, Modeling and Simulation, Butterworth-Heinemann.

537

Gao, W., Alemany, L., Ci, L. and Ajayan, P. (2009) New insights into the structure and reduction

538

of graphite oxide. Nature Chemistry 1(5), 403-408.

539

Geim, A.K. (2009) Graphene: Status and Prospects. Science 324(5934), 1530-1534.

540

Geim, A.K. and Novoselov, K.S. (2007) The rise of graphene. Nat Mater 6(3), 183-191.

541

Gengler, R.Y.N., Badali, D.S., Zhang, D., Dimos, K., Spyrou, K., Gournis, D. and Miller, R.J.D.

542

(2013) Revealing the ultrafast process behind the photoreduction of graphene oxide. Nat

543

Commun 4.

544

Hartono, T., Wang, S., Ma, Q. and Zhu, Z. (2009) Layer structured graphite oxide as a novel

545

adsorbent for humic acid removal from aqueous solution. Journal of Colloid and Interface

546

Science 333(1), 114-119.

547

Hou, W.-C., BeigzadehMilani, S., Jafvert, C.T. and Zepp, R.G. (2014) Photoreactivity of

548

Unfunctionalized Single-Wall Carbon Nanotubes Involving Hydroxyl Radical: Chiral

549

Dependency and Surface Coating Effect. Environmental Science & Technology 48(7), 3875-

550

3882.

551

Hou, W.-C., Chowdhury, I., Goodwin, D.G., Henderson, M., Fairbrother, D.H., Bouchard, D.

552

and Zepp, R.G. (2015) Photochemical transformation of graphene oxide in sunlight.

553

Environmental Science & Technology, Article ASAP.

AC C

EP

TE D

M AN U

SC

RI PT

531

25

ACCEPTED MANUSCRIPT

Hou, W.-C. and Jafvert, C.T. (2009) Photochemistry of Aqueous C60 Clusters: Evidence of 1O2

555

Formation and its Role in Mediating C60 Phototransformation. Environmental Science &

556

Technology 43(14), 5257-5262.

557

Hou, W.-C., Kong, L., Wepasnick, K.A., Zepp, R.G., Fairbrother, D.H. and Jafvert, C.T. (2010)

558

Photochemistry of Aqueous C60 Clusters: Wavelength Dependency and Product

559

Characterization. Environmental Science & Technology 44(21), 8121-8127.

560

Hu, X. and Zhou, Q. (2013) Health and Ecosystem Risks of Graphene. Chemical Reviews

561

113(5), 3815-3835.

562

Hummers, W.S. and Offeman, R.E. (1958) Preparation of Graphitic Oxide. Journal of the

563

American Chemical Society 80(6), 1339-1339.

564

Koinuma, M., Ogata, C., Kamei, Y., Hatakeyama, K., Tateishi, H., Watanabe, Y., Taniguchi, T.,

565

Gezuhara, K., Hayami, S., Funatsu, A., Sakata, M., Kuwahara, Y., Kurihara, S. and Matsumoto,

566

Y. (2012) Photochemical Engineering of Graphene Oxide Nanosheets. The Journal of Physical

567

Chemistry C 116(37), 19822-19827.

568

Kong, L., Mukherjee, B., Chan, Y.F. and Zepp, R.G. (2013) Quenching and Sensitizing

569

Fullerene Photoreactions by Natural Organic Matter. Environmental Science & Technology

570

47(12), 6189-6196.

571

Kong, L., Tedrow, O.N., Chan, Y.F. and Zepp, R.G. (2009) Light-Initiated Transformations of

572

Fullerenol in Aqueous Media. Environmental Science & Technology 43(24), 9155-9160.

573

Li, D., Muller, M.B., Gilje, S., Kaner, R.B. and Wallace, G.G. (2008) Processable aqueous

574

dispersions of graphene nanosheets. Nat Nano 3(2), 101-105.

575

Lowry, G.V., Gregory, K.B., Apte, S.C. and Lead, J.R. (2012) Transformations of Nanomaterials

576

in the Environment. Environmental Science & Technology 46(13), 6893-6899.

AC C

EP

TE D

M AN U

SC

RI PT

554

26

ACCEPTED MANUSCRIPT

Matsumoto, Y., Koinuma, M., Ida, S., Hayami, S., Taniguchi, T., Hatakeyama, K., Tateishi, H.,

578

Watanabe, Y. and Amano, S. (2011) Photoreaction of Graphene Oxide Nanosheets in Water. The

579

Journal of Physical Chemistry C 115(39), 19280-19286.

580

Matsumoto, Y., Koinuma, M., Kim, S.Y., Watanabe, Y., Taniguchi, T., Hatakeyama, K.,

581

Tateishi, H. and Ida, S. (2010) Simple Photoreduction of Graphene Oxide Nanosheet under Mild

582

Conditions. ACS Applied Materials & Interfaces 2(12), 3461-3466.

583

Qu, X., Hwang, Y.S., Alvarez, P.J.J., Bouchard, D. and Li, Q. (2010) UV Irradiation and Humic

584

Acid Mediate Aggregation of Aqueous Fullerene (nC60) Nanoparticles. Environmental Science

585

& Technology 44(20), 7821-7826.

586

Radich, J.G., Krenselewski, A.L., Zhu, J. and Kamat, P.V. (2014) Is Graphene a Stable Platform

587

for Photocatalysis? Mineralization of Reduced Graphene Oxide With UV-Irradiated TiO2

588

Nanoparticles. Chemistry of Materials 26(15), 4662-4668.

589

Reviakine, I., Johannsmann, D. and Richter, R.P. (2011) Hearing What You Cannot See and

590

Visualizing What You Hear: Interpreting Quartz Crystal Microbalance Data from Solvated

591

Interfaces. Analytical Chemistry 83(23), 8838-8848.

592

Sauerbery, G. (1959) Verwendung von Schwingquarzen zur Wagung dunner schichten und zur

593

mikrowagung. Zeitschrift Fur Physik 155(2), 206-222.

594

Segal, M. (2009) Selling graphene by the ton. Nat Nano 4(10), 612-614.

595

Si, Y. and Samulski, E.T. (2008) Synthesis of Water Soluble Graphene. Nano Letters 8(6), 1679-

596

1682.

597

Stankovich, S., Dikin, D.A., Dommett, G.H.B., Kohlhaas, K.M., Zimney, E.J., Stach, E.A.,

598

Piner, R.D., Nguyen, S.T. and Ruoff, R.S. (2006) Graphene-based composite materials. Nature

599

442(7100), 282-286.

AC C

EP

TE D

M AN U

SC

RI PT

577

27

ACCEPTED MANUSCRIPT

Wu, L., Liu, L., Gao, B., Muñoz-Carpena, R., Zhang, M., Chen, H., Zhou, Z. and Wang, H.

601

(2013) Aggregation Kinetics of Graphene Oxides in Aqueous Solutions: Experiments,

602

Mechanisms, and Modeling. Langmuir 29(49), 15174-15181.

603

Yang, Y., Asiri, A.M., Tang, Z., Du, D. and Lin, Y. (2013a) Graphene based materials for

604

biomedical applications. Materials Today 16(10), 365-373.

605

Yang, Z., Yan, H., Yang, H., Li, H., Li, A. and Cheng, R. (2013b) Flocculation performance and

606

mechanism of graphene oxide for removal of various contaminants from water. Water Research

607

47(9), 3037-3046.

608

Yi, P. and Chen, K.L. (2011) Influence of Surface Oxidation on the Aggregation and Deposition

609

Kinetics of Multiwalled Carbon Nanotubes in Monovalent and Divalent Electrolytes. Langmuir

610

27(7), 3588-3599.

611

Zhou, X., Zhang, Y., Wang, C., Wu, X., Yang, Y., Zheng, B., Wu, H., Guo, S. and Zhang, J.

612

(2012) Photo-Fenton Reaction of Graphene Oxide: A New Strategy to Prepare Graphene

613

Quantum Dots for DNA Cleavage. ACS Nano 6(8), 6592-6599.

SC

M AN U

TE D

AC C

616

EP

614 615

RI PT

600

28

ACCEPTED MANUSCRIPT

List of Tables and Figures No Table 1

Title Summary of XPS results and release of photo-transformed GO from SRHA-coated

Figure 1

RI PT

surface upon injection of DI water

AFM images of A) GO control, B) GO phototransformed aerobically after 11 h of irradiation, and C) GO phototransformed anaerobically after 11 h of irradiation. The

SC

size distribution of GO for different conditions are provided in Figures D, E and F, respectively.

(A) Hydrodynamic diameter and (B) Electrophoretic mobilities (EPM) of photo-

M AN U

Figure 2

transformed GO as a function of irradiation time. (C) EPM and (D) Hydrodynamic diameter of photo-transformed GO as a function of NaCl concentration. GO concentration was maintained 2.5 mg/L TOC at pH 5.5 ± 0.3. Initial aggregation rates of photo-transformed GO in the presence of (A) 50 mM NaCl

TE D

Figure 3

without and with 1 mg/L SRHA and (B) 1 mM CaCl2 without and with 1 mg/L SRHA. GO concentration was maintained 2.5 mg/L TOC at pH 5.5 ± 0.3. Deposition of photo-transformed GO on SRHA-coated surface using QCM-D in 30

EP

Figure 4

mM NaCl. (A) Frequency shift at the third overtone (∆f3) and (B) Initial deposition

AC C

rates. Step i: formation of SRHA layer on silica surface; Step ii: injection of

phototransformed GO; Step iii: injection of background electrolyte; and Step iv: injection of DI water. GO concentration was maintained 1 mg/L TOC at pH 5.5 ± 0.3.

Figure 5

Thickness of GO sheets as a function of irradiation time under aerobic and anaerobic conditions. Thickness was measured from AFM images using Nanoscope Analysis software.

29

ACCEPTED MANUSCRIPT

Table 1. Summary of XPS results and release of photo-transformed GO from SRHA-coated surface upon injection of DI water Conditions

Release rates

RI PT

0.17 ± 0.01 0.16 ± 0.03 0.12 ± 0.01 NR NR NR NR NR NR

SC

0.58 43 Control 0.55 36 Air-1h 0.54 31 Air -3h 0.48 26 Air-11h 0.46 20 Air-61h 0.36 15 Air-187h 0.56 35 N2-3h 0.50 25 N2-11h 0.40 17 N2-61h NR: Negligible (<1%) amount of release

EP

TE D

Data reproduced from our previous study (Hou et al. 2015).

AC C

a

(1/min)

%C-Oa

M AN U

O:Ca

% Release

30

13.39 ± 1.25 10.03 ± 0.79 4.40 ± 0.47 NR NR NR NR NR NR

ACCEPTED MANUSCRIPT

GO Control

GO-Air 11 h

A

GO-N2 11h

C

Diameter: 123.01 nm Area: 17237.16 nm2

25

25

D

10 5

TE D

15

15 10

EP

5

0

0 100 200 300 400 500 600 700 800

Effective Diameter (nm)

0

AC C

0

25

E

20

GO (%)

GO (%)

20

Diameter: 130.60 nm Area: 22554.28 nm2

F

20

GO (%)

Diameter: 202.92 nm Area: 50409.81 nm2

M AN U

SC

RI PT

B

15 10 5 0

100 200 300 400 500 600 700 800

Effective Diameter (nm)

0

100 200 300 400 500 600 700 800

Effective Diameter (nm)

Figure 1. AFM images of A) GO control, B) GO phototransformed aerobically after 11 h of irradiation, and C) GO phototransformed anaerobically after 11 h of irradiation. The size distribution of GO for different conditions are provided in Figures D, E and F, respectively.

31

200 150 100 50 0 0

100

200

300

400

500

-3 -4 Control Air-3h Air-11h Air-61 h N2-3h

EP

-5 -6

N2-11h

-7 0

20

40

60

80

100

-4 -5 -6 -7 100

120

200

300

400

500

D

Control Air-3h Air-11h Air-61h N2-3h

2000 1500

600

Time (h)

2500

TE D

-2

Hydrodynamic Diameter (nm)

C

AC C

EPM (10-8 m2 V-1 s-1)

Time (h) -1

-3

0

600

B

GO-air GO-N2

-2

RI PT

GO-air GO-N2

-1

SC

A

M AN U

250

EPM (10-8 m2 V-1 s-1)

Hydrodynamic Diameter (nm)

ACCEPTED MANUSCRIPT

N2-11h

1000 500 0 0

20

40

60

80

100

120

NaCl Concentration (mM)

NaCl Concentration (mM)

Figure 2. A) Hydrodynamic diameter and (B) Electrophoretic mobilities (EPM) of photo-transformed GO as a function of irradiation time. (C) EPM and (D) Hydrodynamic diameter of photo-transformed GO as a function of NaCl concentration. GO concentration was maintained 2.5 mg/L TOC at pH 5.5 ± 0.3. 32

MAggregation Rates, k (nm/min) AN US CR IP T

200

200

A

50 mM NaCl 50 mM NaCl + SRHA

a

150

B

1 mM CaCl2 1 mM CaCl2 +SRHA

150 100

100

50

50

TE D

N2 -3 h N2 -1 1h N2 -6 1h

ai r-1 1h Ai r6 1 h ai r-1 87 h

r-3 h

ai

r-1 h

ai

C

h r-1 87 h N2 -3 h N2 -1 1h N2 -6 1h ai

r6 1 Ai

r-1 1h ai

ai r-3 h

ai r-1 h

on tro l

on tro l

0

0 C

Aggregation Rates, ka (nm/min)

ACCEPTED MANUSCRIPT

AC C

EP

Figure 3. Initial aggregation rates of photo-transformed GO in the presence of (A) 50 mM NaCl without and with 1 mg/L SRHA and (B) 1 mM CaCl2 without and with 1 mg/L SRHA. GO concentration was maintained 2.5 mg/L TOC at pH 5.5 ± 0.3.

33

A

-20

2.5 2.0

-40 -60

1.5

Control Air-1 h Air-3h Air-11h Air-61 h N2- 11 h

-80 -100 -120

1.0

N2-61 h

-140

0.5

i

ii

iii

-160 0

2000

4000

6000

iv

8000 10000 12000 14000 16000 18000

TE D

C

Time (s)

0.0

on tro l ai r-1 h ai r-3 h ai r-1 1h ai r-6 1h ai r-1 87 h N2 -3 h N2 -1 1h N2 -6 1h

Frequency Shift (Hz)

0

MDeposition Rates (Hz/min) AN US CR IP T

ACCEPTED MANUSCRIPT

AC C

EP

Figure 4. Deposition of photo-transformed GO on SRHA-coated surface using QCM-D in 30 mM NaCl. (A) Frequency shift at the third overtone (∆f3) and (B) Initial deposition rates. Step i: formation of SRHA layer on silica surface; Step ii: injection of phototransformed GO; Step iii: injection of background electrolyte; and Step iv: injection of DI water. GO concentration was maintained 1 mg/L TOC at pH 5.5 ± 0.3.

34

B

ACCEPTED MANUSCRIPT

RI PT

4

Air N2

3

SC

2 1 0 0h

1h

M AN U

Thickness (nm)

5

3h

11 h

61 h

TE D

Irradiation Time (h)

AC C

EP

Figure 5. Thickness of GO sheets as a function of irradiation time under aerobic and anaerobic conditions. Thickness was measured from AFM images using Nanoscope Analysis software.

35

ACCEPTED MANUSCRIPT

Highlights.

RI PT

SC

M AN U TE D

•

EP

• • •

Aggregation and deposition of graphene oxide (GO) can be significantly influenced by sunlight. Sunlight exposure can increase aggregation of GO Photo-transformation can reduce deposition of GO on natural organic matter surface Aggregation and deposition of GO is a function of oxygen level during phototransformation Remobilization of deposited GO from surfaces can be reduced due to phototransformation

AC C

•

1

ACCEPTED MANUSCRIPT

Appendix A: Supplementary Information

RI PT

Sunlight Affects Aggregation and Deposition of Graphene Oxide in Aquatic Environment

National Research Council Associate, National Exposure Research Laboratory, Ecosystems Research Division, U. S. Environmental Protection Agency, Athens, GA, 30605

M AN U

†

SC

Indranil Chowdhury*,†, Wen-Che Hou†,┴, David Goodwin‡, Matthew Henderson§, Richard G. Zepp§, Dermont Bouchard*,§

┴

Department of Environmental Engineering, National Cheng Kung University, Tainan City, Taiwan, 70101 ‡

§

Department of Chemistry, Johns Hopkins University, Baltimore, MD

National Exposure Research Laboratory, Ecosystems Research Division, U. S. Environmental Protection Agency, Athens, GA, 30605

TE D

*Address correspondence to either author. Phone: 706-355-8341 (I.C.); 706-355-8333 (D.C.B.)

AC C

EP

E-mail: [email protected] (I.C.); [email protected] (D.C.B.)

S1

ACCEPTED MANUSCRIPT

Irradiation Study. The sunlight experiments were carried out in an Atlas SunTest CPS+ solar simulator

RI PT

equipped with a 1kW xenon arc lamp (Hou et al. 2015). Reaction vessels were 24 mL quartz tubes to which 20 mL of reaction solution was added. Samples were not pH-buffered but the pH was monitored before and after irradiation and stayed in the range of 4.0-4.3. The sample tubes were sealed with open-top caps lined with gastight PTFE septa, submerged in a thermostated

SC

water bath (25oC) during irradiation. The incident light intensity at the tube surface, summed

M AN U

from 290 to 700 nm, was 0.065 W/cm2. For kinetic studies, a series of tubes were prepared for irradiation. At specific time periods during irradiation, one tube was removed from the reactor and sacrificed for chemical analysis. Dark control tubes were covered by aluminum foil and irradiated concurrently and analyzed in the same manner. Therefore, any differences observed could be attributed solely to light irradiation. We examined the effect of dissolved oxygen (DO)

TE D

level on GO’s photoreactivity under air-equilibrated and O2-deficient conditions. To create O2deficient samples, we removed dissolved O2 from GO samples by bubbling pure N2 gas into the tubes for 1 h prior to irradiation. N2 bubbling was performed by penetrating a needle through the

EP

septa and below the liquid surface; another needle above the liquid surface equated pressure in

AC C

the tubes with the atmosphere above the septa.

Formation of Suwannee River Humic Acid (SRHA)-coated Surface on QCM-D Sensor A silica-coated crystal sensor (QSX 303, Q-Sense, Västra Frölunda, Sweden) was utilized for deposition and release studies for SRHA-coated surfaces. Prior to deposition experiments, the aluminum oxide and silica crystal sensors were cleaned following the protocol described in the S2

ACCEPTED MANUSCRIPT

QCM-D manual, and referenced elsewhere (Chang and Bouchard 2013, Chowdhury et al. 2013). SRHA-coated surfaces were prepared following a published protocol (Chen and Elimelech 2008, Chowdhury et al. 2014).

First, silica surfaces were coated with cationic poly-L-lysine

RI PT

hydrobromide (PLL, molecular weight 70,000-150,000 by viscosity, P-1274, Sigma Aldrich, St. Louis, MO) (Chowdhury et al. 2013, Chen and Elimelech 2008, de Kerchove and Elimelech 2006, Yi and Chen 2011). Next, 2 mL of HEPES solution was flowed over the PLL layer

SC

followed by 2 mL of 1 mM NaCl. Then, 30 mg/L of SRHA prepared in 1 mM NaCl was injected until the frequency shift stabilized. Adsorption of SRHA led to a sharp frequency shift

M AN U

of about 10 Hz. After the SRHA layers formed over the PLL-coated silica surface, 1 mM NaCl

AC C

EP

TE D

was flowed through QCM-D crystal for at least 20 min to remove any unadsorbed SRHA.

S3

ACCEPTED MANUSCRIPT

Table S1. Initial aggregation rates (ka) of phototransformed GO

EP

TE D

M AN U

2.32 2.45 3.87 7.07 10.14 8.81 29.41 120.50 167.71

AC C

Control air-1h air-3h air-11h air 61 h air-187h N2-3h N2-11h N2-61h

1 mM CaCl2 + SRHA 0.12 0.18 0.20 1.43 0.71 0.49 0.56 6.11 81.37

RI PT

50 mM NaCl

Initial Aggregation Rates (nm/min) 50 mM NaCl + 1 mM CaCl2 SRHA NA 11.63 NA 21.82 NA 42.48 0.05 55.49 0.08 85.52 0.23 55.68 NA 48.78 1.04 128.57 51.63 191.75

SC

Conditions

S4

1000

0 0

500

1000

1500

2000

1000

C

Control AIr-1h Air-3h Air-11h Air-61h Air-187h N2-11h N2-61h

TE D

800 600

EP

400

0 0

500

1000

Time (s)

AC C

200

1500

1500 1000 500 0

M AN U

500

B

2000

Control Air-1h Air-3h Air-11h Air-61h Air-187h N2-11h N2-61h

RI PT

1500

2000

SC

Control Air-1h Air-3h Air-11h Air-61h Air-187h N2-11h N2-61h

A

Time (s)

Hydrodynamic Diameter (nm)

Hydrodynamic Diameter (nm)

2000

Hydrodynamic Diameter (nm)

Hydrodynamic Diameter (nm)

ACCEPTED MANUSCRIPT

0

1000

500

1000

1500

2000

Time (s)

D

Control Air-1h Air-3h Air-11h Air-61h Air-187h N2-11h N2-61h

800 600 400 200 0

0

500

1000

1500

2000

Time (s)

Figure S1. Aggregation kinetics of phototransformed GO in A) 50 mM NaCl and B) 1 mM CaCl2. In the presence of SRHA, aggregation kinetics of phototransformed GO in C) 50 mM NaCl, and D) 1 mM CaCl2.

S5

ACCEPTED MANUSCRIPT

A

RI PT

2 0 -2

-6 Control GO-air- 3 h GO-air- 61 h

-8 -10 0

1000 2000 3000 4000 5000 6000

B

TE D

0 -2 -4

EP

-6 -8 -10

AC C

Frequency Shift (Hz)

Time (s)

2

SC

-4

M AN U

Frequency Shift (Hz)

4

0

2000

GO-air-187 h- 30 mM GO-air-187h- 50 mM GO-air-187h-100 mM GO-N2-11 h- 50 mM GO-H2O2- 7.75h - 50 mM

4000

6000

8000

Time (s)

Figure S2. Frequency shifts at the third overtone (∆f3) while GO was depositing on the silica surface A) in the presence of 10 mM NaCl, and B) as a function of NaCl concentration.

S6

RI PT

ACCEPTED MANUSCRIPT

M AN U

SC

GO Control

GO N2 3h

AC C

EP

TE D

GO Air 3 h

GO Air 61 h

GO N2 61 h

Figure S3. 3D images of photo-transformed GO as determined by atomic force microscope (AFM). 3D images were generated using Nanoscope Analysis Software.

S7

ACCEPTED MANUSCRIPT

References. Hou, W.-C., Chowdhury, I., Goodwin, D.G., Henderson, M., Fairbrother, D.H., Bouchard, D.

RI PT

and Zepp, R.G. (2015) Photochemical transformation of graphene oxide in sunlight. Environmental Science & Technology, Article ASAP.

Chang, X. and Bouchard, D.C. (2013) Multiwalled Carbon Nanotube Deposition on Model

SC

Environmental Surfaces. Environmental Science & Technology 47(18), 10372-10380.

Chowdhury, I., Duch, M.C., Mansukhani, N.D., Hersam, M.C. and Bouchard, D. (2013)

M AN U

Deposition and Release of Graphene Oxide Nanomaterials Using a Quartz Crystal Microbalance. Environmental Science & Technology 48(2), 961-969.

Chen, K.L. and Elimelech, M. (2008) Interaction of Fullerene (C60) Nanoparticles with Humic Acid and Alginate Coated Silica Surfaces: Measurements, Mechanisms, and Environmental Implications. Environmental Science & Technology 42(20), 7607-7614.

TE D

Chowdhury, I., Duch, M.C., Mansukhani, N.D., Hersam, M.C. and Bouchard, D. (2014) Interactions of Graphene Oxide Nanomaterials with Natural Organic Matter and Metal Oxide Surfaces. Environmental Science & Technology 48(16), 9382-9390.

EP

de Kerchove, A.J. and Elimelech, M. (2006) Formation of Polysaccharide Gel Layers in the

121.

AC C

Presence of Ca2+ and K+ Ions:  Measurements and Mechanisms. Biomacromolecules 8(1), 113-

Yi, P. and Chen, K.L. (2011) Influence of Surface Oxidation on the Aggregation and Deposition Kinetics of Multiwalled Carbon Nanotubes in Monovalent and Divalent Electrolytes. Langmuir 27(7), 3588-3599.

S8