Heteroaggregation of Graphene Oxide with Nanometer- and

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Heteroaggregation of Graphene Oxide with Nanometerand Micrometer-Sized Hematite Colloids: Influence on Nanohybrid Aggregation and Microparticle Sedimentation Yiping Feng, Xitong Liu, Khanh An Huynh, J. Michael McCaffery, Liang Mao, Shixiang Gao, and Kai Loon Chen Environ. Sci. Technol., Just Accepted Manuscript • Publication Date (Web): 15 May 2017 Downloaded from http://pubs.acs.org on May 20, 2017

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

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

Page 1 of 29

Environmental Science & Technology

1 2

Heteroaggregation of Graphene Oxide with Nanometer- and

3

Micrometer-Sized Hematite Colloids: Influence on Nanohybrid

4

Aggregation and Microparticle Sedimentation

5 6 7 8

Revised manuscript submitted to:

9 10


11 12

May 7, 2017

13

Yiping Feng,1, 2, 3, *, Xitong Liu,1, Khanh An Huynh,4 J. Michael

14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

McCaffery,5 Liang Mao,2 Shixiang Gao,2 and Kai Loon Chen1

⊥

1

2

⊥

Department of Environmental Health and Engineering, Johns Hopkins University, Baltimore, Maryland 21218-2686, United States

State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing, Jiangsu 210023, China 3

Institute of Environmental Health and Pollution Control, School of Environmental Science and Engineering, Guangdong University of Technology, Guangzhou 510006, China

4

5

Faculty of Environment and Natural Resources, Ho Chi Minh City University of Technology, Ho Chi Minh City, Vietnam

The Integrated Imaging Center, Department of Biology, Johns Hopkins University, Baltimore, Maryland 21218-2686, United States

* Corresponding author: Yiping Feng, E-mail: [email protected], Phone: 86-20-39322547.

1

ACS Paragon Plus Environment


33

Abstract

34

Heteroaggregation of graphene oxide (GO) with nanometer- and micrometer-sized

35

hematite colloids, which are naturally present in aquatic systems, is investigated in this

36

study. The heteroaggregation rates between GO and hematite nanoparticles (HemNPs)

37

were quantified by dynamic light scattering, while the heteroaggregation between GO

38

and micrometer-sized hematite particles (HemMPs) was examined through batch

39

adsorption and sedimentation experiments. The heteroaggregation rates of GO with

40

HemNPs first increased and then decreased with increasing GO/HemNP mass

41

concentration ratios. The conformation of GO–HemNP heteroaggregates at different

42

GO/HemNP mass concentration ratios was observed through transmission electron

43

microscopy imaging. Initially, GO underwent heteroaggregation with HemNPs through

44

electrostatic attraction to form primary heteroaggregates, which were further bridged by

45

GO to form bigger clusters. At high GO/HemNP mass concentration ratios where GO

46

outnumbered HemNPs, heteroaggregation resulted in the formation of stable GO–

47

HemNP nanohybrids that have a critical coagulation concentration of 308 mM NaCl at

48

pH 5.2. In the case of HemMPs, GO adsorbed readily on the microparticles and, at an

49

optimal GO/HemMP ratio of ca. 0.002, the sedimentation of HemMPs was fastest, most

50

likely due to the formation of “electrostatic patches” which led to favorable aggregation

51

of the microparticles.

52 53

Introduction

54

Graphene oxide nanosheets (GO), due to their outstanding physicochemical

55

properties,1, 2 are expected to have many potential applications.2-4 During the production

56

and use of GO-containing consumer products as they enter the market (e.g., GO-based

2


Page 2 of 29

Page 3 of 29


57

sensors for cancer detection, thermal paste for cooling CPU, and electrodes for lithium

58

ion batteries),5-7 it is inevitable that GO will be released into the environment, possibly

59

causing adverse impacts on human health and ecosystems.8-14 Due to their layered

60

structure and the distribution of oxygen-containing functional groups on the basal planes

61

and sheet edges,2, 15, 16 GO has excellent dispersibility and high mobility in the aqueous

62

environment. Dispersed GO has been reported to have higher cytotoxicity than the

63

aggregated ones because the dispersed nanosheets have more opportunities to interact

64

with cells directly.11 Therefore, it is essential to investigate the aggregation behavior of

65

GO in aquatic environments in order to have a better understanding about their impact

66

and potential risk to human health and aquatic organisms.

67

In natural waters, the concentration of GO is likely to be several orders of magnitude

68

lower than the concentration of natural colloids.17, 18 Thus, the heteroaggregation of GO

69

with natural colloids (e.g., metal oxides, clay particles, and macromolecular organic

70

matter), rather than the homoaggregation of GO, is expected to be the dominant process

71

that governs the fate and transport of GO.17-21 To date, only a few studies have focused

72

on the heteroaggregation of GO in aquatic environments. Negatively charged GO has

73

been observed to interact favorably with positively charged particles via electrostatic

74

attraction.22-24 For example, Zhao et al.24 investigated the heteroaggregation of GO with

75

different minerals through adsorption studies.

76

irreversibly adsorbed to positively charged goethite, but not to negatively charged

77

montmorillonite or kaolinite.24 Nevertheless, no research has been conducted to quantify

78

the heteroaggregation rates between GO and natural colloids. In particular, the difference

79

in the concentration ratio of GO to natural colloids in various environmental settings

80

(such as high ratio near the point where GO-containing wastewater is discharged and

3

The authors found that GO was



81

lower ratio downstream in the river) can result in different heteroaggregation rates

82

between GO and natural colloids. In addition, the heteroaggregation between GO and

83

HemNPs can lead to the formation of GO-HemNP nanohybrids, the colloidal stability of

84

which remains largely unknown. Moreover, it is expected that the wide size distribution

85

of natural colloids in natural waters (i.e., from nanometer to micrometer size range) can

86

influence the stability of GO through different mechanisms.25 Therefore, the study of GO

87

heteroaggregation at environmentally relevant conditions is needed in order to better

88

understand and predict the environmental fate and transport of these nanosheets.

89

The objective of this work is to investigate the heteroaggregation behavior of GO

90

with nanometer- and micrometer-sized hematite particles (HemNPs and HemMPs,

91

respectively), both chosen as representative natural colloids. The heteroaggregation rate

92

of GO with HemNPs was determined over a broad range of GO/HemNP mass

93

concentration ratios through time-resolved dynamic light scattering (DLS). The

94

conformation of GO–HemNP heteroaggregates was directly observed under a

95

transmission electron microscope (TEM). The effect of GO on the sedimentation of

96

HemMPs was also investigated through sedimentation and batch adsorption experiments,

97

as well as electrophoretic mobility measurements.

98

heteroaggregation between GO and HemNPs or HemMPs for different mass

99

concentration ratios were proposed based on our experimental results.

Finally, the mechanisms for

100 101

Experiment Section

102

Preparation of GO and Hematite Colloids. Single-layered GO nanosheets

103

(SKU-GO-005) are purchased from Graphene Supermarket (Reading, MA). Details for

104

the preparation of GO stock suspensions are provided in the Supporting Information (SI).

4


Page 4 of 29

Page 5 of 29


105

Both hematite colloids, HemNPs and HemMPs, were synthesized through the forced

106

hydrolysis of FeCl3.26, 27 The method used to synthesize hematite colloids is presented in

107

SI.

108

Characterization of GO and Hematite Colloids. The hydrodynamic diameters

109

of GO and HemNPs in the stock suspensions were measured through DLS. These

110

measurements were performed using a BI-200SM goniometer (Brookhaven, Holtsville,

111

NY) equipped with a 488 nm argon laser at a scattering angle of 90°.28 The shape and

112

size distribution of HemMPs were determined through TEM imaging (Philips EM 420

113

TEM with an accelerating voltage of 100 kV). The electrophoretic mobilities (EPMs) of

114

GO and hematite colloids were measured with a ZetaPALS analyzer (Brookhaven,

115

Holtsville, NY) at 0.1 mM NaCl and pH 5.2 ± 0.2.

116

Solution Chemistry. A NaCl stock solution was prepared by dissolving ACS-

117

grade NaCl in deionized (DI) water (18 MΩ•cm, Milli-Q system, Millipore, MA) and

118

then filtered through a 0.1-µm PVDF syringe filter (Millipore, MA). All aggregation

119

experiments were performed at an unadjusted pH of 5.2 ± 0.2.

120

Determination of Homoaggregation Kinetics and Heteroaggregation Rates.

121

Time-resolved DLS measurements were used for the determination of homoaggregation

122

kinetics and heteroaggregation rates. Homoaggregation kinetics of GO and HemNPs

123

were determined to evaluate the stability of these nanomaterials under different

124

concentrations of NaCl.

125

efficiencies, α. Detail procedures for the determination of α29, 30 and critical coagulation

126

concentrations (CCCs)31 of GO and HemNPs are presented in the SI.

Homoaggregation kinetics was quantified using attachment

127

Heteroaggregation experiments between GO and HemNPs were performed at a

128

constant HemNP concentration of 0.44 mg/L and at a NaCl solution of 0.1 mM (pH 5.2)

5



Page 6 of 29

129

using the method reported in our previous publication.32 This solution chemistry was

130

chosen to avoid the homoaggregation of both GO and HemNP.32 In addition, the pH of

131

5.2 lies within the typical pH range of 5–9 for most natural waters.33, 34 The HemNP

132

concentration of 0.44 mg/L is representative of typical concentrations of iron in river

133

waters.35, 36 GO concentrations were varied from 0.9 to 440.0 µg TOC/L to achieve the

134

GO/HemNP mass concentration ratios in the range of 0.002–1.000.

135

heteroaggregation rates of GO and HemNPs were determined through time-resolved DLS

136

measurements. Notably, the scattered light intensity from the HemNP suspension (0.44

137

mg/L) was significantly higher (> 18 times) than the intensities of the GO suspensions

138

(0.9–440.0 µg TOC/L). During the process of heteroaggregation, the increase in the

139

hydrodynamic diameter (Dh) was recorded over 5–60 min for the accurate derivation of

140

the initial heteroaggregation rates. The linear regression of heteroaggregation rates was

141

performed over a time period in which Dh,t reached 1.3 times of the initial hydrodynamic

142

diameter of HemNPs, Dh,0. When the heteroaggregation was very slow (i.e., Dh,t failed to

143

reach 1.3 Dh,0), the regression would be performed over a duration of 60 min. This

144

approach to calculating the rates of aggregation has been employed in our previous

145

publications.28, 30 In all cases, the y-intercepts of the fitted lines did not exceed 8 nm in

146

excess of the initial hydrodynamic diameter of HemNPs (Dh,0 = 90 nm).29

The

147

Homoaggregation Kinetics and EPMs of GO–HemNP Nanohybrids. GO and

148

HemNP suspensions were mixed together at 0.1 mM NaCl and pH 5.2 to obtain a 10-mL

149

GO–HemNP mixture containing 12.4 mg TOC/L GO and 1.24 mg/L HemNPs. GO and

150

HemNPs were allowed to undergo heteroaggregation for 30 min. At these concentrations,

151

the HemNPs were expected to be completely coated with GO to form GO–HemNP

152

nanohybrids (to be discussed in Results and Discussion section). The suspension was

6


Page 7 of 29


153

then centrifuged at 3,650 g (Beckman Coulter, CA) for 1 h to separate the GO–HemNP

154

nanohybrids from the suspension. The supernatant (90 % of total volume) was carefully

155

decanted and replaced with an equal volume of deionized (DI) water (Millipore, MA).

156

This cleaning process was repeated once.

157

suspension was sonicated for 10 min in order to obtain a monodispersed nanohybrid

158

suspension. The homoaggregation kinetics of these nanohybrids was determined as a

159

function of NaCl concentration at pH 5.2 through time-resolved DLS.

160

measurements, the nanohybrid suspension was diluted ca. 5 times with 0.1 mM NaCl

161

solution, resulting in a GO concentration of 2.33 mg TOC/L and a HemNP concentration

162

of 0.23 mg/L.

The obtained GO–HemNP nanohybrid

For EPM

163

Sedimentation Rates of HemMPs and HemNPs in the Presence of GO. Due

164

to the fast sedimentation associated with the large HemMPs (~ 2 µm), we were not able

165

to perform DLS measurements on the HemMP suspensions.

166

experiments were conducted on HemMPs under varying concentrations of GO to provide

167

insights into the heteroaggregation behavior between HemMPs and GO.

168

sedimentation of HemMPs was examined at 0.1 mM NaCl and pH 5.2.

169

experiments, the HemMP concentration was fixed at 150 mg/L, while the GO

170

concentrations were 0.00, 0.25, 4.00, and 8.00 mg TOC/L. Immediately after being

171

prepared in polystyrene cuvettes (BI-SCP, Brookhaven), the GO–HemMP suspensions

172

were mixed using a rotating disk (TC-7, New Brunswick Scientific, 30 rpm). After 30

173

min of mixing, the absorbance values of the mixtures at a wavelength of 760 nm, at

174

which the absorption of GO was negligible (Figure S1a), were recorded over 60 min

175

using a UV–Vis spectrometer (UV-1800, Shimadzu, Japan). The adsorption of GO on

176

HemMPs was also determined to elucidate its effects on the sedimentation of HemMPs.

7


Instead, sedimentation

The

For these


177

Page 8 of 29

Detailed procedures of the adsorption experiments are provided in the SI.

178

For the sedimentation experiments with HemNPs, 30 mg/L HemNP was mixed

179

with varying concentrations of GO in polystyrene cuvettes.

180

suspensions were mixed using a shaking incubator (1585, VWR) at 150 rpm for 20 min.

181

Next, the cuvettes were placed in a UV-Vis spectrometer and the absorbance at 416 nm

182

(where the contribution of GO to the total absorbance was insignificant) was recorded

183

over 60 min to determine the sedimentation rates of HemNPs.

184

Transmission Electron Microscopy.

The GO–HemNP

The conformation of GO–HemNP

185

heteroaggregates formed at GO/HemNP mass concentration ratios of 0.004, 0.060, and

186

0.400 was observed through TEM imaging (Philips EM 420).

187

suspensions were prepared in the same solution chemistry as that used for

188

heteroaggregation experiments (i.e., 0.1 mM NaCl and pH 5.2).

189

heteroaggregation, 5–10 µL of heteroaggregate suspensions was deposited on carbon-

190

coated copper TEM grids and the heteroaggregates were subsequently stained with 2%

191

uranyl acetate solutions before imaging. Details on sample preparation can be found in

192

the SI.

The heteroaggregate

After 15 min of

193 194

Results and Discussion

195

Homoaggregation of GO and HemNPs. The hydrodynamic diameters of the

196

GO and HemNP stock suspensions were measured to be 125–149 nm and 87–95 nm,

197

respectively. At 0.1 mM NaCl and pH 5.2, EPM measurements showed that GO was

198

negatively charged and HemNPs were positively charged (Figure 1a). The ionization of

199

the carboxylic groups on GO is primarily responsible for the negative surface charge of

200

GO at this solution chemistry.1, 37 HemNPs were positively charged at pH 5.2 due to the

8


Page 9 of 29

201


dominance of –OH2+ surface functional groups (isoelectric point = 9.3).27

202 203

FIGURE 1 Homoaggregation kinetics of GO and HemNPs were systematically examined

204

over a wide range of NaCl concentrations.

205

HemNPs, which were calculated from their respective aggregation profiles (Figure S2),

206

increased with increasing NaCl concentrations before reaching a maximum value at the

207

CCC. The attachment efficiencies of GO and HemNPs at different NaCl concentrations

208

were shown in Figure 1b. The distinct reaction-limited (α < 1) and diffusion-limited (α =

209

1) regimes indicate that the homoaggregation of GO and HemNPs is qualitatively in

210

agreement with DLVO theory.30, 31 The CCCs of GO and HemNP were 221.6 and 45.8

211

mM NaCl, respectively, indicating that both nanoparticles are stable to homoaggregation

212

at 0.1 mM NaCl, the salt concentration at which the GO–HemNP heteroaggregation

213

experiments were conducted.

214

The homoaggregation rates of GO and

Heteroaggregation Rates of GO with HemNPs.

At 0.1 mM NaCl, the

215

homoaggregation of GO and HemNPs was inhibited by strong electrostatic repulsion

216

(Figure S3).

217

concentration ratio of 0.06, the hydrodynamic diameter measured from this mixture

218

significantly increased, indicating that heteroaggregation between the oppositely charged

219

GO and HemNPs occurred favorably (Figure S3). Similar observations were made in our

220

earlier studies on the heteroaggregation between HemNPs and multiwalled carbon

221

nanotubes (MWCNTs),32,

222

nanoparticles.39

However, when GO were mixed with HemNPs at GO/HemNP mass

38

and also between HemNPs and citrate-coated silver

223

Figure 2a shows representative GO–HemNP heteroaggregation profiles obtained

224

at GO/HemNP mass concentration ratios of 0.000–0.400. The heteroaggregation rates

9



225

obtained over a wide range of GO/HemNP mass concentration ratios are shown in Figure

226

2b.

227

heteroaggregation was very slow and the heteroaggregation rate was even lower than the

228

maximum HemNP homoaggregation rate under diffusion-limited conditions (ca. 0.07

229

nm/s, indicated by dashed line in Figure 2b). As the GO/HemNP ratio was gradually

230

increased from 0.006 to 0.090, the heteroaggregation rate increased correspondingly.

231

The maximum heteroaggregation rate (0.40 nm/s) was obtained at the optimal

232

GO/HemNP ratio of 0.090 and this rate was ca. 5 times higher than the maximum

233

homoaggregation rate of HemNPs. As the GO/HemNP ratios were further increased

234

above 0.090, the heteroaggregation rate continually decreased until it approached zero at

235

GO/HemNP ratios of 0.400 and higher.

236 237

It is apparent that at extremely low GO/HemNP ratios (0.002–0.004),

FIGURE 2 TEM Imaging of Heteroaggregates.

TEM was employed to observe the

238

conformation of GO–HemNP heteroaggregates at low, near optimal, and high

239

GO/HemNP mass concentration ratios. In the following discussion, small, intermediate,

240

and big heteroaggregates are defined as heteroaggregates containing 1–3, 4–8, and > 8

241

HemNPs, respectively. Figure 3 shows representative TEM images of different sized

242

GO–HemNP heteroaggregates, all formed at a GO/HemNP mass concentration of 0.06

243

after ca. 15 min of heteroaggregation. It was generally observed that one GO can bind

244

with either one or several HemNPs to form a primary heteroaggregate.40 These primary

245

heteroaggregates were then bridged together by GO to form bigger clusters. It is worth

246

nothing that some HemNPs were observed to be in contact with one another in Figure 3.

247

It is possible that GO, with the flexible sheet and large surface area, can bridge several

248

HemNPs close together despite the electrostatic repulsion between them.

10


Page 10 of 29

Page 11 of 29


249

FIGURE 3

250

The distributions of heteroaggregates comprising between 1 and > 8 HemNPs per

251

heteroaggregate were determined at GO/HemNP mass concentration ratios of 0.004,

252

0.060, and 0.400 (Figure 4).

253

concentration ratios were 82, 130, and 142, respectively. At the low GO/HemNP ratio of

254

0.004, more than 80 % of heteroaggregates were small and most HemNPs did not

255

undergo heteroaggregation with GO. This observation was in agreement with our DLS

256

results, which showed that HemNPs remained relatively stable in the presence of low GO

257

concentrations (around several µg TOC/L).

258

percentage of small heteroaggregates decreased (< 35 %) while the percentages of

259

intermediate and big heteroaggregates increased. This change in the heteroaggregate size

260

distribution indicated that GO underwent significant and rapid heteroaggregation with

261

HemNPs, which was also consistent with the DLS results (Figure 2b).

The numbers of heteroaggregates examined at these

At the GO/HemNP ratio of 0.060, the

262

FIGURE 4

263

At the high GO/HemNP ratio of 0.400 (Figure 4), small heteroaggregates again

264

became dominant while intermediate and big heteroaggregates were rarely observed (
8 4-8 1-3

80 60 40 20 0

0.004

0.06

0.4

GO/HemNP Ratio

Figure 4. Size distributions of heteroaggregates at different GO/HemNP mass concentration ratios. More than 80 heteroaggregates were observed for each concentration ratio.

26


Page 27 of 29


1.0

GO/HemMP Ratio 0 0.0017 0.027 0.053

C/C0

0.8 0.6 0.4 0.2 0.0

(a) 0

10

20

30

40

50

60

Time (min) 1.0

C/C0

0.8 0.6 GO/HemNP Ratio

0.4

0 0.001 0.01 0.3

0.2 0.0

(b) 0

10

20

30

40

50

60

Time (min) Figure 5. Sedimentation of (a) HemMPs and (b) HemNPs in the absence and presence of GO at 0.1 mM NaCl and pH 5.2. Absorbance of the HemMP and HemNP suspensions was measured at 760 and 416 nm, respectively.

27



1.4

(a)

Incubation Time (hour) 0.25 48

1.2

Page 28 of 29

q (mg/g)

1.0 0.8 0.6 0.4 0.2

(b)

4 2

-8

2

EPM (10 m /Vs)

0.0

0 -2 -4

0

1

2

3

4

5

GO Concentration (mg/L) Figure 6. (a) Adsorption of GO on HemMPs at 0.1 mM NaCl and pH 5.2 at incubation times of 15 min and 48 hours. (b) Electrophoretic mobility (EPM) measurements of GO mixed with HemMPs at 0.1 mM NaCl and pH 5.2. The dashed line represents the EPM of GO. Error bars represent the standard deviations from triplicate experiments.

28


Page 29 of 29


TOC art

29


Heteroaggregation of Graphene Oxide with Nanometer- and

Recommend Documents