Content uploaded by Alice Pradel
Author content
All content in this area was uploaded by Alice Pradel on Dec 16, 2024
Content may be subject to copyright.
1Effect of the Surface Hydrophobicity−Morphology−Functionality of
2Nanoplastics on Their Homoaggregation in Seawater
3CloéVeclin, CloéDesmet, Alice Pradel, Andrea Valsesia, Jessica Ponti, Hind El Hadri, Thomas Maupas,
4Virginie Pellerin, Julien Gigault, Bruno Grassl,*and Stéphanie Reynaud
Cite This: https://doi.org/10.1021/acsestwater.1c00263
Read Online
ACCESS Metrics & More Article Recommendations *
sıSupporting Information
5ABSTRACT: The way nanoplastics aggregate in the environment is one of
6the key properties that control their final fate and impact on the
7environment. In the present work, to better predict their transportation
8pathways, nanoplastic homoaggregation was studied in saltwater to predict
9the behavior in seawater. We designed nanoplastic models that are free of
10 additives with a chemical control of the surface to model surface weathering.
11 The samples present a wide distribution of relevant surface properties such
12 as functionality (ionizable carboxylic group, 0.10 to 1.7 mmol g−1),
13 hydrophobicity (surface energy, 2.20 to 37.5 mJ m−2), surface morphology
14 (smooth or “raspberry-textured),”zeta potential (−31 to −21 mV), and
15 anisotropy in shape. The critical coagulation concentration (CCC)
16 measurements demonstrate that spherical nanoparticles are more stable in
17 seawater (CCC > 600 mmol L−1) than anisotropic nanoplastics (CCC ∼100 mmol L−1). The results highlight the importance of
18 considering the surface properties and shape when assessing the behavior of nanoplastics in the environment.
19 KEYWORDS: nanoplastic, aggregation, seawater, colloid, stability
1. INTRODUCTION
20 The environmental fate and impact of nanoplastics have raised
21 considerable and urgent concerns for several years. It is now
22 well-accepted knowledge that plastic debris degrades by
23 various physical, chemical, and biological processes to pieces
24 smaller than the microscale, down to nanoscale-sized pieces
25 known as nanoplastics.
1−4
To better predict the possible
26 impact of nanoplastics, it is crucial to investigate their
27 transportation and accumulation pathways,
5
which are
28 governed by their aggregation.
6
The aggregation of nano-
29 plastics is caused by particle−particle interactions between
30 particles of the same type (homoaggregation) or different types
31 (heteroaggregation).
7
The interactions (repulsions vs attrac-
32 tions) of nanoparticles are controlled by their own properties
33 such as hydrophobicity (i.e., Hamaker constant), size, and
34 surface properties.
6
35 Recently, significant efforts were made to describe the
36 stability of anionic-modified nanoplastics based on polystyrene
37 (PS), polyethylene terephthalate (PET), and polyethylene
38 (PE) by investigating homoaggregation rates and CCCs in
39 monovalent and divalent salt solutions.
8−21
40 In general, the data show that the particle−particle
41 attachment efficiency increases with the ionic strength; this is
42 due to the compression of the electrical double layer, in
43 agreement with the Deryaguin−Landau−Verwey−Overbeek
44 (DLVO) theory. The importance of valence is shown; divalent
45 cations (Ca2+) destabilize nanoplastics at lower concentrations
46
than monovalent cations (Na+), following the Schulze−Hardy
47rule (CCC ∝1/zn, with 2 < n< 6).
22
48
At the same pH and for an equivalent size, the large spread
49
of the provided CCC data can be explained by the
50
nanoplastics’commercial origin. Manufacturers mainly use
51
surfactants and are often unwilling to inform the user that
52
surfactants were employed and how much remained in the
53
latex dispersion, or about details of the surface charge density.
54
There is usually significant batch-to-batch variability and
55
differences between manufacturers, as well, which inevitably
56leads to differences in CCC experimental data.
57
Compared with commercial nanoplastics, those obtained
58
from laser ablation or mechanical grinding are polydisperse
59
with irregular shapes and complex surface chemistry, and they
60are free from residual surfactants.
19,23
61
Creating models that are environmentally relevant is the
62
principal scientific challenge concerning the determination of
63
the nanoplastics’fate and impact on laboratory.
24
Nanoplastic
64
models used in experimental studies must mimic environ-
65mental nanoplastics as closely as possible.
19,23,25−27
A wide
Received: July 26, 2021
Revised: December 2, 2021
Accepted: December 16, 2021
Articlepubs.acs.org/estwater
© XXXX American Chemical Society A
https://doi.org/10.1021/acsestwater.1c00263
ACS EST Water XXXX, XXX, XXX−XXX
*Unknown *|ACSJCA |JCA11.2.5208/W Library-x64 |manuscript.3f (R5.1.i4:5009 |2.1) 2021/10/27 08:51:00 |PROD-WS-120 |rq_5145980 |12/22/2021 21:42:35 |8|JCA-DEFAULT
66 variety of natural processes form nanoplastics in the environ-
67 ment, so no optimal strategy to prepare models has been
68 settled on. To date, material models have been attempted with
69 either bottom-up
28−30
or top-down approaches,
1,19,23,31
with a
70 focus on the composition, size distribution, and the potential
71 to be traceable. The used model offers the advantage of
72 knowing the shape, the chemical composition, the function-
73 ality, the size, the dispersity, the formulation process, and the
74 composition of the additives to highlight the main parameters
75 governing the behavior of nanoplastics in the environment.
76 The perfect model does not exist; it is important to choose
77 which parameters are important for the study and well-
78 characterized the model used.
79 The present work investigates the effects of underestimated
80 and relevant parameters such as the surface morphology,
81 hydrophobicity, and oxidation of nanoplastics on their
82 aggregation behavior in synthetic seawater. First, monodis-
83 persed nanoplastic models were produced using a bottom-up
84 process to control the surface characteristics, without the use
85 of any stabilizing agents, which might bias the environmental
86 relevancy of the model.
32
This method is based on a soap-free
87 emulsion polymerization
27
and allowed us to produce a range
88 of surface functionalities, hydrophobicities, and morphologies
89 to mimic natural oxidation in environmental systems. These
90 models were compared to polydisperse and anisotropic
91 nanoplastic samples that have so far been the best models to
92 mimic the size and shape properties of environmental
93 nanoplastics.
23
The aggregation kinetics of these models in
94 various compositions of saltwater show that the colloidal
95 stability of the nanoplastics does correlate to their surface
96 hydrophobicity and morphology.
33
2. MATERIALS AND METHODS
97 2.1. Materials. Styrene (S, ≥99%), acrylic acid (AA, 97%),
98 methacrylic acid (MAA, 99%), ammonium persulfate (APS,
99 ≥99%), sodium chloride (NaCl), sodium sulfate (Na2SO4),
100 potassium chloride (KCl), sodium tetraborate decahydrate
101 (Na2B4O7, 10H2O), magnesium chloride hexahydrate (MgCl2,
102 6H2O), calcium chloride dihydrate (CaCl2,2H
2O), and
103 sodium hydrogen carbonate (NaHCO3) were purchased
104 from Aldrich. Potassium bromide (KBr) was obtained from
105 VWR. Deionized water (18 MΩcm) was supplied by a
106 Millipore water purification system (Merck, Darmstadt,
107 Germany) and was used for all sample preparations, dilutions,
108 and purifications.
109 2.2. Nanoplastic Model Preparation. 2.2.1. Soap-Free
110 Emulsion Polymerization. Soap-free PS nanoplastic samples
111 were prepared by emulsion copolymerization. The synthesis
112 was adopted from Pessoni et al.
27
In this study, polymer
113 samples synthesized from S and AA or MAA monomers are
114 named PSAA2, PSAA7, PSAA36, PSAA56, and PSMAA2. The
115 samples have been named according to the number of acidic
116 groups (−COOH) on the surface of the nanoplastic particles,
117 for example, the PSAA2 contains 2 COOH per nm2.
118 Experimental conditions for soap-free nanoplastic synthesis
119 and the final solid content of purified samples (3.1 to 9.2%)
t1 120 measured using gravimetric methods are reported in Table 1.
121 2.2.2. Mechanical Fragmentation. A nanoplastic model
122 with irregular and polymorphic shapes, named PS-m, was
123 produced by the mechanical abrasion of industrial-grade
124 polystyrene pellets (PS LACQRENE 1160), as described by
125 El Hadri et al.
23
The pellets are composed of primary PS,
126which contains no additives and has not been aged. A stock
127dispersion at a concentration of 100 mg L−1was prepared.
1282.3. Sample Preparation. The purified nanoplastic sample
129was diluted with deionized water to the desired concentration
130before use. A concentrated electrolyte solution with the salt
131composition of artificial seawater (150 g L−1, ionic strength 3.0
132mol L−1) was prepared according to the United States
133Environmental Protection Agency protocol
34
with NaCl (104
134gL
−1), Na2SO4(17 g L−1), KCl (2.6 g L−1), Na2B4O7(0.08 g
135L−1), MgCl2(22 g L−1), CaCl2(5 g L−1), NaHCO3(0.91 g
136L−1), and KBr (0.42 g L−1) in deionized water and filtered
137through a 0.1 μmcelluloseacetatemembrane.Asalt
138concentration of 35 g L−1with the same ionic composition
139was used to mimic seawater.
1402.4. Sample Characterizations. 2.4.1. Size and Mor-
141phology Characterization. Hydrodynamic diameters were
142determined using a contactless (in situ) diffusion light
143scattering (DLS) probe (Vasco flex) from Cordouan
144Technologies, at an angle of 170°. The intensity fluctuation
145as a function of time was processed as an autocorrelation
146function. Cumulant algorithms were used to fit this function to
147obtain a size distribution. Cumulants allow for the determi-
148nation of the average diameter (Dz‑average).
149Atomic force microscopy (AFM) images were obtained
150using a MultiMode 8 Atomic Force Microscope from Bruker in
151PeakForce QNM (Quantitative NanoMechanics) mode.
152Nanoplastics were analyzed in 150 μL samples deposited by
153spin-coating (2000 rpm, 60 s) onto a silica wafer. Transmission
154electron microscopy (TEM) images were obtained using a
155transmission electron microscope from JEOL at an accel-
156eration voltage of 200 kV. A drop (4 μL) of the sample was
157placed onto ultrathin Formvar-coated 200-mesh copper grids
158(Tedpella Inc.) and dried at room temperature.
159ImageJ was used to measure the minimum (Fmin) and
160maximum (Fmax) Feret diameter of each fragmented nano-
161plastic (PS-m). The number-average values of Fmin and Fmax
162are, respectively, 374 ±211 and 683 ±445 nm, and the
163corresponding aspect ratio (Fmin/Fmax) is equal to 0.55.
1642.4.2. Surface Charge Characterization. The surface
165charge of the particles was evaluated using a Wallis zetameter
166from Cordouan Technologies (Pessac, France). The Wallis
167zetameter measures the electrophoretic mobility of colloidal
168particles by laser Doppler electrophoresis. Then, a model
169(Smoluchowski) was used to calculate the zeta potential (ζ).
170The analysis was performed in phosphate buffer (PB) at 10
171mmol L−1. Each measurement was performed 10 times at room
172temperature (25 °C). The zeta potential was also measured at
173different concentrations of the electrolyte solution (0.1 to 10 g
174L−1).
175The number of acidic groups (−COOH) on the surface of
176the nanoplastic particles was determined by titration with a
Table 1. Experimental Conditions for Soap-Free
Nanoplastic Synthesis with S: Styrene, Acidic Monomers:
AA: Acrylic Acid, MAA: Methacrylic Acid
a
ratio nAA/nSratio nMAA/nSsolid content (wt %)
PSAA2 0.092 3.1
PSAA7 0.129 7.9
PSAA36 0.182 8.3
PSAA56 0.235 7.5
PSMAA2 0.216 5.8
a
nAA/nSand nMAA/nSas the Mol ratio.
ACS ES&T Water pubs.acs.org/estwater Article
https://doi.org/10.1021/acsestwater.1c00263
ACS EST Water XXXX, XXX, XXX−XXX
B
177 Titrando 888 equipped with stirrer 728 and a combined pH
178 glass electrode (pH: 0−14; temperature: 0−80 °C; [KCl]: 3
179 mol L−1) purchased from Metrohm. The software used to
180 collect the data was Tiamo 2.5. The pH of the nanoplastic
181 solution was increased to reach pH >11 with an NaOH
182 solution (0.10 mol L−1) before titration with HCl at 6.58
183 mmol L−1.
184 2.4.3. Surface Hydrophobicity. The surface hydrophobicity
185 of the nanoplastic was determined by studying the nanoplastic
186 adsorption on functionalized collectors by dark-field micros-
187 copy (DFM), as described elsewhere.
35,36
Briefly purified
188 solutions of nanoplastics were diluted in a PB solution (10
189 mmol L−1). The adsorption rates of nanoplastics on three
190 different collectors were measured by recording DFM videos
191 for 20 min. Automatic frame-by-frame analysis of videos was
192 performed with the TrackMate plugin of the open-source
193 graphics software ImageJ, to assess the binding velocity of the
194 nanoplastics on different collectors.
195 According to the XDLVO (eXtended Derjaguin−Landau−
196 Verwey−Overbeek) theory,
37−39
the total interaction energy
197 Gtot can be expressed as follows:
=+ +GGG G
tot el AB LW
198 (1)
199 where Gel,GAB,andGLW are energies relative to the
200 electrostatic, acid−base, and Lifshitq−van der Waals inter-
201 actions, respectively. The three potentials depend on the
202 distance between the nanomaterials and the surface. GAB
203 includes all forces involving the structural reorganization of
204 the water molecules around two surfaces, depending on the
205 degree of wettability of the surface involved. The polar
206 component of the surface free energy (γN
AB)wasthen
207 quantified. Finally, γN
ABwas converted to a value of an equivalent
208 contact angle with water using the Owens−Wendt equation:
39
γγ γ γ γ+=+
(
)( ) (cosCA1)/2
N
D
l
D1/2
N
AB
l
AB 1/ 2
leq
209 (2)
210 where γN
D,γN
AB are the dispersive (i.e., Van der Waals
211 interactions) and polar components of the surface free energy
212 of the solid (nanoplastics in this case), respectively; γl
D,γl
AB are
213 the dispersive and polar components of the surface tension of
214 the liquid, respectively; γl=γl
D+γl
AB is the total surface free
215 energy of the liquid; and CAeq is the equivalent contact angle.
216 An equivalent contact angle (CAeq) was calculated using eq
217 3for each found γN
AB, using an estimated value of the dispersive
218 component of the surface free energy of polystyrene of 35 mJ
219 m−2.
220 2.5. Determination of the CCC. Time-resolved DLS was
221 used for the measurement of the hydrodynamic diameter as a
222 function of time. The concentration of purified nanoplastics
223 was adjusted to 200 mg L−1with deionized water, and a
224 concentrated electrolyte solution was added to obtain a salt
225 concentration ranging between 0.1 and 90 g L−1. The total
226 volume of the sample (10 mL) was stirred for 30 s. Then, the
227 nanoplastic size was recorded by DLS each 60 s for 1 h.
228 At the beginning of the aggregation process, the evolution of
229 the hydrodynamic diameter as a function of time follows a
230 linear regression to the primary particle diameter and the
231 aggregation rate constant (k).
14
The aggregation kinetics of
232 nanoplastics are characterized by the attachment efficiency α
233 that is determined by normalizing the aggregation rate
234 constant kfor a solution to the rate constant under the fastest
235 aggregation condition kfast. The CCC value was obtained by
236 fitting the profile with eq 2 defined by Grolimund et al.
40
and
237used by Liu et al.
10
to determine the CCC of UV-irradiated
238nanoplastics.
α
=
+
β
()
1
1C
CCC
s239
(3)
240where Csis the molar concentration of the electrolyte and βis
241the slope of dlog(α)/dlog(Cs).
3. RESULTS AND DISCUSSION
2423.1. Sample Characterization. The process schematic is
243 f1shown Figure 1. Polyacrylic acid (PAA) and polymethacrylic
244acid (PMAA) behave in aqueous solution as weak polyelec-
245trolytes because of the ionization of the carboxyl group on the
246backbone with increasing pH. AA and MAA have been chosen
247as co-units of the polymer backbone because (i) the carboxylic
248functional groups allow us to customize their concentration
249embedded on the nanoplastic to mimic the oxidation state of
250the aged nanoplastic
2,41
and (ii) the methyl group of MAA is
251responsible for varying the conformational behavior and
252hydrophobicity of PAA and PMAA in aqueous solution.
42
253 t2Table 2 reports the physicochemical characteristics found
254for five synthesized samples with diameters ranging from 220
255to 390 nm. Their polydispersity indices (PDI) are below 0.05.
256All nanoplastic models present a zeta potential at pH = 7 (PB,
25710 mmol L−1) ranging between −31 and −20 mV, which is
258characteristic of colloidal stability in an aqueous dispersion.
259The control of the AA monomer content led to nanoplastics
260having from 80 to 1700 μmol g−1of carboxylic functional
261groups at the nanoplastic surface (i.e., from 2 to 56 COOH/
262nm2). Beyond AA, nanoplastics were also synthesized with
263MAA. Comparable nanoplastics in terms of carboxylic
264functional groups per unit area were obtained by either
265incorporating AA or MAA. However, a higher content of MAA
266is needed to reach similar hydrophobic surface properties in
267that case. PSAA2 and PSMAA2 are interesting to study
268because they have a similar size and contain the same
269carboxylic functional groups per unit area (2 COOH/nm2, that
270is, 80 and 100 μmol g−1respectively); PSAA2 was made with
271hydrophilic AA and PSMAA2 was made with hydrophobic AA.
272The results are reported in Table 2 with those of the
273fragmented sample (PS-m).
274The fragmented polystyrene-based nanoplastics (PS-m,
275Table 2) have an average size of 370 nm and have 390 μmol
276g−1carboxylic functional groups per unit area, which is in the
277range of the models obtained from the bottom-up approach.
278Because of the fragmentation process, the latter remains shows
279polydispersity with the aspect ratio (Fmin/Fmax) equal to 0.55
Figure 1. Nanoplastic model synthesis: soap-free emulsion polymer-
ization with the control of hydrophilicity and anionicity at the
nanoplastic/water interface.
ACS ES&T Water pubs.acs.org/estwater Article
https://doi.org/10.1021/acsestwater.1c00263
ACS EST Water XXXX, XXX, XXX−XXX
C
280 (Figure S3) and the anisotropic morphology (Figure 3F),
281 while emulsion polymerization gives spherical and mono-
282 disperse samples (Figures 3 and S3).
23
283 We determined the nanoplastic surface hydrophobicity with
284 a recently developed method for the direct characterization of
285 the hydrophobicity of manufactured nanomaterials.
35,36
The
286 method involves the measurement of the affinity of nano-
287 plastics for different engineered surfaces, or collectors, with
288 specific surface charges and surface energy components. The
289 method analyzes the binding efficiency of nanoplastics onto
290 different collectors. This property is evaluated vs time and is
291 correlated to their hydrophobic properties to give access to a
292 quantitative result. Other methods, such as hydrophobic
293 interaction chromatography or water−octanol partitioning,
294 only provide qualitative values and are inadequate for
295 nanomaterials.
43,44
A combination of different studies
39,45,46
296 also led to the description of the relationship between the
297 dispersive and polar components of the surface free energy of a
298 solid and its equivalent contact angle, as presented in eq 3.
299 As reported in Table 2, the polar component of the surface
300 free energy γN
ABof nanoplastics obtained using the polymer-
301 ization process varies significantly from 2.20 to 37.5 mJ m−2,
302 and the CAeq ranges from 87 to 16°. PSAA2 shows a γN
AB and
303 CAeq close to a hydrophobic reference polystyrene particle, and
304 γN
AB increases with the surface charge for the sample series
f2 305 based on the AA comonomer.
35
Figure 2 shows that
306γN
ABincreases rapidly with the carboxylic group content to
307reach a plateau. Both PSAA2 and PSAA7 follow this first stage
308of the behavior, while PSAA36 and PSAA56 exhibit an average
309γN
ABof 36.7 mJ m−2.
310The microscopy and AFM images illustrate the influence of
311the AA unit incorporation on the morphology of the
312 f3nanoplastic surface (Figures 3 and S1). The lowest function-
313alized samples (PSAA2 and PSAA7) present a smooth surface
314morphology, while the highest functionalized samples, PSAA36
315and PSAA56, exhibit a rough surface similar to that of a
316“raspberry.”This change in morphology may be explained by
317the number of COOH molecules likely to be at the nanoplastic
318surface. During the synthesis, AA units are preferentially
319incorporated on the surface of the particles.
47
For the
320molecules with the highest AA content, the roughness of the
321particles helps minimize the stress on their surface, as
322generated by the high number of acrylic functional groups.
323This suggests that there is a saturated level of AA on the
324surface. These results correlate with those obtained in a
325previous study for the series of AA-based nanoplastics.
27
326For the MAA-based sample, which is more hydrophobic
327than AA, the physicochemical characteristics in terms of
328surface charge, morphology, and surface tension are different
329from these AA-based homologues with the same ratio of
330monomer involved in the synthesis (nAA/nS= 0.24 and nMAA/
331nS= 0.22). PSAA56 and PSMAA2, respectively, have
332functionalities of 56 and 2 COOH/nm2, surface morphology
333of raspberry and smooth type, surface tension of 37.5 and 22.6
Table 2. Nanoplastic Properties with the Average Diameter, Dz‑average, the PDI, the Polar Component of the Surface Free
Energy, γN
AB,and the Equivalent Contact Angle, CAeq, with Water
Dz‑average (nm) PDI
[COOH]
zeta (mV) pH = 7, 10 mM PB
CCC
γN
AB (mJ m−2) CAeq. (°) surface morphology
mmol g−1(g L−1)
COOH/nm2(mmol L−1)
PSAA2 270 ±10 0.08 ±0.02 −31 ±2 43 ±2 2.2 87 smooth
0.007 2730 ±37
PSAA7 350 ±20 0.19 ±0.02 −26 ±2 n.d 28.4 36 smooth
0.008 7
PSAA36 390 ±20 0.92 ±0.1 −21 ±3 n.d 35.9 21 raspberry
0.003 36
PSAA56 330 ±20 1.7 ±0.2 −20 ±3 38 ±1 37.5 16 raspberry
0.008 56 665 ±6
PSMAA2 220 ±10 0.10 ±0.02 −28 ±3 36 ±1 22.6 46 smooth
0.009 2630 ±24
PS-m 370 ±20 0.39 ±0.02N.A. −22 ±35±2 31.8 30 anisotropic
0.09 94 ±20
Figure 2. Correlation between the polar component of the surface
energy of the nanoplastic (γN
AB) (which represents its hydrophilicity)
and functionality of the nanoplastic.
Figure 3. TEM images of (A) PSAA2, (B) PSAA7, (C) PSAA36, (D)
PSAA56, and (E) PSMAA2 are obtained via an emulsion process. (F)
PS-m is elaborated via a fragmentation process.
ACS ES&T Water pubs.acs.org/estwater Article
https://doi.org/10.1021/acsestwater.1c00263
ACS EST Water XXXX, XXX, XXX−XXX
D
334 mJ nm−2, that is, a more hydrophobic behavior for PSMAA2.
335 This is not possible to determine if the MAA monomer is less
336 incorporated at the surface or removed during the synthesis
337 process compared to AA. Nevertheless, the MAA monomer
338 allows for the generation of the nanoplastics of similar size with
339 physicochemical surface characteristics significantly different to
340 the PSAA series. The comparison between PSAA2 and
341 PSMAA2 is interesting; they have an identical surface
342 functionality (2 COOH/nm2) and a smooth morphology for
343 a significantly different surface tension. According to surface
344 tension values, PSAA2 can be classified as hydrophobic and
345 PSMAA2 as hydrophilic (2.2 and 22.6 mJ m2). In this case, we
346 can consider that the surface hydrophobicity is the only
347 different physicochemical characteristic. This difference in
348 surface tension was not expected, given the hydrophobicity of
349 the monomer. Indeed, according to its chemical structure,
350 MAA should have generated more surface hydrophobicity than
351 AA; our results show the opposite. This may be explained by
352 the effect of MAA hydrophobicity during the synthesis, where
353 this acidic monomer will be preferentially incorporated into
354 the core of the polystyrene particle (Figure 1). Considering the
355 total polymerization of AA and MAA, this preference greatly
356 reduces the level of ionizable carboxylic functions at the surface
357 at an identical monomer amount in the synthetic formulation,
358 but at a higher ratio (nMAA/nS= 0.22 and nAA/nS= 0.092), this
359 increases the rate of the nonionizable carboxylic group at the
360 surface.
361 Acidic comonomers were chosen for their ability to stabilize
362 the particle with its amphiphilic properties and because it
363 mimics the oxidation state of environmental nanoplastics.
12
364 The surface energy may be considered to be a representative
365 parameter to describe the surface polarity of the samples, and
366 therefore, the state of nanoplastic aging in the environment.
367 The surface energy should be considered to be a more reliable
368 indicator than the level of ionizable carboxylic function.
48
369 3.2. Stability/Homoaggregation. Similar-sized nano-
370 plastic particles of PSAA2, PSAA56, PSMAA2, and PS-m
371 exhibit different surface characteristics, morphology, COOH
372 functionality per surface area, and surface energy. The CCC is
373 the point at which αreaches unity, beyond which the
374 aggregation rate is no longer sensitive to an increase in ionic
375 strength.
376 In aqueous media, synthesized nanoplastics and fragmented
377 samples (PS-m) aggregate (Table 2) with salt concentrations
378 of 675 ±51 and 94 ±20 mmol L−1, respectively. This implies
379 that the fragmented sample is less stable than these two in 600
380 mmol L−1seawater. For comparison, note that CCC values
381 were previously reported
10,12
for commercial carboxylated PS
382 latex (PSL) in NaCl and were equal to 613 and 462 mmol L−1
383 at pH = 7.5 and 8, respectively, while they were equal to 1095
384 and 890 mmol L−1for ultraviolet (UV)-aged PS under the
385 same conditions.
386 We used better representative environmental conditions for
387 this research, with artificial seawater and with nanoplastics of
388 strictly known and controlled chemical composition. It should
389 be also mentioned that all the nanoplastics we obtained from
390 the polymerization process were pure of both ionic and
391 nonionic surfactants. These surfactants can bias the study, but
392 they are often contained in commercial samples.
393 All the isotropic nanoplastics reported in Table 2 and
394 obtained via soap-free emulsion polymerization show un-
395 expectedly good stability in seawater, regardless of their surface
f4 396 characteristics. Figure 4 reports the evolution of the zeta
397potential as a function of the electrolyte concentration, a
398representative aggregation profile of nanoplastic in electrolyte
399solution, and the attachment efficiency of nanoplastics as a
400function of the electrolyte concentration. The measurements of
401the zeta potential in seawater, as seen in Figure 4A, show
402destabilization above 10 mmol L−1, which is lower than that of
403CCCs. Particles with a zeta potential between ±0 and 10 mV
404are often considered in the literature to be particles with a high
405probability to aggregate.
49
Figure 4. Influence of artificial seawater on the electrokinetic
properties and aggregation kinetics of nanoplastics synthesized with
AA containing 2 and 56 COOH/nm2(PSAA2 and PSAA56), MAA
containing 2 COOH/nm2(PSMAA2), and obtained using a
fragmentation method (PS-m). (A) ζpotential as a function of
electrolyte concentration. The lines are for eye guidance. (B)
Representative aggregation profiles at different electrolyte concen-
trations on PSAA56. (C) Attachment efficiencies as a function of
electrolyte concentration. The line represents empirical fitting to eq 2.
The complete data for PS-m are reported in Figure S2.
ACS ES&T Water pubs.acs.org/estwater Article
https://doi.org/10.1021/acsestwater.1c00263
ACS EST Water XXXX, XXX, XXX−XXX
E
406 The XDLVO theory does not take into account the surface
407 heterogeneity in terms of surface roughness. However, Lan et
408 al.
50
recently demonstrated that roughness is an important
409 parameter for particle stability in water with high salinity and
410 that “raspberry-textured”colloids are more stable because of
411 higher repulsive interactions between particles compared to
412 smooth PS particles. In this work, spherical models with
413 different surface morphologies, amount of ionizable function,
414 and hydrophilicity/hydrophobicity ratio used are stable in
415 seawater, and none of these three parameters seems to
416 dominate.
417 The CCC of PS-m is 94 mmol L−1in seawater; it is
418 significantly lower than their spherical isotropic homologue. Of
419 all parameters studied, the shape appears to have the greatest
420 influence on homoaggregation behavior. It is important to
421 consider this heterogeneity, which is certainly more
422 representative of the secondary nanoplastic morphology
423 present in the environment. In addition, spherical nanoplastics
424 exhibit stability in seawater and a wide variety of surface
425 properties that can offer new opportunities in the development
426 of analytical methods, as recently published,
51
adsorption of
427 metallic compounds,
27
or transfer in porous media.
25
Spherical
428 nanoplastic stability associated with the absence of additives
429 may be useful to examine the potential routes of exposure to
430 nanoplastics, biological effects of these particles in mammalian
431 cells, factors influencing toxicity, and the probable mechanisms
432 of cytotoxicity.
52
4. CONCLUSIONS
433 In this work, we used nanoplastic models with a controlled
434 surface morphology, surface hydrophilicity, surface carboxyl
435 function density, and shape to mimic nanoplastics found in the
436 environment. Soap-free emulsion polymerization is used to
437 produce samples that are spherical and homogeneous in size.
438 The rate of ionizable carboxylate functional groups at the
439 surface (0.1−1.7 mmol L−1) and the rate of the surface energy
440 can be modified while controlling the size of the particles. The
441 nanoplastics thus formed exhibit a surface morphology ranging
442 from smooth to raspberry-textured, a surface energy
443 representative of the hydrophilicity of 2.20 to 37.5 mJ m−2.
444 All of these parameters modulate the zeta potential of
445 nanoplastics as a function of the ionic strength of seawater,
446 and this stability has been studied in seawater. Spherical
447 nanoplastics with CCCs greater than the seawater salt
448 concentration (>600 mmol L−1) were very stable. This work
449 is completed with anisotropic nanoplastics (PS-m), which
450 exhibit an anisotropic surface morphology, an ionizable
451 function rate at the surface of 390 μmol per gram of
452 nanoplastic, and a surface energy of 31.8 mJ m−2. The
453 heterogeneity in terms of morphology on anisotropic nano-
454 plastics, which have surface characteristics similar to those of
455 spherical samples, is the parameter which should affect the
456 stability of nanoplastics in terms of homoaggregation with
457 strong destabilization in seawater (CCC = 94 mmol L−1).
458 Further studies may include heteroaggregation and/or organic
459 matter to getting closer to environmental conditions.
460 ■ASSOCIATED CONTENT
461 *
sıSupporting Information
462 The Supporting Information is available free of charge at
463 https://pubs.acs.org/doi/10.1021/acsestwater.1c00263.
464AFM topography images of PSAA2, PSAA56, and
465PSMAA2 and the complement data of attachment
466efficiency for PS-m (PDF)
467
■AUTHOR INFORMATION
468Corresponding Author
469Bruno Grassl −Universite de Pau et des Pays de l’Adour, E2S
470UPPA, CNRS, IPREM, Pau 64053, France; orcid.org/
4710000-0002-1554-1411; Email: bruno.grassl@univ-pau.fr
472Authors
473Cloé Veclin −Universite de Pau et des Pays de l’Adour, E2S
474UPPA, CNRS, IPREM, Pau 64053, France
475Cloé Desmet −European Commission Joint Research Centre
476(JRC), Ispra 21020, Italy
477Alice Pradel −Géosciences Rennes, UMR 6118, CNRS
478Universitéde Rennes, 35000 Rennes CEDEX, France;
479TAKUVIK Laboratory, CNRS/UniversitéLaval, Québec
480City, Québec G1V 0A6, Canada; orcid.org/0000-0002-
4812078-9016
482Andrea Valsesia −European Commission Joint Research
483Centre (JRC), Ispra 21020, Italy
484Jessica Ponti −European Commission Joint Research Centre
485(JRC), Ispra 21020, Italy
486Hind El Hadri −Universite de Pau et des Pays de l’Adour,
487E2S UPPA, CNRS, IPREM, Pau 64053, France
488Thomas Maupas −Universite de Pau et des Pays de l’Adour,
489E2S UPPA, CNRS, IPREM, Pau 64053, France
490Virginie Pellerin −Universite de Pau et des Pays de l’Adour,
491E2S UPPA, CNRS, IPREM, Pau 64053, France
492Julien Gigault −TAKUVIK Laboratory, CNRS/Université
493Laval, Québec City, Québec G1V 0A6, Canada;
494orcid.org/0000-0002-2988-8942
495Stéphanie Reynaud −Universite de Pau et des Pays de
496l’Adour, E2S UPPA, CNRS, IPREM, Pau 64053, France;
497orcid.org/0000-0001-9048-0842
498Complete contact information is available at:
499https://pubs.acs.org/10.1021/acsestwater.1c00263
500Notes
501The authors declare no competing financial interest.
502
■ACKNOWLEDGMENTS
503The project leading to this publication was funded by the
504Excellence Initiative of Université de Pau et des Pays de
505l’Adour −I-Site E2S UPPA, a French “Investissements
506d’Avenir”program.
507
■REFERENCES
(1) 508Enfrin, M.; Lee, J.; Gibert, Y.; Basheer, F.; Kong, L.; Dumée, L.
509F. Release of hazardous nanoplastic contaminants due to microplastics
510fragmentation under shear stress forces. J. Hazard. Mater. 2020,384,
511No. 121393.
(2) 512Andrady, A. L. The plastic in microplastics: A review. Mar.
513Pollut. Bull. 2017,119, 12−22.
(3) 514Gigault, J.; Pedrono, B.; Maxit, B.; Ter Halle, A. Marine plastic
515litter: the unanalyzed nano-fraction. Environ. Sci.: Nano 2016,3, 346−
516350.
(4) 517Jahnke, A.; Arp, H. P. H.; Escher, B. I.; Gewert, B.; Gorokhova,
518E.; Ku
hnel, D.; Ogonowski, M.; Potthoff, A.; Rummel, C.; Schmitt-
519Jansen, M.; Toorman, E.; Mac Leod, M. Reducing Uncertainty and
520Confronting Ignorance about the Possible Impacts of Weathering
521Plastic in the Marine Environment. Environ. Sci. Technol. Lett. 2017,4,
52285−90.
ACS ES&T Water pubs.acs.org/estwater Article
https://doi.org/10.1021/acsestwater.1c00263
ACS EST Water XXXX, XXX, XXX−XXX
F
(5)
523 Nguyen, B.; Claveau-Mallet, D.; Hernandez, L. M.; Xu, E. G.;
524 Farner, J. M.; Tufenkji, N. Separation and Analysis of Microplastics
525 and Nanoplastics in Complex Environmental Samples. Acc. Chem. Res.
526 2019,52, 858−866.
(6)
527 Hotze, E. M.; Phenrat, T.; Lowry, G. V. Nanoparticle
528 aggregation: challenges to understanding transport and reactivity in
529 the environment. J. Environ. Qual. 2010,39, 1909−1924.
(7)
530 Wiesner, M. R.; Bottero, J.-Y., Environmental Nanotechnology:
531 Applications and Impacts of Nanomaterials. 2nd ed.; McGraw-Hill
532 Education: New York, 2017.
(8)
533 Singh, N.; Bhagat, J.; Tiwari, E.; Khandelwal, N.; Darbha, G. K.;
534 Shyama, S. K. Metal oxide nanoparticles and polycyclic aromatic
535 hydrocarbons alter nanoplastic’s stability and toxicity to zebrafish. J.
536 Hazard. Mater. 2021, 407.
(9)
537 Singh, N.; Tiwari, E.; Khandelwal, N.; Darbha, G. K.
538 Understanding the stability of nanoplastics in aqueous environments:
539 effect of ionic strength, temperature, dissolved organic matter, clay,
540 and heavy metals. Environ. Sci.: Nano 2019,6, 2968−2976.
(10)
541 Liu, Y.; Hu, Y.; Yang, C.; Chen, C.; Huang, W.; Dang, Z.
542 Aggregation kinetics of UV irradiated nanoplastics in aquatic
543 environments. Water Res. 2019,163, No. 114870.
(11)
544 Liu, Y.; Huang, Z.; Zhou, J.; Tang, J.; Yang, C.; Chen, C.;
545 Huang, W.; Dang, Z. Influence of environmental and biological
546 macromolecules on aggregation kinetics of nanoplastics in aquatic
547 systems. Water Res. 2020, 186.
(12)
548 Mao, Y.; Li, H.; Huangfu, X.; Liu, Y.; He, Q. Nanoplastics
549 display strong stability in aqueous environments: Insights from
550 aggregation behaviour and theoretical calculations. Environ. Pollut.
551 2020,258, No. 113760.
(13)
552 Sun, H.; Jiao, R.; Wang, D. The difference of aggregation
553 mechanism between microplastics and nanoplastics: Role of Brownian
554 motion and structural layer force. Environ. Pollut. 2021, 268.
(14)
555 Yu, S.; Shen, M.; Li, S.; Fu, Y.; Zhang, D.; Liu, H.; Liu, J.
556 Aggregation kinetics of different surface-modified polystyrene nano-
557 particles in monovalent and divalent electrolytes. Environ. Pollut.
558 2019,255, No. 113302.
(15)
559 Li, X.; He, E.; Xia, B.; Van Gestel, C. A. M.; Peijnenburg, W. J.
560 G. M.; Cao, X.; Qiu, H. Impact of CeO2 nanoparticles on the
561 aggregation kinetics and stability of polystyrene nanoplastics:
562 Importance of surface functionalization and solution chemistry.
563 Water Res. 2020,186, No. 116324.
(16)
564 Nolte, T. M.; Hartmann, N. B.; Kleijn, J. M.; Garnæs, J.; van de
565 Meent, D.; Jan Hendriks, A.; Baun, A. The toxicity of plastic
566 nanoparticles to green algae as influenced by surface modification,
567 medium hardness and cellular adsorption. Aquat. Toxicol. 2017,183,
568 11−20.
(17)
569 Shams, M.; Alam, I.; Chowdhury, I. Aggregation and stability of
570 nanoscale plastics in aquatic environment. Water Res. 2020,171,
571 No. 115401.
(18)
572 Pradel, A.; Ferreres, S.; Veclin, C.; El Hadri, H.; Gautier, M.;
573 Grassl, B.; Gigault, J. Stabilization of Fragmental Polystyrene
574 Nanoplastic by Natural Organic Matter: Insight into Mechanisms.
575 ACS ES&T Water 2021, 1198−1208.
(19)
576 Magrì, D.; Sánchez-Moreno, P.; Caputo, G.; Gatto, F.;
577 Veronesi, M.; Bardi, G.; Catelani, T.; Guarnieri, D.; Athanassiou,
578 A.; Pompa, P. P.; Fragouli, D. Laser ablation as a versatile tool to
579 mimic polyethylene terephthalate nanoplastic pollutants: Character-
580 ization and toxicology assessment. ACS Nano 2018,12, 7690−7700.
(20)
581 Dong, S.; Cai, W.; Xia, J.; Sheng, L.; Wang, W.; Liu, H.
582 Aggregation kinetics of fragmental PET nanoplastics in aqueous
583 environment: Complex roles of electrolytes, pH and humic acid.
584 Environ. Pollut. 2021, 268.
(21)
585 Wang, J.; Zhao, X.; Wu, A.; Tang, Z.; Niu, L.; Wu, F.; Wang, F.;
586 Zhao, T.; Fu, Z. Aggregation and stability of sulfate-modified
587 polystyrene nanoplastics in synthetic and natural waters. Environ.
588 Pollut. 2021, 268.
(22)
589 Oncsik, T.; Trefalt, G.; Csendes, Z.; Szilagyi, I.; Borkovec, M.
590 Aggregation of Negatively Charged Colloidal Particles in the Presence
591 of Multivalent Cations. Langmuir 2014,30, 733−741.
(23) 592
El Hadri, H.; Gigault, J.; Maxit, B.; Grassl, B.; Reynaud, S.
593
Nanoplastic from mechanically degraded primary and secondary
594
microplastics for environmental assessments. Nano Impact 2020,17,
595No. 100206.
(24) 596
Gigault, J.; Halle, A. T.; Baudrimont, M.; Pascal, P.-Y.; Gauffre,
597
F.; Phi, T.-L.; El Hadri, H.; Grassl, B.; Reynaud, S. Current opinion:
598What is a nanoplastic? Environ. Pollut. 2018,235, 1030−1034.
(25) 599
Pradel, A.; Hadri, H. E.; Desmet, C.; Ponti, J.; Reynaud, S.;
600
Grassl, B.; Gigault, J. Deposition of environmentally relevant
601
nanoplastic models in sand during transport experiments. Chemo-
602sphere 2020,255, No. 126912.
(26) 603
Mitrano, D. M.; Beltzung, A.; Frehland, S.; Schmiedgruber, M.;
604
Cingolani, A.; Schmidt, F. Synthesis of metal-doped nanoplastics and
605
their utility to investigate fate and behaviour in complex environ-
606mental systems. Nat. Nanotechnol. 2019,14, 362−368.
(27) 607
Pessoni, L.; Veclin, C.; Hadri, H. E.; Cugnet, C.; Davranche,
608
M.; Pierson-Wickmann, A.-C.; Gigault, J.; Grassl, B.; Reynaud, S.
609
Soap- and metal-free polystyrene latex particles as a nanoplastic
610model. Environ. Sci.: Nano 2019,6, 2253−2258.
(28) 611
Balakrishnan, G.; Déniel, M.; Nicolai, T.; Chassenieux, C.;
612
Lagarde, F. Towards more realistic reference microplastics and
613
nanoplastics: preparation of polyethylene micro/nanoparticles with a
614biosurfactant. Environ. Sci.: Nano 2019,6, 315−324.
(29) 615
Tian, L.; Chen, Q.; Jiang, W.; Wang, L.; Xie, H.; Kalogerakis,
616
N.; Ma, Y.; Ji, R. A carbon-14 radiotracer-based study on the
617
phototransformation of polystyrene nanoplastics in water versus in air.
618Environ. Sci.: Nano 2019,6, 2907−2917.
(30) 619
Rodríguez-Hernández, A. G.; Mun
oz-Tabares, J. A.; Aguilar-
620
Guzmán, J. C.; Vazquez-Duhalt, R. A novel and simple method for
621
polyethylene terephthalate (PET) nanoparticle production. Environ.
622Sci.: Nano 2019,6, 2031−2036.
(31) 623
Astner, A. F.; Hayes, D. G.; O’Neill, H.; Evans, B. R.; Pingali, S.
624
V.; Urban, V. S.; Young, T. M. Mechanical formation of micro- and
625
nano-plastic materials for environmental studies in agricultural
626ecosystems. Sci. Total Environ. 2019,685, 1097−1106.
(32) 627
Pikuda, O.; Xu, E. G.; Berk, D.; Tufenkji, N. Toxicity
628
Assessments of Micro- and Nanoplastics Can Be Confounded by
629
Preservatives in Commercial Formulations. Environ. Sci. Technol. Lett.
6302019,6, 21−25.
(33) 631
Kokalj, A. J.; Hartmann, N. B.; Drobne, D.; Potthoff, A.;
632
Ku
hnel, D. Quality of nanoplastics and microplastics ecotoxicity
633
studies: Refining quality criteria for nanomaterial studies. J. Hazard.
634Mater. 2021,415, No. 125751.
(34) 635
USEPA. Methods for measuring the acute toxicity of effluents and
636
receiving waters to freshwater and marine organisms. USEPA, Office of
637Water: Washington DC, 2002.
(35) 638
Desmet, C.; Valsesia, A.; Oddo, A.; Ceccone, G.; Spampinato,
639
V.; Rossi, F.; Colpo, P. Characterisation of nanomaterial hydro-
640phobicity using engineered surfaces. J. Nanopart. Res. 2017,19, 117.
(36) 641
Valsesia, A.; Desmet, C.; Ojea-Jiménez, I.; Oddo, A.;
642
Capomaccio, R.; Rossi, F.; Colpo, P. Direct quantification of
643nanoparticle surface hydrophobicity. Commun. Chem. 2018,1, 53.
(37) 644
van Oss, C. J. Acidbase interfacial interactions in aqueous
645media. Colloids Surf., A 1993,78,1−49.
(38) 646
Voïtchovsky, K.; Kuna, J. J.; Contera, S. A.; Tosatti, E.;
647
Stellacci, F. Direct mapping of the solid-liquid adhesion energy with
648subnanometre resolution. Nat. Nanotechnol. 2010,5, 401−405.
(39) 649
Owens, D. K.; Wendt, R. C. Estimation of the surface free
650energy of polymers. J. Appl. Polym. Sci. 1969,13, 1741−1747.
(40) 651
Grolimund, D.; Elimelech, M.; Borkovec, M. Aggregation and
652
deposition kinetics of mobile colloidal particles in natural porous
653media. Colloids Surf., A 2001,191, 179−188.
(41) 654
Ter Halle, A.; Jeanneau, L.; Martignac, M.; Jardé, E.; Pedrono,
655
B.; Brach, L.; Gigault, J. Nanoplastic in the North Atlantic subtropical
656gyre. Environ. Sci. Technol. 2017,51, 13689−13697.
(42) 657
Dobrynin, A. V.; Rubinstein, M. Theory of polyelectrolytes in
658solutions and at surfaces. Prog. Polym. Sci. 2005,30, 1049−1118.
ACS ES&T Water pubs.acs.org/estwater Article
https://doi.org/10.1021/acsestwater.1c00263
ACS EST Water XXXX, XXX, XXX−XXX
G
(43)659 Crandon, L. E.; Boenisch, K. M.; Harper, B. J.; Harper, S. L.
660 Adaptive methodology to determine hydrophobicity of nanomaterials
661 in situ. PLoS One 2020,15, No. e0233844.
(44)
662 Praetorius, A.; Tufenkji, N.; Goss, K.-U.; Scheringer, M.; von
663 der Kammer, F.; Elimelech, M. The road to nowhere: equilibrium
664 partition coefficients for nanoparticles. Environ. Sci.: Nano 2014,1,
665 317−323.
(45)
666 Good, R. J.; Girifalco, L. A.; Kraus, G. A theory for estimation
667 of interfacial energies. II. Application to surface thermodynamics of
668 teflon and graphite. J. Phys. Chem. 1958,62, 1418−1421.
(46)
669 Fowkes, F. M. Attractive forces at interfaces. Ind. Eng. Chem.
670 1964,56, 40−52.
(47)
671 Goodall, A. R.; Wilkinson, M. C.; Hearn, J. Mechanism of
672 emulsion polymerization of styrene in soap-free systems. J. Polym. Sci.,
673 Part A: Polym. Chem. 1977,15, 2193−2218.
(48)
674 Sorasan, C.; Edo, C.; González-Pleiter, M.; Fernández-Pin
as, F.;
675 Leganés, F.; Rodríguez, A.; Rosal, R. Generation of nanoplastics
676 during the photoaging of low-density polyethylene. Environ. Pollut.
677 2021,289, No. 117919.
(49)
678 Bhattacharjee, S. DLS and zeta potential −What they are and
679 what they are not? J. Controlled Release 2016,235, 337−351.
(50)
680 Lan, Y.; Caciagli, A.; Guidetti, G.; Yu, Z.; Liu, J.; Johansen, V.
681 E.; Kamp, M.; Abell, C.; Vignolini, S.; Scherman, O. A.; Eiser, E.
682 Unexpected stability of aqueous dispersions of raspberry-like colloids.
683 Nat. Commun. 2018,9, 3614.
(51)
684 Jiménez-Lamana, J.; Marigliano, L.; Allouche, J.; Grassl, B.;
685 Szpunar, J.; Reynaud, S. A Novel Strategy for the Detection and
686 Quantification of Nanoplastics by Single Particle Inductively Coupled
687 Plasma Mass Spectrometry (ICP-MS). Anal. Chem. 2020,92, 11664−
688 11672.
(52)
689 Banerjee, A.; Shelver, W. L. Micro- and nanoplastic induced
690 cellular toxicity in mammals: A review. Sci. Total Environ. 2021,755,
691 No. 142518.
ACS ES&T Water pubs.acs.org/estwater Article
https://doi.org/10.1021/acsestwater.1c00263
ACS EST Water XXXX, XXX, XXX−XXX
H