Accepted Manuscript

Effects of inorganic ions and natural organic matter on the aggregation of

nanoplastics

Li Cai, Lingling Hu, Huahong Shi, Junwei Ye, Yunfei Zhang, Hyunjung Kim

PII: S0045-6535(18)30060-2
DOI: 10.1016/j.chemosphere.2018.01.052
Reference: CHEM 20633

To appear in: ECSN

Received Date: 7 November 2017

Revised Date: 10 January 2018
Accepted Date: 11 January 2018

Please cite this article as: Cai, L., Hu, L., Shi, H., Ye, J., Zhang, Y., Kim, H., Effects of inorganic ions
and natural organic matter on the aggregation of nanoplastics, Chemosphere (2018), doi: 10.1016/
j.chemosphere.2018.01.052.

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
Graphical Abstract

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Revised manuscript CHEM50247 submitted to Chemosphere

PT
Effects of Inorganic Ions and Natural Organic Matter on the

RI
Aggregation of Nanoplastics

U SC
Li Cai a,∗, Lingling Hu b, Huahong Shi b,Junwei Ye a, Yunfei Zhang a, Hyunjung
AN
Kim c

a
Natural History Research Center, Shanghai Natural History Museum,
M

Shanghai Science and Technology Museum, Shanghai 200127, P. R. China

D

b
State Key Laboratory of Estuarine and Costal Research, East China Normal
TE

University, Shanghai 200062, P. R. China

c
Department of Mineral Resources and Energy Engineering, Chonbuk National
EP

University, 567 Baekje-daero, Deokjin-gu, Jeonju, Jeonbuk 54896, Republic of

C

Korea
AC

*
Corresponding author: Tel: +86 2168622000; E-mail address: [email protected].
1
 ACCEPTED MANUSCRIPT

1 Abstract

2 The aggregation of nanoplastics (NPs) is a key issue in understanding the

3 dynamic nature of NPs in the environment. The aggregation of NPs under various

4 environmental conditions has not yet been studied. We investigated the influences of

PT
5 inorganic ions and natural organic matter (NOM) on polystyrene (PS) NPs

RI
6 aggregation in solutions. Results showed that PS NPs remained stable in wide ionic

7 strength solutions of NaCl (1 to 100 mM) and CaCl2 (0.1 to 15 mM), and only in low

SC
8 ionic strength FeCl3 solutions (0.01 mM). However, obvious PS NPs aggregation was

U
9 observed in FeCl3 solutions with an increase in ionic strength (0.1 and 1 mM).
AN
10 Moreover, NOM had a negligible effect on PS NPs aggregation in all ionic strengths

11 of NaCl and CaCl2 solutions and in low ionic strength FeCl3 solutions (0.01 mM).
M

12 However, NOM reduced PS NPs aggregation in an intermediate ionic strength FeCl3

D

13 (0.1 mM) solution and increased aggregation in a high ionic strength FeCl3 (1 mM)
TE

14 solution. Based on the theoretical analysis of interaction forces among PS NPs, the

15 Derjaguin–Landau–Verwey–Overbeek force was a contributor governing PS NPs

EP

16 aggregation either in the absence or presence of NOM. In addition, other factors,

C

17 including electrostatic heterogeneity of PS NPs surfaces, steric repulsion induced by

AC

18 NOM, and clusters formed via bridging effect in the presence of NOM also

19 contributed to altered PS NPs aggregation under selected conditions. The PS

20 NPs-NOM clusters were directly observed using a cryogenic scanning electron

21 microscope.

22 Keywords: nanoplastic, aggregation, inorganic ions, NOM, mechanism

2
 ACCEPTED MANUSCRIPT

23 1. Introduction

24 Microplastics (MPs) are synthetic polymer particles with diameters of 1 µm to 5

25 mm, and are derived from a wide range of sources (Cheung and Fok, 2016; Jemec et

26 al., 2016; Napper and Thompson, 2016). The extensive use of plastics inevitably

PT
27 releases MPs into natural environments (Eerkes-Medrano et al., 2015; Li et al., 2015;

RI
28 Yang et al., 2015; Su et al., 2016; Ziajahromi et al., 2017). The size distribution of

29 MPs in past ocean surface field surveys highlights the possibility that millimeter-scale

SC
30 debris might fragment to form nanoplastics (NPs) (Cózar et al., 2014; Mattsson et al.,

U
31 2015). Furthermore, a solar reactor was used to illustrate that NPs can be formed by
AN
32 the fragmentation of weathered polyethylene and polypropylene MPs collected from

33 marine waters (Gigault et al., 2016). Moreover, very recently, studies showed that NPs
M

34 can be found in both personal care product (Hernandez et al., 2017) and North
D

35 Atlantic subtropical gyre (Ter Halle et al., 2017).

TE

36 Limited information had been reported on the fate and transport of NPs in

37 aquatic systems until recently. For example, Wegner et al. (2012) found that 30 nm
EP

38 nanopolystyrene can rapidly form millimeter-sized aggregates in seawater. More

C

39 recently, it was reported that NPs have a tendency to aggregate with other substances,
AC

40 such as natural colloids and suspended solids, resulting in the formation of NP

41 aggregates in eco-corona, faeces, and marine snows (Long et al., 2015; Galloway et

42 al., 2017). Complex environmental factors, such as pH, inorganic ions, natural organic

43 matter (NOM), suspended clay particles, and bacteria, play critical roles in NPs

44 aggregation. However, to date, the effect of environmental factors on NPs aggregation

3
 ACCEPTED MANUSCRIPT

45 has not yet been identified. Given their particulate nature, the fate and transport of

46 NPs in surrounding environments is largely based on theories developed in

47 colloid/particle science (Hüffer et al., 2017).

48 Inorganic ions and NOM are two important factors affecting the stability of

PT
49 engineered nanoparticles. Because of the similarities between engineered

RI
50 nanoparticles and NPs (Hüffer et al., 2017), we postulate that inorganic ions and

51 NOM can also influence the aggregation behavior of NPs, and thus need further

SC
52 investigation (Petosa et al., 2010; Alimi et al., 2017). The current consensus drawn

U
53 from the literature is that higher concentrations or higher valences of ions (calcium
AN
54 ion rather than sodium ion) can enhance the aggregation of engineered nanoparticles.

55 Furthermore, attributed to different mechanisms, dispersion or aggregation of

M

56 engineered nanoparticles in the presence of NOM under various solution conditions is

D

57 likely to be present. In fact, it is known that besides sodium and calcium ions, Fe (III)
TE

58 is also ubiquitous in natural aquatic environments, especially those containing oxygen

59 (Ilbert and Bonnefoy, 2013). Moreover, elevated dissolved Fe (III) concentration

EP

60 (reach to ~1.2 mM ionic strength) can occur in estuaries (Hawkings et al., 2014; Yang
C

61 et al., 2017), which happen to be the places suffering from serious MPs pollution
AC

62 (Sadri and Thompson, 2014; Zhao et al., 2014a). However, limited work has been

63 conducted on the aggregation of NPs under such environmental relevant conditions.

64 Hence, this study is designed to systematically investigate the effects of different

65 inorganic ions and NOM on NPs aggregation in solutions. We used Suwannee River

66 humic acid (SRHA) and Suwannee River fulvic acid (SRFA) as representative

4
 ACCEPTED MANUSCRIPT

67 dissolved humic substances to study the aggregation of polystyrene (PS) in solutions

68 with different inorganic ions. It was worth pointing out that PS is one of the most

69 commonly detected plastics in the environment (Alimi et al., 2017) and used as

70 nanopalstics in many previous studies (e.g., Ma et al., 2016; Pitt et al., 2018). Possible

PT
71 mechanisms by which inorganic ions and NOM affected PS NPs aggregation were

RI
72 proposed and discussed.

73

SC
74 2. Material and Methods

U
75 2.1 Chemicals
AN
76 NaCl, CaCl2, and FeCl3 salts were purchased from Sinopharm Chemical Reagent
M

77 Co., Ltd., and used as received without further purification.

78 The yellow-green fluorescent PS nanoparticles with diameter of 0.1 µm

D

79 (Thermo-Fisher Corp., USA) were used as received without surface modification. The
TE

80 PS NPs stock suspension had a particle concentration of 10 g L-1. PS NPs are

EP

81 hydrophilic, with a density of 1.05 g cm-3, and were electrostatically stabilized

82 (according to the manufacturer). We diluted the PS NPs stock solution in electrolyte

C
AC

83 solutions (NaCl, CaCl2 or FeCl3) to achieve the desired target concentrations of PS

84 NPs.

85 SRHA (Cat#2S101H, Standard II, International Humic Substance Society) and

86 SRFA (Cat#2S101F, Standard II, International Humic Substance Society) were used

87 as model humic substances (Franchi and O'Melia, 2003; Kumpulainen et al., 2008).

88 We prepared SRHA and SRFA stock solutions by dissolving 25.0 mg of SRHA and

5
 ACCEPTED MANUSCRIPT

89 SRFA powders in 50 mL Milli-Q water (Q-Gard 1, Millipore Inc., MA) and stirring

90 the solution for 24 h. We then filtered the solutions through 0.22 µm cellulose acetate

91 membranes. A TOC-meter (TOC-VCPN, Shimadzu, Japan) was used to determine the

92 total organic carbon (TOC) content of the SRHA and SRFA stock solutions. The TOC

PT
93 of SRHA stock solution was 421.2 mg L−1 and the TOC of SRFA stock solution was

172.9 mg L−1.

RI
94

2.2 Measurements of zeta potentials of PS NPs

SC
95

96 A Zetasizer Nano ZS90 (Malvern Instruments, UK) was used to determine the

97
U
zeta potentials of PS NPs in the absence or presence of NOM in NaCl, CaCl2, and
AN
98 FeCl3 solutions of different ionic strengths, respectively. Each measurement was
M

99 performed at least 9–12 times at room temperature (25oC).

D

100 2.3 Measurements of aggregation of PS NPs

TE

101 Time-resolved dynamic light scattering (DLS) was used to measure the change

102 of number-weighted averaged hydrodynamic size of PS NPs as a function of time.

EP

103 The size of PS NPs was recorded every 20 s, with the first record at 20s after the
C

104 beginning of the DLS experiments. The experiments were conducted on Zetasizer
AC

105 Nano ZS90 (Malvern Instruments, UK), operating with a He–Ne laser at a wavelength

106 of 633 nm and at a scattering angle of 90°. Each DLS measurement took 10 min.

107 For DLS experiments, the concentration of PS NPs were 10 or 50 mg L−1. For

108 DLS experiments with NOM suspensions, a certain amount of NOM stock solution

109 was added into PS NPs suspensions to prepare mixed solutions. Electrolyte solution

6
 ACCEPTED MANUSCRIPT

110 (NaCl, CaCl2 or FeCl3) were then added into PS NPs suspensions (with or without

111 NOM). The total volume of the final suspension in the DLS cuvette was 1 mL. For the

112 sake of accurate detection, the concentration of PS NPs was higher than the value

113 detected in the environment, which was also used in previous study (Besseling et al.,

PT
114 2017) and might also occur in some circumstances (e.g. landfill leachates and disposal

RI
115 of industrial and domestic sludge etc.) (Nizzetto et al., 2016; Horton et al., 2017). The

116 concentration of NOM used in this study was 10 mg L−1 for both SRHA and SRFA,

SC
117 which was within the range of NOM concentration (from a few hundred µg L−1 to a

U
118 few hundred mg L−1) encountered in natural environments (Thurman, 1985). We
AN
119 started monitoring the change of hydrodynamic sizes of PS NPs immediately after

120 adding electrolyte solution to the cuvette. Due to the highly negative charge of PS
M

121 NPs across a wide range of pH (Fig. S1), DLS measurements for PS NPs in the
D

122 absence and presence of NOM were carried out at unadjusted pH conditions (~5.4) to
TE

123 eliminate the interference induced by pH adjustment in wide range of ionic strength of

124 NaCl (1, 10, 50, and 100 mM), CaCl2 (0.1, 1, 5, and 15 mM), and FeCl3 (0.01, 0.1 and
EP

125 1 mM) solutions (Lin et al., 2016; Yang et al., 2017). The densities of different
C

126 solutions were gravimetrically determined to be from 1.000–1.002 g cm-3.

AC

127 2.4 Measurements of aggregation state and morphology of PS NPs

128 To further investigate the effect of NOM on PS NPs aggregation at selected

129 conditions, we took fluorescence microscope images at ~10 min (similar as the

130 duration of DLS measurements) after mixing PS NPs and NOM (DM4000b, Leica,

131 Germany).

7
 ACCEPTED MANUSCRIPT

132 To visualize the interactions between NOM and PS NPs in FeCl3 solutions, we

133 used a cryogenic scanning electron microscope (Cryo-SEM) to identify the

134 morphology of PS NPs in solutions. For Cryo-SEM imaging, a cold stage for sample

135 freezing and fracturing was located outside of the SEM. Inside the SEM was a cold

PT
136 stage that kept samples frozen during observation and allowed for controlled

RI
137 sublimation. Rapid freezing of the samples maintained their morphology or

138 configuration among various components in suspensions. By keeping the sample

SC
139 frozen on a cold stage in the SEM, it was possible to image the insitu morphology of

U
140 PS NPs and NOM. To the best of our knowledge, Cryo-SEM is optimal for observing
AN
141 the insitu morphology of colloids in solutions as opposed to using air-dried and

142 vacuum freeze-dried specimens for SEM imaging (Gauer et al., 2009). Specifically, 6
M

143 µL of selected samples were dropped onto a Cu sample stub and quickly frozen in
D

144 liquid nitrogen at about −200oC. The specimen was then put inside a SEM
TE

145 (S-3400N-II, Hitachi, Japan) equipped with a Gatan cold stage and cryo-transfer

146 system (ALTO1000, Gatan, USA). The specimen was then coated with 3–6 nm of
EP

147 platinum by sputter deposition and examined at 3–5 kV at ~−160oC.

C

148 2.5 Theoretical prediction on aggregation of PS NPs

AC

149 The Derjaguin–Landau–Verwey–Overbeek (DLVO) theory has been used to

150 study the PS NPs aggregation (Derjaguin, 1941; Verwey et al., 1948). The DLVO

151 theory states that particles are subject to van der Waals interaction forces and

152 electrostatic repulsion forces, and is used to calculate the total interaction energy of

153 particles as a function of separation distance and predict the aggregation behavior of

8
 ACCEPTED MANUSCRIPT

154 particles. The particle–particle interaction forces of PS NPs are determined by treating

155 the particle–particle system as sphere–sphere interaction. Classic DLVO theory has

156 also been used in previous studies to investigate the aggregation of TiO2 nanoparticles

157 and soil colloids with NOM in solutions (Romanello and Fidalgo de Cortalezzi, 2013;

PT
Zhao et al., 2014b). The van der Waals interaction energy ( Φ
vdW
158 ) and electric double

layer interaction energy ( Φ

EDL

RI
159 ) can be calculated according to equations (1) and (2)

160 (Gregory, 1975, 1981), respectively:

SC
A  2a 2 2a 2  d ( 4a + d )  
161 Φ vdW = −  + + ln 
2 
(1)
6  d ( 4a + d ) ( 2a + d ) 2
 ( 2a + d ) 
162

U
32πεε 0 k 2T 2 a zeψ
163 Φ EDL = 2 2
× tanh 2 ( ) × exp(−κd ) (2)
AN
e z 4kT
164 Detailed information on the parameters in the above equations is provided in the

Supplementary Information (Text S1).

M

165
D
TE
C EP
AC

9
 ACCEPTED MANUSCRIPT

166 3. Results

167 3.1 Influences of NaCl and NOM on the aggregation of PS NPs

168 The hydrodynamic sizes of PS NPs (10 mg L−1) in NaCl solutions at 1, 10, 50,

and 100 mM ionic strengths were monitored to study the effect of NaCl on the

PT
169

170 aggregation behavior of PS NPs in solutions. The change of hydrodynamic sizes of PS

RI
171 NPs (10 mg L−1) in Milli-Q water were measured as control (Fig. S2). It indicated that

SC
172 the PS NPs can disperse well in Milli-Q water with size of ~100 nm. Similarly, the

173 hydrodynamic sizes of PS NPs remained stable, with a size of ~100 nm at ionic

174
U
strengths from 1 to 50 mM (Fig. 1a–c). With an increase of ionic strength to 100 mM
AN
175 NaCl, we observed a slight aggregation of PS NPs (reaching ~120 nm) (Fig. 1d).
M

176 The effects of ubiquitous NOM in aquatic ecosystems on the aggregation

177 behavior of PS NPs in NaCl solutions (1–100 mM) were also characterized by the
D

178 change of hydrodynamic sizes of PS NPs in the presence of NOM (either SRHA or
TE

179 SRFA, 10 mg L−1). Interestingly, the hydrodynamic sizes of PS NPs in the presence of
EP

180 NOM in suspensions were equivalent to those in the absence of NOM in 1 to 50 mM

181 NaCl solutions (Fig. 1a–c). The observations clearly show that the presence of NOM,
C
AC

182 both SRHA and SRFA, does not affect PS NPs aggregation in NaCl solutions from 1

183 to 50 mM. However, with NOM in suspension at 100 mM NaCl, the hydrodynamic

184 sizes of PS NPs decreased (Fig. 1d).

185 Fig. 1

186 To study the effect of concentration on the aggregation behavior of PS NPs, we

10
 ACCEPTED MANUSCRIPT

187 also investigated the change of hydrodynamic sizes of 50 mg L−1 PS NPs both in the

188 absence and presence of NOM. The change of hydrodynamic sizes of PS NPs (50 mg

189 L−1) in Milli-Q water were also measured as control (Fig. S2). Similar as that of 10

190 mg L−1 PS NPs, the PS NPs can also disperse well in Milli-Q water with size of ~100

PT
191 nm. The DLS results for PS NPs in CaCl2 solutions are presented in Fig. 1e–h. In

RI
192 general, the hydrodynamic sizes kept stable with the original size of ~100 nm at ionic

193 strengths ranging from 1 to 100 mM, and neither SRHA nor SRFA affected PS NPs

SC
194 sizes across all ionic strengths (Fig. 1e–h). It was worth pointing out that unlike 10

U
195 mg L−1 PS NPs, no obvious aggregation was observed in 100 mM NaCl for 50 mg L−1
AN
196 PS NPs although the difference is minimal (Fig. 1d vs. Fig. 1h).

3.2 Influences of CaCl2 and NOM on the aggregation of PS NPs

M

197

198 To examine whether the effect of divalent CaCl2 on the aggregation behavior of
D

199 PS NPs would be similar to that of NaCl solutions, we conducted DLS measurements
TE

200 of PS NPs (both 10 and 50 mg L−1) in 0.1, 1, 5, and 15 mM CaCl2 solutions. Results
EP

201 showed that the hydrodynamic sizes of PS NPs did not significantly change in CaCl2

202 solutions, with ionic strengths ranging from 0.1 to 15 mM. Only a very slight
C

aggregation of PS NPs in 15 mM CaCl2 solutions was observed for 10 mg L−1 PS NPs

AC

203

204 (reaching ~110 nm). This indicated that CaCl2 also had negligible effect on PS NPs

205 aggregation in both 10 and 50 mg L−1 concentrations (Fig. 2).

206 Fig. 2

207 Furthermore, close inspection of the hydrodynamic sizes of PS NPs in the

11
 ACCEPTED MANUSCRIPT

208 presence of NOM (SRHA and SRFA, 10 mg L−1) in suspensions clearly showed that

209 SRHA and SRFA did not affect the sizes of PS NPs in CaCl2 solutions with different

210 ionic strengths (Fig. 2). This held true for both 10 and 50 mg L−1 PS NPs. In general,

211 the aggregation of PS NPs in CaCl2 solutions was similar to that in NaCl solutions. As

PT
212 stated above, inorganic salts (NaCl and CaCl2) and NOM (SRHA and SRFA) had

RI
213 negligible effects on PS NPs aggregation in solutions at environmentally relevant

214 solution conditions.

SC
215 3.3 Influences of FeCl3 and NOM on the aggregation of PS NPs

216
U
Because of the negligible difference of aggregation behavior between 10 and 50
AN
217 mg L−1 PS NPs in NaCl and CaCl2 solutions, we investigated the change of
M

218 hydrodynamic sizes of PS NPs for 50 mg L−1 PS NPs in FeCl3 solutions at ionic

219 strengths of 0.01, 0.1, and 1 mM. We performed DLS measurements in the absence
D

220 and presence of NOM (SRHA and SRFA, 10 mg L−1). The speciation of Fe (III) with
TE

221 and without NOM in suspension was calculated using MINTEQ software (Kinniburgh
EP

222 et al., 1996; Lu et al., 2017). The calculations demonstrated that the dominant

223 speciation of Fe (III) in FeCl3 solutions was dissolved Fe(OH)2+ and Fe3(OH)45+ in the
C

absence of NOM and Fe(OH)2+, phenolic-, and carboxylic- Fe (III) in the presence of
AC

224

225 NOM, without precipitation of Fe (III). Results showed that only in very low ionic

226 strength FeCl3 solutions (0.01 mM), minimal change of hydrodynamic sizes was

227 obtained (Fig. 3a). However, the sizes of PS NPs increased significantly over time in

228 FeCl3 solutions (reaching ~350 nm) with ionic strengths of 0.1 and 1 mM (Fig. 3b and

229 c). The ionic strengths of FeCl3 were much lower than those of the NaCl and CaCl2

12
 ACCEPTED MANUSCRIPT

230 solutions investigated.

231 Furthermore, in the 0.01 mM FeCl3 solution, the presence of NOM did not alter

232 the sizes of PS NPs (Fig. 3a). However, the presence of NOM in suspension in FeCl3

233 solutions altered the hydrodynamic sizes of PS NPs at relatively high ionic strengths

PT
234 (0.1 and 1 mM) regardless of the type of NOM (Fig. 3b and c). Specifically, it

RI
235 decreased the sizes of PS NPs in 0.1 mM FeCl3 solution (Fig. 3b). On the contrary, the

236 presence of NOM significantly increased the sizes of PS NPs in 1 mM FeCl3 solution

SC
237 (Fig. 3c). The size of PS NPs reached ~1200 nm after 10 min in 1 mM FeCl3 solution

U
238 in the presence of either SRHA or SRFA.
AN
239 Fig. 3

To gain more insight, fluorescence microscope images (Fig. 4A–D) and

M

240

241 Cryo-SEM images (Fig. 4a–d) were used to characterize the aggregation state of PS
D

242 NPs in the absence and presence of NOM in 0.1 or 1 mM FeCl3 solutions. In the
TE

243 presence of NOM, aggregate sizes decreased with either SRHA or SRFA in 0.1 mM
EP

244 FeCl3 solution, resulting in more homogeneous dispersion of PS NPs (Fig. 4A and C).

245 Instead, obvious clusters of PS NPs were observed in the presence of either SRHA or
C

246 SRFA in 1 mM FeCl3 solution (Fig. 4B and D). Accordingly, no obvious aggregates of
AC

247 PS NPs in 0.1 mM FeCl3 solution (Fig. 4a and c) and large clusters of PS NPs

248 aggregates with NOM in 1 mM FeCl3 solution were observed (Fig. 4b and d).

249 Fluorescence microscope and Cryo-SEM images showed good agreement with DLS

250 measurement findings for PS NPs in the presence of NOM in 0.1 and 1 mM FeCl3

251 solutions.

13
 ACCEPTED MANUSCRIPT
Fig. 4
252

253 4. Discussion

254 4.1 Influence of inorganic ions on the aggregation of PS NPs

Previous studies have shown that surface charge alterations induced by inorganic

PT
255

256 ions are the main factor controlling nanoparticle aggregation behavior (French et al.,

RI
257 2009; Zhang et al., 2013). Our results showed that the zeta potentials of PS NPs were

SC
258 all negative in both NaCl and CaCl2 solutions (Table S1), and the zeta potential

259 became less negative as ionic strength increased. The DLVO interaction energy

260
U
profiles among PS NPs under all examined ionic strengths of NaCl and CaCl2
AN
261 solutions are shown in Figs. S3 and S4. It was clear that repulsive forces were present
M

262 in all but the highest ionic strengths of NaCl and CaCl2 solutions (100 mM NaCl and

263 15 mM CaCl2) for 10 mg L−1 PS NPs. These predictions are qualitatively consistent
D

264 with the observed stability behavior of PS NPs (Figs. 1 and 2); relatively greater
TE

265 stability of PS NPs was observed when energy barriers were present. Consistently, our
EP

266 previous study also reported that PS NPs can disperse well in solutions due to the high

267 negative surface charge in 1 and 10 mM NaCl solutions (Cai et al., 2016). However,
C
AC

268 in contrast to the good stability of PS NPs in CaCl2 solutions even at ionic strength as

269 high as 15 mM in the present study, engineered nanoparticles (e.g., CNT, TiO2 and Ag)

270 have been reported to undergo obvious aggregation in CaCl2 or MgCl2 solutions with

271 low ionic strength, even at ~2 mM, might due to the abundant functional groups

272 (Saleh et al., 2010; Huynh and Chen, 2011) and the surface chemical heterogeneity

273 (Liu et al., 2011; Bouchard et al., 2012) of engineered nanoparticles.

14
 ACCEPTED MANUSCRIPT

274 In the presence of Fe (III), the zeta potentials of PS NPs were still highly

275 negative in 0.01 mM FeCl3, whereas they became positive in either 0.1 or 1 mM

276 FeCl3 solutions (Table S1). Due to the high negative charge of PS NPs in 0.01 mM

277 FeCl3, a repulsive force was therefore expected among PS NPs (Fig. S5a), which

PT
278 would then result in the dispersion of PS NPs. With the positive charge of PS NPs in

RI
279 0.1 mM and 1 mM FeCl3 solutions, the DLVO interaction energy calculation predicts

280 a repulsive force among PS NPs (Fig. S5b and c). However, contrary to the interaction

SC
281 energy prediction, obvious aggregation of PS NPs was observed in both conditions

U
282 (Fig. 3b and c). A plausible explanation on this discrepancy is as follows: the charge
AN
283 reversal of PS NPs observed from of the zeta potential results in 0.1 mM and 1 mM

284 FeCl3 solutions clearly indicates that the specific adsorption of ferric ions occurred
M

285 onto the surface of PS NPs (Table S1), which would render the NPs’ surface
D

286 electrostatically heterogeneous. Note that the DLVO interaction energy prediction
TE

287 only uses the mean value of surface charge, which neglects the effect of the

288 heterogeneity on the interacting surfaces. Furthermore, it is expected that the extent of
EP

289 the heterogeneity effect on the interaction varies with the degree of the heterogeneity
C

290 existing on the interacting surfaces. Several studies, in fact, reported the deviation of
AC

291 the DLVO interaction energy prediction from experimental observation due to the

292 surface charge heterogeneity (Sennato et al., 2009; Valmacco et al., 2016). Therefore,

293 PS NPs could easily approach each other in 0.1 and 1 mM FeCl3. This observation

294 also agrees well with findings on the aggregation of latex induced by FeCl3 coagulant

295 due to the charge neutralization mechanism (Oriekhova and Stoll, 2014).

15
 ACCEPTED MANUSCRIPT

296 4.2 Influence of NOM on the aggregation of PS NPs

297 Results showed that the influence of NOM on PS NPs aggregation differed in the

298 presence of different inorganic ions in suspensions. Interactions of NOM and PS NPs

299 were characterized to interpret the mechanisms affecting the aggregation behavior of

PT
300 PS NPs in the presence of NOM. In general, the zeta potentials of PS NPs in the

RI
301 presence of NOM became more negative than those in the absence of NOM in both

NaCl and CaCl2 solutions except for high ionic strengths of CaCl2 (5 and 15 mM) for

SC
302

303 50 mg L−1 PS NPs (Table S1). As a result, the energy barrier heights of PS NPs

U
304 increased in the presence of NOM compared to those in the absence of NOM in both
AN
305 NaCl and CaCl2 solutions (Figs. S3 and S4), except for high ionic strengths of CaCl2

(5 and 15 mM) for 50 mg L−1 PS NPs. For 50 mg L−1 PS NPs in 5 and 15 mM CaCl2
M

306

307 solutions, the change of DLVO interaction energy is minimal. Collectively, with the
D

308 presence of NOM in all ionic strengths of NaCl and CaCl2 solutions, the electrostatic
TE

309 repulsive forces were present among PS NPs. Therefore, electrostatic repulsive
EP

310 force-induced dispersion of PS NPs was obtained with NOM in suspensions in both

311 NaCl and CaCl2 solutions.

C

312 For 0.01 mM FeCl3 solution, the function of NOM was similar to that in both
AC

313 NaCl and CaCl2 solutions. Negligible change was observed for DLVO interaction

314 energy among PS NPs with NOM in suspensions (Fig. S5a), thus, the size of PS NPs

315 did not change. However, charge reversal of PS NPs from positive (~+8 mV) to

316 negative (~-34 mV) in the presence of NOM appeared in 0.1 mM FeCl3 solution

317 (Table S1). Because of the relatively high negative charge of PS NPs in the presence

16
 ACCEPTED MANUSCRIPT

318 of NOM, the DLVO interaction energy barriers among PS NPs increased in the

319 presence of NOM in 0.1 mM FeCl3 solution (Fig. S5b). Additionally, the plentiful

320 NOM surrounding nearby PS NPs, as observed in Cryo-SEM images, is also proposed

321 to induce the steric repulsion in a 0.1 mM FeCl3 solution (Fig. 4a and c), repelling PS

PT
322 NPs when they approach each other. The polymer-induced steric repulsion is also

RI
323 reported in many previous studies (Kim et al., 2010; Tang and Cheng, 2018).

324 Collectively, with NOM in 0.1 mM FeCl3 solutions, the hydrodynamic sizes of PS

SC
325 NPs decreased significantly.

U
326 For PS NPs in 1 mM FeCl3 solution, the positive charge of PS NPs was markedly
AN
327 decreased in the presence of NOM (Table S1). Thus, the DLVO interaction energy

barriers among PS NPs decreased significantly or even was eliminated with SRFA
M

328

329 (Fig. S5c), which would result in increased aggregation of PS NPs in theory. However,
D

330 close inspection of DLVO interaction energy profiles revealed that even though the
TE

331 energy barriers among PS NPs decreased in the presence of SRHA, significant
EP

332 magnitude of the energy barrier (>60 kT) was still present among PS NPs.

333 Nevertheless, increased aggregation of PS NPs was obtained in the presence of SRHA
C

334 in 1 mM FeCl3 solution (sizes of PS NPs increased to ~1200 nm), which is not
AC

335 explained by the DLVO theory. Therefore, this clearly suggests that other mechanisms

336 other than the DLVO interaction are likely involved in the abnormal stability behavior

337 of PS NPs with SRHA in suspensions. Cryo-SEM images clearly showed that the

338 attachment of PS NPs onto NOM (both SRHA and SRFA) in the formation of PS

339 NPs-NOM clusters induced by the bridging effect was another factor leading to

17
 ACCEPTED MANUSCRIPT

340 increased PS NPs aggregation in 1 mM FeCl3 solution (Fig. 4b and d). In terms of the

341 type of NOM in NPs aggregation, the enhanced aggregation of PS NPs was more

342 pronounced in the presence of SRFA than in SRHA in high ionic strength FeCl3

343 solution. This demonstrates that the type of NOM affects the aggregation of NPs

PT
344 under specific solution conditions. Consistently, a previous study also identified that

RI
345 SRHA is more effective than SRFA in suppressing fullerene aggregation (enhanced

346 dispersion) in MgSO4 solutions due to the structural and conformational

SC
347 characteristics of the adsorbed NOM layers (Zhang et al., 2013).

U
348 Fig. 5
AN
349 In conclusion, both increased DLVO interaction energy and steric repulsion
M

350 induced by NOM contributed to decreased PS NPs aggregation in the presence of

351 NOM in 0.1 mM FeCl3 solution. Meanwhile, both decreased DLVO interaction
D

352 energy and the bridging effect between NOM and PS NPs led to enhanced PS NPs
TE

353 aggregation in the presence of NOM in 1 mM FeCl3 solution. The process that
EP

354 occurred for PS NPs in the presence of NOM in both 0.1 and 1 mM FeCl3 solutions is

355 schematically illustrated in Fig. 5.

C
AC

356 4.3 Aggregation of NPs influenced by complex environmental factors

357 Although more studies are required to clarify the NPs aggregation under

358 complex environmental factors, specific solution conditions, such as ionic valence,

359 ionic strength, NOM type and concentration, and co-present chemicals, are proposed

360 to affect the stabilization or destabilization of NPs. In the present study, we found that

18
 ACCEPTED MANUSCRIPT

361 severe aggregation of NPs with Fe (III) precedes Na (I) and Ca (II). Moreover, our

362 findings on the effects of NOM in solutions are far different from the simple

363 prediction that increased stabilization of NPs in previous review articles (Galloway et

364 al., 2017; Hüffer et al., 2017). Furthermore, our findings on the effects of NOM on

PT
365 NPs are also different from the increased dispersion of nanoparticles with humic acid

RI
366 in suspensions in our previous studies (Han et al., 2014; Han et al., 2017). We

367 demonstrated that the aggregation behavior of PS NPs is influenced by complex

SC
368 environmental factors. Results of this study highlight the importance of relative ratios

U
369 of the concentration of NOM to Fe (III). In general, a high relative NOM to Fe (III)
AN
370 ratio resulted in stabilization (decreased aggregation) of PS NPs, whereas a low

371 relative ratio led to the enhanced aggregation of PS NPs.

M

372 When NPs remain stable (with nanosizes), as in NaCl and CaCl2 solutions, even
D

373 with NOM in suspensions, their potential consumption by plankton and subsequent
TE

374 uptake across cell membranes will take place. Well stabilized NPs have shown more

375 adverse effects on Daphnia magna than MPs (Ma et al., 2016). The severe
EP

376 aggregation of PS NPs occurring in the presence of NOM at high ionic strength FeCl3
C

377 solutions shows the reasonability and possibility of the formation of MPs in
AC

378 eco-corona and marine snow in natural environments (Long et al., 2015; Galloway et

379 al., 2017). Moreover, the complexities associated with NPs aggregates, such as NOM

380 and other chemicals, can sink from surface water to deep water and benthos by being

381 repacked into NPs aggregates (Cole et al., 2016). This will therefore play an important

382 role in nutrient and contaminant cycling in aquatic ecosystems.

19
 ACCEPTED MANUSCRIPT

383 5. Conclusions

384 In brief, compared to the negligible effects of NaCl and CaCl2 over a wide range

385 of ionic strength on PS NPs aggregation, Fe (III) with far lower ionic strengths in

386 waters could induce a rapid and obvious aggregation of PS NPs. Moreover, NOM had

PT
387 negligible effects on the aggregation of PS NPs in both NaCl and CaCl2 solutions;

RI
388 however, NOM could decrease or increase the aggregation of PS NPs in FeCl3

389 solutions at different ionic strengths. This indicates that the aggregation of PS NPs

SC
390 with Fe (III) precedes that with Na (I) and Ca (II), and the effect of ubiquitous NOM

U
391 on PS NPs aggregation behavior is subject to specific solution conditions. Although
AN
392 the experiments were performed for one type and shape of NPs, the aforementioned

393 findings clearly showed that the effects of solution chemistry on NPs aggregation are
M

394 complex. Therefore, we need to set our sights on how best to provide the essential
D

395 information on the aggregation of NPs in a complex environment to further explain

TE

396 the consumption, ingestion, and excretion of NPs by organisms. Some examples could

397 be NPs type, NPs shape, NPs concentration and the naturally occurring coatings on
EP

398 NPs surfaces.

C

399
AC

400 Acknowledgment

401 This work was supported by the National Nature Science Foundation of China

402 (41601518, 41571467, 31600301). We thank Professor Xiangke Wang from North

403 China Electric Power University for reviewing of the manuscript. We acknowledge

404 the editor and the reviewers for their very helpful comments.

20
 ACCEPTED MANUSCRIPT

405 Appendix A. Supplementary data

406 Zeta potentials and average hydrodynamic sizes of PS NPs, detailed information

407 on the parameters for the DLVO interaction energy calculation, zeta potentials of PS

408 NPs as a function of pH, the hydrodynamic sizes of PS NPs in Milli-Q water as

PT
409 control and the results of DLVO calculations are provided in the Supplementary

RI
410 Information.
411

SC
412 References
413 Alimi, O., Farner Budarz, J., Hernandez, L.M., Tufenkji, N., 2017. Microplastics and nanoplastics

U
414 in aquatic environments: Aggregation, deposition, and enhanced contaminant transport.
415 Environ. Sci. Technol. DOI: 10.1021/acs.est.7b05559
AN
416 Besseling, E., Quik, J.T.K., Sun, M., Koelmans, A.A., 2017. Fate of nano- and microplastic in
417 freshwater systems: A modeling study. Environ. Pollut. 220, 540-548.
418 Bouchard, D., Zhang, W., Powell, T., Rattanaudompol, U. S., 2012. Aggregation kinetics and
M

419 transport of single-walled carbon nanotubes at low surfactant concentrations. Environ. Sci.
420 Technol. 46, 4458-4465.
D

421 Cózar, A., Echevarría, F., González-Gordillo, J.I., Irigoien, X., Úbeda, B., Hernández-León, S.,
422 Palma, Á.T., Navarro, S., García-de-Lomas, J., Ruiz, A., Fernández-de-Puelles, M.L., Duarte,
TE

423 C.M., 2014. Plastic debris in the open ocean. Proc. Natl. Acad. Sci. 111, 10239-10244.
424 Cai, L., Peng, S., Wu, D., Tong, M., 2016. Effect of different-sized colloids on the transport and
EP

425 deposition of titanium dioxide nanoparticles in quartz sand. Environ. Pollut. 208, 637-644.
426 Cheung, P.K., Fok, L., 2016. Evidence of microbeads from personal care product contaminating
427 the sea. Mar. Pollut. Bull. 109, 582-585.
C

428 Cole, M., Lindeque, P.K., Fileman, E., Clark, J., Lewis, C., Halsband, C., Galloway, T.S., 2016.
AC

429 Microplastics alter the properties and sinking rates of zooplankton faecal pellets. Environ. Sci.
430 Technol. 50, 3239-3246.
431 Derjaguin, B., 1941. Theory of the stability of strongly charged lyophobic sols and the adhesion of
432 strongly charged particles in solutions of electrolytes. Acta Physicochim. USSR 14, 633-662.
433 Eerkes-Medrano, D., Thompson, R.C., Aldridge, D.C., 2015. Microplastics in freshwater systems:
434 A review of the emerging threats, identification of knowledge gaps and prioritisation of
435 research needs. Water Res. 75, 63-82.
436 Franchi, A., O'Melia, C.R., 2003. Effects of natural organic matter and solution chemistry on the
437 deposition and reentrainment of colloids in porous media. Environ. Sci. Technol. 37,
438 1122-1129.

21
 ACCEPTED MANUSCRIPT
439 French, R.A., Jacobson, A.R., Kim, B., Isley, S.L., Penn, R.L., Baveye, P.C., 2009. Influence of
440 ionic strength, pH, and cation valence on aggregation kinetics of titanium dioxide
441 nanoparticles. Environ. Sci. Technol. 43, 1354-1359.
442 Galloway, T.S., Cole, M., Lewis, C., 2017. Interactions of microplastic debris throughout the
443 marine ecosystem. Nat. Ecol. Evol. 1, 0116.
444 Gauer, C., Jia, Z., Wu, H., Morbidelli, M., 2009. Aggregation kinetics of coalescing polymer
445 colloids. Langmuir 25, 9703-9713.

PT
446 Gigault, J., Pedrono, B., Maxit, B., Ter Halle, A., 2016. Marine plastic litter: The unanalyzed
447 nano-fraction. Environ. Sci.: Nano 3, 346-350.

RI
448 Gregory, J., 1975. Interaction of unequal double layers at constant charge. J. Colloid Interf. Sci. 51,
449 44-51.
450 Gregory, J., 1981. Approximate expressions for retarded van der waals interaction. J. Colloid

SC
451 Interf. Sci. 83, 138-145.
452 Hüffer, T., Praetorius, A., Wagner, S., von der Kammer, F., Hofmann, T., 2017. Microplastic

U
453 exposure assessment in aquatic environments: Learning from similarities and differences to
454 engineered nanoparticles. Environ. Sci. Technol. 51, 2499-2507.
AN
455 Han, Y., Hwang, G., Park, S., Flores, A.G., Jo, E., Eom, I., Tong, M., Kim, H.J., Kim, H., 2017.
456 Stability behavior of carboxyl-functionalized carbon black nanoparticles: Role of solution
457 chemistry and humic acid. Environ. Sci.: Nano 4, 800-810.
M

458 Han, Y., Kim, D., Hwang, G., Lee, B., Eom, I., Kim, P.J., Tong, M., Kim, H., 2014. Aggregation
459 and dissolution of ZnO nanoparticles synthesized by different methods: Influence of ionic
D

460 strength and humic acid. Colloids Surf., A 451, 7-15.

461 Hawkings, J.R., Wadham, J.L., Tranter, M., Raiswell, R., Benning, L.G., Statham, P.J., Tedstone,
TE

462 A., Nienow, P., Lee, K., Telling, J., 2014. Ice sheets as a significant source of highly reactive
463 nanoparticulate iron to the oceans. Nat. Commun. 5, 3929.
EP

464 Hernandez, L.M., Yousefi, N., Tufenkji, N., 2017. Are there nanoplastics in your personal care
465 products? Environ. Sci. Technol. Lett. 4, 280-285.
466 Horton, A.A., Walton, A., Spurgeon, D.J., Lahive, E., Svendsen, C., 2017. Microplastics in
C

467 freshwater and terrestrial environments: Evaluating the current understanding to identify the
AC

468 knowledge gaps and future research priorities. Sci. Total Environ. 586, 127-141.
469 Huynh, K.A., Chen, K.L., 2011. Aggregation kinetics of citrate and polyvinylpyrrolidone coated
470 silver nanoparticles in monovalent and divalent electrolyte solutions. Environ. Sci. Technol.
471 45, 5564-5571.
472 Ilbert, M., Bonnefoy, V., 2013. Insight into the evolution of the iron oxidation pathways. BBA -
473 Bioenergetics 1827, 161-175.
474 Jemec, A., Horvat, P., Kunej, U., Bele, M., Kržan, A., 2016. Uptake and effects of microplastic
475 textile fibers on freshwater crustacean Daphnia magna. Environ. Pollut. 219, 201-209.
476 Kim, H.N., Walker, S.L., Bradford, S.A., 2010. Macromolecule mediated transport and retention
477 of Escherichia coli O157:H7 in saturated porous media. Water Res. 44, 1082-1093.

22
 ACCEPTED MANUSCRIPT
478 Kinniburgh, D.G., Milne, C.J., Benedetti, M.F., Pinheiro, J.P., Filius, J., Koopal, L.K., Van
479 Riemsdijk, W.H., 1996. Metal ion binding by humic acid:  Application of the NICA-Donnan
480 model. Environ. Sci. Technol. 30, 1687-1698.
481 Kumpulainen, S., von der Kammer, F., Hofmann, T., 2008. Humic acid adsorption and surface
482 charge effects on schwertmannite and goethite in acid sulphate waters. Water Res. 42,
483 2051-2060.
484 Li, J., Yang, D., Li, L., Jabeen, K., Shi, H., 2015. Microplastics in commercial bivalves from

PT
485 China. Environ. Pollut. 207, 190-195.
486 Lin, D., Drew Story, S., Walker, S.L., Huang, Q., Cai, P., 2016. Influence of extracellular

RI
487 polymeric substances on the aggregation kinetics of TiO2 nanoparticles. Water Res. 104,
488 381-388.
489 Liu, X., Chen, G., Su, C., 2011. Effects of material properties on sedimentation and aggregation of

SC
490 titanium dioxide nanoparticles of anatase and rutile in the aqueous phase. J. Colloid Interf.
491 Sci. 363, 84-91.

U
492 Long, M., Moriceau, B., Gallinari, M., Lambert, C., Huvet, A., Raffray, J., Soudant, P., 2015.
493 Interactions between microplastics and phytoplankton aggregates: Impact on their respective
AN
494 fates. Mar. Chem. 175, 39-46.
495 Lu, J., Shi, Y., Ji, Y., Kong, D., Huang, Q., 2017. Transformation of triclosan by laccase catalyzed
496 oxidation: The influence of humic acid-metal binding process. Environ. Pollut. 220,
M

497 1418-1423.
498 Ma, Y., Huang, A., Cao, S., Sun, F., Wang, L., Guo, H., Ji, R., 2016. Effects of nanoplastics and
D

499 microplastics on toxicity, bioaccumulation, and environmental fate of phenanthrene in fresh

500 water. Environ. Pollut. 219, 166-173.
TE

501 Mattsson, K., Hansson, L.A., Cedervall, T., 2015. Nano-plastics in the aquatic environment.
502 Environ. Sci. Proc. Impacts 17, 1712-1721.
EP

503 Napper, I.E., Thompson, R.C., 2016. Release of synthetic microplastic plastic fibres from
504 domestic washing machines: Effects of fabric type and washing conditions. Mar. Pollut. Bull.
505 112, 39-45.
C

506 Nizzetto, L., Futter, M., Langaas, S., 2016. Are agricultural soils dumps for microplastics of urban
AC

507 origin? Environ. Sci. Technol. 50, 10777-10779.

508 Oriekhova, O., Stoll, S., 2014. Investigation of FeCl3 induced coagulation processes using
509 electrophoretic measurement, nanoparticle tracking analysis and dynamic light scattering:
510 Importance of pH and colloid surface charge. Colloids Surf., A 461, 212-219.
511 Petosa, A.R., Jaisi, D.P., Quevedo, I.R., Elimelech, M., Tufenkji, N., 2010. Aggregation and
512 deposition of engineered nanomaterials in aquatic environments: Role of physicochemical
513 interactions. Environ. Sci. Technol. 44, 6532-6549.
514 Pitt, J.A., Kozal, J.S., Jayasundara, N., Massarsky, A., Trevisan, R., Geitner, N., Wiesner, M.,
515 Levin, E.D., Di Giulio, R.T., 2018. Uptake, tissue distribution, and toxicity of polystyrene
516 nanoparticles in developing zebrafish (Danio rerio). Aquat. Toxicol. 194, 185-194.

23
 ACCEPTED MANUSCRIPT
517 Romanello, M.B., Fidalgo de Cortalezzi, M.M., 2013. An experimental study on the aggregation
518 of TiO2 nanoparticles under environmentally relevant conditions. Water Res. 47, 3887-3898.
519 Sadri, S.S., Thompson, R.C., 2014. On the quantity and composition of floating plastic debris
520 entering and leaving the Tamar Estuary, Southwest England. Mar. Pollut. Bull. 81, 55-60.
521 Saleh, N.B., Pfefferle, L.D., Elimelech, M., 2010. Influence of biomacromolecules and humic acid
522 on the aggregation kinetics of single-walled carbon nanotubes. Environ. Sci. Technol. 44,
523 2412-2418.

PT
524 Sennato, S., Truzzolillo, D., Bordi, F., Sciortino, F., Cametti, C., 2009. Colloidal particle
525 aggregates induced by particle surface charge heterogeneity. Colloids Surf., A 343, 34-42.

RI
526 Su, L., Xue, Y., Li, L., Yang, D., Kolandhasamy, P., Li, D., Shi, H., 2016. Microplastics in Taihu
527 Lake, China. Environ. Pollut. 216, 711-719.
528 Tang, Z., Cheng, T., 2018. Stability and aggregation of nanoscale titanium dioxide particle (nTiO2):

SC
529 Effect of cation valence, humic acid, and clay colloids. Chemosphere 192, 51-58.
530 Ter Halle, A., Jeanneau, L., Martignac, M., Jardé, E., Pedrono, B., Brach, L., Gigault, J., 2017.

U
531 Nanoplastic in the North Atlantic subtropical gyre. Environ. Sci. Technol. 51, 13689-13697.
532 Thurman, E.M., 1985. Organic geochemistry of natural waters. Springer Netherlands.
AN
533 Valmacco, V., Elzbieciak-Wodka, M., Herman, D., Trefalt, G., Maroni, P., Borkovec, M., 2016.
534 Forces between silica particles in the presence of multivalent cations. J. Colloid Interf. Sci.
535 472, 108-115.
M

536 Verwey, E.J.W., Overbeek, J.T.G., Van Nes, K., 1948. Theory of the stability of lyophobic colloids:
537 the interaction of sol particles having an electric double layer. Elsevier: Amsterdam.
D

538 Wegner, A., Besseling, E., Foekema, E.M., Kamermans, P., Koelmans, A.A., 2012. Effects of
539 nanopolystyrene on the feeding behavior of the blue mussel (Mytilus edulis L.). Environ.
TE

540 Toxicol. Chem. 31, 2490-2497.

541 Yang, D., Shi, H., Li, L., Li, J., Jabeen, K., Kolandhasamy, P., 2015. Microplastic pollution in
EP

542 table salts from China. Environ. Sci. Technol. 49, 13622-13627.
543 Yang, R., Su, H., Qu, S., Wang, X., 2017. Capacity of humic substances to complex with iron at
544 different salinities in the Yangtze River estuary and East China Sea. Sci. Rep. 7, 1381.
C

545 Zhang, W., Rattanaudompol, U.s., Li, H., Bouchard, D., 2013. Effects of humic and fulvic acids
AC

546 on aggregation of aqu/nC60 nanoparticles. Water Res. 47, 1793-1802.

547 Zhao, S., Zhu, L., Wang, T., Li, D., 2014a. Suspended microplastics in the surface water of the
548 Yangtze Estuary System, China: first observations on occurrence, distribution. Mar. Pollut.
549 Bull. 86, 562-568.
550 Zhao, W., Walker, S.L., Huang, Q., Cai, P., 2014b. Adhesion of bacterial pathogens to soil
551 colloidal particles: Influences of cell type, natural organic matter, and solution chemistry.
552 Water Res. 53, 35-46.
553 Ziajahromi, S., Neale, P.A., Rintoul, L., Leusch, F.D.L., 2017. Wastewater treatment plants as a
554 pathway for microplastics: Development of a new approach to sample wastewater-based
555 microplastics. Water Res. 112, 93-99.

24
 ACCEPTED MANUSCRIPT

Figure Captions

Fig.1. Change of hydrodynamic size over time for 10 (a–c) and 50 (e–h) mg L−1 PS
NPs in 1, 10, 50, and 100 mM NaCl solutions in the absence of NOM (-NOM), in the
presence of SRHA (+SRHA), and in the presence of SRFA (+ SRFA) in suspensions.
Replicated experiments were performed under all conditions (n≥2). In consideration

PT
of clarity, error bars are not shown.

RI
Fig.2. Change of hydrodynamic size over time for 10 (a–c) and 50 (e–h) mg L−1 PS

SC
NPs in 0.1, 1, 5, and 15 mM CaCl2 solutions in the absence of NOM (- NOM), in the
presence of SRHA (+ SRHA), and in the presence of SRFA (+ SRFA) in suspensions.

U
Replicated experiments were performed under all conditions (n≥2). In consideration
of clarity, error bars are not shown.
AN
Fig. 3. Change of hydrodynamic size over time for 50 mg L−1 PS NPs in 0.01 mM (a),
M

0.1 mM (b), and 1 mM (c) FeCl3 solutions in the absence of NOM (- NOM), in the
D

presence of SRHA (+ SRHA), and in the presence of SRFA (+ SRFA) in suspensions.

Replicated experiments were performed under all conditions (n≥2). In consideration
TE

of clarity, error bars are not shown.

EP

Fig.4. Fluorescence images (A–D) and Cryo-SEM images (a–d) of PS NPs in 0.1 mM
(left) and 1 mM (right) FeCl3 solutions in the presence of SRHA or SRFA in
C

suspensions (images were taken after ~10 min upon mixing PS NPs, NOM, and
AC

FeCl3).

Fig. 5. Schematic diagram illustrating the interactions of PS NPs and NOM in 0.1
mM (a) and 1 mM (b) FeCl3 solutions.
 ACCEPTED MANUSCRIPT

-NOM + SRHA + SRFA

-1
10 mg L PS NPs
200
(a) 1 mM NaCl (b) 10 mM NaCl
150

100
Hydrodynamic Size (nm)

PT
50

0
200

RI
(c) 50 mM NaCl (d) 100 mM NaCl
150

SC
100

50

U
AN
0
-1
50 mg L PS NPs
200
(e) 1 mM NaCl (f) 10 mM NaCl
M

150

100
D
Hydrodynamic Size (nm)

50
TE

0
200
(g) 50 mM NaCl (h) 100 mM NaCl
EP

150

100
C

50
AC

0
0 100 200 300 400 500 600 0 100 200 300 400 500 600
Time (s) Time (s)

Fig.1.
 ACCEPTED MANUSCRIPT

-NOM +-1 SRHA + SRFA

10 mg L PS NPs
200
(a) 0.1 mM CaCl2 (b) 1 mM CaCl2
150

100
Hydrodynamic Size (nm)

PT
50

RI
200
(c) 5 mM CaCl2 (d) 15 mM CaCl2
150

SC
100

U
50 AN
0
-1
50 mg L PS NPs
200
(e) 0.1 mM CaCl2 (f) 1 mM CaCl2
M

150

100
D
Hydrodynamic Size (nm)

50
TE

0
200
(g) 5 mM CaCl2 (h) 15 mM CaCl2
EP

150

100
C

50
AC

0
0 100 200 300 400 500 600 0 100 200 300 400 500 600
Time (s) Time (s)

Fig.2.
 ACCEPTED MANUSCRIPT

- NOM + SRHA + SRFA

-1
50 mg L PS NPs
200
(a) 0.01 mM FeCl3
150

100

50

PT
0
500
Hydrodynamic Size (nm)

RI
(b) 0.1 mM FeCl3
400

300

SC
200

100

U
AN
2000
(c) 1 mM FeCl3
1600

1200
M

800
D

400

0
TE

0 100 200 300 400 500 600

Time (s)

Fig. 3.
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP

Fig.4.
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D

Fig. 5.
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Highlights

Effects of inorganic ions and NOM on the aggregation of NPs were first studied.

High valence of ions is liable to induce NPs aggregation than low valence of ions.

Effect of NOM on NPs aggregation is dependent on specific solution conditions.

PT

DLVO force is a contributor affecting the aggregation of NPs.

RI

Cryo-SEM was used to directly observe the in-situ morphology of NPs in

suspensions.

U SC
AN
M
D
TE
C EP
AC